The 50th Annual Meeting of the European Society for Blood and Marrow Transplantation: Physicians - Oral Session (O009-O154) (2024)

Roni Shouval ¹, Myriam Labopin², Joshua Fein³, Jurjen Versluis⁴, Matthias Stelljes⁵, Nicolaus Kröger⁶, Igor Wolfgang Blau⁷, Thomas Schroeder⁸, Matthias Eder⁹, Didier Blaise¹⁰, Ibrahim Yakoub-Agha¹¹, Eva Maria Wagner-Drouet¹², Régis Peffault de Latour¹³, Johannes Schetelig¹⁴, Tobias Gedde-Dahl¹⁵, Peter Dreger¹⁶, Anne Huynh¹⁷, Fabio Ciceri¹⁸, Mohamad Mohty², Arnon Nagler¹⁹

¹Memorial Sloan Kettering Cancer Center, New York, United States, ²Sorbonne University, Saint-Antoine Hospital, AP-HP, Paris, France, ³Weill Cornell Medicine, New York, United States, ⁴Erasmus MC, Rotterdam, Netherlands, ⁵University of Muenster, Muenster, Germany, ⁶University Hospital Eppendorf, Hamburg, Germany, ⁷Medizinische Klinik m. S. Hämatologie, Onkologie und Tumorimmunologie, Berlin, Germany, ⁸University Hospital | Essen, Essen, Germany, ⁹Hannover Medical School, Hannover, Germany, ¹⁰Programme de Transplantation & Therapie Cellulaire, Marseille, France, ¹¹CHU de Lille, Lille, France, ¹²University Medical Center Mainz, Mainz, Germany, ¹³Saint-Louis Hospital, Paris, France, ¹⁴University Hospital, TU Dresden, Dresden, Germany, ¹⁵Oslo University Hospital, Rikshospitalet, Oslo, Norway, ¹⁶University of Heidelberg, Heidelberg, Germany, ¹⁷CHU - Institut Universitaire du Cancer Toulouse, Toulouse, France, ¹⁸IRCCS San Raffaele Scientific Institute, Milan, Italy, ¹⁹Chaim Sheba Medical Center, Ramat Gan, Israel

Background: The Disease Risk Stratification System (DRSS) serves as a vital prognostic tool in determining the outcomes of allogeneic hematopoietic cell transplantation (HCT) candidates. Initially developed and validated in a substantial cohort exceeding 47,000 patients transplanted between 2012 and 2016 (Shouval et al., Lancet Haem 2021), the DRSS relies on disease-related factors encompassing histological aspects, genetic profiles, and treatment responses. Recognizing the evolving landscape of transplantation practices and demographic shifts over time, this study aimed to validate and refine the DRSS within a contemporary cohort of acute leukemia patients.

Methods: We investigated 31,425 adult HCT recipients with acute myeloid leukemia (AML) or acute lymphoblastic leukemia (ALL) transplanted between 2017 and 2021, as reported to The European Society for Blood and Marrow Transplantation (EBMT). We assessed the original DRSS’s prognostic impact and refined the risk levels based on individual disease feature combinations and their association with overall survival (OS), ultimately generating a refined version of the system (rDRSS).

Results: Patients had a median age of 54 years (interquartile range 40-63), with AML being the primary indication for HCT (76%). Most patients received myeloablative conditioning (54%) and donors were predominantly unrelated to recipients (57%). Post-transplantation cyclophosphamide for graft-versus-host disease (GVHD) prophylaxis in 25% of patients. The original DRSS categorized patients into OS risk groups of low (26%), intermediate 1 (34%), intermediate 2 (14%), high (11%), and very high-risk (15%). Corresponding 2-year OS probability was 66%, 61%, 51%, 44%, and 34%. Subsequent multivariable Cox regression models, accounting for key patient and transplant characteristics, confirmed the association between DRSS levels and a stepwise increase in risks of mortality, leukemia-free survival (LFS), and relapse following HCT.

Refinements implemented in the rDRSS resulted in an increased proportion of patients classified as low-risk (36%) and a decreased number in the very high-risk category (6%). With the refined risk groups, the 2-year OS ranged from 33% in the very-high risk group to 73% in the low-risk group. rDRSS was consistently associated with mortality, LFS, and relapse. Furthermore, within a subset of patients with available pre-HCT measurable residual disease (MRD; n = 5148), the rDRSS maintained its prognostic impact, irrespective of MRD status.

Finally, we categorized patients based on the Disease Risk Index (Armand et al., Blood 2014). Notably, the bulk of patients (59%) were categorized as intermediate-risk and had a projected 2-year OS of 71% (95%CI 70–71). Among this group, rDRSS offered finer prognostic stratification, revealing substantial variability in survival outcomes across its risk categories, ranging from a 59% to 73% OS probability at 2 years after HCT.

Conclusions: This study successfully validated the DRSS in a large contemporary acute leukemia cohort. Additionally, we propose a refined model (rDRSS), updated to reflect recent HCT practice in acute leukemia. Both the DRSS and rDRSS can serve as benchmarks for future studies, facilitating interpretation and disease adjustment within heterogeneous cohorts, and enabling trial designs with more inclusive populations.

Disclosure: None

19: Acute Leukaemia

Ali Bazarbachi ¹, Myriam Labopin², Iman Abou Dalle¹, Ibrahim Yakoub-Agha³, Régis Peffault de Latour⁴, Thomas Schroeder⁵, Didier Blaise⁶, Xavier Poiré⁷, Marie Balsat⁸, Urpu Salmenniemi⁹, Nicolaus Kröger¹⁰, Alexander Kulagin¹¹, Eva Maria Wagner-Drouet¹², Depei Wu¹³, Eolia Brissot¹⁴, Arnon Nagler¹⁵, Sebastian Giebel¹⁶, Fabio Ciceri¹⁷, Mohammad Mohty¹⁴

¹American University of Beirut Medical Center, Beirut, Lebanon, ²EBMT Statistical Unit, Sorbonne University, Paris, France, ³CHU de Lille, Lille, France, ⁴Saint-Louis Hospital, Paris, France, ⁵University Hospital Essen, Essen, Germany, ⁶Transplantation and Cellular Immunotherapy Program, Marseille, France, ⁷Cliniques Universitaires St. Luc, Brussels, Belgium, ⁸Centre Hospitalier Lyon Sud, Lyon, France, ⁹Comprehensive Cancer Center, Stem Cell Transplantation Unit, Helsinki University Hospital, Helsinki, Finland, ¹⁰University Medical Center, Hamburg, Germany, ¹¹Alexander Kulagin, Saint-Petersburg, Russian Federation, ¹²University Medical Center Mainz, Mainz, Germany, ¹³First Affiliated Hospital of Soochow University, Suzhou, China, ¹⁴Sorbonne University, Paris, France, ¹⁵Chaim Sheba Medical Center, Tel-Hashomer, Israel, ¹⁶Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice, Poland, ¹⁷IRCCS Ospedale San Raffaele, Milano, Italy

Background: Allogeneic hematopoietic cell transplantation (allo-HCT) remains an important curative modality for patients with Philadelphia-positive acute lymphoblastic leukemia (Ph+ ALL) in first complete remission (CR1). Recent results using the combination of blinatumomab and second or third-generation tyrosine kinase inhibitors (TKI) have challenged the role of allo-HCT in CR1.

Methods: Using a large dataset from the EBMT registry, we identified 3292 adult patients (45% female; median age 45 years, range 18-76) with Ph+ ALL allografted between 2001 and 2020 in CR1 from a matched sibling (38%), unrelated (54%) or haploidentical donor (8%). At transplant, 41% of patients were measurable disease (MRD) positive (pos). Conditioning was myeloablative (MAC) in 77% of patients, and included total body irradiation (TBI) in 64% of patients. Median follow-up for live patients was 56 months (interquartile range [IQR] 28-32).

Results: We compared changes in patient and transplant characteristics over time in 245 patients transplanted in 2001-2005, 679 patients transplanted in 2006-2010, 1035 patients transplanted in 2011-2015, and 1333 patients transplanted in 2016-2020. Patients transplanted in recent years were older, were less likely to be MRDpos, and more likely to receive peripheral blood stem cells (PBSC) and in vivo T cell depletion (TCD). The 3-year cumulative incidence of relapse (CIR) gradually and significantly decreased from 41% to 32%, 27%, and 19% over the 4 time periods (p = 0.001), and non-relapse mortality (NRM) significantly decreased as well from 25% to 24%, 23% and 17% (p = 0.001). The 3-year leukemia-free survival (LFS) and overall survival (OS) gradually and significantly improved over time from 34% to 44%, 50%, and 64% (p = 0.001) and from 47% to 56%, 65%, and 75% (p = 0.001), respectively. Finally, graft-versus-host disease (GVHD)-free, relapse-free survival (GRFS) improved from 26% to 34%, 37%, and 49% (p = 0.001). In a Cox regression multivariate analysis (MVA), a progressive and significant improvement in all transplant outcomes was noted including reduced acute and chronic GVHD, reduced CIR and NRM, and increased LFS, OS, and GRFS. This improvement in post-transplant outcomes over time was observed both in MRDpos and in MRD negative (neg) patients. In MRDneg patients, 3-year CIR decreased from 34% to 30%, 24%, and 17% (p = 0.001) over the 4 time periods whereas LFS improved from 41% to 46%, 52%, and 66% (p = 0.001). Similarly, in MRD-positive patients, CIR decreased from 48% to 34%, 32%, and 23% (p = 0.001) over the 4 time periods whereas LFS improved from 27% to 42%, 47% and 60% (p = 0.001).

Conclusions: In patients with Ph+ ALL, we observed an impressive improvement over time in post-transplant outcomes with decreased CIR and NRM and improved LFS, OS, and GRFS, both in MRDpos and MRDneg patients. These large-scale, real-world data can serve as a benchmark for future studies in this setting, including those testing the combination of TKI and blinatumomab as an alternative to transplant.

Disclosure: Nothing to Disclose

19: Acute Leukaemia

Jean El Cheikh ¹, Maud Ngoya², Jacques-Emmanuel Galimard², Péter Reményi³, Alexander Kulagin⁴, Mahmoud Aljurf⁵, Ashrafsadat Mousavi⁶, Depei Wu⁷, Mutlu Arat⁸, Urpu Salmenniemi⁹, Cristina Castilla-Llorente¹⁰, Gerard Socie¹¹, Grzegorz Helbig¹², Thomas Schroeder¹³, Ioanna Sakellari¹⁴, Alessandro Rambaldi¹⁵, Ben Carpenter¹⁶, Alessandro Busca¹⁷, Helene Labussiere-Wallet¹⁸, Matthias Stelljes¹⁹, Eolia Brissot²⁰, Sebastien Giebel²¹, Zina Peric²², Arnon Nagler^23,24, Ali Bazarbachi²⁵, Fabio Ciceri²⁶, Mohamad Mohty²⁰

¹AUBMC, Beyrouth, Lebanon, ²EBMT ALWP Saint Antoine Hospital Sorbonne University, Paris, France, ³Dél-Pesti Centrumkórház – Országos Hematológiai és Infektológiai Intézet, Budapest, Hungary, ⁴RM Gorbacheva Research Institute, Pavlov University, Saint Petersburg, Russian Federation, ⁵c. King Faisal Specialist Hospital & Research Centre, Oncology (Section of Adult Haematolgy/BMT), Riyadh,, Saudi Arabia, ⁶Shariati Hospital, Hematology-Oncology and BMT Research, Teheran, Iran, Islamic Republic of, ⁷First Affiliated Hospital of Soochow University, Suzhou, China, ⁸Demiroglu Bilim University Istanbul Florence Nightingale Hospital, Hematopoietic SCT Unit, Istanbul, Turkey, ⁹HUCH Comprehensive Cancer Center, Stem Cell Transplantation Unit, Helsinki, Finland, ¹⁰Gustave Roussy Cancer Campus, BMT Service Villejuif, Paris, France, ¹¹Hopital St. Louis, Paris, France, ¹²Silesian Medical Academy, Katowice, Poland, ¹³University Hospital, Essen, Germany, ¹⁴George Papanicolaou General Hospital, Thessaloniki., Greece, ¹⁵University of Milan and Azienda Socio-Sanitaria Territoriale Papa Giovanni XXIII, Bergamo, Italy, ¹⁶University College London Hospital, London, United Kingdom, United Kingdom, ¹⁷S.S.C.V.D Trapianto di Cellule Staminali, A.O.U Citta della Salute e della Scienza di Torino, Torino., Italy, ¹⁸Centre Hospitalier Lyon Sud, Service Hematologie, Lyon, France, ¹⁹University of Muenster, Muenster, Germany, ²⁰Sorbonne University, Saint Antoine Hospital, INSERM UMRs 938, Paris, France, ²¹Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice, Poland, ²²School of Medicine, University Hospital Centre Zagreb, University of Zagreb, Zagreb, Croatia, ²³Sheba Medical Center, Tel Aviv University, Tel Hashomer, Israel, ²⁴Tel Aviv University, Ramat Gan, Israel, ²⁵Bone Marrow Transplantation Program, American University of Beirut Medical Center, Beirut, Lebanon, ²⁶Ospedale San Raffaele s.r.l. Hematology and BMT, Milano, Italy

Background: T-cell acute lymphoblastic leukemia (T-ALL) predominantly affects individuals in late childhood and young adulthood. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a curative modality particularly in the setting of poor risk genetics and/or persistent minimal residual disease. Limited studies have directly explored the impact of patient- and transplant-related factors on post-transplant outcomes in T-ALL.

Methods: Using a large dataset from the European Society for Blood and Marrow Transplantation registry, we identified 1907 adult T-ALL patients (70% male) who underwent their first allo-HSCT in first complete remission (CR1) from matched sibling donors (MSD; 45%), unrelated donors (UD; 43%) or haploidentical donors (12%) between 2010 and 2021.

Results: The median age at transplant was 33.4 years (18.1-75). The median follow up was 2.9 years. Most patients underwent total body irradiation (TBI)-based myeloablative conditioning (69%). The 2-year overall survival (OS) was 69.4%, and leukemia-free survival (LFS) was 62.1%. In multivariate analysis, advanced age at transplant negatively affected LFS (for each 10-year increment, HR = 1.11, p = 0.004), GVHD-free, relapse-free survival (GRFS) (HR = 1.06, p = 0.04), OS (HR = 1.12, p = 0.002), and non-relapse mortality (NRM) (HR = 1.23, p = 0.001). A later year of HSCT was associated with improved GFRS (For each 3 year increment, HR = 0.89, p = 0.001), OS (HR = 0.9, p = 0.02), and decreased NRM (HR = 0.82, p = 0.008). TBI improved LFS (HR = 0.79, p = 0.02), GRFS (HR = 0.83, p = 0.04), and relapse incidence (RI) (HR = 0.65, p = 0.001). Female-to-male transplant negatively affected GFRS (HR = 1.21, p = 0.02) and OS (HR = 1.23, p = 0.048). In vivo T-cell depletion significantly improved GFRS.

Multivariate Analysis of Prognostic Factors Influencing Patient Outcomes Following Allogeneic Hematopoietic Stem Cell Transplantation.

Conclusions: This large study identified prognostic factors, such as age at transplant conditioning regimen, in influencing post-transplant in adult T-ALL patients undergoing allo-HSCT. Importantly, a significant improvement over time was noted. These findings hold great promise for new adapted treatment strategies and can serve as a benchmark for future studies in that setting.

Disclosure: no disclosure

19: Acute Leukaemia

Lin Li ¹, Yishan Ye¹, Jacques-Emmanuel Galimard², Myriam Labopin², Depei Wu³, Jia Chen³, Nicolaus Kröger⁴, Jakob Passweg⁵, Urpu Salmenniemi⁶, Maija Itäla-remes⁷, Xavier Poiré⁸, Matthias Eder⁹, Johan Maertens¹⁰, David Burns¹¹, Henrik Sengeloev¹², Gitte Olesen¹³, Didier Blaise¹⁴, Jürgen Finke¹⁵, Alain Gadisseur¹⁶, Ali Bazarbachi¹⁷, Eolia Brissot², Arnon Nagler¹⁸, Yi Luo¹, Jimin Shi¹, Fabio Ciceri¹⁹, Mohamad Mohty², He Huang¹, Norbert Claude Gorin²

¹The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China, ²EBMT Paris Office, Hôpital Saint Antoine 184, rue du faubourg Saint Antoine 75571 Paris cedex 12, Paris, France, ³First Affiliated Hospital of Soochow University, Suzhou, China, ⁴University Hospital Eppendorf, Hamburg, Germany, ⁵University Hospital Basel, Basel, Switzerland, ⁶HUCH Comprehensive Cancer Center, Helsinki, Finland, ⁷Turku University Hospital, Turku, Finland, ⁸Cliniques Universitaires St. Luc, Brussels, Belgium, ⁹Hannover Medical School, Hannover, Germany, ¹⁰University Hospital Gasthuisberg, Leuven, Belgium, ¹¹University Hospital Birmingham NHS Trust, Stoke, United Kingdom, ¹²Rigshospitalet, Copenhagen, Denmark, ¹³Aarhus University Hospital Denmark, Aarhus, Denmark, ¹⁴Programme de Transplantation & Therapie Cellulaire, Marseille, France, ¹⁵University of Freiburg, Freiburg, Germany, ¹⁶Antwerp University Hospital (UZA), Antwerp E, Belgium, ¹⁷American University of Beirut, Beirut, Lebanon, ¹⁸Hematology Division, Chaim Sheba Medical Center, Tel-Hashomer, Israel, ¹⁹Unit of Hematology and Stem Cell Transplantation, Ospedale San Raffaele, University Vita-Salute San Raffaele, Milano, Italy

Background: Mutations in TET2 present in about 15% of adult AML. However the impact of allogeneic hematopoietic cell transplantation (allo-HCT) for the treatment of these patients remains unclear.

Methods: 644 adult AML patients with TET2 mutations receiving first non ex-vivo depleted allo-HCT from 2013-2022 in 127 centers were analyzed. All patients achieved first complete remission (CR1) before allo-HCT.

Results: The median age of the 644 AML patients was 59.4 (range, 18.1-86.3) years. 556 patients (86.3%) had de novo, and 88 (13.7%) had secondary AML, respectively. 367 (57.2%) patients were male. The median interval from diagnosis to allo-HCT was 4.9 (IQR, 3.8-6.2) months. Patients were categorized into favorable -(N = 23, 3.9%), intermediate-(N = 475, 79.8%) and adverse (N = 97, 16.3%) risk categories according to cytogenetic characteristics. Conditioning regimen was myeloablative (MAC) in 46.7% of patients. 149 patients (23.1%) received a matched-sibling donor, 176 (27.3%) a haploidentical donor (Haplo), 268 (41.6%) a 10/10 unrelated donor (UD) and 51 (7.9%) a 9/10 UD allo-HCT, respectively.For the entire cohort, with a median follow-up of 1.9 years, the 30-day cumulative incidence of engraftment was 97.6%. The 100-day cumulative incidence of grade II-IV aGVHD was 22.9% and the 2-year cumulative incidence of extensive cGVHD was 13.7%. In addition, the 2-year relapse incidence and NRM were 23.2% and 12.8%, respectively. Finally, the 2-year OS, LFS and GRFS were 69.8%, 63.9% and 49.8%, respectively.In multivariable analyses, a donor type 9/10 UD was associated with a higher RI (HR = 2.33, 95% CI 1.21-4.48; p = 0.01), a lower LFS (HR = 2.20, 95% CI 1.31-3.70; p < 0.01) and OS (HR = 2.29, 95% CI 1.33-3.96; p < 0.01) as compared to MSD as reference. Compared to MSD, 10/10 UD and Haplo had similar outcomes. The adverse-cytogenetic group was associated with higher RI (HR = 1.85, 95% CI 1.19-2.86; p < 0.01), and higher incidence of grade II-IV acute GVHD (HR = 1.74, 95% CI 1.14-2.64; p = 0.01) as compared to other groups combined as reference. Compared to de novo AML (reference), secondary AML was associated with lower LFS (HR = 1.49, 95% CI 1.03-2.15; p < 0.05) and OS (HR = 1.68, 95% CI 1.15-2.45; p < 0.01). When age increased, patients faced higher RI (HR for 10y increment=1.24, 95% CI 1.01-1.53; p < 0.05) and extensive chronic GVHD (HR = 1.34, 95% CI 1.02-1.77; p < 0.05), which translated into higher NRM (HR = 1.66, 95% CI 1.23-2.24; p < 0.01), lower LFS (HR = 1.36, 95% CI 1.14-1.62; p < 0.01) and OS (HR = 1.60, 95% CI 1.31-1.96; p < 0.01). Later year of transplantation was associated with lower incidence of II-IV acute GVHD (HR for 2y increment=0.76, 95% CI 0.63-0.92; p < 0.01), chronic GVHD (HR = 0.67, 95% CI 0.54-0.83; p < 0.01) and extensive chronic GVHD (HR = 0.72, 95% CI 0.52-0.99; p < 0.05). Female to male, patient CMV status and interval between diagnosis and HCT did not affect significantly the transplant outcomes in multivariable analysis.

Conclusions: No significant difference was observed in outcomes of AML patients with TET2 mutations following MSD, HAPLO and 10/10 UD allo-HCTs, while 9/10 UD was associated with higher relapse incidence and lower survival outcomes. Higher patient age, secondary AML and adverse cytogenetics predicted worse transplant outcomes. GVHD prevention improved with time for these patients.

Disclosure: Nothing to declare.

19: Acute Leukaemia

Arnon Nagler ¹, Maud Ngoya², Jacques-Emmanuel Galimard², Myriam Labopin^3,4, Igor Wolfgang Blau⁵, Nicolaus Kröger⁶, Tobias Gedde-Dahl⁷, Thomas Schroeder⁸, David Burns⁹, Urpu Salmenniemi¹⁰, Alessandro Rambaldi¹¹, Goda Choi¹², Régis Peffault de Latour¹³, Jan Vydra¹⁴, Henrik Sengeloev¹⁵, Matthias Eder¹⁶, Stephan Mielke¹⁷, Edouard Forcade¹⁸, Sergey Bondarenko¹⁹, Fabio Ciceri²⁰, Mohamad Mohty^3,4

¹Sheba Medical Center, Division of Hematology, Tel Hashomer, Israel, ²EBMT Statistical Unit, Paris, France, ³EBMT Paris study office; Saint Antoine Hospital, Paris, France, ⁴Sorbonne University, Saint-Antoine Hospital, AP-HP, INSERM UMRs 938, Paris, France, ⁵Medizinische Klinik m. S. Hämatologie, Onkologie und Tumorimmunologie, Berlin, Germany, ⁶University Hospital Eppendorf, Eppendorf, Germany, ⁷Oslo University Hospital, Rikshospitalet, Norway, ⁸University Hospital of Essen, Essen, Germany, ⁹University Hospital Birmingham NHSTrust, Birmingham, United Kingdom, ¹⁰HUCH Comprehensive Cancer Center, Helsinki, Finland, ¹¹ASST Papa Giovanni XXIII, Bergamo, Italy, ¹²University Medical Center Groningen (UMCG), Groningen, Netherlands, ¹³Saint-Louis Hospital, BMT Unit, Paris, France, ¹⁴Institute of Hematology and Blood Transfusion, Prague, Czech Republic, ¹⁵Rigshospitalet, Herlev, Denmark, ¹⁶Hannover Medical School, Hannover, Germany, ¹⁷Karolinska University Hospital, Stockholm, Sweden, ¹⁸CHU Bordeaux, Hôpital Haut-Leveque, Pessac, France, ¹⁹RM Gorbacheva Research Institute, Pavlov University, Petersburg, Russian Federation, ²⁰IRCCS Osspedale San Raffaele, Vita-Salute San Raffaele University Hematology and BMT, Milano, Italy

Background: Post-transplant cyclophosphamide (PTCy) reduces chronic (c) graft-versus-host disease (GVHD) and thus may suppress the graft-versus-leukemia (GVL) effect leading to increased relapse (RI).

Methods: The study aim was to compare RI post 9-10/10 unrelated donor transplantations (UD-HSCT) with PTCy vs. no PTCy (in vivo T-cell depletion or calcineurin inhibitor-based) GVHD prophylaxis, in adult patients (pts) with AML in CR1 transplanted between 2014 and 2021 with peripheral blood grafts. Engraftment was a secondary endpoint. Statistical tests included a multivariate analysis (MVA) using a Cox proportional-hazards regression model for main outcomes.

Results: In total, 7049 pts were included: 707 received PTCy, and 6342 did not (no PTCy). Median follow-up was 2.8 (range, 2.6-2.9) years(y). The pts in the PTCY group were younger, median age was 52.7 (range, 18-77.5) vs. 56.6 (range, 18-80.3) y (p < 0.001). Cytogenetic risk category (ELN2017 classification) did not differ between the groups, (p = 0.54) (missing data-1196 pts). There were more HLA 9/10 donors in the PTCy group 33.8% vs. 16.4% (p < 0.001). More pts in the PTCy group received myeloablative conditioning, 61.7% vs. 50.2%, respectively (p < 0.001). In the no PTCy group, 87.7% of pts received in vivo T-cell depletion.

Day 30 neutrophil (ANC > 0.5x10⁹/L) and platelet ( > 20x10⁹/L) engraftment was achieved in 93.8% and 80.9% vs. 97.6% and 92.6% of the pts in the PTCy and no PTCy groups, respectively. Median days of engraftment were +17 and +20 vs. + 14 and + 18, respectively (p < 0.001). In the MVA, RI was not significantly different in the PTCy vs. the no PTCy group, with a hazard ratio (HR) of 1.11 (95% CI 0.9-1.37) (p = 0.31). Relapse was the cause of death in 58.8% vs. 54.3% of the pts that died. Myeloid and platelet engraftment were significantly lower in the PTCy vs. the no PTCy group, HR = 1.67 (95% CI 1.47-1.92, p < 0.001) and HR = 2.44 (95% CI 2.17-2.78, p < 0.001). Day 180 acute (a) GVHD (grades II-IV and severe grades III-IV) were significantly lower with PTCy compared to the no PTCy group, with HR of 0.74 (95% CI 0.59-0.92, p = 0.007) and HR = 0.56 (95% CI 0.38-0.83, p = 0.004), as were 2-y cGVHD (total and extensive) with HRs of 0.5 (95% CI 0.41-0.62, p = 0.001) and HR = 0.31 (95% CI 0.22-0.42, p = 0.001). NRM was significantly lower with PTCy compared to the no PTCy group with an HR of 0.67 (95% CI 0.5-0.91, p = 0.007). GRFS was higher in the PTCY vs the no PTCy group, with a HR = 0.69 (95% CI 0.59-0.81, p = 0.001). LFS and OS did not differ significantly between the groups, with HRs of 0.93 (95% CI 0.78-1.1, p = 0.39) and HR = 0.93 (95% CI 0.77-1.12, p = 0.45), respectively.

Conclusions: In this registry-based, retrospective analysis of UD-HSCT with PTCy vs. no PTCy in pts with AML in CR1, we observed non-significantly different RI, OS, and LFS. Incidence of GVHD and NRM were significantly lower, while GRFS was significantly higher with PTCy compared to no PTCy anti-GVHD prophylaxis. Engraftment was significantly lower with PTCy. Clinically, these results do not support PTCy negating the GVL effect in UD-HSCT.

Disclosure: Nothing to declare

19: Acute Leukaemia

Yishan Ye ¹, Myriam Labopin², Socié Gérard³, Ibrahim Yakoub-Agha⁴, Igor Wolfgang Blau⁵, Mahmoud Aljurf⁶, Edouard Forcade⁷, Tobias Gedde-Dahl⁸, David Burns⁹, Jan Vydra¹⁰, Khaled Halahleh¹¹, Rose-Marie Hamladji¹², Ali Bazarbachi¹³, Arnon Nagler¹⁴, Eolia Brissot², Lin Li¹, Yi Luo¹, Yanmin Zhao¹, Fabio Ciceri¹⁵, He Huang¹, Mohamad Mohty^2,16, Norbert Claude Gorin^2,16

¹The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China, ²EBMT Paris Office, Paris, France, ³Saint-Louis Hospital, Paris, France, ⁴CHU de Lille, Lille, France, ⁵Medizinische Klinik m. S. Hämatologie, Berlin, Germany, ⁶King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia, ⁷CHU Bordeaux, Hopital Haut-Leveque, Bordeaux, France, ⁸Oslo University Hospital, Oslo, Norway, ⁹University Hospital Birmingham NHSTrust, Birmingham, United Kingdom, ¹⁰Institute of Hematology and Blood Transfusion, Prague, Czech Republic, ¹¹King Hussein Cancer Centre Adult BMT Program, Amman, Jordan, ¹²Centre Pierre et Marie Curie, Alger, Algeria, ¹³American University of Beirut Medical Center, Beirut, Lebanon, ¹⁴Chaim Sheba Medical Center, Tel-Hashomer, Israel, ¹⁵Ospedale San Raffaele s.r.l., Haematology and BMT, Milano, Italy, ¹⁶Hospital Saint Antoine, Paris, France

Background: Although core-binding factor acute myeloid leukemia (CBF-AML) generally has a favorable prognosis, relapse occurs in ~40% of cases. Allogeneic hematopoietic cell transplantation (allo-HCT) is recommended when patients achieve second complete remission (CR2). However, transplant outcomes and donor preference for these patients remain unclear. We compared in an EBMT global multi-center registry-based analysis the outcomes following allo-HCT using either haploidentical donors (HAPLO), matched siblings donors (MSD), or 10/10 matched unrelated donors (MUD).

Methods: Data from 865 de novo adult CBF AML patients in CR2 receiving allo-HCT in 227 EBMT centers from 2010-2022 were analyzed. Transplants from mismatched UD ( < 10/10), umbilical cord blood or grafts with ex-vivo manipulation were excluded. Univariate analyses and Cox regression models were used.

Results: Results from 329 MSD, 374 MUD, and 162 HAPLO HCTs were analyzed. For the entire cohort, median age was 43y (range: 18-76). Inv(16)/CBFBMYH11 was detected in 503 patients (58%) and t(8;21)/RUNX1-RUNX1T1 was detected in 362 patients (42%). Peripheral blood was the major graft source (86.1%), and myeloablative conditioning was used for the majority (60.1%) of all allo-HCTs. 83% of patients in the HAPLO group received post transplant cyclophosphamide (PTCy) as graft-versus-host disease (GVHD) prophylaxis. Engraftment rates were higher in the MSD (98.8%) and MUD (98.9%) than in the HAPLO (94.9%) group (p < 0.01). Original disease was the major cause of death for the MSD and MUD cohorts, while in the HAPLO group, it was infection.

On univariate analysis, HAPLO HCT was associated with a higher 2-year non-relapse mortality (NRM) (MSD 17.1%, MUD 15.3%, HAPLO 25.7%). The 180-day cumulative incidence of grade II-IV acute GVHD was higher in HAPLOs (14.8%) than in MSD (6.8%) or MUD (7.9%) recipients, respectively. No significant difference was observed among the three cohorts for 2-year leukemia-free survival (LFS), overall survival (OS) and GVHD-free/relapse free survival (GRFS). CBF-AML with inv(16) was associated with lower relapse incidence RI (18.1%) than with t(8;21) (28.3%), which contributed to better OS (75.7% vs 59.8% at 2 years), LFS (65.8% vs 51.2%) and GRFS (50.8% vs 38.5%).

On multivariate analysis, no significant difference was observed among HAPLO, MSD and MUD allo-HCTs on NRM. HAPLO HCT was associated with lower RI compared to MSD (hazard ratio [HR] = 0.56, 95% CI 0.32-0.97; p < 0.05) and MUD (HR = 0.57, 95% CI: 0.33-0.99, p < 0.05) while there was no significant differences between the 3 types of donors on LFS, OS and GRFS. CBF-AML with t(8;21) was associated with both higher RI (HR = 1.79, 95% CI 1.3-2.47; p < 0.01) and higher NRM (HR = 1.58, 95% CI 1.1-2.27; p < 0.01) than CBF-AML with inv(16), which led to worse LFS, OS and GRFS. Higher age (per 10 year) was associated with higher risk of NRM (HR = 1.29, 95% CI 1.11-1.5; p < 0.01) and worse OS (HR = 1.13, 95% CI 1.01-1.27; p = 0.03).

Conclusions: For CBF-AML patients in CR2, haploidentical HCTs were associated with a lower RI compared to MSD and MUD allo-HCTs. HAPLO achieved similar LFS, OS and GRFS to MSD or MUD, respectively. In addition, CBF AML with inv(16) had better outcomes than with t(8;21) after allo-HCT.

Disclosure: Nothing to declare

19: Acute Leukaemia

Arnon Nagler ¹, Myriam Labopin^2,3, Jurjen Versluis⁴, Jaime Sanz⁵, Tobias Gedde-Dahl⁶, David Burns⁷, Michel Schaap⁸, Hélène Labussière-Wallet⁹, Peter von dem Borne¹⁰, Gwendolyn Van Gorkom¹¹, Nathalie Contentin¹², Andreas Neubauer¹³, Eva Maria Wagner-Drouet¹⁴, Nicolaus Kröger¹⁵, Jordi Esteve¹⁶, Fabio Ciceri¹⁷, Mohamad Mohty^2,3

¹Sheba Medical Center, Division of Hematology, Tel Hashomer, Israel, ²EBMT Paris study office; Saint Antoine Hospital, Paris, France, ³Sorbonne University, Saint-Antoine Hospital, AP-HP, INSERM UMRs 938, Paris, France, ⁴Erasmus MC Cancer Institute, Rotterdam,, Netherlands, ⁵University Hospital La Fe, Valencia, Spain, ⁶Oslo University Hospital, Rikshospitalet,, Norway, ⁷University Hospital Birmingham NHSTrust, Birmingham, United Kingdom, ⁸Nijmegen Medical Centre, Nijmegen, Netherlands, ⁹Hopital Lyon Sud, Hospices Civils de Lyon, Pierre Bénite, France, ¹⁰Leiden University Hospital, Leiden, Netherlands, ¹¹University Hospital Maastricht, Maastricht, Netherlands, ¹²Centre Henri Becquerel, Rouen, France, ¹³Philipps Universitaet Marburg, Marburg, Germany, ¹⁴University Medical Center Mainz, Mainz, Germany, ¹⁵University Hospital Eppendorf, Eppendorf, Germany, ¹⁶Hospital Clínic of Barcelona, IDIBAPS, Barcelona, Spain, ¹⁷IRCCS Osspedale San Raffaele, Vita-Salute San Raffaele University Hematology and BMT, Milano, Italy

Background: Patients (pts) formerly diagnosed as AML with myelodysplasia-related changes are currently recategorized according to their underlying genetic events in three different AML subgroups, namely AML with mutated TP53, AML with myelodysplasia-related gene mutations (MR-GM), and AML with myelodysplasia-related cytogenetic abnormalities (MR-CG) (International Consensus Classification 2022) (Arber AD, Blood 2022). The relevance of this new classification for transplantation (HSCT) outcome is largely unknown.

Methods: We analyzed outcomes of AML pts with TP53-mutation, pts with any of the 9 MR-GM (Arber AD, Blood 2022), and pts with MR-CG without TP53 and or MR-GM transplanted in CR1 between 2010-2022 from matched siblings, unrelated or haplo donors. Pts were hierarchically assigned to these genetic categories after excluding any pre-defined AML with recurrent genetic abnormalities (ICC 2022). The TP53 pts were subdivided into those that harbor in addition to the TP53 mutation complex karyotype (CK) (TP53 mut.CK) and those with no CK (TP53 mut. notCK).

Results: 1152 pts were included: TP53 -(n = 527: CK -328; notCK -199), MR-GM (n = 379), and MR-CG (n = 246). Median FU was 24 mos. [IQR, 21-24]. Median age was 60.3 years (range 18.4-78.2). FLT3-ITD was more prevalent in the MR-GM group at 17.1% vs.7.4 % and 10.7% in the other 2 groups (p = 0.001). A higher proportion of pts in the TP53 group had KPS < 90 32.4% vs. 23.2% and 27.7% in the MR-GM and MR-CG groups (p < 0.012). A higher proportion of pts in the MR-CG group received MAC at 46.5% vs. 37.8% and 37% in the TP53 and MR-GM groups (p = 0.036). Finally, more pts in the MR-GM group received PTCy as GVHD prophylaxis 38.8% vs. 24.7% and 29.1% of the pts in the TP-53 and MR-GM groups (p < 0.0001). On MVA, pts in the MR-GM and MR-CG groups showed a better outcome compared to. the TP53 group with lower RI: HR = 0.24 (95% CI 0.18-0.33, p < 0.0001) and HR = 0.4 (95% CI 0.29-0.54, p < 0.0001) and better LFS HR = 0.37 (95% CI 0.3-0.47, p < 0.0001) and HR = 0.52 (95% CI 0.41-0.66, p < 0.0001), OS HR = 0.37 (95% CI 0.39-0.46, p < 0.0001) and HR = 0.5 (95% CI 0.39-0.64, p < 0.0001) and GRFS HR = 0.47 (95% CI 0.38-0.57, p < 0.0001) and HR = 0.64 (95% CI 0.51-0.79, p < 0.0001), respectively. Outcome was the worse for the TP53 with CK group. NRM, acute, and chronic GVHD did not differ significantly. In addition, pts with MR-CG showed worse outcome vs. those with MR-GM with higher RI HR = 1.67 (95% CI 1.14-2.43, p = 0.008) and lower LFS HR = 1.39 (95% CI 1.05-1.84, p < 0.021), OS HR = 1.36 (95% CI 1.01-1.82, p < 0.043) and GRFS HR = 1.37 (95% CI 1.07-1.75, p < 0.012). Of note, patients with sAML did not show a worse outcome than those with de novo AML after adjustment for genetic features, on MVA.

Conclusions: Analysis of HSCT outcomes in pts with myelodysplasia-related features per ICC 2022 classification demonstrated worse outcomes in pts with TP53 compared to those with MR-GM or MR-CG. Notably, the outcomes of pts with MR-GM were better than those with MR-CG. This is to our knowledge the first validation of the ICC2022 classification in AML pts with myelodysplasia-related changes undergoing transplantation.

Disclosure: Nothing to declare

19: Acute Leukaemia

Yeqian Zhao^1,2,3,4, Jimin Shi^1,2,3,4, Yi Luo^1,2,3,4, Guifang Ouyang⁵, Lieguang Chen⁵, Jian Yu^1,2,3,4, Yamin Tan⁶, Xiaoyu Lai^1,2,3,4, Lizhen Liu^1,2,3,4, Huarui Fu^1,2,3,4, Yishan Ye^1,2,3,4, Luxin Yang^1,2,3,4, Congxiao Zhang^1,2,3,4, He Huang^1,2,3,4, Xiaoxia Hu^1,2,3,4, Yanmin Zhao ^1,2,3,4

¹Bone Marrow Transplantation Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China, ²Institute of Hematology, Zhejiang University, Hangzhou, China, ³Liangzhu Laboratory, Zhejiang University Medical Center, Hangzhou, China, ⁴Zhejiang Province Engineering Research Center for Stem Cell and Immunity Therapy, Hangzhou, China, ⁵Ningbo First Hospital, Ningbo, China, ⁶Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Hangzhou, China

Background: Post-transplant sorafenib maintenance could reduce relapse in patients with FLT3-ITD acute myeloid leukemia (AML), however, a subset of patients still experienced relapse after allogeneic hematopoietic stem-cell transplantation (allo-HSCT) even in the era of tyrosine kinase inhibitor (TKI). This study aimed to identify specific subgroups of FLT3-ITD AML experiencing poorer outcomes after allo-HSCT in the context of sorafenib maintenance.

Methods: This multi-center retrospective cohort study included patients with AML who underwent allo-HSCT at three hospitals between January 2016 and December 2022. The study’s inclusion criteria were (1) aged from 14 to 70 years, (2) diagnosed as FLT3 -ITD AML, (3) achieving complete remission (CR) post-transplantation and subsequently receiving sorafenib as part of their post-HSCT maintenance. A comprehensive profiles of cytogenetic and concomitant gene mutation were screened to identify the genetic risk factors influencing prognosis in FLT3-ITD AML patients.

Results: A total of 171 patients diagnosed with FLT3-ITD AML receiving allo-HSCT were included in the analysis. The median age of patients was 45 (range: 14-70) years, with approximately 30% manifesting refractory/relapse (R/R) diseases (Table 1). At the time of transplantation, 88% had achieved CR/CRi. The median follow-up period after transplantation was 20 months, 20% of the cohort experienced relapse, with a median time to relapse was 18 months (IQR 10-33 months) post-transplantation. The 2-year overall survival (OS) was 85.5% (95% CI 80.0%-91.3%). The 2-year relapse-free survival (RFS) and cumulative incidence of relapse were 79.1% (95% CI 72.8%-86.0%) and 20.3% (95% CI 14.7%-27.6%), respectively. Based on the European LeukemiaNet 2022 risk stratification (ELN2022), 25% of the patients were classified into the adverse-risk group, among whom 12% harbored myelodysplasia-related (MR) gene mutations, including ASXL1, BCOR, EZH2, RUNX1, SF3B1, SRSF2, STAG2, U2AF1, and ZRSR2. Univariate analyses have suggested that MR gene mutations, DNMT3A mutations and non-remission status at transplantation significantly correlated with poorer OS, whereas R/R disease, along with aforementioned factors, indicated significantly inferior RFS. The FLT3-ITD allelic ratio, karyotype and application of pre-transplant TKIs have no impact on OS and RFS. Among patients in CR at transplant, there were no differences in prognosis between those who had minimal residual disease (MRD) negative, and MRD positive (2-year OS: 88.70% versus 83.30%, p = 0.930). Multivariate analysis by Cox proportional hazard model revealed that MR gene mutations were independently associated with adverse outcomes (OS: HR 11.51, 95%CI 1.44 -91.80, p = 0.021, RFS: HR 2.50, 95% CI 1.09-5.80, p = 0.031). Additionally, DNMT3A mutation was also identified as an independent risk factor affecting RFS (HR 2.7, 95%CI 1.30-5.60, p = 0.008). More interestingly, among the ELN2022 adverse-risk individuals, patients with MR gene mutations exhibited a significantly inferior OS and RFS, compared with those without the mutations, while the ELN2022 adverse risk individuals without MR gene mutations yielded survival outcomes similar to their ELN2022 intermediate risk counterparts.

Table 1. Baseline characteristics of 171 AML patients with FLT3-ITD receiving sorafenib maintenance post-transplantation

Conclusions: Among FLT3-ITD AML patients undergoing sorafenib maintenance post-transplantation, patients in CR at transplant will have similar outcomes regardless of MRD status. Further, those with MR gene mutations experienced inferior outcomes after HSCT. This particular subset of FLT3-ITD AML warrants further investigation.

Disclosure: The authors declare that they have no competing interests.

19: Acute Leukaemia

Ali Bazarbachi ¹, Myriam Labopin², Nour Moukalled¹, Nicolaus Kröger³, Christina Rautenberg⁴, Johannes Schetelig⁵, Jürgen Finke⁶, Igor Wolfgang Blau⁷, Didier Blaise⁸, Matthias Stelljes⁹, Matthias Eder¹⁰, Uwe Platzbecker¹¹, Peter Dreger¹², Wolfgang Bethge¹³, Johanna Tischer¹⁴, David Burns¹⁵, Henrik Sengeloev¹⁶, Eolia Brissot¹⁷, Sebastian Giebel¹⁸, Arnon Nagler¹⁹, Fabio Ciceri²⁰, Mohamad Mohty²¹

¹American University of Beirut Medical Center, Beirut, Lebanon, ²EBMT Statistical Unit, Sorbonne University, Saint-Antoine Hospital, Paris, France, ³University Medical Center, Hamburg, Germany, ⁴University Hospital Essen, Essen, Germany, ⁵Universitaetsklinikum Dresden, Dresden, Germany, ⁶University of Freiburg, Freiburg, Switzerland, ⁷Medizinische Klinik m. S. Hämatologie, Onkologie und Tumorimmunologie, Berlin, Germany, ⁸Instititut Paoli Calmettes, MSC Lab, Aix Marseille University, Marseille, France, ⁹University Hospital Muenster, Muenster, Germany, ¹⁰Hannover Medical School, Hannover, Germany, ¹¹Medical Clinic and Policinic 1, Leipzig, Germany, ¹²University of Heidelberg, Heidelberg, Germany, ¹³University Hospital Tuebingen, Tuebingen, Germany, ¹⁴LMU University Hospital, Munich, Germany, ¹⁵University Hospital Birmingham NHSTrust, Stoke, United Kingdom, ¹⁶Rigshospitalet, Herlev, Denmark, ¹⁷Sorbonne University, Paris, France, ¹⁸Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice, Poland, ¹⁹Hematology Division, Chaim Sheba Medical Center, Tel-Hashomer, Israel, ²⁰IRCCS Ospedale San Raffaele, Milano, Italy, ²¹Sorbonne University, Saint-Antoine Hospital, Paris, France

Background: Acute myeloid leukemia (AML) is a disease of older patients. Progress in allogeneic hematopoietic cell transplantation (allo-HCT) allowed the delivery of allo-HCT to older patients. However, little information is available about the global impact of the current standard of care for older AML patients after allo-HCT and about the predictive factors for post-transplant outcomes. To address these challenges, we assessed real-world changes over time in transplant characteristics and post-transplant outcomes in older patients with AML (>65 years), using a large dataset from the EBMT registry.

Methods: We identified 7215 adult patients (40% female; median age 68 years, range 65-80) with AML (secondary AML in 25%) allografted between 2000 and 2021 from a matched sibling donor (MSD; 23%), unrelated donor (UD; 63%) or haploidentical donor (Haplo; 14%). Karyotype risk was intermediate in 71% and adverse in 25%. At transplant, 64% of patients were in first complete remission (CR1), 14% in CR2+ and 22% had active disease. Comorbidity index was zero in 43% of patients, 1-2 in 26% and >3 in 31%, while the Karnofsky score was <90 in 35%. Conditioning was reduced intensity (RIC) in 82% of patients. Median follow up was 40 months.

Results: Patients transplanted in recent years (65%) were older, were more likely to be in CR1 and to have an adverse-risk karyotype. Patients transplanted more recently were less likely to have had an MSD but more likely to have had a Haplo, peripheral blood stem cells and in vivo T cell depletion. The 3-year cumulative relapse incidence (RI) gradually and significantly decreased from 37% to 31%, then to 30% (p = 0.001) over the 3 time periods (2000-2009; 2010-2014; 2015-2021) whereas NRM decreased from 31% and 31% to 27% (p = 0.003). The 3-year leukemia free survival (LFS) and overall survival (OS) gradually and significantly improved from 32% to 38%, and then to 44% (p = 0.001) and from 37% to 42%, and then to 49% (p = 0.001), respectively. In multivariate analysis, significant improvement in the RI, LFS and OS were noted after 2015 whereas NRM was not significantly affected. This improvement of post-transplant outcomes over time was observed for patients in CR1 with LFS and OS increasing from 39% to 44% and then to 47% (p = 0.005) and from 43% to 48%, and then to 52% (p = 0.006) as well as for patients in CR2+ with LFS and OS increasing from 29% to 40% and then to 47% (p = 0.001) and from 37% to 44% and then to 52% (p = 0.001), respectively.

Conclusions: In older AML patients, we observed an impressive improvement over time in post-transplant outcomes, mostly attributed to decreased RI rather than decreased NRM, and regardless of disease status at transplant. These large-scale, real-world data can serve as a benchmark for future studies in this setting and indicate that the opportunity for transplant for the elderly should be mandatory and no longer an option.

Disclosure: Nothing to declare

19: Acute Leukaemia

Kai Yu¹, Jakob D. Devos², Mei-Jie Zhang², Kristin Page², Nelli Bejanyan³, Partow Kebriaei⁴, Filippo Milano⁵, Rachel Cook^6,7, Brian J. Druker^6,7, Katrin Ericson⁸, Paul Liu¹, Wael Saber², Lea Cunningham ^1,9

¹National Human Genome Research Institute/National Institutes of Health, Bethesda, United States, ²Center for International Blood and Marrow Transplant Research, Milwaukee, United States, ³Moffitt Cancer Center, Tampa, United States, ⁴MD Anderson Cancer Center, Houston, United States, ⁵Fred Hutchinson Cancer Center, Seattle, United States, ⁶Knight Cancer Institute, Oregon Health & Science University, Portland, United States, ⁷Division of Hematology & Medical Oncology, Oregon Health & Science University, Portland, United States, ⁸RUNX1 Research Program, Santa Barbara, United States, ⁹National Cancer Institute/National Institutes of Health, Bethesda, United States

Background: RUNX1 mutations have been reported in 10% of large de novo AML cohorts of which 27-30% were germline^1,2 and 16% were deleterious in subsequent analysis³. RUNX1 mutations are associated with poor prognosis and patients are usually referred for HSCT. A systematic analysis comparing outcomes for patients with germline vs somatic RUNX1 mutations has not been reported. We hypothesized that germline RUNX1 mutation status may affect HSCT outcomes. The primary objectives of this retrospective study were to estimate the prevalence of germline RUNX1 mutations and describe overall survival (OS), disease-free survival (DFS), transplant-related mortality (TRM), and relapse for patients with germline RUNX1 mutations who underwent allogeneic HSCT for AML. The secondary objective was to compare post-HSCT outcomes in AML patients with germline RUNX1 mutations to those with somatic RUNX1 mutations.

Methods: The CIBMTR registry was queried for individuals with RUNX1 mutations who underwent first allogeneic HSCT during 2013-2020 with banked pre-HSCT blood samples. Extracted DNA was used for targeted sequencing to detect RUNX1 small variants in all coding exons and flanking regions of RUNX1, to the coverage of >500x, using Illumina MiSeq protocol. A variant allele frequency (VAF) threshold in the range of 0.3-0.7 was used to categorize germline variants. A subset of variants that met pathogenic or likely pathogenic classification according to ACMG guidelines were deemed RUNX1 mutations.

Results: A total 330 patients (median follow up: 60.8 months (range, 3.3-115) from 98 centers were included. Of 330 patients, 6.7% (n = 22) patients had P/LP germline RUNX1 mutations and 93.3% (n = 308) had somatic mutations per the identified criteria. Patients with germline mutations were more likely to be older, (median 61 vs 49.1 years, p < 0.02), with primary induction failure (40.9% vs 33%, p < 0.01), with AML transformed from MDS rather than de novo (40.9% vs 13.3%, p < 0.01) and detectable measurable residual disease (MRD) pre-HSCT (84.6% vs 37.6%, p < 0.01). Patients with germline RUNX1 mutations had lower 2-year OS (40.9% vs 64.1%, p < 0.033) and DFS (36.4% vs 60%, p < 0.027) compared to patients with somatic RUNX1 mutations. TRM (13.8% vs 15.4%, p = 0.884) and relapse (23.6% vs 38.5%, p = 0.308) at 2 years were not different between the 2 groups. Within the subset of patients receiving HSCT in complete remission (n = 13), no differences in OS and DFS were observed between germline and somatic RUNX1 mutations. Sex, ethnicity, PS, HCT-CI, donor type, graft source, conditioning intensity, GVHD prophylaxis, year of HSCT, engraftment, GVHD were also not different between the two groups.

Conclusions: In this study, germline RUNX1 mutations are associated with greater likelihood of AML arising from MDS rather than de novo AML, primary induction failure, and MRD positivity at time of HSCT as compared to somatic RUNX1 mutations. While germline RUNX1 mutations results in lower OS and DFS after HSCT, if transplanted in CR the survival outcomes are similar to patients with somatic RUNX1. This study is not without limitations because true germline testing from cultured skin fibroblasts for germline confirmation was not possible. These results should be validated with a larger number of participants in multivariate analysis.

Disclosure: This study was supported by the Intramural Research programs of NHGRI, NCI of NIH, and by CIBMTR. Funding for DNA extraction from whole blood samples was also provided by the RUNX1 Research Program (501c non-profit).

19: Acute Leukaemia

Michael Loschi ¹, Jacques-Emmanuel Galimard², Matthieu Duchmann³, Laetitia Largeaud⁴, Pascal Turlure⁵, Stéphanie Nguyen Quoc⁶, Flore Sicre de Fontbrune⁷, Marie Robin⁸, Ibrahim Yakoub Agha⁹, Etienne Daguindau¹⁰, Sylvain Chantepie¹¹, Amandine Charbonnier¹², Sebastien Maury¹³, Helene Labussiere-Wallet¹⁴, Anne Huynh⁴, Edouard Forcade¹⁵, Cristina Castilla-Llorente¹⁶, Thomas Cluzeau¹, Marie-Thérèse Rubio¹⁷, Raynier Devillier¹⁸, Natacha Maillard¹⁹, Herve Dombret⁷, Christian Recher⁴, Arnaud Pigneux¹⁵, Eric Delabesse⁴, Nicolas Duployez²⁰, Regis Peffault de Latour⁸, Marie Sébert⁸

¹Nice University Hospital, Cote d’Azur University, Nice, France, ²EBMT, Statistical Unit, Paris, France, ³Saint Louis Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP); Université de Paris, INSERM and Centre National de la Recherche Scientifique (CNRS), Paris, France, ⁴CHU Toulouse, IUCT-Oncopole, Toulouse, France, ⁵CHU Dupuytren Limoges, Limoges, France, ⁶Pitie-Salpetriere Hospital, AP-HP, Sorbonne Université, Paris, France, ⁷Hôpital Saint Louis, APHP, Paris, France, ⁸Hôpital Saint Louis, APHP, Université Paris Cité, Paris, France, ⁹CHU de Lille, Univ Lille, INSERM, Lille, France, ¹⁰CHU Besançon, Besançon, France, ¹¹Caen University Hospital, Caen, France, ¹²Amiens University Hospital, Amiens, France, ¹³Hôpital Henri Mondor, Assistance Publique-Hôpitaux de Paris (AP-HP), Institut Mondor de Recherche Biomédicale, Université Paris Est, Créteil, France, ¹⁴Hopital Lyon Sud, Hospices Civils de Lyon, Pierre Bénite, France, ¹⁵CHU Bordeaux, Bordeaux, France, ¹⁶Gustave Roussy Cancer Campus, Villejuif, France, ¹⁷Nancy University Hospital, Vandoeuvre les Nancy, France, ¹⁸Institut Paoli-Calmettes; Aix-Marseille Université, Marseille, France, ¹⁹CHU de Poitiers, Poitiers, France, ²⁰CHU de Lille, Lille, France

Background: Pathogenic germline mutations of DDX41 (DDX41^mut) are the most common genetic predisposition to myelodysplastic syndromes (MDS) and acute myeloid leukemias (AML). Even if some recent studies reported DDX41^mut neoplasms as a specific entity with favourable outcomes compared to DDX41^wt, there are scarce data on the outcome after allogeneic hematopoietic stem cell transplantation (aSCT) in this population. We report a pair matched analysis comparing the outcome of DDX41^mut and DDX41^wt AML patients after aSCT in first complete remission (CR1).

Methods: In our previous study on 1690 AML patients treated with intensive chemotherapy in ALFA and FILO prospective trials, we reported detailed germline DDX41^mut identification (Duployez et al., Blood 2022). In the present work, we focused on the 304 patients who received aSCT in CR1. The primary endpoint was the leukemia free survival (LFS). In order to take into account, the differences between the DDX41mut and DDX41wt groups, a pair-match analysis have been performed. Exact matching have been applied on the donor type, source of cells, patient sex and ELN2017 classification. Propensity score matching have been applied with age at transplant, donor female to male recipient, conditioning regimen and use of ATG. We were eventually able to pair matched 32 DDX41^mut with 77 DDX41^wt patients.

Results: Patients characteristics are displayed in Table 1. With a median age at aSCT of 61.5 yo for DDX41^mut and 58.9 yo for DDX41^wt, the median follow-up in the entire pair match cohort was 6.1 years (95%CI, 5-6.9).

There was no significant difference in LFS nor Overall Survival (OS) between DDX41^mut and DDX41^wt (6y LFS: 47.7% (95%CI, 27.6 – 65.4) and 52.1% (95% CI, 40.1 – 60.4) HR: 1.06 (95%CI 0.59-1.91), p = 0.86; and 6y OS: : 56.8% (95%CI, 36.5 - 72.8) and 54.2 (95% CI, 42 - 64.8), HR: 1.04 (95%CI: 0.48-1.94), p = 0.92) respectively). The 6y relapse incidence (RI) was 21% (95%CI: 7-40) and 28% (95%CI: 18-38) for DDX41^mut and DDX41^wt respectively. The timing of the relapse however wasn’t similar between the two groups. Indeed, even if those results are not significant, there was a trend for a lower RI in the 1y post-transplant in DDX41^mut (HR: 0.29 (95%CI: 0.07-1.23), p = 0.09), while the RI of DDX41^mut 1y after aSCT was higher as compared to DDX41^wt(HR: 6.69 (95%CI: 0.73-61.5), p = 0.09). NRM at 6y was not significantly higher in DDX41^mut than DDX41^wt group, (31.2% (95%CI: 16.1 - 47.6) for DDX41^mut vs 20.2% (95%CI: 11.9 - 30.2) for DDX41^WT, (HR: 1.63% (95%CI: 0.75-3.55), p = 0.22). Of note, aGVHD was not different regarding DDX41 status (p = 0.33) and accounted for 30% of the NRM in the DDX41^mut patients.

Conclusions: Our study revealed no significant difference in LFS, OS, and NRM between DDX41^mut and DDX41^wt AML patients who received an aSCT in CR1 after intensive chemotherapy. However, regarding DDX41 status, we observed specificities in the chronology of post-transplantation events, especially for the time to relapse and the cause of NRM. This suggests that DDX41^mut patients may require dedicated pre and post-transplant management.

Disclosure: Authors have nothing to declare

19: Acute Leukaemia

Erik G J. von Asmuth ¹, Liesbeth C. de Wreede¹, Constantijn J M. Halkes¹, Dirk-Jan A. Eikema², Riccardo Saccardi³, Mette D. Hazenberg⁴, María Paula Busto², John A. Snowden⁵, Anna Sureda⁶, Arjan C. Lankester¹, Hein Putter¹

¹Leiden University Medical Center, Leiden, Netherlands, ²The European Society for Blood and Marrow Transplantation, Leiden, Netherlands, ³Careggi University Hospital, Firenze, Italy, ⁴Amsterdam UMC, Amsterdam, Netherlands, ⁵Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom, ⁶Institut Català d’Oncologia, Barcelona, Spain

Background: When investigating determinants of outcome after allogeneic HSCT (alloHSCT), such as when comparing centers in benchmarking, adjustment for patient- and transplant-specific variables (casemix variables) is needed, which can be done by using a prediction model. Existing models often do not account for non-linear effects or interactions. We aimed to create a more accurate model by using machine learning which can incorporate these aspects as well as disease-specific factors.

Methods: We included patients undergoing a first alloSCT between 2017 and 2021 using the same inclusion criteria as EBMT benchmarking (N = 54.457), and modelled overall survival (OS) based on patient, donor and transplant characteristics not freely chosen by the physician, so including matching and donor variables, but excluding conditioning-, country and hospital-related variables.

We coded complex data into categories using the following scores: DRSS for disease, stage and cytogenetics, HCT-CI for comorbidity, IPSS-R and ELN2017 for cytogenetics in MDS and AML. HLA mismatches were coded using molecular matching on 10 alleles.

We created an XGBoost model, and fitted a normal Cox model on the same dataset for comparison. Within the XGBoost model, hierarchical parameterization of disease was used, and disease-specific covariates were added. These weren’t added to the normal Cox model, as it cannot process these parameters in the same way.

Predictive performance was quantified using the AUC at 1 and 2 years, where 0.5 means no better than random and 1 means perfect predictions. We used a 70-30 train test split, and 10-fold cross-validation within the train cohort. Results are from the N = 38.103 train set, reported results including risk group assignments are based on cross-validated predictions.

Results: The XGBoost model had an AUC of 0.691 at 1 year, while the Cox model performed significantly worse as assessed using AUC (AUC 0.675, delta AUC 0.015, P < 0.001) and Brier score (delta Brier score 0.002, P < 0.001).

Disease was the main determinant of outcome, with age, HLA match, performance score and comorbidities being secondary. Donor relation was of little importance to the model, indicating that when adjusting for HLA match and donor age, donor relation does not impact outcome.

To create risk groups, we plotted each cross-validated risk percentile, which revealed good continuous ordering between percentiles from low to high risk, with two discrete groups at the low- and high end. We then created the following risk groups: very low (≤7th percentile), low (7-25th percentile), low-intermediate (25-50th percentile), high-intermediate (50-75th percentile), high (75-97th percentile), and very high (>97th percentile). The most common diagnosis in the very low risk group was a hemoglobinopathy, while the majority of the very high risk group had high-risk malignant disease. A trend of higher age, lower performance score, more HLA mismatched donors and more comorbidities can be seen according to risk groups (table 1).

Conclusions: Using machine learning to account for nonlinear effects and interactions can improve accuracy of survival predictions compared to the standard Cox model, leading to better adjustment of casemix effects, which could be applied in benchmarking.

Disclosure: Nothing to declare

19: Acute Leukaemia

Wolfgang Bethge ¹, Sarah Flossdorf², Franziska Hanke³, Christoph Schmid⁴, Mark Ringhoffer⁵, Stefan Klein⁶, Bernd Hertenstein⁷, Johannes Schetelig⁸, Matthias Stelljes⁹, Thomas Schroeder¹⁰, Igor Blau¹¹, Francis Ayuk¹², Matthias Eder¹³, Robert Zeiser¹⁴, Katharina Fleischhauer¹⁰, Nicolaus Kröger¹², Peter Dreger¹⁵

¹University Hospital Tuebingen, Tuebingen, Germany, ²DRST and University Hospital Essen, Essen, Germany, ³DRST - Deutsches Register für Stammzelltransplantationen e. V., Ulm, Germany, ⁴University Hospital Augsburg, Augsburg, Germany, ⁵Hospital Karlsruhe, Karlsruhe, Germany, ⁶Universtity Hospital Mannheim, Mannheim, Germany, ⁷Central Hospital Bremen, Bremen, Germany, ⁸University Hospital Dresden, Dresden, Germany, ⁹University Hospital Muenster, Muenster, Germany, ¹⁰University Hospital Essen, Essen, Germany, ¹¹University Hospital Charité Berlin, Berlin, Germany, ¹²University Hospital Hamburg, Hamburg, Germany, ¹³University Hospital Hannover, Hannover, Germany, ¹⁴University Hospital Freiburg, Freiburg, Germany, ¹⁵University Hospital Heidelberg, Heidelberg, Germany

Background: While disease- and procedure-related outcome predictors in alloHCT for AML are well characterized, the impact of center-specific variables on outcomes are still a matter of debate. A CIBMTR analysis has suggested a superior outcome for patients transplanted in high-volume centers, defined by ≥ 40 transplants/year. The purpose of the present study was to investigate the effects of numbers of alloHCT procedures per year, center experience, program duration, academic status, and JACIE accreditation along with common disease- and transplant-specific parameters on survival after alloHCT in AML using registry data provided by the DRST, the German national partner of the EBMT.

Methods: Eligible were adult patients who had received a first alloHCT for AML in any disease stage from a matched or mismatched related or unrelated donor (>= 9/10) between 2015 and 2021 and were registered with the DRST. The primary endpoint was overall survival (OS) and secondary endpoint was competing risk relapse and non-relapse mortality (NRM) in the first 12 months after HCT. Univariable and multivariable analyses of the impact of baseline characteristics on survival endpoints were performed using Cox regression and Fine & Gray modeling, respectively. Best subset selection was used to select relevant variables for multivariable analysis. In further analysis, center volume (as a non-linear covariate) was modelled using splines.

Results: Of the 5328 patients meeting all eligibility criteria, 83% had been transplanted in a high-volume center (≥40 alloHCT/year); 90% in a university hospital; 90% in a center performing alloHCT for ≥10 years; and 73% in a JACIE-accredited center. Median age was 58y (range: 18-83y). 54% of the patients were in CR1 at alloHCT; 64% and 33% had an ECOG performance status (PS) of 0 and 1, respectively; and 42% had a documented HCT-CI < 3. Of 4162 patients (78%) with ELN risk information available, risk was adverse in 37% and intermediate in 42%. Donor distribution was matched related (MRD) 29%, matched unrelated (MUD) 49%, and mismatched related (MMRD) and unrelated (MMUD) each 11%. On multivariable analysis, center-specific factors predicting adverse OS (12 month since HCT) were program duration <5-10 years (hazard ratio (HR) 1.23, 95% CI [1.02; 1.49]); center volume <40/y (HR 1.21, [1.02; 1.45]); and non-academic hospital (HR 1.21, [0.98; 1.49]), whereas JACIE accreditation did not remain in the final model. Of note, the survival effect of the center-related adverse factors was largely driven by excess non-relapse mortality rather than relapse/progression. The volume effect could be shown for any cutoff between 30 and 70 alloHCT/year. Spline modeling suggested that the volume effect becomes non-significant beyond 45 allotransplants per year. Other parameters that remained in the final models were age, sex, disease status at HCT, stem cell source, donor type, conditioning, ECOG PS, HCT-CI and ELN risk.

Conclusions: Center volume, center experience, academic status, but not JACIE accreditation have an impact of alloHCT outcomes in adult patients with AML in Germany.

Disclosure: No conflict of interests to disclose

19: Acute Leukaemia

Eolia Brissot ¹, Myriam Labopin², Eleni Tholouli³, Tobias Gedde-Dahl⁴, Mahmoud Aljurf⁵, Gérard Socié⁶, David Burns⁷, Ibrahim Yakoub-Agha⁸, Matthias Stelljes⁹, Thomas Schroeder¹⁰, Nicolaus Kröger¹¹, Edouard Forcade¹², Claude Eric Bulabois¹³, Urpu Salmenniemi¹⁴, Alessandro Rambaldi¹⁵, Ali Bazarbachi¹⁶, Arnon Nagler¹⁷, Simona Piemontese¹⁸, Jaime Sanz¹⁹, Mohamad Mohty², Fabio Ciceri¹⁸

¹Sorbonne University, Inserm U_938S, Service d’hématologie et thérapie cellulaire, Hôpital Saint-Antoine, APHP, Paris, France, ²Service d’hématologie et thérapie cellulaire, Hôpital Saint-Antoine, APHP, Paris, France, ³Manchester Royal Infirmary, Manchester, United Kingdom, ⁴Oslo University Hospital, Rikshospitalet, Oslo, Norway, ⁵King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia, ⁶Saint-Louis Hospital, BMT Unit, Paris, France, ⁷Birmingham Centre for Cellular Therapy and Transplant (BCCTT), Stoke, United Kingdom, ⁸CHU de Lille, Univ Lille, INSERM U1286, Infinite, Lille, France, ⁹University of Muenster, Muenster, Germany, ¹⁰University Hospital | Essen, Essen, Germany, ¹¹University Hospital Eppendorf, Hamburg, Germany, ¹²Service d’Hématologie Clinique et Thérapie Cellulaire, CHU Bordeaux, Bordeaux, France, ¹³CHU Grenoble Alpes - Université Grenoble Alpes, Grenoble, France, ¹⁴HUCH Comprehensive Cancer Center, Helsinki, Finland, ¹⁵University of Milan and ASST Papa Giovanni XXIII, Bergamo, Italy, ¹⁶American University of Beirut-Medical Center, Bone Marrow Transplantation Program, Beirut, Lebanon, ¹⁷Chaim Sheba Medical Center, Tel-Hashomer, Israel, ¹⁸Hematology and Bone Marrow Transplantation Unit, IRCCS San Raffaele Hospital, San Raffaele, Italy, ¹⁹University Hospital La Fe, University of Valencia, Valencia, Spain

Background: In acute myeloid leukemia (AML), the presence of FLT3-ITD mutation is associated with early relapse and an overall poor prognosis. The availability of active FLT3 inhibitors has improved the survival of FLT3-mutated patients with AML. In the 2022 European LeukemiaNet (ELN) classification, patients with FLT3-ITD mutation are classified in the intermediate-risk category, regardless of NPM1 mutation, and allogeneic hematopoietic stem cell transplantation (allo-HCT) for FLT3-ITD AML patients in first complete remission (CR1) is still recommended. However, the impact of donor type has not yet been well established. Therefore, the current study aimed to compare allo-HCT from matched sibling donors (MSD) with those from matched 10/10 unrelated donors (MUD) and T-replete haploidentical (Haplo) donors in FLT3-ITD + AML in CR1.

Methods: Patients with FLT3-ITD + AML in CR1 who underwent an allo-HCT from an MSD, MUD or Haplo donor between 2017 and 2022, and were reported to the registry of the EBMT ALWP, were included. The major endpoints were overall survival (OS), leukemia-free survival (LFS), relapse incidence (RI), non-relapse mortality (NRM), and graft-versus-host disease (GVHD)-free, relapse-free survival (GRFS). Univariate comparisons were performed using the log-rank test for LFS, OS, and GRFS, and Gray’s test for cumulative incidences of NRM and RI. A Cox proportional-hazards model was used for multivariate regression.

Results: We identified 2113 patients: 721 patients received a transplant from an MSD, 1015 from an MUD, and 377 from a Haplo donor. The median follow-up was 24.4 months. Median age was lower in the MSD group compared to the other groups (50.8 years, p < 10^-4). Time from diagnosis to transplant was higher in the MUD group. The proportion of intermediate- and adverse-risk cytogenetics was similar in the 3 groups, as well as the proportion of mutated-NPM1 versus non-mutated NPM1. Measurable residual disease at transplant was negative in 64.2% patients with no statistical difference between the 3 groups (p = 0.1). The HCT-CI was higher in the MUD group. Significantly more patients received a myeloablative conditioning regimen in the MSD group compared to the MUD and Haplo groups (62%, p < 10^-4). At 2 years, for the whole population, OS was 70.6% (95% CI: 68.3-72.8), LFS was 62.7% (95% CI:60.3-64.9), RI was 25.9% (95% CI, 23.9-28), NRM was 11.4% (95% CI:9.9-12.9), and GRFS was 50.9% (95% CI: 48.5-53.3). Acute GVHD II–IV was higher in the MUD compared to MSD group. No difference in chronic GVHD was observed. RI was significantly lower in the Haplo group compared to the MSD group (hazard ratio [HR] = 0.6, 95% CI: 0.42-0.89, p = 0.01). NRM was higher in MUD and in Haplo recipients than in the MSD group. LFS, OS and GRFS did not differ significantly between donor types.

Conclusions: Our study results show that allo-HCT may rescue two thirds of patients with FLT3-ITD + AML in CR1. Importantly, the donor type did not have any impact on patients’ OS, LFS, or GRFS. Moreover, the significant decrease in RI in the Haplo group underlines the potential interest in this donor type, taking into account the need for good timing and feasibility of SCT in this patient population

Disclosure: Nothing to declare

19: Acute Leukaemia

Nihar Desai ¹, Carol Chen¹, Ivan Pasic^1,2, Wilson Lam^1,2, Arjun Law^1,2, Armin Gerbitz^1,2, Auro Viswabandya^1,2, Fotios Michelis^1,2, Rajat Kumar^1,2, Jonas Mattsson^1,2, Igor Novitzky-Basso^1,2, Dennis Dong Hwan Kim^1,2

¹Hans Messner Allogeneic Transplant Program, Division of Medical Oncology and Hematology, Princess Margaret Hospital, Toronto, Canada, ²University of Toronto, Toronto, Canada

Background: Leukemia relapse following allogeneic hematopoietic stem cell transplantation (HCT) in acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS) still remains a significant challenge. Loss of heterozygosity (LOH) in major histocompatibility complex (MHC) gene or down-expression of MHC gene have been identified as a potential mechanism leading to relapse post-HCT. A Recent work (Toffalori et al, ASH 2023 abstract 362) suggested the genotoxicity of ganciclovir as one of the mechanisms of LOH-MHC, leading to post-HCT relapse. It has never been replicated whether ganciclovir or valganciclovir are associated with an increased risk of relapse. We attempted to replicate the hypothesis that the use of (val)ganciclovir is associated with increased relapse incidence in a large single-centre cohort.

Methods: Patients with a diagnosis of AML/MDS who underwent HCT at Princess Margaret Cancer Centre between January 2015 and March 2023 were evaluated. Cumulative incidence of relapse (CIR) and non-relapse mortality (NRM) was determined considering competing events, respectively. Overall Survival (OS) was determined using the Kaplan-Meier method. A p-value of <0.05 was considered statistically significant.

Results: A total of 727 patients with AML/MDS underwent HCT, of whom 624 (85.8%) were CMV seropositive, constituting the primary group for analysis. Baseline characteristics are summarized in Table 1. More than half of the patients were aged >60 years. The median follow-up duration was 24.9 months (range, 8-54.5) among survivors, and 146 (23.4%) patients relapsed at a median of 6 months (range, 2.8-10.2) post-HCT.

Of the 350 episodes (56.1%) of CMV viremia, 317 were treated with (Val)ganciclovir. Patients receiving (Val)ganciclovir exhibited a lower CIR at 24 months, 19.6% (95% confidence interval: 15.4-24.2) compared to controls, 25.8% (20.8–31.1, p = 0.03). Regarding specific donor types, CIR for mismatched unrelated donor (MMUD) transplants was 33% (20.5-46.1) for patients who received (Val)ganciclovir compared to 35.6% (16.1-55.7) in controls (p = 0.95); the corresponding numbers for haploidentical transplants were 21.5% (12.7-31.9) and 24.6 (11-41), respectively (p = 0.68).

The OS rate at 24 months was 56.4% (50.7-61.7) for patients who received (Val)ganciclovir and 68% (62.1-73.1) for controls (p = 0.006), whereas the NRM rate at 2 years was 28.1% (23.2-33.2) in those who received (Val)ganciclovir and 14.6% (10.8-18.9) in the control group (p < 0.001).

On univariate analysis, exposure to (Val)ganciclovir (HR: 0.60; p = 0.003), and the use of myeloablative conditioning (MAC) (HR: 0.49; p = 0.004) were associated with decreased incidence of relapse. Conversely, a high-very high disease risk index (HR: 2.52; p < 0.001) and the use of MMUD (HR: 1.72; p = 0.01) increased the likelihood of relapse. Multivariable analysis confirmed that the exposure to (Val)ganciclovir does not increase the risk of relapse, rather reducing the relapse risk (Table 1).

Table 1: Patient characteristics and multivariable analysis

Conclusions: We could not replicate that (Val)ganciclovir treatment is associated with an increased risk of relapse post HCT in patients with AML/MDS, including MMUD and haploidentical transplant recipients. Further study is warranted to define the genotoxic effects of ganciclovir, and its impact on LOH-MHC and relapse post-HCT in a larger cohort.

Disclosure: Conflict of interest: Nothing to declare

Funding: Nothing to declare

19: Acute Leukaemia

Gesine Bug ¹, Kerstin Schäfer-Eckart², Carina Rave³, Henning Baldauf³, Fabian Lang¹, Knut Wendelin², Katja Sockel⁴, Christian Reicherts⁵, Vera Schlipfenbacher¹, Julia-Annabell Georgi⁴, Christoph Schliemann⁵, Marek Werth¹, Katharina Egger-Heidrich⁴, Julia Marx⁵, Simon Call⁵, Heike Pfeifer¹, Christian Thiede⁶, Johannes Schetelig⁴, Matthias Stelljes⁵

¹Goethe University, Frankfurt, Germany, ²Klinikum Nürnberg, Nürnberg, Germany, ³DKMS gemeinnützige GmbH, Dresden, Germany, ⁴TU Dresden, Dresden, Germany, ⁵University Hospital Münster, Münster, Germany, ⁶AgenDix Applied Genetic Diagnostics, Dresden, Germany

Background: Patients with NPM1-mutated AML have a good prognosis after intensive chemotherapy, but most patients with persistence or recurrence of measurable residual disease (MRD) will eventually relapse. Since molecular relapse detected by RT-qPCR of NPM1 precedes hematological relapse by 2-4 months, many centers provide allogeneic hematopoietic cell transplantation (HCT) upon detection of MRD, assuming better transplant results than with HCT after overt relapse. Aim of this retrospective study was to describe the outcome of patients with NPM1-mutated AML at different disease stages, and to identify prognostic variables.

Methods: We included 227 consecutive adult patients with NPM1-mutated AML who received a first allogeneic HCT in 2010-2020 with molecular persistence or relapse detected by NPM1-specific RT-qPCR within 60 days prior to conditioning, or primary refractory or relapsed disease. Exclusion criteria included use of post-transplant cyclophosphamide for GVHD prophylaxis outside of haploidentical HCT and in vivo T-cell depletion other than ATG. Median follow-up was 4.5 years.

Results: Patients had a median age of 56 years (range, 24-76), 67% had a Karnofsky score ≥90 and 85% an HCT-comorbidity index <3. Karyotype was normal in 69% of patients and 47% harbored an FLT-3 mutation. Transplant characteristics included myeloablative, reduced intensity, sequential and non-myeloablative conditioning in 21%, 37%, 41% and 2% of patients, respectively. Patients primarily received peripheral blood stem cells (93%). Grafts were from an HLA-identical sibling (21%), haploidentical family donor (5%), 10/10 HLA-matched unrelated donor (MUD, 54%) and 9/10 MUD (17%). GVHD prophylaxis included CSA/MTX (35%), CSA/MMF (53%) or Tacrolimus/MMF (4%), plus ATG (57%).

At four years, overall survival (OS) was 61% (95%-CI: 54%-68%), the cumulative incidence of relapse/progression was 26% (95%-CI: 21%-33%) and of non-relapse mortality (NRM) 22% (95%-CI: 17%-28%). The 4y-OS rate was most favorable and not significantly different (p = 0.7, log-rank test) in patients transplanted in MRD-positive CR1 (82%, 95%-CI: 71%-94%, n = 46), molecular relapse (84%, 95%-CI: 74%-96%, n = 45) or MRD-negative CR2 (73%, 95%-CI: 53%-100%, n = 15). Notably, 35/45 patients (78%) with molecular relapse had not been treated prior to start of conditioning. In patients transplanted with hematological relapse or refractory AML (untreated relapse, n = 40; refractory relapse (r/r AML), n = 44; primary refractory disease, n = 24) 4y-OS was 34%-45%, not significantly different (p = 0.3, log-rank test) from that of patients transplanted in MRD-positive CR2 (54%, 95%-CI: 33%-89%, n = 13). Poor outcome in patients with overt untreated or refractory relapse was attributable to high 4y-NRM (27% and 35%) and relapse/progression (33% and 36%), respectively.

In multivariable analysis, transplant outcome depended primarily on disease status, with significantly worse OS and event-free survival (EFS) in patients transplanted with overt disease. In addition, 9/10 MUD transplantation and reduced intensity conditioning negatively impacted OS and EFS, respectively. Stem cell source, age, performance status, GVHD prophylaxis and FLT-3 mutation status had no significant impact on overall and event-free survival.

Conclusions: A favorable outcome in patients experiencing molecular recurrence in CR1 depends on being transplanted before progressing to hematological relapse, with no need for additional therapy aimed at inducing MRD-negativity at HCT. In contrast, transplantation with overt disease is associated with poor outcome, warranting experimental treatment approaches.

Disclosure: Gesine Bug: Honoraria from Novartis, Jazz, BMS, Gilead. Travel grants from Jazz, Gilead and Neovii. Johannes Schetelig: Participation at Advisory Boards organized by Abbvie, AstraZeneca, BeiGene, BMS, Janssen, and MSD. Lecture Fees from Astellas, AstraZeneca, BeiGene, BMS, Novartis, Eurocept, Medac and Janssen.

19: Acute Leukaemia

Troy Quigg ¹, Neena Kapoor², Jemily Malvar², Elizabeth Gourdine², Amy Keating^3,4, Leslie Lehmann³, Kris Mahadeo^5,6, Sajad Khazal^7,6, Jodi Skiles⁸, Dana Salzberg⁹, Rachel Phelan¹⁰, Nahal Lalefar¹¹, Emi Caywood¹², Rabi Hanna¹³, Shalini Shenoy¹⁴, Heather Stefanski¹⁵, Jordan Milner¹⁶, Kevin McNerney^17,18, Christine Higham¹⁹, Ulrich Duffner¹, Joseph Chewning²⁰, Niketa Shah^21,22, Jeffrey Huo²³, Ibrahim Ahmed²⁴, Michael Pulsipher²⁵, Hisham Abdel-Azim⁷

¹Section of Pediatric Blood and Marrow Transplantation and Cellular Therapy, Helen DeVos Children’s Hospital, Grand Rapids, United States, ²Children’s Hospital of Los Angeles, Los Angeles, United States, ³Dana-Farber/Boston Children’s Cancer & Blood Disorders Center, Boston, United States, ⁴Division of Pediatric Hematology-Oncology-BMT, University of Colorado, Aurora, United States, ⁵Division of Transplant and Cellular Therapy, Duke University School of Medicine, Durham, United States, ⁶Pediatric Stem Cell Transplantation and Cellular Therapy, The University of Texas MD Anderson Cancer Center, Children’s Cancer Hospital, Houston, United States, ⁷Loma Linda University School of Medicine, Cancer Center, Children Hospital and Medical Center, Loma Linda, United States, ⁸Riley Hospital for Children, Indianapolis, United States, ⁹Phoenix Children’s Hospital, Phoenix, United States, ¹⁰Division of Pediatric Hematology, Oncology, and Blood and Marrow Transplantation, Medical College of Wisconsin, Milwaukee, United States, ¹¹University of California San Francisco Benioff Children’s Hospital Oakland, Oakland, United States, ¹²Nemours/Alfred I duPont Hospital for Children, Wilmington, United States, ¹³Taussig Cancer Institute, Cleveland Clinic, Cleveland, United States, ¹⁴Washington University School of Medicine, St. Louis, United States, ¹⁵CIBMTR (Center for International Blood and Marrow Transplant Research), National Marrow Donor Program/Be The Match, Minneapolis, United States, ¹⁶Division of Hematology/Oncology, University of Florida, UF Health Shands Children’s Hospital, Gainesville, United States, ¹⁷Lurie Children’s Hospital, Northwestern University School of Medicine, Chicago, United States, ¹⁸Johns Hopkins All Children’s Hospital, St. Petersburg, United States, ¹⁹Division of Pediatric Allergy, Immunology, and Bone Marrow Transplant, Benioff Children’s Hospital, University of California San Francisco, San Francisco, United States, ²⁰Pediatric Blood and Marrow Transplant Program, University of Alabama at Birmingham, Birmingham, United States, ²¹Rutgers Cancer Institute, Pediatric Bone Marrow Transplant/Stem Cell Transplant and Pediatric Cellular Therapy Program, New Brunswick, United States, ²²Yale New Haven Children’s Hospital, New Haven, United States, ²³Pediatric Cellular Therapies, Cancer and Blood Disorders, Atrium Health Levine Children’s Hospital, Charlotte, United States, ²⁴Pediatric Hematology, Oncology and BMT, Children’s Mercy Hospital Kansas City, Kansas City, United States, ²⁵Division of Hematology and Oncology, Intermountain Primary Children’s Hospital, Huntsman Cancer Institute, Spencer Fox Eccles School of Medicine, University of Utah, Salt Lake City, United States

Background: Our updated analysis with extended follow up shows key insights into the predictive ability of BCR (B-cell) vs TCR (T-cell) [beta (B) and gamma (G)] receptor rearrangements for NGS-MRD analysis and excellent relapse-free (RFS) and overall (OS) survival in children with pre-HCT NGS-MRD negative B-ALL undergoing allogeneic HCT.

Methods: 86 B-ALL patients enrolled on the observational arm of PTCTC ONC1701 trial (NCT03509961) were included. All patients were in complete morphological remission (CR) at time of HCT (CR1 n = 32; CR2 n = 40; ≥CR3 n = 14). Detectable bone marrow pre-HCT NGS-MRD, regardless of detection level, was considered positive. All patients received myeloablative conditioning, including TBI and non-TBI regimens, based on center standard/preference.

Results: Median age at HCT was 12.1 (range= 0.4-25.3) years. Pre-HCT NGS-MRD by trackable BCR rearrangements, regardless of residual TCR rearrangements, best predicts relapse-free survival (RFS) for allogeneic HCT in B-ALL. RFS was significantly better for pre-HCT NGS-MRD negative patients (based on BCR) for the whole cohort (p = 0.005) and for patients in CR1/2 compared to ≥ CR3 (p = 0.045). Patients ≥ CR3 and pre-HCT NGS-MRD negative have higher RFS (>75%) than historically reported in literature. Of 14 patients with detectable pre-HCT NGS-MRD by TCRG+ alone, RFS was 100%. None of the patients in this cohort had detectable TCRB on pre-HCT NGS-MRD. More than half of pre-HCT NGS-MRD positive patients can be salvaged with HCT. Of patients who received TBI-based conditioning (n = 68; 79.1%), 16.2% (n = 11) relapsed with 77.9% (n = 53) alive at a at a median follow-up of 452 (range= 19-1465) days post-HCT. Of patients who received non-TBI conditioning (n = 18; 20.9%), 16.7% (n = 3) relapsed with 77.8% (n = 14) alive at a median follow-up of 705 (range= 73-1406) days post-HCT. RFS of pre-HCT NGS-MRD negative patients who received TBI vs non-TBI conditioning were similar, as was overall survival. RFS for patients who had pre-HCT CART to achieve negative pre-HCT NGS-MRD was lower (p = 0.036) compared to patients who achieved negative MRD without CART which may reflect the molecular biology of the disease, previous lines of treatments, later remission status, or earlier prior relapse and currently the subject of ongoing analysis.

Conclusions: Pre-HCT NGS-MRD negative B-ALL patients continue to have excellent RFS and OS with allogeneic HCT. Trackable BCR rearrangements offer the most reliable and predictive measure of relapse risk for B-ALL undergoing allogeneic HCT while isolated TCRG + , in the face of BCR-, unlikely represent residual disease. The interaction and outcomes of patients who required pre-HCT CART, to achieve MRD negativity, appear inferior, and further data analysis is needed to better understand factors impacting those outcomes. While sample size is limited, outcomes of TBI and non-TBI allogeneic HCT remain similar. Further data from the study observational and non-TBI treatment arms will help define the role of TBI for pre-HCT NGS-MRD negative patients.

Clinical Trial Registry: NCT03509961

Disclosure: Research is supported by Gateway for Cancer Research and Adaptive Biotechnologies (Hisham Abdel-Azim).

Hisham Abdel-Azim has research funding from Adaptive Biotechnologies to support this research.

Michael Pulsipher has received research funding from Adaptive Biotechnologies.

All other authors have nothing to declare related to this work.

19: Acute Leukaemia

Yongping Zhang ¹, Hujun Li¹, Hui Zhang¹, Qi Ji¹, Shaoyan Hu¹

¹Children’s Hospital of Soochow University, Suzhou, China

Background: In the bone marrow micro-environment of AML, the coordinated interaction of various immune cells and immune regulatory factors results in the dysfunction of T cells. In this process, naïve CD8 + T cells, upon activation by antigen presentation from APCs such as DCs, differentiate into effector cells and memory cells. However, a considerable proportion of the remaining CD8 + T cells exhibit diminished effector functions due to exhaustion or senescence. While numerous studies have been conducted on the mechanisms and intervention strategies related to the exhaustion or senescence, the functional role of naïve CD8 + T cells in AML remains unclear. Therefore, our aim is to identify the correlation between the functional status of naïve CD8 + T cells in AML and to investigate its impact on the survival prognosis of AML patients.

Methods: Collecting fresh bone marrow samples from pediatric AML patients below the age of 14 at the time of diagnosis. Flow-cytometry, transcriptome sequencing and in vitro culture models were adapted to identify the quantity, activation, and differentiation of CD45RO-CCR7+ naïve CD8 + T cells(CD8+Tnaïve). Single-cell RNA sequencing was utilized to reveal the key genes and transcription factors and CRISPR-Cas9 technology was employed to perform gene editing on naïve CD8 + T cells to study the impact on exhaustion. Conducting correlation analysis with chemotherapy assessment and survival data, the role of naïve CD8 + T cells in AML treatment and prognosis was elucidated.

Results: Compared to 21 age-matched healthy donors (HDs) of HSCT candidates, 56 pediatric AML patients showed a significantly lower proportion of CD45RO-CCR7 + CD8+Tnaïve cells in BM (mean: HDs, 60%; AML, 45%; p < 0.05). CD45RO-CCR7 + CD8+Tnaïve cells in AML exhibited significantly higher expression of inhibitory molecules VISTA and CTLA-4 (HDs, < 5%; AML, 10% to 20%; p < 0.05). After sorting and in vitro culture, the CD69-CD25- non-activated fraction of CD45RO-CCR7 + CD8+Tnaïve cells was significantly higher in AML compared to HDs (HDs, 20%; AML, 60%; p < 0.05). RNA-seq and GSEA revealed significantly activated bile-acid metabolism pathway but down-regulated MTORC1 signaling in CD45RO-CCR7 + CD8+Tnaïve cells in AML. Single-cell RNA-seq demonstrated significantly higher expression of FCER1G, ID2 in CD8+Tnaïve cells which highly expressed CCR7, SELL, LTB and TCF1 in AML. Over-expression of FCER1G in CD45RO-CCR7 + CD8+Tnaïve cells using CRISPR-Cas9 led to significant decrease in memory T cells but increase in effector T cells with elevated expressions of PD-1 and GZMB. Paired analysis of CD45RO-CCR7 + CD8+Tnaïve cells’ proportion before and after the first induction chemotherapy showed an increasing trend in 11 complete remission (CR) patients, but a decreasing trend in 4 non-CR patients among 15 pediatric AML patients. Using the mean proportion (60%) of CD45RO-CCR7 + CD8+Tnaïve cells in HDs as the cutoff point, the 47 AML patients were divided into high and low expression groups (high : 10 cases; low : 37 cases). The 2-year overall survival was significantly lower in the high group (about 70%) compared to the low group (95%) (p = 0.026, log-rank test).

Conclusions: In pediatric AML patients, CD45RO-CCR7+ naïve CD8 + T cells exhibited impaired activation and memory-formation, and serve as a risk factor for 2-year overall survival.

Disclosure: Nothing to declare

19: Acute Leukaemia

Thomas Coats ¹, Toby Winter², Grace Walburn², Rob Bennett², Pramila Krishnamurthy³, Victoria Potter³, Justin Loke^4,5, Steve Knapper⁶, Julia Lee⁷, Rachel Pearce⁷, Ram Malladi⁸, Robin Adlam², Charles Craddock⁹

¹Royal Devon University Hospital Foundation Trust, Exeter, United Kingdom, ²Lone Star Analysis UK, Lincoln, United Kingdom, ³King’s College Hospital NHS Foundation Trust, London, United Kingdom, ⁴University of Birmingham, Birmingham, United Kingdom, ⁵Dana Farber Cancer Institute, Boston, United States, ⁶Cardiff University, Cardiff, United Kingdom, ⁷British Society of Blood and Marrow Transplantation and Cellular Therapy (BSBMTCT), London, United Kingdom, ⁸Cambridge University Hospitals NHS Foundation Trust, Cambridge, United Kingdom, ⁹Queen Elizabeth Hospital Birmingham, Birmingham, United Kingdom

Background: Allogeneic stem cell transplant (allo-SCT) is a centrally important treatment in adults with acute myeloid leukaemia (AML) in first complete remission (CR1). Despite advances in transplant technology and increased donor availability, concerns remain that this potentially curative modality is under-utilised. We have developed an explainable AI model to estimate the under-referral of patients with AML in CR1 who could gain an overall survival (OS) advantage from allo-SCT.

Methods: We modelled a typical annual UK cohort of newly diagnosed patients. Five senior AML clinicians independently estimated relevant patient variables, including age, performance status, co-morbidities (HCT-CI), disease biology and response rates.

For each variable, the experts supplied five data points, representing 0^th, 10^th, 50^th, 90^th, and 100^th percentiles. Aggregated distributions were fed through a Bayesian Inference Network (built with Lone Star Analysis), representing each individual patient pathway. Running Monte-Carlo simulations, 50,000 cases were derived from the experts’ answers to identify an expected range of patients in CR.

The clinicians provided a threshold of HCT-CI score for recommending transplants (based on: ELN risk group, MRD status, age [</>40] and performance status [ECOG < 2/2/ > 2)]). Allo-SCT is rarely performed in patients >75yr old so this demographic was excluded from transplant eligibility analysis. These thresholds were applied to the simulated patients to identify patients with potential OS benefit from CR1 allo-SCT, this being compared to the number of AML CR1 allografts actually performed (data obtained from the British Society of Blood and Marrow Transplantation and Cellular Therapy (BSBMTCT), 2016-2020). Given that BSBMTCT data do not differentiate whether patients had intensive chemotherapy (IC) or non-intensive chemotherapy (N-IC) to achieve CR1, we asked our experts what proportion of N-IC-treated patients treated achieving CR would be predicted to gain OS benefit from allo-SCT.

Sensitivity analyses were performed to determine the impact of uncertainty of each variable on the model output.

Results: Sensitivity analyses identified highest variability in clinicians’ estimates of i) number of AML cases/year and ii) age distribution at diagnosis. To improve accuracy of the model, clinician estimates were replaced with published incidence data (Cancer UK) and age distribution from National Cancer Registry and Analysis Service data (Hanhua et al, Blood 2023). All other variables had less than or equal to 10% variation above or below the model prediction.

Our updated model predicts a median of 569 UK adult patients/year should derive improved OS from allo-SCT in CR1 (95% CI range 477-669), 443 following induction with IC and 126 following N-IC. This is significantly higher than the UK allo-SCT rate of 318/year.

The proportion of patients predicted to be transplant-eligible who did not have allo-SCT in CR1 was highest in the older age groups; 36% aged 18-54, 33% aged 55-64, 44% aged 65-69, 75% aged 70-74.

Conclusions: Our innovative AI model suggests that the annual transplant rate in the UK should be almost 80% higher than observed. There is substantial inter-expert agreement on the size of the transplant eligible population to support these findings. These data provide the basis for further work to examine why there is ongoing under-transplantation.

Disclosure: Nothing to declare.

19: Acute Leukaemia

Christina Rautenberg ¹, Friedrich Stoelzel², Christoph Roellig³, Matthias Stelljes⁴, Verena Gaidzik⁵, Oliver Kriege⁶, Mareike Verbeek⁷, Julia Unglaub⁸, Felicitas Thol⁹, Stefan Krause¹⁰, Mathias Haenel¹¹, Charlotte Neuerburg¹², Vladan Vucinic¹³, Christian-Friedrich Jehn¹⁴, Maxi Weiss¹⁵, Lars Fransecky², Jens Chemnitz¹⁶, Kerstin Schäfer-Eckart¹⁷, Josephine Schröder¹⁸, Sabrina Kraus¹⁹, William Krüger²⁰, Ulrich Kaiser²¹, Sebastian Scholl²², Guido Kobbe²³, Paul Jäger²³, Nael Alakel³, Malte von Bonin³, Katja Sockel³, Maher Hanoun¹, Uwe Platzbecker¹³, Tobias Holderried¹², Anke Morgner¹¹, Michael Heuser⁹, Tim Sauer⁸, Katharina Götze⁷, Eva Wagner⁶, Konstanze Döhner⁵, Hartmut Döhner⁵, Christoph Schliemann⁴, Johannes Schetelig³, Martin Bornhäuser³, Ulrich Germing²³, Thomas Schroeder¹, Moritz Middeke³

¹University Hospital Essen, Essen, Germany, ²UK Kiel, Kiel, Germany, ³University Hospital Dresden, Dresden, Germany, ⁴University Hospital Muenster, Muenster, Germany, ⁵UK Ulm, Ulm, Germany, ⁶UK Mainz, Mainz, Germany, ⁷Klinikum rechts der Isar, Muenchen, Germany, ⁸University Hospital Heidelberg, Heidelberg, Germany, ⁹MHH Hannover, Hannover, Germany, ¹⁰UK Erlangen, Erlangen, Germany, ¹¹Städtische Kliniken Chemnitz, Chemnitz, Germany, ¹²UK Bonn, Bonn, Germany, ¹³UK Leipzig, Leipzig, Germany, ¹⁴Asklepios St. Georg Hamburg, Hamburg, Germany, ¹⁵Klinik St. Georg, Leipzih, Germany, ¹⁶Klinikum Koblenz, Koblenz, Germany, ¹⁷Klinikum Nürnberg, Nuernberg, Germany, ¹⁸Helios Berlin buch, Berlin, Germany, ¹⁹UK Würzburg, Würzburg, Germany, ²⁰UK Greifswald, Greifswald, Germany, ²¹Krankenhaus Bernward, Bernward, Germany, ²²UK Jena, Jena, Germany, ²³UK Düsseldorf, Düsseldorf, Germany

Background: CPX-351, the liposomal formulation of AraC and daunorubicin, has been approved for the treatment of therapy-related AML (tAML) and AML myelodysplasia-related (AML-MR) without age-specific limiitations based on the results from a phase-3 trial performed in patients aged 60 to 75 years. Three European real-world analyses including one from our group (Rautenberg et al., 2021) generally confirmed the results following CPX-351 first-line treatment, but all with short median follow-up <1 year. Furthermore, although all reports included patients below the age of 60 years, a detailed analysis focusing specifically on younger patients (<60 years) is still lacking.

Methods: To address this age aspect as well as the outcome after subsequent allo-SCT and to provide longer follow-up, we updated and reanalyzed the data of the initially published 188 patients, of which 122 (67%) received allo-SCT after in-label CPX-351 induction. Focusing on the age issue, results were compared between patients aged <60 years and >60 years for the purpose of this analysis.

Results: Overall, we were able to update the data of 182 (97%) of the 188 patients with AML-MR (70%) or t-AML (29%) thereby providing a median follow-up of 35.6 months in all patients from CPX-351 start. Median overall survival (OS) in the entire cohort was 32.3 months and 2-year OS rate was 52%. In univariate analysis, OS was significantly longer in younger patients (<60 years; hazard ratio 0.48 [0.31-0.75], p = 0.008), although baseline characteristics were comparable between the two age groups. This survival benefit was not related to a higher likelihood of response (CR/CRi/MLFS) to CPX-351 based induction (<60 years 67% vs ≥60 years 71%, n. s.), but rather to a higher frequency of subsequent allo-SCT (<60 years 85% vs ≥60 years 61%, p = 0.003) in younger patients. Accordingly, allo-SCT (<0.001) but not younger age was associated with longer OS in multivariate analysis, while prior use of hypomethylating agents (p = 0.09) and adverse European LeukemiaNet 2017 genetic risk (p < 0.002) were associated with lower OS. In line with this 2-year OS rate, relapse incidence and non-relapse mortality in those 122 patients (67%) undergoing allo-SCT were 64%, 20% and 20%, respectively, with no significant difference between younger and older patients.

Conclusions: These updated real-world data with extended follow-up of almost 3 years confirm that CPX-351 is an efficient treatment for high-risk AML patients, facilitating allo-HCT in many patients with promising post-transplantation outcome. Our results suggest that this also applies to younger patients (<60 years), which were initially not included in the phase-III trial, but this requires confirmation in a prospective trial.

Disclosure: CR: travel support and lecture fees from JAZZ

TS: travel support and lecture fees and research funding from JAZZ

19: Acute Leukaemia

Yibo Wu ^1,2,3,4, Xiaolin Yuan^1,2,3,4, Xiaoyu Lai^1,2,3,4, Lizhen Liu^1,2,3,4, Lihong Ni⁵, Shanshan Hu⁶, Jimin Shi^1,2,3,4, Weiyan Zheng^1,2,3,4, Jian Yu^1,2,3,4, Yanmin Zhao^1,2,3,4, Jie Sun^1,2,3,4, Xiujin Ye^1,2,3,4, Zhen Cai^1,2,3,4, He Huang^1,2,3,4, Shanshan Pei^1,2,3,4, Yi Luo^1,2,3,4

¹Bone Marrow Transplantation Center, the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China, ²Liangzhu Laboratory, Hangzhou, China, ³Institute of Hematology, Zhejiang University, Hangzhou, China, ⁴Zhejiang Province Engineering Research Center for Stem Cell and Immunity Therapy, Hangzhou, China, ⁵Jinhua Hospital, Zhejiang University School of Medicine, Jinhua, China, ⁶Jinhua Hospital of Wenzhou Medical University, Jinhua, China

Background: AML was a disease remains challenging with a dismal prognosis. Recent advance in the single-cell transcriptomics reveals the polyclonal and evolving heterogeneity of AML. The differentiation hierarchies of myeloid malignancies may directly impact the drug responses, clinical predictions. AML with t (8;21) contributes one of the most recurring genetic subtypes of AML, consisting 4%-8% of whole AML patients. It was a unique entity in AML with similar characteristics classified as favorable group. However, only 45%-70% of patients achieve long-term disease-free survival. Approximately 30-40% of patients relapse after achieving complete remission. As we know the potential impact of leukemia cell differentiation status on drug responses, clinical predictions. AML with t (8;21) was an ideal model to find whether phenotypically-defined stages of leukemia may impact mutations and outcomes in AML. Here, we did a phenotypic, cytogenetic, molecular and therapeutic response correlative analysis of a large cohort of AML with t (8;21) patients to capture the stage of leukemia arrest, specific genomic features and clinical outcomes.

Methods: We enrolled the AML patients diagnosed with recurring genetic subtypes of t (8;21) treated at three centers of hematology in China. We classify AML blasts into two stages of AML differentiation-arrest categories primitive cells (Prim) and Monocyte (Mon) group based on CD34, CD117, CD13, CD33, CD38, MPO, CD11b, CD15, CD64 expression.

Results: A of total of 338 AML with t (8;21) patients were enrolled. 170(50.30%) patients accepted chemotherapy only and 141(41.72%) patients bridge to allo-HSCT treatment. There were 68(24.46%), 49(17.63%), 43(15.47%), 83(29.86%), 35(12.59%) patients classified as MPP, LMPP, CMP, GMP, GP/MP. Thus, 160(57.55%) patients were classified as Prim group and 118(42.45%) patients were classified as Mono group. The most gene mutations were KIT(42.4%), FLT3(17.0%), NRAS(12.2%) in whole cohort.

Mono group was associated with lower 2-year OS and higher 2-year cumulative incidence of relapse (CIR) than Prim group based immunophenotypic stratification. Patients with KIT mutation had worse 2-year OS and higher 2-year CIR than patients without KIT mutation. Mono and Prim group with KIT mutation had lower 2-year OS than Mono and Prim group without KIT mutation respectively (P < 0.001). Mono group with or without KIT mutation had higher 2-year CIR than Prim group with or without KIT mutation respectively (P < 0.001).

In allo-HSCT cohort, Mono group was associated with lower 2-year OS and higher 2-year CIR than Prim group. Gene mutations and loss of chromosome didn’t significantly impact both OS and CIR. Differentiation stage of leukemia cells itself significantly impact the outcomes after transplantation. Mono group with or without KIT mutation had higher 2-year CIR than Prim group with or without KIT mutation respectively. GO enrichment analysis of RNA-seq of patients’primary samples showed lower expression of antigen associate genes in Mono group than Prim group. It indicates that immune escape of GVL effect may mediate relapse after allo-HSCT in Mono group.

Conclusions: We demonstrate that AML with t (8;21) arrested at specific stages of myeloid differentiation that significantly correlate with AML genetic lesions, clinical presentation, response to therapy, and outcomes.

Clinical Trial Registry: none

Disclosure: There are no conflicts of interest to report.

20: Aplastic Anaemia

Cjm Halkes ¹, Eas Koster¹, Ejm Bogers¹, Lgm Daenen², Sk Klein³, Smc Langemeijer⁴, F. Moenen⁵, E. Nur^6,7, Mhg Raaijmakers⁸, Tjf Snijders⁹, Jennifer ML Tjon¹, L.C. de Wreede¹

¹Leiden University Medical Centre, Leiden, Netherlands, ²Utrecht UMC, Utrecht, Netherlands, ³UMC Groningen, Groningen, Netherlands, ⁴Radboudumc, Nijmegen, Netherlands, ⁵Maastricht UMC, Maastricht, Netherlands, ⁶Amsterdam UMC, Amterdam, Netherlands, ⁷Sanquin Research, Amsterdam, Netherlands, ⁸EramsusMC, Rotterdam, Netherlands, ⁹Medical Spectrum Twente, Enschede, Netherlands

Background: The preferred first-line treatment options for adult severe aplastic anemia (AA) patients are allogeneic stem cell transplantation (alloSCT) and anti-thymocyte globulin (ATG)-based intensive immune suppressive treatment (IST). Analysis of long-term and detailed real-world results of these treatment modalities can guide personalized treatment decisions. Long-term favorable outcome of IST can be hampered by relapse AA, development of PNH disease or AML/MDS and the need for long-term treatment with immunosuppressive drugs. We developed a new endpoint to define treatment success after IST in AA: Transplantation-, Transfusion- and Treatment- (for AA, PNH disease or AML/MDS) free survival (TTT-free survival) and analyzed which baseline characteristics are associated with the probability of reaching this endpoint.

Methods: We analyzed data from a real-world prospective multi-center cohort study of 127 adult AA patients in the Netherlands who all received IST with ATGAM and Cyclosporine within a standardized treatment protocol as first-line therapy. We used a multi-state model to model the disease-recovery process after start of IST, taking into account the individual start and stop dates of different transfusion and treatment episodes. The impact of risk factors on becoming transfusion-free and becoming non-transplant therapy-free was modelled by means of two Cox proportional hazards models. The probability of TTT-free survival over time was estimated by adding up the (time-dependent) probabilities to be in any of the relevant states

Results: Median age was 54 years (range 18-79; 40% >60, 32% <41) and 64% was male. Disease severity was non-severe, severe and very severe in 33%, 43% and 24%, respectively. PNH clone at diagnosis was present (>1%) in 45% of patients. In the total cohort, overall survival (OS) at 5 years was 79% (95%-CI 70-87), while the 5-year TTT-free survival was only 42% (95%-CI 33-54). Patients aged <41 years had a 5-year OS of 94% (CI 87-100) and a 5-years TTT-free survival of 58% (95%-CI 43-80), in patients older than 60 years this was only 66% (95%-CI 49-83) and 33% (95%-CI 20-54), respectively. The presence of a PNH-clone at diagnosis was associated with a higher chance of response to IST (HR 2.2, 95%-CI 1.4-3.4). In contrast, higher age and higher severity of AA were associated with lower chance of becoming transfusion independent after IST. Patients aged 41 to 60 and with an age above 60 were less likely to become transfusion-independent after IST compared to younger patients (HR for both age groups 0.4, 95%-CI 0.2-0.7 and 0.3-0.7, respectively). The same was observed for patients with (very) severe AA compared to patients with non-severe AA (HR for severe and very severe AA both 0.4, 95%-CI 0.3-0.7 and 0.3-0.8, respectively). We used these three distinguishing baseline characteristics to calculate model-based predictions of treatment success for reference patients. The Figure shows an example for 2 reference patients with severe AA but different baseline characteristics.

Conclusions: TTT-free survival can be used as a treatment goal for IST in AA patients. Higher age, (very-)severe AA and the absence of a PNH clone at diagnosis are all associated with a lower probability to reach this positive endpoint.

Clinical Trial Registry: Not applicable

Disclosure: No potential conflicts of interest

20: Aplastic Anaemia

M.Luz Uria-Oficialdegui ¹, Victor Quintero², M.Isabel Benitez-Carabante¹, Maria Travazo³, David Bueno², Marta Gonzalez-Vicent⁴, Monica Lopez-Duarte⁵, Jaime Verdu⁶, Carolina Fuentes⁷, Graciela Gomez⁸, Laura Murillo-SanJuan¹, Laura Alonso¹, Melissa Panesso¹, Crsitina Diaz-de-Heredia¹

¹Hospital Universitari Vall d´Hebron, Barcelona, Spain, ²Hospital Universitario La Paz, Madrid, Spain, ³Hospital Sant Joan de Deù, Barcelona, Spain, ⁴Hospital Universitario Niño Jesus, Madrid, Spain, ⁵Hospital Universitario Marqués de Valdecilla, Santander, Spain, ⁶Hospital Clinico Universitario, Valencia, Spain, ⁷Hospital Universitario La Fe, Valencia, Spain, ⁸Hospital Clinico de Santiago, Santiago, Spain

Background: The recommended treatment in pediatric patients with Severe Idiopathic Aplastic Anemia (SAA) and Refractory Cytopenia of Childhood (RCC) in first-line is hematopoietic stem cell transplantation (HSCT) from a matched sibling/family donor (MSD). When MSD is unavailable, immunosuppressive therapy (IST) with horse antithymocyte globulin (ATG) plus cyclosporin is indicated. Nevertheless, IST has a high failure rate attributed to lack of response, relapse, and clonal evolution. The outcomes of matched unrelated donor (MUD) HSCT following IST are favorable but inferior when compared to MSD. Front-line MUD-HSCT may improve outcomes compared to IST and rescue MUD-HSCT after IST failure. The aim of this study is to assess the role of upfront MUD-HSCT for pediatrics patients with SAA/RCC who lack a MSD.

Methods: We performed a retrospective multicenter observational study of pediatric patients diagnosed with either SAA or RCC and treated with MUD-HSCT as first line in the absence of MSD. All patients were treated in six hospitals in Spain affiliated to the Spanish Group for HSCT in children.

Results: 19 patients (16 males/3 females) with SAA (17) and RCC (2) underwent an upfront MUD-HSCT (10/10). Median age at diagnosis was 11 years (1-17). No one had cytogenetic anomalies. Median time from diagnosis to HSCT was 102 days (61-258). Most patients had a good performance status (Lansky >70). Two had fungal infections before HSCT. Conditioning regimen consisted of fludarabine plus cyclophosphamide in SAA-patients (n = 17), and busulfan/treosulfan, fludarabine +/- thiotepa in RCC-patients (n = 2). All received lymphodepletion with thymoglobulin. GVHD prophylaxis was cyclosporine/methotrexate in most cases. Bone marrow (BM) was the prevalent source. Median cell dose for BM was 4x10⁸ CNT/kg (1.72-8.8) and for peripheral blood 13.5x10⁶ CD34 + /kg (7.3-12.3). All patients engrafted. Fifteen had complete donor chimerism, four SAA-patients had mixed chimerism, and one experienced secondary graft failure 11 months after-HSCT. The remaining three maintained stable mixed chimerism. Four developed poor graft function (PGF) and eight immune cytopenias post-HSCT. Eleven experienced bacterial infections, two with severe septic shock requiring Pediatric Intensive Care Unit admission. No cases of post –HSCT fungal infection occurred, and while viral reactivations were common, no patient developed viral disease. Endothelial complications included veno-occlusive disease (n = 3) and transplant-associated microangiopathy (n = 2). Five presented grade 2 acute GVHD and two developed chronic GVHD (one mild, one moderate), none developed severe chronic GVHD. No transplant-related mortality was registered. Median duration of post-transplant immunosuppressive treatment was 15 months (2-23). After a median follow up of 20 months (6-105), the overall survival (OS) for the entire cohort was 100% and the event-free survival (EFS) was 94.7% (CI 84-100).

Conclusions: In our experience, upfront MUD-HSCT yields outstanding outcomes in terms of OS, EFS, chronic GVHD, and transplant-related complications. The main complications were bacterial infections, viral reactivation, and immune cytopenias. Managing mixed chimerism and PGF is a critical aspect. A more extended follow-up is required to assess late sequelae and substantiate these findings. Our data strongly support the preference for upfront MUD-HSCT in pediatric patients newly diagnosed with SAA and RCC.

Disclosure: none

20: Aplastic Anaemia

Andrea Bacigalupo¹, Nicola Mordini², Francesco Zallio³, Irene Cavattoni⁴, Luca Castagna⁵, Anna Paola Iori⁶, Patrizia Chiusolo⁷, Federica Sora’⁷, Sabrina Giammarco ⁷, Elisabetta Metafuni⁷, Alessandro Busca⁸, Alessandra Carotti⁹, Fabio Benedetti¹⁰, Giorgia Saporiti¹¹, Simona Sica⁷

¹Fondazione Policlinico Universitario Gemelli IRCCS, Roma, Italy, ²Ospedale Santa Croce, Cuneo, Italy, ³Ospedale Civile, Alessandria, Italy, ⁴Ospedale di Bolzano, Bolzano, Italy, ⁵Ospedale Cervello, Palermo, Italy, ⁶Policlinico Umberto1, Roma, Italy, ⁷Policlinico Gemelli, Roma, Italy, ⁸Ospedale Le Molinette, Torino, Italy, ⁹Ospedale Perugia, Perugia, Italy, ¹⁰Ospedale Borgo Roma, Verona, Italy, ¹¹Policlinico Ca Granda, Milano, Italy

Background: The Baltimore group has introduced a new transplant platform for patients with aplastic anemia (AA) undergoing a haploidentical transplantation (HAPLO) (De Zern et al 2020); the conditioning regimen is based on fludarabine cyclophosphamide (FLU-CY) TBI ; GvHD prophylaxis is based on 4 drugs : ATG on day -9-8-7, post transplant CY (PTCY) 50 mg/kg on days +3 + 4 and tacrolimus mycophenolate. The Baltimore group reports 37 patients, with 35 surviving without GvHD. The Brasilian group has confirmed these results in a multicenter study on 87 AA patients, with survival in excess of 90%, especially for those receiving an intensified TBI dose of 4Gy (instead of 2 Gy) and an increased dose of CY pre transplant (50 instead of 30 mg/kg) (BBMT, 2020; e222-e226). If this platform allows the engraftment of an HLA HAPLO mismatched marrow, with little or no GvHD, then the same platform should be successful in patients with AA undergoing an unrelated donor (UD) transplant, or in elderly patients grafted from matched siblings.

Methods: We have grafted 25 patients with AA with this platform. The conditioning was as follows: CY 14.5 mg/kg days -6-5, Fludarabine 30 mg/m^2 days -6-5-4-3-2, TBI 2 Gy day -1 and unmanipulated bone marrow on day 0 (8 patients were grafted with PB cells). GvHD prophylaxis was ATG 0.5, 2 and 2 mg/kg days -9-8-7, PTCY 50 mg/kg day +3 + 4, cyclosporine (CSA) on day +5 onward and mycophenolated (MMMF) day 5 day 35. The median age was 40 years (range 19-70). The donor was unrelated, matched (8/8, n = 16), or mismatched (7/8, n = 6), HAPLO (n = 2), or a matched sibling (n = 1). In the latter this transplant platform was chosen because of the patients age (60 years).

Results: All patients had 100% donor chimerism within 30 days from transplant. There were 4 early deaths within day +100 (2 infections, 1 cardiac failure, 1 brain hemorrhage). One patient (UD 7/8 matched donor) (4%) experienced a rejection on day +88; underwent a second transplant from the same UD with peripheral blood as a stem cell source, on day +154 from the first transplant; the patient is currently alive and well with trilineage recovery on day +901. The median time to a neutrophil count of 0.5x10^9/L was 19 days (15-27) and for 20x10^9/L platelets, it was 22 days (13-38). Acute GvHD grade I was recorded in 3 patient (11%), minimal chronic GvHD was reported in 2 patients (7%); moderate severe chronic GvHD was not reported. EBV reactivation was seen in 5 patients and was treated in 2 patients with rituximab. CMV reactivation was seen in 3 patients. With a median follow up for surviving patients of 490 days (30-1400), the one year actuarial survival is 85% (92% vs 75% for patients younger or older than 40 years).

Conclusions: This initial experience with the Baltimore quadruple GvHD prophylaxis, in patients undergoing unrelated donor transplants for AA, confirms a very high degree of engraftment, low early mortality, low rate of rejection, and little or no GvHD.

Clinical Trial Registry: NA

Disclosure: The AUthors declare no conflict of interest

21: Autoimmune Diseases

Denis Fedorenko ¹, Vladimir Melnichenko¹, Anatoly Rukavitsin¹, Nikolai Vasilev¹, Ilya Nikolaev¹, Tatiana Nikitina², Natalia Porfirieva³, Tatiana Ionova²

¹Pirogov National Medical Surgical Center, Moscow, Russian Federation, ²Saint-Petersburg State University Hospital, Saint-Petersburg, Russian Federation, ³Multinational Center for Quality of Life Research, Saint-Petersburg, Russian Federation

Background: Multiple sclerosis (MS) is a major inflammatory and demyelinating autoimmune disease of the central nervous system, which affects mainly young people and can lead to severe and irreversible clinical disability. Modern disease-modifying treatment does not provide satisfactory control of MS. At present HDIT + AHSCT may be considered as an optimistic therapeutic option for patients with MS. We aimed to evaluate outcomes of HDIT + AHSCT in patients with various types of MS at long-term follow-up.

Methods: Patients with various types of MS who underwent HDIT + AHSCT from October 2006 to January 2022 were enrolled in a single-center study. Three low intensity conditioning regimens were used – BEAM-like, Cyclophosphamide, and Fludarabine+Cyclophosphamide. Transplantation-related mortality (TRM) definition included every death occurring within 100 days of transplantation. For clinical outcomes neurological assessment and MRI scans were performed. Neurological assessment using EDSS was performed at baseline, at discharge, at 3, 6, and 12 months after transplantation, every 6 months thereafter up to 48 months, and then at yearly intervals. Study end-points were disability defined by the EDSS score and event-free survival (EFS). EDSS decrease of 1.0 or greater was considered significant improvement and an increase of 1.0 or greater is considered significant worsening. For comparisons paired t-test, Wilcoxon test and ANOVA were used. EFS in terms of relapse-free survival (RFS) and progression-free survival (PFS) after AHSCT were evaluated using Kaplan-Meyer method. For comparison of survival log-rank and Tarone-Ware criterions were applied.

Results: In total, 730 patients with relapsing-remitting (RRMS, n = 387, 53%), primary progressive MS (PPMS, n = 120, 16%), and secondary progressive MS (SPMS, n = 223, 31%) were enrolled. Median age – 40 years (Q1;Q3 – 32;48), 39% were males. Mean (SD) disease duration was 6.8 (5.9) yrs. BEAM-like regimen was used in 31%, Cyclophosphamide – 60%, and Fludarabine+Cyclophosphamide - 9%. There were 6 cases of TRM (RRMS - 3, SPMS - 2, PPMS - 1; median EDSS 4.5). In all cases Cyclophosphamide was used. TRM in whole group was 0.82%. Median follow-up after AHSCT was 21.5 months (Q1;Q3 – 12;47). The vast majority of patients responded to treatment and exhibited EDSS improvement or were stable during the entire period of follow-up. Significant improvement in EDSS for the entire group at all-time intervals after transplantation as compared with baseline was observed (p < 0.001). At 12 months after AHSCT the decrease of EDSS in RRMS took place from median 2.0 to 1.5, in SPMS – from median 6.0 to 5.0, and in PPMS – from 5.5 to 4.0. At 5-years the estimated EFS probability for the entire group was 82.8% (95%CI 78.2-87.3), RFS for RRMS – 87.8% (95%CI 81.5-94.2), PFS for SPMS – 83.6% (95%CI 76.5-90.8), and for PPMS – 85.6% (95%CI 75.9-95.1). No differences in EFS were found depending on conditioning regimen (log-rank, p = 0.128, Tarone-Ware, p = 0.129).

Conclusions: The results of our study demonstrate that HDIT + AHSCT with low-intensity conditioning regimens is safe and effective in patients with different types of MS. The vast majority of patients improved and clinical improvement was persistent at long-term follow-up. The outcomes were similar across patients who received different low intensity conditioning regimens.

Clinical Trial Registry: No

Disclosure: Nothing to declare

21: Autoimmune Diseases

Majid Kazmi ¹, Paolo Muraro², Varun Mehra³, Ian Gabriel⁴, Eleonora De Matteis², Gavin Brittain⁵, Richard Nicholas⁴, Eli Silber³, Julia Lee⁶, Rachel Pearce⁶, Ruth Paul⁶, Maria Pia Sormani⁷, Victoria Potter³, Eduardo Olavarria⁴, Ram Malladi⁸, Basil Sharrack⁵, John Snowden⁵

¹Guys & St Thomas’ NHS Trust, London, United Kingdom, ²Imperial College London, London, United Kingdom, ³Kings College Hospital NHS Trust, London, United Kingdom, ⁴Imperial College Healthcare Trust, London, United Kingdom, ⁵Sheffield Teaching Hospitals NHS Foundation Trust, Sheffield, United Kingdom, ⁶British Society of Blood and Marrow Transplantation and Cellular Therapy, London, United Kingdom, ⁷University of Genova, Genova, Italy, ⁸Cambridge University Hospitals NHS Trust, Cambridge, United Kingdom

Background: Autologous Haematopoietic Stem Cell Transplant (HSCT) has been used as a treatment for Multiple Sclerosis (MS) since 1995 with over 2000 procedures now reported to the EBMT registry, making it the most common indication for any autoimmune disease. The UK activity has increased significantly since 2016 and it is now one of the most active countries performing HSCT for MS in the EBMT registry. We report on the UK experience of HSCT in MS in terms of UK activity (2002 to 2023), transplant related mortality (TRM) and efficacy of HSCT for MS as defined by expanded disability status scale (EDSS) Progression Free Survival (PFS) at 2 years and 4 years (EDSS-PFS).

Methods: Retrospective cohort study on patients with MS (pwMS) treated with HSCT at 14 transplant centres across the UK, between 2002 and 2023. Patients in active trials were excluded. EDSS-PFS was defined as 6 months confirmed EDSS progression compared to pre-treatment baseline (0.5 point increase if the baseline score was ≥6.0 and 1.0 for baseline scores <6.0). TRM was defined as all patients who had started conditioning therapy for HSCT regardless of whether they received stem cell re-infusion.

Results: 363 pwMS from 14 UK transplant centres were identified; 129 patients (38%) had progressive MS (primary (PPMS) or secondary (SPMS)) and 209 (62%) had relapsing-remitting MS (RRMS). By cohort analysis 22 % of patients were transplanted prior to 2016 and 88% post 01/2016. 97% of patients had cyclophosphamide/ATG based conditioning. Overall median EDSS at time of HSCT was 6.0 (range 0-9) and duration of disease 10 years (range 4-34) [table 1].TRM was 1.1%. Median EDSS for those that died was 6.5 at time of HSCT. EBV reactivation occurred in 76.8 % of patients where EBV results were reported (239/311). EDSS-PFS was 86% at 2 years post HSCT and 71% at 4 years (Fig.1). At 4 years patients with PPMS had lower EDSS-PFS (56%) compared to SPMS (71%) and RRMS (75%). When split by EDSS at time of HSCT; 4 year EDSS-PFS was 80% for those with EDSS 4.5 or less and 66 % for those with EDSS above 4.5.

Table 1.

Fig 1.

Conclusions: This is the largest UK report on HSCT for pwMS. Most activity occurred since Jan 2016 in line with published evidence and commissioning approval. This study included 38% progressive MS patients historically excluded from randomised studies of HSCT and many patients had advanced disability at time of HSCT (median EDSS 6.0). Despite that, in this real-world cohort, 71% of patients remain free of progression at 4 years post HSCT. Higher disability at time of HSCT and primary progressive MS is associated with worse EDSS-PFS. TRM was 1.1% with all those patients having advanced disability. EBV reactivation remains a significant issue and routine monitoring is mandatory. Careful patient selection is important to minimise risk and maximise benefit. The STAR MS trial is currently examining the efficacy of HSCT vs high efficacy Disease modifying therapy in relapsing-remitting MS.

Clinical Trial Registry: BSBMT CTCR 18-01.

Disclosure: Nothing to declare

21: Autoimmune Diseases

Alice Mariottini ¹, Chiara Nozzoli², Alessandro Barilaro², Riccardo Boncompagni², Ilaria Cutini², Antonella Gozzini², Chiara Innocenti², Anna Maria Repice², Luisa Pastò², Maria Pia Amato¹, Luca Massacesi¹, Riccardo Saccardi²

¹University of Florence and Careggi University Hospital, Florence, Italy, ²Azienda Ospedaliero Universitaria Careggi, Florence, Italy

Background: Autologous haematopoietic stem-cell transplantation (AHSCT) is a treatment option in aggressive relapsing-remitting (RR-) multiple sclerosis (MS) refractory to conventional disease-modifying treatments (DMTs). Here we report the long-term outcomes from our 25-year experience at the Careggi University Hospital in Florence.

Methods: We performed a monocentric retrospective study including relapsing (R-)MS patients (primary-progressive MS excluded) treated with AHSCT in Florence between 1998 and 2023. Peripheral blood hematopoietic stem cells were mobilized with chemotherapy (Cyclophosphamide [Cy] 4 g/sqm) followed by G-CSF; the conditioning protocol used was BEAM + ATG in all the patients. Variables are reported as median (range) and number (frequency), as appropriate. Efficacy outcomes include survival free from relapses (RFS), EDSS worsening (progression-free survival, PFS) and No Evidence of Disease Activity (NEDA-3), defined as the absence of all the following: relapses, EDSS worsening, and new inflammatory activity at brain magnetic resonance imaging (MRI). Primary safety endpoints explored were 100-days transplant-related mortality (TRM) and hospital readmission for any cause.

Results: One hundred and ten R-MS patients were included. TRM at 100 days was 1% and 4 (3%) patients required hospital readmission, mainly for infective episodes. Efficacy outcomes are currently available for 93 cases (Table 1). Follow-up duration was 57 months (1 – 262). PFS at 3, 5 and 10 years was 71%, 69% and 62%, respectively, being higher in RR-MS than in secondary-progressive (SP-)MS (at year 5: 88% vs 48%, respectively; p < 0.0001). RFS was 98% at 5 years, due to 3 relapsed patients in the RR-MS group and 1 in the SP-MS group. NEDA-3 survival at 3, 5 and 10 years was 70%, 68%, and 52%, respectively, being higher in RR-MS than in SP-MS (at year 5: 86% vs 48%, respectively; p < 0.0001). NEDA-3 failure was mostly caused by EDSS progression, with sparse events of new focal inflammatory activity, being 6 events (4 relapses + 2 new asymptomatic brain lesions) observed over 575 patient-years of follow-up.

Table 1: Baseline clinical-demographic characteristics of the R-MS patients included in the efficacy analyses

Conclusions: Our data firmly support that AHSCT is a safe and effective treatment for patients with aggressive R-MS, likely offering better long-term efficacy outcomes compared to high-efficacy DMTs. However, limitations arise from the lack of prospective comparative data with DMTs and of comparison between different intensity conditioning regimens.

Disclosure: Nothing to declare.

21: Autoimmune Diseases

Riccardo Saccardi ¹, Marta Ponzano², Manuela Badoglio³, Grzegorz Helbig⁴, Marek Smilowsky⁴, Alice Mariottini¹, Joachim Burman⁵, Kristina Carlson⁵, Majid Kazmi⁶, Paolo Muraro⁷, Jane Apperley⁸, Barbara Withers⁹, Jennifer Massey⁹, Riccardo Varaldo¹⁰, Matilde Inglese², Jaime Sanz¹¹, Sara Garcia Gil-Perotin¹¹, Basil Sharrack¹², Tobias Alexander¹³, Raffaella Greco¹⁴, John Snowden¹²

¹Azienda Ospedaliero Universitaria Careggi, Florence, Italy, ²University of Genoa, Genoa, Italy, ³EBMT Paris Study Unit, Paris, France, ⁴Silesian Medical Academy, Katowice, Poland, ⁵University Hospital | Uppsala, Uppsala, Sweden, ⁶Kings College Hospital, London, United Kingdom, ⁷Imperial College London and Imperial College Healthcare Trust, London, United Kingdom, ⁸Imperial College Hammersmith London, London, United Kingdom, ⁹St. Vincent’s Hospital, Sydney, Australia, ¹⁰IRCCS Ospedale Policlinico San Martino, Genoa, Italy, ¹¹University Hospital La Fe, Valencia, Spain, ¹²Sheffield Teaching Hospitals, Sheffield, United Kingdom, ¹³Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin, Germany, ¹⁴IRCCS San Raffaele Scientific Institute, Vita-Salute San Raffaele University, Milan, Italy

Background: Multiple Sclerosis (MS) is the most frequent Autoimmune Disease (AD) treated with Autologous HSCT. Starting from 1985 more than 2000 patients underwent the procedure and were reported to the EBMT Registry, with a constant increasing trend in the last 10 years. BEAM/ATG was the most frequent conditioning regimen up to 2013, whilst Cyclophosphamide (CYC)/ATG predominated in the following years. Both regimens are included in the guidelines and recommendations from the EBMT Autoimmune Diseases Working Party (Sharrack et al 2020). Intensity of conditioning regimen results in different toxicity and, at least in the oncological setting, in different efficacy. There are no prospective trials comparing these two regimens: we report here a retrospective analysis of the EBMT database aimed to assess the risk/benefit ratio of the two regimens.

Methods: We selected 1114 MS patients from the EBMT database who were conditioned with either BEAM/ATG or CYC/ATG between 1985 and 2023. Primary endpoint was time to fail NEDA (Non-Evidence of Disease Activity), which is lack of relapses, no progression of the disability and lack of new lesions at MRI scan. Kaplan Meier estimates, log rank test and multivariable Cox regression models were used to study time to events (engraftment, death and NEDA).

Results: Data about engraftment and 100-days mortality were available in 1114 patients (442 BEAM/ATG, 672 CYC/ATG). PMN engraftment in BEAM/ATG- and CYC/ATG-treated patients occurred at a median time of 11 (95% CI: 11; 11) and 10 (95%CI: 10; 11) days after HSCT, respectively, p = 0.079. Nine out 442 (2.0%) patients conditioned with BEAM/ATG and 7 out 672 (1.04%) with CYC/ATG died within 100 days from HSCT (p = 0.191).

Exhaustive data MS about the primary endpoint (time to fail NEDA) were available from 616 patients (295 received BEAM/ATG, 321 CYC/ATG). Details about this patient subset are reported in table 1. Overall incidence of NEDA failure in BEAM/ATG and CYC/ATG at 5 years was 42.3% (95% CI: 36.3%-48.8%) and 44.2% (95% CI: 31.0%-60.2%), respectively (p = 0.145). No statistically significant differences between the two treatments (HR = 0.95, p-value = 0.809, 95% CI = 0.65; 1.41) were confirmed, when adjusting for disease type (progressive vs relapsing), EDSS at baseline and year of transplant. KM curves reporting time to NEDA failure are reported in graph 1.

Table 1

Graph 1

Conclusions: Despite the relevant number of patients included in the analysis, a clear difference between the regimens is lacking, both in terms of toxicity and efficacy, possibly due also to the different median follow-up interval in the two groups. Type of disease (relapsing/remitting vs progressive) is still the major determinant of neurological outcome. A prospective comparative trial is likely necessary to assess the risk/benefit ratio of the two regimens. However, a tailored approach according to patients and disease characteristics might be considered.

Disclosure: Nothing to declare.

21: Autoimmune Diseases

María de Lourdes Pastelín-Martínez¹, Moisés Manuel Gallardo-Pérez², Andres Gómez-De-León³, Juan Carlos Olivares-Gazca², Edgar Jared Hernández-Flores⁴, Daniela Sánchez-Bonilla², Merittzel Abigail Montes-Robles¹, Max Robles-Nasta², Guillermo Ocaña-Ramm⁴, Silvia Soto-Olvera⁴, David Gómez-Almaguer³, Guillermo José Ruiz-Delgado ^2,4, Guillermo José Ruiz-Arguelles^2,4

¹Universidad Anáhuac Puebla, Puebla, Mexico, ²Centro de Hematología y Medicina Interna, Clínica Ruiz, Puebla, Mexico, ³Servicio de Hematología del Hospital Universitario “Dr. José Eleuterio González”, Monterrey, Mexico, ⁴Universidad Popular Autónoma del Estado de Puebla, Puebla, Mexico

Background: Multiple sclerosis (MS) is a disabling disease that affects young adults. Treatments for MS have increased exponentially in number, efficacy and risk. Autologous hematopoietic stem cell transplantation (aHSCT) can change the natural history of the disease. To analyze if aHSCT should be done early in the course of the disease or after failing of other therapies, we have studied the long-term results of aHSCT in a cohort of persons with MS.

Methods: Patients with MS referred to our center for aHSCT between June 2015 and July 2023 were prospectively entered in the study. All phenotypes of MS were included. The follow up was assessed with the patient reported EDSS score in an online form; only patients followed by three or more years were included in the analysis. The progression of MS was assessed by changes in the expanded disability status scale (EDSS).The response rate was analyzed for four variables: Disease-modyfing treatment (DMT) given or not before the HSCT and delay in the diagnosis and treatment: More (delayed diagnosis DD), or less (non-delayed diagnosis NDD) than 8 months after the onset of symptoms.

Results: 1518 subjects were prospectively enrolled. In a subset of 103 patients, detailed pre- and post-transplant evolution could be recorded. Two groups of patients were analyzed according to the time of evolution between the onset of symptoms and the definite diagnosis of MS: More than 8 months, (delayed diagnosis, DD) or less than 8 months (non-delayed diagnosis, NDD). The time elapsed between the onset of symptoms and the correct diagnosis was lower for the NDD group (1.55 vs 35.87 months, p < 0.05). The EDSS at the time of the transplant was higher in the DD group (4.5 vs 3.0, p = 0.3) and the response of the EDSS score to the transplant was significantly better for the NDD group, the last censored values for the EDSS being 2.5 vs 4.25 (p = 0.03). The two groups of patients responded to aHSCT by diminishing the EDSS, but the response was significantly better in the NDD group. 74 patients were followed for more than 36 months, the response rate (RR = improvement + stabilization) at 12, 24 and 36 mo was 84%, 84% and 58% respectively for patients not receiving prior disease modifying treatments (DMT) and 72%, 90% and 67% for patients receiving DMT. The EDSS score was on average worsening in patients before the aHSCT, but the transplant stabilized the EDSS score at 3 years in patients with prior exposure to DMT, whereas in persons not given DMT, the transplant resulted in a significant decrease (p = .01) of the EDSS score.

Conclusions: These data indicate that both the pre-transplant progression of the disease and the response to aHSCT were significantly worse in persons with a delayed diagnosis. The response to aHSCT was better for persons not exposed to immunosuppressive DMT before the transplant. Additional studies are needed to further analyze the impact of the use of DMT before aHSCT in MS, as well as the timing of the procedure.

Disclosure: Nothing to declare

21: Autoimmune Diseases

Sebastian Bittner ¹, Thomas Hehlgans¹, Markus Feuerer¹

¹Leibniz Institute for Immunotherapy, Regensburg, Germany

Background: Engineered regulatory T cell (Treg cell) therapy is a promising strategy to treat patients suffering from inflammatory diseases, autoimmunity and transplant rejection. However, in human autoimmune diseases, the implicated autoantigens, which could serve as potential targets for TCR- or CAR-engineered Treg cells, are often unknown or vary strongly between individual patients. In order to circumvent this problem, we invented a new class of artificial receptors for Treg cells that sense and respond mediators of inflammation instead of tissue-specific antigens.

Methods: We genetically modify Treg cells with new synthetic receptors, named artificial immune receptors (AIRs), that recognize inflammatory ligands and trigger Treg activation resulting in improved Treg-mediated protection in a major MHC mismatch model of graft versus host disease (GvHD).

Results: In many autoimmune and inflammatory diseases there is a lack of knowledge about relevant disease-driving antigens and often also multiple organs and tissues are affected in individual patients. Therefore, potential uniform target antigens that could be selected for TCR- or CAR–Treg cell therapy are unknown. In contrast to this, mediators of inflammation show a high redundancy as well as functional importance for the development of various inflammatory diseases, including autoimmunity and alloreactivity. Especially, cytokines of the tumor necrosis factor (TNF) superfamily are involved in different inflammatory and autoimmune diseases. For example, therapeutic intervention with TNF-inhibitors is an important treatment option for several inflammatory diseases. These considerations led us to develop a concept for engineered Treg cell therapy by generating artificial immune receptors (AIRs) that target these inflammatory mediators instead of tissue-specific antigens.

In our work, we introduce a new class of synthetic biosensors, named AIRs, for murine and human Treg cells. AIRs consist of three domains and have an extracellular binding domain of a tumor necrosis factor (TNF)-receptor superfamily member, an intracellular costimulatory signaling domain of CD28, and T cell receptor signaling domain of CD3-ζ chain. AIR receptors equip Treg cells with an inflammation-sensing machinery and translate this environmental information into a CD3-ζ chain–dependent TCR-activation program. Different AIRs were generated, recognizing the inflammatory ligands of the TNF-receptor superfamily, including LIGHT, TNFα, CD40L/ CD154 and TNF-like ligand 1A (TL1A), leading to activation, differentiation, and proliferation of AIR–Treg cells. In preclinical GvHD models, AIR-expressing Treg cells, which can be activated by the ligands LIGHT or CD40L, protect significantly better than control Treg cells. Expression and signaling of the corresponding human AIR in human Treg cells prove that this concept can be translated. AIRs as a synthetic tool for cellular therapy allow Treg cells to sense and target these environmental inflammatory signals and translate them into a CD3-ζ chain–dependent TCR program, enabling Treg cells to fulfill their suppressive and tissue-protective functions, independent of a CAR- or TCR-specific antigen.

Conclusions: Engineering AIR-Treg cells that target inflammatory ligands leading to TCR signaling and activation might be used as a Treg cell–based therapy approach for a broad range of inflammation-driven diseases.

Disclosure: S.B., T.H., and M.F. are inventors on patent application (patent application no. WO/2023/227521) based on the presented technology.

21: Autoimmune Diseases

Giorgio Orofino ¹, Angela Genchi¹, Maria Teresa Lupo-Stanghellini¹, Matteo Doglio¹, Maddalena Noviello¹, Alessandra Mandelli¹, Sara Mastaglio¹, Simona Piemontese¹, Elena Tassi¹, Francesca Ferrua¹, Roberto Furlan¹, Gianvito Martino^1,2, Massimo Filippi^1,2, Massimo Bernardi¹, Consuelo Corti¹, Jacopo Peccatori¹, Francesca Tortorelli¹, Cristina Toffalori¹, Luca Vago^1,2, Chiara Bonini^1,2, Lucia Moiola¹, Fabio Ciceri^1,2, Raffaella Greco¹

¹IRCCS Ospedale San Raffaele Scientific Institute, Milan, Italy, ²San Raffaele Vita-Salute University, Milan, Italy

Background: Here we report the long-term follow-up analysis of the first 2 patients treated with allogeneic Hematopoietic Stem Cell Transplantation (HSCT) for refractory neuromyelitis optica (NMO), an aggressive neurological autoimmune disease with poor prognosis, respectively 14 and 13 years ago, with a sustained disease remission.

Methods: Both patients exhibited serum positivity for pathogenic anti-aquaporin 4 antibodies (AQP4-Ab). They had a severe disease course with disability accumulation, despite several lines of immunosuppressive treatment, including autologous HSCT. UPN#1 underwent allogeneic HSCT from his HLA-identical sibling, while UPN#2 from Matched Unrelated Donor (MUD). Conditioning regimen consisted of treosulfan-fludarabine. GvHD prophylaxis combined anti-T lymphocyte globulin (ATLG) with cyclosporine-methotrexate (Upn#1) or mycophenolate-sirolimus (Upn#2); B cell depletion was obtained by rituximab. We performed sequential clinical follow-up visits, with laboratory and neuroradiological analyses. Immune-reconstitution was assessed at 3, 4, 6 and 13 years after HSCT.

Results: Hematopoietic recovery occurred within day 30, accompanied by full donor chimerism. None of the patients experienced Graft-versus-Host-Disease (GvHD). Both patients demonstrated a marked and sustained improvement of neurological functions (Expanded Disability Status Scale dropped from 6 to 2.5 in UPN#1, and from 8.5 to 7.5 in UPN#2). At the latest follow-up both patients are alive; clinical and neuroradiological assessments showed disease remission, along with the persistent disappearance of pathogenetic AQP4-Ab.

Immune-reconstitution analyses showed a normal recovery of circulating lymphocytes, except for a mild impairment of B-Cell Immune Reconstitution in #UPN1. This patient developed a post-transplant Immunoglobuline (Ig)-deficiency, with recurrent upper-tract respiratory symptoms, requiring Ig supplementation. To investigate this immunodeficiency, we performed exome sequencing on the patient’s Peripheral Blood Mononuclear Cells (PBMC), finding a mutation on the NFkB gene with a variant of uncertain significancy (VUS), and performed lineage specific microchimerism analysis, demonstrating full donor chimerism in B, T and myeloid cells.

In both patients, circulating B cells were predominantly naïve.

All the main differentiation subsets for CD4 + T-cells were present, including early differentiated ones (naive, stem-memory and central-memory T-cells). The expression of the CD31 marker on naive CD4 + T-cells, identifying bona fide recent thymic emigrants, indicates a robust thymic output.

Moreover, we observed a resting phenotype of CD4 + T-Cells (HLA-DR-), with only a minority of them expressing checkpoints inhibitors on their surface.

Upon activation with polyclonal stimuli T-cells proved to be polyfunctional with sustained cytokine production at all time-points.

Similar findings were confirmed among the CD8 + T-cell compartment.

We documented the presence and the changes of the T-Reg compartment and it’s subsets across time, potentially contributing to long-term beneficial effects.

Conclusions: The effectiveness of allogeneic HSCT in achieving sustained remission in NMO patients is noteworthy. Our immunological results suggest that allogeneic HSCT can provide long-term disease control in refractory NMO through several concurring mechanisms: the eradication of autoreactive cell clones by high-dose chemotherapy and by in vivo T and B cell depletion, elimination of long-lived plasma cells producing AQP4-Ab, the re-establishment of thymic central tolerance and renewal of the immune repertoire.

Disclosure: I declare no disclosures

4: CAR-based Cellular Therapy – Clinical

Alejandro Luna de Abia ¹, Sean M. Devlin¹, Jessica Flynn¹, Magdalena Corona de Lapuerta¹, Efrat Luttwak¹, Ivan Landego¹, Parastoo B. Dahi¹, Richard J. Lin¹, Allison Parascondola¹, M.Lia Palomba¹, Gunjan L. Shah¹, Michael Scordo¹, Brandon S. Imber¹, Ana Alarcón Tomás¹, Gilles Salles¹, Jae H. Park¹, Miguel-Ángel Perales¹, Roni Shouval¹

¹Memorial Sloan Kettering Cancer Center, New York, United States

Background: Extranodal (EN) disease is an established poor prognostic factor in large B-cell lymphoma (LBCL). Whether EN sites serve as potential sanctuaries fostering resistance against CD19-directed chimeric antigen T-cell (CAR-T) activity remains unclear. This study aims to elucidate disease dynamics before and after CD19-CAR-T therapy across anatomical sites of LBCL involvement.

Methods: In this single-center retrospective study, adults with LBCL treated with CD19-CAR-T were included; primary central nervous system (CNS) disease was an exclusion criterion. We retrospectively reviewed 921 18F-fluorodeoxyglucose positron emission tomography/computed tomography (PET) scans performed before apheresis, after apheresis, at the time of best response, and relapse; CNS and bone marrow involvements were evaluated when available.

Results: Among 274 patients with LBCL receiving CD19-CAR-T cells (axi-cel 54%, tisa-cel 26%, liso-cel 20%), 38% had transformed lymphoma, and the majority received bridging therapy (78%). EN involvement, with or without nodal (ND) disease, before apheresis was observed in 68% of cases, with 32% having ND disease exclusively. In the last disease assessment before CAR-T, the most common EN sites involved were bone/soft tissue (41%) and gastrointestinal/peritoneum (22%); 8% had CNS/orbital involvement. Poor prognostic factors, including lower performance status (p = 0.015) and higher metabolic tumor volume (p < 0.001) were enriched in the EN group.

Patients with EN disease pre-CAR-T exhibited higher grade ≥ 2 cytokine release syndrome (CRS) (45% vs. 32%, p = 0.046), grade ≥ 2 immune effector cell-associated neurotoxicity syndrome (ICANS) (27% vs. 6%, p < 0.001), and lower complete response rates (55% vs. 78%, p < 0.001) than those with exclusive ND disease. ND disease pre-CAR-T correlated with improved progression-free survival (PFS) (HR 0.51 [95%CI: 0.34-0.77]) and overall survival (OS) (HR 0.49 [0.29-0.83]), in multivariable Cox-regression models adjusting for relevant factors.

EN site-specific response rates to CAR-T were lowest for adrenal/genitourinary (12/32, 38%), gastrointestinal/peritoneum (23/68, 34%), lungs/pleura/pericardium (19/56, 34%) and CNS/orbital (8/25, 32%). In a multivariable Cox-regression model, pre-CAR-T involvement of adrenal/genitourinary (HR 2.45 [1.59-3.77]) and hepatobiliary/pancreas (HR 1.9 [95%CI 1.23-2.94]) was associated with greater risk of relapse or progression at any site after CAR-T.

Finally, the 1 year-OS rate landmarked from the time of relapse post-CAR-T was worse in EN involvement (32% [95%CI: 24-43]) than in exclusive ND involvement (63% [95%CI: 46-86]).

Conclusions: This extensive analysis highlights the significant impact of EN disease on CAR-T therapy outcomes in LBCL. It emphasizes the necessity for tailored management strategies specific to EN involvement, advocating for organ-specific approaches in future studies. Additionally, further investigations should delve into the underlying biology driving sanctuary sites to enhance our understanding and refine therapeutic interventions.

Disclosure: Palomba: Garuda Therapeutics: Honoraria; Smart Immune: Honoraria; Thymofox: Honoraria; Novartis: Honoraria; Ceramedix: Honoraria; Seres Therapeutics: Honoraria, Patents & Royalties; Rheos: Honoraria; Juno: Honoraria, Patents & Royalties; Cellectar: Honoraria; BMS: Honoraria; MustangBio: Honoraria; Kite: Honoraria; Pluto Immunotherapeutics: Honoraria; Synthekine: Honoraria. Park: Intella: Consultancy; Servier: Consultancy, Research Funding; Allogene: Consultancy, Membership on an entity’s Board of Directors or advisory committees; Minerva Bio: Consultancy; BeiGene: Consultancy; Incyte: Research Funding; Bright Pharmacetuicals: Consultancy; Pfizer: Consultancy; Be Biopharma: Consultancy; Sobi: Consultancy, Research Funding; Takeda: Consultancy, Research Funding; Amgen: Consultancy; Fate Therapeutics: Research Funding; Autolus Therapeutics: Research Funding; Genentech, Inc.: Research Funding; Kite: Consultancy; Curocell: Consultancy; Artiva Biotherapeutics: Consultancy, Current holder of stock options in a privately-held company, Membership on an entity’s Board of Directors or advisory committees; Affyimmune: Consultancy; GC Cell: Membership on an entity’s Board of Directors or advisory committees. Salles: Ipsen: Consultancy, Research Funding; Loxo/Lilly: Consultancy; Genmab: Consultancy; Merck: Consultancy, Honoraria; Nordic Nanovector: Consultancy; Kite/Gilead: Consultancy; Genentech, Inc./F. Hoffmann-La Roche Ltd: Consultancy, Research Funding; Orna: Consultancy; AbbVie: Consultancy, Honoraria; ATB Therapeutics: Consultancy; Molecular Partners: Consultancy; Incyte: Consultancy; Janssen: Consultancy, Research Funding; EPIZYME: Consultancy; Owkin: Current holder of stock options in a privately-held company; Debiopharm: Consultancy; BMS/Celgene: Consultancy; BeiGene: Consultancy; Novartis: Consultancy; Nurix: Consultancy. Scordo: Medscape, LLC: Honoraria; CancertNetwork (Intellisphere LLC): Honoraria; Omeros Corporation: Consultancy, Research Funding; Amgen, Inc.: Research Funding; Angiocrine Bioscience, Inc.: Research Funding. Shah: ArcellX: Other: DSMB; Janssen: Research Funding; Beyond Spring: Research Funding; BMS: Research Funding; Amgen: Research Funding. Perales: Allogene: Research Funding; Omeros: Consultancy, Current equity holder in publicly-traded company, Honoraria; MorphoSys: Consultancy, Honoraria; Novartis: Consultancy, Honoraria, Research Funding; BMS: Consultancy, Honoraria; Astellas: Consultancy, Honoraria; Orcabio: Consultancy, Current equity holder in publicly-traded company, Honoraria; AbbVie: Consultancy, Honoraria; Caribou: Consultancy, Honoraria; Syncopation: Honoraria; Cidara Therapeutics: Consultancy, Other; NexImmune: Consultancy, Current equity holder in publicly-traded company; Medigene: Consultancy, Other; Equillium: Consultancy, Honoraria; Miltenyi Biotec: Consultancy, Honoraria, Research Funding; Vor Biopharma: Consultancy, Honoraria; Allovir: Consultancy; Kite: Consultancy, Honoraria, Research Funding; Servier: Other; Miltenyi Biotec: Honoraria; DSMB: Other; Incyte: Consultancy, Honoraria, Research Funding; Nektar Therapeutics: Consultancy, Honoraria, Research Funding; Merck: Consultancy, Honoraria; Adicet: Honoraria; Celgene: Honoraria; Karyopharm: Consultancy, Honoraria; Exevir: Consultancy, Honoraria; Sellas Life Sciences: Consultancy; VectivBio AG: Consultancy, Honoraria; Takeda: Consultancy, Honoraria. Imber: research funding (to institution) from Novartis, Bayer, Kazia, GT Medical Technologies, received honorarium from GT Medical Technologies.

4: CAR-based Cellular Therapy – Clinical

Larisa Shelikhova¹, Olga Molostova¹, Arina Rakhteenko¹, Margarita Perminova¹, Maria Dunaykina¹, Elena Kurnikova¹, Yakov Muzalevsky¹, Dmitriy Pershin¹, Maria Fadeeva¹, Olga Illarionova¹, Alexander Popov¹, Daria Kobyzeva¹, Natalya Miakova¹, Galina Novichkova¹, Alexei Maschan ¹, Michael Maschan¹

¹Dmitry Rogachev Federal Research Centre of Pediatric Hematology, Oncology and Immunology, Moscow, Russian Federation

Background: High disease burden is associated with CAR-T cells toxicity and high relapse rate among children with r/r BCP-ALL. In many cases long-term remission is achieved only with HSCT as post-CAR-T consolidation. We reported on the new approach with simultaneous infusion of the allogenic αβ-T cells depleted grafts and CAR-T cells, derived from the donor memory T-cells fraction. Experience suggested that this approach can be a safe treatment.

Methods: A total of 20 pts with r/r BCP-ALL (n = 16), B-NHL (n = 3) and MPAL (n = 1) (9 female, 11 male, median age 7,6 y) were enrolled. All patients had relapse after multiple lines of treatment including previous HSCT(n = 10), blinatumomab (n = 12), inotuzumab (n = 1) and CAR-T cell infusion(n = 8). At the time of HSCT application the disease burden was MRD (n = 6), AD(n = 12) and two pts had MRD negative CR2. PBMS used to produce CAR T cells were provided by the patient’s transplant donor.

The CliniMACS Prodigy T cell transduction (TCT) process was used to produce CD19 and СD19/22CAR-T cells, CD3/CD28 stimulation with MACS GMP T Cell TransAct, transduction with lentiviral (second generation CD19.4-1BB zeta, vector (Lentigen, Miltenyi Biotec company) and expansion over 7 days.

Nine (45%) pts received treosulfan-based myeloablative preparative regimen, while TBI-based regimen was used in 11(55%) pts. GvHD prophylaxis included tocilizumab at 8 mg/kg on day -1 and abatacept at 10 mg/kg on day -1, +7, +14, +28.

Final product was administered without cryopreservation to the patients: 19 pts received allogeneic CAR T cell with haplo and one with MR TCRαβ+ depletion graft (CD19 n = 3 and CD19/22 n = 17. The cell products were administered at a dose of 0,1*10⁶/kg of CAR-T cells in haploidentical donors and 0,5 *10⁶/kg in MRD. The median dose of CD34+ cells was 8.5 x10⁶/kg, αβ T cells - 56x10³/kg.

Results: Primary engraftment was achieved in 19 of 20 pts, the median time to neutrophil and platelet recovery was 13 and 14 days, respectively, all engrafted patients achieved CR at day +28 after CAR-T cell therapy. The median time to CAR-T cell peak expansion was 14 days. The median time of CAR-T persistence was 100 days and B cell aplasia was 150 days.

CRS occurred in 10 (50%) patients and only one had grade 3. Ten patients had neurologic events (ICANS grade 1 n = 5, grade 2 n = 4 and grade 3 n = 1). 5 patients had the signs of aGvHD, grade 2 (n = 4), grade 3(n = 1), 2 patients had mild chGvHD.

Eleven (55%) patients are alive in CR with a median follow-up of 3 years, 9 patients died (5 due to relapse (CD19 pos (n = 4), CD19 neg(n = 1), the median time of relapse was day 131; 4 in CR due to VOD, viral (ADV + COVID19) and fungal infection.

Conclusions: Our early experience suggests that allogenic CAR-T cells can be safely infused simultaneously with the HST graft on the platform of ab T cell depletion. The infusions did not compromise engraftment and GvHD control, while specific CAR-T toxicity was mild and manageable. We have documented CAR-T expansion and persistence. Prospective testing of the approach is warranted.

Disclosure: M. Maschan - Myltenyi Biotec - Honoraria

4: CAR-based Cellular Therapy – Clinical

Emily Liang ^1,2, Aya Albittar¹, Andrew Portuguese^1,2, Jennifer Huang^1,2, Natalie Wuliji^1,2, Qian Wu¹, Joseph De Los Reyes¹, Nikki Pin¹, Aiko Torkelson¹, Delaney Kirchmeier¹, Abigail Chutnik¹, Barbara Pender¹, Joshua Hill^1,2, Noam Kopmar^1,2, Rahul Banerjee^1,2, Andrew Cowan^1,2, Damian Green^1,2, Ajay Gopal^1,2, Christina Poh^1,2, Mazyar Shadman^1,2, Alexandre Hirayama^1,2, Brian Till^1,2, Erik Kimble^1,2, Lorenzo Iovino^1,2, Aude Chapuis^1,2, Folashade Otegbeye^1,2, Ryan Cassaday^1,2, Filippo Milano^1,2, Cameron Turtle³, David Maloney^1,2, Jordan Gauthier^1,2

¹Fred Hutchinson Cancer Center, Seattle, United States, ²University of Washington, Seattle, United States, ³University of Sydney, Sydney, Australia

Background: The EHA/EBMT immune effector cell-associated hematotoxicity (ICAHT) grading system was recently developed for hematologic toxicity after CAR T-cell therapy. We assessed the application of unsupervised time-series clustering to longitudinal ANC data, relative to the ICAHT grading system.

Methods: Adults who underwent CAR T-cell therapy for hematologic malignancies with commercial or investigational products at our center between 2013 and 2023 were included (n = 454). Log10-transformed ANC trajectories were clustered using non-supervised longitudinal k-means based on Euclidean distances (latrend and kml packages in R). Overall survival (OS) was modeled using Cox regression.

Results: The most common disease types were aggressive NHL (n = 216; 48%), ALL (n = 82; 18%), and MM/PCL (n = 61; 13%) (Table). The most common CAR T-cell products were investigational CD19 CAR T-cell products (n = 180; 40%), axi-cel (n = 114; 25%), and liso-cel (n = 58; 13%). The median follow-up was 17 months (IQR, 1-18).

The longitudinal ANC data clustered into 4 distinct trajectories (“clusters”): 1) very good (high nadir followed by rapid recovery), n = 335 (74%); 2) good (low nadir followed by rapid recovery), n = 92 (20%); 3) poor (low nadir followed by intermittent recovery), n = 18 (4%); 4) very poor (aplastic phenotype), n = 9 (2%). Grade 1, 2, 3, and 4 ICAHT occurred in 141 (31%), 177 (39%), 47 (10%), and 30 (7%), respectively. Sixty-five percent of grade 3-4 ICAHT patients were in the very good or good clusters.

Clusters were strongly associated with OS (poor vs. very good, HR = 3.5, 95% CI, 1.8-6.9, p = 0.0003; very poor vs. very good, HR = 6.6, 95% CI, 3.4-12.7, p = <0.0001), as were ICAHT categories (grade 3 vs. 0, HR = 2.7 = 9, 95% CI, 1.5-5.8, p = 0.006; grade 4 vs. 0, HR = 6.8, 95% CI, 3.3-14.0, p < 0.0001) with comparable discrimination (C-index: 0.59 vs. 0.62 for the 4-cluster solution and ICAHT grades, respectively). However, clusters demonstrated less intra-cluster variability and greater inter-cluster distinction compared to ICAHT grades (Dunn index: 0.11 vs. 0.03; average silhouette width: 0.29 vs. -0.04). The 3- and 18-month OS in patients in the poor/very poor clusters was 42% and 9%, respectively, compared to 69% and 24% in patients with grade 3-4 ICAHT.

We next evaluated non-concordant trajectories within each ICAHT grade relative to the identified clusters. We found that 14% and 51% of grade 3 patients were in the very good and good clusters, respectively; these patients exhibited distinct and more favorable patterns of ANC recovery (e.g., stable ANC recovery by day 15 after CAR T-cell infusion) compared to poor/very poor clusters.

Table. Demographic, disease, and treatment characteristics (N = 454)

Conclusions: Unsupervised time-series clustering identified distinct patterns of hematotoxicity that are strongly associated with OS in keeping with ICAHT grades. Our approach further stratified grade 3 ICAHT patients into clusters with distinct trajectories of ANC recovery. Our findings suggest opportunities to refine the EHA/EBMT ICAHT grading system to better capture patterns of intermittent ANC recovery.

Clinical Trial Registry: Unsupervised time-series clustering identified distinct patterns of hematotoxicity that are strongly associated with OS in keeping with ICAHT grades. Our approach further stratified grade 3 ICAHT patients into clusters with distinct trajectories of ANC recovery. Our findings suggest opportunities to refine the EHA/EBMT ICAHT grading system to better capture patterns of intermittent ANC recovery.

Disclosure: Joshua Hill: allovir: Consultancy, Research Funding; moderna: Consultancy; deverra: Research Funding. Rahul Banerjee: BMS: Consultancy; Janssen: Consultancy; Genentech: Consultancy; SparkCures: Consultancy; Sanofi: Consultancy; Caribou: Consultancy; Pfizer: Consultancy; Pack Health: Research Funding. Andrew Cowan: BMS, Adaptive: Consultancy; Adaptive Biotechnologies, Harpoon, Nektar, BMS, Janssen, Sanofi, Abbvie: Research Funding. Damian Green: Cellectar Biosciences: Research Funding; SpringWorks Therapeutics: Research Funding; Celgene: Consultancy; GlaxoSmithKline: Membership on an entity’s Board of Directors or advisory committees; Sanofi: Research Funding; Janssen Biotech: Consultancy, Research Funding; Ensoma: Consultancy; Seattle Genetics: Consultancy, Research Funding; Juno Therapeutics A BMS Company: Patents & Royalties, Research Funding. Ajay Gopal: Compliment Corporation: Current holder of stock options in a privately-held company; Incyte, Kite, Morphosys/Incyte, ADCT, Acrotech, Merck, Karyopharm, Servier, Beigene, Cellectar, Janssen, SeaGen, Epizyme, I-Mab bio, Gilead, Genentech, Lilly, Caribou, Fresenius-Kabi: Consultancy; Merck, I-Mab bio, IgM Bio, Takeda, Gilead, Astra-Zeneca, Agios, Janssen, BMS, SeaGen, Teva, Genmab: Research Funding. Christina Poh: BeiGene: Consultancy; Seattle Genetics: Consultancy; Incyte: Research Funding; Acrotech: Consultancy. Mazyar Shadman: Fate Therapeutics: Consultancy; Genmab: Consultancy, Research Funding; Vincerx: Research Funding; MorphoSys/Incyte: Consultancy, Research Funding; Eli Lilly: Consultancy; Bristol Myers Squibb: Consultancy, Research Funding; Genentech: Consultancy, Research Funding; Kite, a Gilead Company: Consultancy; AbbVie: Consultancy, Research Funding; Mustang Bio: Consultancy, Research Funding; ADC therapeutics: Consultancy; BeiGene: Consultancy, Research Funding; AstraZeneca: Consultancy, Research Funding; Pharmacyclics: Consultancy, Research Funding; Janssen: Consultancy; MEI Pharma: Consultancy; Regeneron: Consultancy; TG Therapeutics: Research Funding. Alexandre Hirayama: Novartis: Honoraria; Bristol Myers Squibb: Honoraria, Research Funding; Nektar Therapeutics: Honoraria, Research Funding; Juno Therapeutics, a Bristol Myers Squibb Company: Research Funding. Till: Mustang Bio: Consultancy, Patents & Royalties, Research Funding; BMS/Juno Therapeutics: Research Funding; Proteios Technology: Consultancy, Current holder of stock options in a privately-held company. Kimble: Juno/BMS: Research Funding. Lorenzo Iovino: Mustang Bio: Current equity holder in publicly-traded company. Aude Chapuis: Juno Therapeutics: Research Funding. Ryan Cassaday: Amgen: Consultancy, Honoraria, Research Funding; Merck: Research Funding; Incyte: Research Funding; Autolus: Membership on an entity’s Board of Directors or advisory committees; Pfizer: Consultancy, Honoraria, Research Funding; Servier: Research Funding; Vanda Pharmaceuticals: Research Funding; Jazz: Consultancy, Honoraria; Kite/Gilead: Consultancy, Honoraria, Research Funding; PeproMene Bio: Membership on an entity’s Board of Directors or advisory committees; Seagen: Other: Spouse was employed by and owned stock in Seagen within the last 24 months. Filippo Milano: ExCellThera Inc.: Research Funding. David Maloney: A2 Biotherapeutics: Consultancy, Current holder of stock options in a privately-held company, Honoraria, Other: Member of the Scientific Advisory Board; Juno Therapeutics: Consultancy, Honoraria, Patents & Royalties: Rights to royalties from Fred Hutch for patents licensed to Juno Therapeutics/BMS, Research Funding; Janssen: Consultancy, Honoraria; Legend Biotech: Consultancy, Honoraria, Research Funding; Mustang Bio: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Lyell Immunopharma: Other: Member, CAR T Steering Committee; Incyte: Consultancy, Honoraria; Gilead Sciences: Consultancy, Honoraria, Other: Member, Scientific Review Committee, Research Scholars Program in Hematologic Malignancies; Kite, a Gilead Sciences: Consultancy, Honoraria, Research Funding; Pharmacyclics: Consultancy, Honoraria; Umoja: Consultancy, Honoraria; Celgene: Consultancy, Honoraria, Membership on an entity’s Board of Directors or advisory committees, Other: Participation on a Data Safety Monitory Board, Research Funding; Genentech: Consultancy, Honoraria, Other: Chair and Member of the Lymphoma Steering Committee; MorphoSys: Consultancy, Honoraria; Bristol Myers Squibb: Consultancy, Honoraria, Other: Member of the JCAR017 EAP-001 Safety Review Committee and Member, CLL Strategic Council, Member of the JCAR017-BCM-03 Scientific Steering Committee under BMS, Research Funding; Amgen: Consultancy, Honoraria; Navan Technologies: Consultancy, Honoraria, Other: Member of the Scientific Advisory Board; Bioline Rx: Consultancy, Membership on an entity’s Board of Directors or advisory committees, Other: Participation on a Data Safety Monitory Board ; Fred Hutch: Other: rights to royalties for patents licensed to Juno; Navan Technologies: Current holder of stock options in a privately-held company; Chimeric Therapeutics: Other: Member of the Scientific Advisory Board; ImmPACT Bio: Other: Member, Clinical Advisory Board, CD19/CD20 bi-specific CAR-T Cell Therapy Program; Interius: Other: Member, Clinical Advisory Board. Jordan Gauthier: Kite Pharma: Consultancy, Honoraria; MorphoSys: Consultancy, Research Funding; Angiocrine Bioscience: Research Funding; Century Therapeutics: Other: Independent data review committee; Celgene (a Bristol Myers Squibb company): Research Funding; Legend Biotech: Consultancy, Honoraria; Janssen: Consultancy, Honoraria; Juno Therapeutics (a Bristol Myers Squibb company): Research Funding; Sobi: Consultancy, Honoraria, Research Funding.

4: CAR-based Cellular Therapy – Clinical

Daniela Guardo ^1,2, Vesna Pavasovic¹, Avijeet Kumar Mishra¹, Kimberly Gilmour¹, Stuart Adams¹, Giorgio Ottaviano¹, Danielle Pinner¹, Jan Chu¹, Martin Sauer³, Ajay Vora¹, Paul Veys¹, Kanchan Rao¹, Waseem Qasim¹

¹Great Ormond Street Hospital for Children NHS Trust, London, United Kingdom, ²IRCCS Istituto Giannina Gaslini, Genoa, Italy, ³Hannover Medical School, Hannover, Germany

Background: Genome-edited allogeneic chimeric antigen receptor (CAR) T cells offer “off-the-shelf” alternatives to autologous CAR T cells. Great Ormond Street Hospital (GOS) has investigated editing strategies to disrupt T cell receptor (TCRαβ) and CD52 expression in early stage human applications. Editing was designed to simultaneously prevent HLA mis-matched donor T cells from mediating graft versus host disease (GVHD) and allowed their persistence in the presence of Alemtuzumab. Initial compassionate experience with TALEN-edited CAR19 T cells in two B-ALL infants was followed by treatment of six children part of a phase 1 clinical trial of UCAR19 (NCT02808442) and a further seven children received CRISPR/Cas9 genome edited TT52CAR19 T cells (NCT04557436). Here we surveyed longer term outcomes of patients who had achieved remission after universal cell therapy and had been bridged to allogeneic haematopoietic stem cell transplantation (HSCT).

Methods: Outcomes of paediatric patients (aged 0.8-16 years) treated at GOS between 2015 and 2022 with either UCART19 (1.1-4.6 x 10⁶) or TT52CAR19 (0.8-2.0 x 10⁶) CAR19 T cells/kg cells were surveyed. Information collected included overall (OS) and disease-free survival (DFS), graft-versus-host-disease (GVHD), chimerism, detection of vector copy number and immune reconstitution. Our GOS augmented lymphodepletion (ALD) protocol comprising (total doses) fludarabine 150 mg/m2, cyclophosphamide 120 mg/kg, and alemtuzumab 1 mg/kg was used in 12/15 children.

Results: Overall, 11/15 patients (73%) achieved complete remission (CR/CRi) with undetectable disease on bone marrow flow cytometry 28 days after cell therapy, with 8/15 (53%) also confirmed as MRD negative by PCR (MRD < 10^-4). Allogenic HSCT was undertaken within the next four weeks, and the majority (64%) received reduced intensity conditionings (RIC) with ATG and low dose (2-4 Gy) TBI. Seven out of eleven transplants were second procedures and 6 of these were from the previous donor.

Sustained molecular remission was obtained in 7/8 (88%) of children who had been transplanted with PCR MRD < 10^-4. Although one subject died from post-transplant complications whilst in remission, 6 subjects have maintained ongoing remission, comprising 40% (6/15) long term survival. OS and DFS are shown in Figure 1. All survivors exhibited evidence of normal immune recovery (CD4 T cells > 300/mm³) within 12 months from SCT and any receiving intra-venous immunoglobulins (IVIG) stopped replacement. Non-persistence of allo-CAR T cells was confirmed in all patients after preparative conditioning and allo-SCT.

Figure 1. OS and DFS. OS, overall survival; DFS, disease-free survival

Conclusions: Longer term data for children treated at a single centre with genome-edited allogeneic CAR19 T cells indicate sustained remission in patients transplanted in molecular (PCR) MRD remission. Overall survival at our centre has been comparable to the wider autologous CAR19 setting, despite multiple lines of previous therapy, including CARs, Bi-specific T-cell engagers (BiTEs) and previous transplant. Notably, surviving subjects had all received augmented lymphodepletion ahead of CAR19 T cells, highlighting the importance of addressing host mediated barriers to mismatched cells for deep and rapid leukemic clearance. The success of subsequent RIC transplantation was consistent with consolidation of donor derived reconstitution rather than additional anti-leukemic chemotherapy effects.

Disclosure: Nothing to declare.

4: CAR-based Cellular Therapy – Clinical

Jaromir Tomasik ¹, Dominik Bilicki¹, Grzegorz Basak¹

¹Medical University of Warsaw, Warsaw, Poland

Background: Large B-cell lymphoma (LBCL) patients who progressed or relapsed after chimeric antigen receptor (CAR)-T therapy have a dismal prognosis with median overall survival (OS) estimated at below six months from the time of progressive disease. Various regimens are administered in a first line salvage therapy after post-CAR-T progression/relapse, however, no consensus regarding the most effective intervention has been established yet.

Methods: The aim of the study was to perform a meta-analysis of response rates to first line treatments after post-CAR-T progression/relapse (P/R) to provide pooled estimates and suggest the most efficient regimens. The population of interest were patients aged 18 or older, treated for LBCL, who experienced post-CAR-T P/R. Any treatment regimen administered as the first line after P/R was considered the intervention. Complete response (CR) rate and overall response rate (ORR) were study endpoints. The meta-analysis was performed using a random-effects model to obtain pooled CR and ORR with 95% confidence intervals (CI) for each intervention group separately. The study was conducted according to the PRISMA statement, the protocol was registered at the PROSPERO register (CRD42023473854).

Results: In the search process, we identified 39 studies meeting the inclusion criteria and providing sufficient data for the meta-analysis. Some studies included several intervention cohorts. The following interventions have been analyzed: bispecific antibodies (BiTEs) – administered in 13 studies, immune checkpoint inhibitors -12, lenalidomide-based regimens – 11, polatuzumab-based regimens – 10, radiotherapy – 9, Bruton’s Tyrosine Kinase inhibitors (BTKi) – 5, non-CD19 CAR-T – 5, CD19 CAR-T retreatment - 2 studies. Infusion of CAR-T cells other than anti-CD19 (e.g., anti-CD22, CD20) yielded the best response rates among all analyzed interventions, with pooled CR rate of 56% (95% CI: 40-71%) and pooled ORR of 80% (95% CI: 66-92%). Repeated, salvage administration of anti-CD19 CAR-T cells gave pooled CR of 30% (95% CI: 7-58%). Only two studies reported this type of intervention, with only one providing ORR. Administration of immune checkpoint inhibitors and BTKi regimens was associated with the lowest pooled CR rates of 11% (95% CI: 4-19%) and 8% (95% CI: 0-23%), respectively. Among other interventions, pooled CR rates were similar and varied from 19% in lenalidomide-treated groups to 28% in BiTEs-treated groups. Distributions of ORR estimates among the intervention groups were comparable with the distribution of CR. BiTEs presented the highest ORR of 48,5% (similar values were achieved in polatuzumab and radiotherapy groups), whereas checkpoint inhibitors and BTKi yielded 29% and 27% respectively.

Conclusions: The response rates vary depending on the intervention used as first line salvage treatment for LBCL patients after post-CAR-T progression/relapse. Infusion of CAR-T cells targeting molecules other than CD19 is the most efficient in achieving complete and overall responses. Regimens such as BiTEs also appear promising. On the contrary, BTK inhibitors and immune checkpoint inhibitors present disappointing performance. Considering the poor prognosis for the patients, the choice of the most efficient regimen is of the utmost importance.

Clinical Trial Registry: PROSPERO register (CRD42023473854)

Disclosure: Novartis and Gilead: honoraria for lectures and advisory boards.

4: CAR-based Cellular Therapy – Clinical

Lei Deng¹, Xiaolin Yu¹, Xiaocheng Song¹, Rui Guan¹, Yan Shao¹, Wenjun Li¹, Fang Zhou ¹

¹The 960th Hospital of The People’s Liberation Army Joint Logistics Support Force, Jinan, China

Background: Autologous chimeric antigen receptor T cells (CAR-T) have exhibited notable effectiveness in cases of relapsed/refractory B-cell acute lymphoblastic leukemia (ALL). Nevertheless, patients experiencing relapse subsequent to allogeneic hematopoietic stem cell transplantation (allo-HSCT) frequently encounter difficulties in acquiring autologous CAR-T products due to diminished lymphocyte counts and accelerated disease progression. To surmount these challenges, CAR-T cells can be sourced from the hematopoietic stem cell donor.

Methods: Peripheral blood mononuclear cells (PBMCs) were obtained from the patient’s hematopoietic stem cell donor for second generation CAR-T culture. As planned, the patient received pretreatment with the FC regimen (Fludarabine and cyclophosphamide) and then received transfusion of donor-derived CAR-T cells.

Results: A total of 14 patients relapsed after allo-HSCT from August 2019 to May 2023. The relapse cases included donor-derived CD19 CAR-T (10 cases), CD22 CAR-T (2 cases), and dual-target CD19-CD22 CAR-T (2 cases). The median age was 22 years (11-43 years), with 9 males and 5 females. Eight patients had bone marrow recurrence, four patients had extramedullary recurrence, and two patients had both bone marrow and extramedullary recurrence. The median follow-up time was 311 days (11-908 days), and the minimum follow-up time excluding death was 221 days. After treatment, 10 (71.4%) cases achieved remission, 2 (14.3%) cases had a partial response, and 2 (14.3%) cases had no response.

Cytokine release syndrome (CRS) occurred in all 14 patients, with grade 1 in 7 (50%) patients, grade 2 in 5 (35.7%) patients, grade 3 in 1 (7.1%) patient, and grade 4 in 1 (7.1%) patient. Of the 14 cases, 4 (28.6%) had severe GVHD (grade III-IV), which improved after treatment. In 13 patients with granulocyte deficiency, the median recovery time (>0.5×10⁹/L) was 10 days (2-107 days), and 1 patient did not recover until death. In 7 patients with severe thrombocytopenia (<20×10⁹/L), the median platelet recovery time was 40 days (4-334 days), and 2 patients did not recover until death. Of the 10 patients in remission, 2 relapsed at 236 days and 495 days after CAR-T treatment. One patient relapsed and died, and 1 achieved remission again after changing the CAR-T target. The 1-year survival rate of 14 patients was 50%. 4 of the 10 patients in remission died, 3 from severe infection and 1 from disease recurrence. Of the 2 patients with a partial response, 1 achieved remission after changing the CAR-T target and 1 died of disease progression. 2 unresponsive patients died from disease progression. (Table 1)

Conclusions: Donor-derived CAR-T cells are an effective method for relapsed acute B-lymphoblastic leukemia after HSCT, improving the remission rate and survival rate of patients. However, the main causes of death are relapse and severe infection after CAR-T therapy. This may be related to GVHD and poor recovery of blood cells after CAR-T therapy. Large-scale studies are needed to confirm.

Disclosure: Nothing to declare.

4: CAR-based Cellular Therapy – Clinical

Valentín Ortiz-Maldonado ¹, Nuria Martinez-Cibrian¹, Julio Delgado¹, Sergi Betriu¹, Leticia Alserawan¹, Ana Triguero¹, Nadia Verbruggen², Maike Spoon³, Marte C. Liefaard³, Anna D.D. van Muyden³, Natalia Tovar¹

¹Hospital Clínic de Barcelona, Barcelona, Spain, ²Galapagos NV, Mechelen, Belgium, ³Galapagos BV, Oegstgeest, Netherlands

Background: An innovative decentralized and automated point-of-care (PoC) manufacturing model was developed to administer fresh, autologous chimeric antigen receptor T-cell (CAR-T) treatments within 7 days of apheresis to patients with relapsed/refractory (R/R) chronic lymphocytic leukemia (CLL) or small lymphocytic lymphoma (SLL), including Richter transformation (RT).

Methods: Euplagia-1 (CTIS:2022-501686-47-00) is an ongoing Phase (Ph)1/2 study of PoC-manufactured anti-CD19/4-1BB CAR-T therapy (GLPG5201) administered as a fresh product following lymphodepleting chemotherapy. Patients with CD19 + R/R CLL/SLL with ≥2 prior therapy lines (including BTKi) are eligible; RT patients are eligible regardless of prior lines. Primary objectives are safety and establishment of a recommended Ph2 dose (RP2D; Ph1), and objective response rate (ORR; Ph2).

Results: As of 06 September 2023, all 15 patients planned for Ph1 were enrolled at dose level (DL)1 (35×10⁶ CAR + T cells, n = 6) or DL2 (100×10⁶ CAR + T cells, n = 9). All enrolled patients received a fresh product after successful PoC-manufacturing. Thirteen of 15 patients (80%) were infused with a median vein-to-vein time of 7 days (range 7–14). All patients were diagnosed with R/R CLL, 9/15 with concurrent RT. The median (range) number of prior treatment lines was 3 (2–10); 14/15 patients received prior BTKi and/or venetoclax. High-risk features were common in this population; at screening: 17p deletion, (n = 3, 23%), TP53 mutation (n = 6, 46%), complex karyotype (n = 3, 50%), unmutated immunoglobulin heavy-chain variable region gene (n = 13, 100%).

Most treatment emergent adverse events were Grade 1–2; most Grade ≥3 events were hematological (Table). Seven (47%) patients had Grade 1–2 cytokine release syndrome (CRS); no CRS Grade ≥3 and no ICANS were reported. One dose limiting toxicity occurred (Grade 4 neutropenia) which was manageable. No unexpected GLPG5201-related toxicities were observed; no deaths occurred while patients were on study.

In efficacy-evaluable patients (n = 14), 13 responded (objective response rate [ORR] 93%). Complete response (CR) confirmed by imaging was observed for eight patients (CR rate [CRR] 57%). ORR and CRR were 83% and 50% at DL1, and 100% and 63% at DL2, respectively. All but one patient with RT responded (ORR 89%); CR was achieved by 6/9 RT patients (CRR 67%). One patient with RT was refractory with CD19-negative disease. All RT patients treated at DL2 (n = 6) responded (ORR 100%, CRR 67%). DL2 was selected as the RP2D. At the time of analysis, 10/13 (77%) patients had ongoing responses. The median duration of response was not reached (median follow-up on study was 6 months, range 1–15). Three patients progressed after an initial response.

Conclusions: Data from 15 patients enrolled in Ph1 of Euplagia-1 show that PoC CAR-T manufacturing with a vein-to-vein time of only 7 days is feasible and effective. There were no unexpected safety signals, and no CRS Grade ≥3 or ICANS occurred. Efficacy data were encouraging, with a best ORR of 93% and CRR of 57%; all patients at RP2D responded (100% ORR). Follow-up is ongoing and Ph2 expansion cohorts will be opened.

Table. TEAEs occurring in ≥3 patients (worst grade per patient)*

Clinical Trial Registry: CTIS:2022-501686-47-00

Disclosure: Valentin Ortiz-Maldonado: Consultancy or Advisory fees: Celgene-BMS, Janssen, Kite, a Gilead company, Miltenyi and Novartis; Honoraria from Celgene-BMS, Janssen and Kite, a Gilead company

Nuria Martinez-Cibrian: Nothing to declare

Julio Delgado: Nothing to declare

Sergi Betriu: Nothing to declare

Leticia Alserawan: Nothing to declare

Ana Triguero: Nothing to declare

Nadia Verbruggen: Employee of, and shareholder in, Galapagos NV, Belgium

Maike Spoon: Employee of Galapagos BV, Netherlands; Shareholder in Galapagos NV, Belgium

Marte C. Liefaard: Employee of Galapagos BV, Netherlands

Anna D. D. van Muyden: Employee of Galapagos BV, Netherlands

Natalia Tovar: Nothing to declare

This study was funded by Galapagos NV (Mechelen, Belgium). Medical writing support was provided by Iain Haslam, PhD, CMPP (Aspire Scientific, Bollington, UK), and funded by Galapagos NV. Publication coordination was provided by John Gonzalez and Slávka Baróniková (Galapagos NV).

4: CAR-based Cellular Therapy – Clinical

Marie José Kersten¹, Kirsten Saevels², Sophie Servais³, Yves Beguin ³, Joost S. P. Vermaat⁴, Eva Santermans⁵, Stavros Milatos⁶, Maike Spoon⁷, Marte C. Liefaard⁷, Claire Vennin⁷, Margot J. Pont⁷, Anna D.D. van Muyden⁷, Maria T. Kuipers¹, Sébastien Anguille²

¹University of Amsterdam, Amsterdam, Netherlands, ²Antwerp University Hospital, Antwerp, Belgium, ³Centre Hospitalier Universitaire de Liège, Liège, Belgium, ⁴Leiden University Medical Center, Leiden, Netherlands, ⁵Galapagos NV, Mechelen, Belgium, ⁶Galapagos GmbH, Basel, Switzerland, ⁷Galapagos BV, Oegstgeest, Netherlands

Background: An innovative decentralized and automated point-of-care (PoC) manufacturing model was developed to administer fresh autologous chimeric antigen receptor T-cell (CAR-T) treatments within 7 days of apheresis. Here we provide an update on the Phase (Ph)1/2 Atalanta-1 trial of GLPG5101 in patients with relapsed/refractory (R/R) non-Hodgkin lymphoma (NHL).

Methods: Atalanta-1 (CTIS: 2022-502661-23-00) is an ongoing Phase (Ph)1/2, multicenter trial of PoC-manufactured anti-CD19 CAR-T therapy. GLPG5101 is administered as a fresh product in patients with R/R diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), marginal zone lymphoma (MZL), mantle cell lymphoma (MCL), Burkitt lymphoma or primary central nervous system lymphoma. Primary Ph1 objectives are safety and establishment of a recommended Ph2 dose. The primary Ph2 objective is efficacy (objective response rate [ORR]). Secondary objectives include PoC manufacturing feasibility, safety, GLPG5101 pharmacokinetics, and additional efficacy endpoints.

Results: As of 01 September 2023, 23 patients (R/R MCL [n = 5], DLBCL [n = 7], FL [n = 9], or MZL [n = 2]) were infused with a fresh product with a median vein-to-vein time of 7 days (range 7–8); one patient received a cryopreserved product with a vein-to-vein time of 13 days. In Ph1, three patients received DL1 instead of intended DL2 due to lower CAR + T-cell yield.

Twenty-three patients were included in the safety set. Most treatment-emergent adverse events were Grade 1–2; the majority of Grade ≥3 events were hematological (Table). In Ph1 and 2, respectively, seven (50%) and three (33%) patients experienced cytokine release syndrome (CRS); immune effector cell-associated neurotoxicity syndrome was seen in six (43%; all Grade 1) and one (11%; Grade 3) patient (s). In Ph1, Grade ≥3 infections occurred in two patients and dose-limiting toxicities were reported for four patients (DL1, n = 1; DL2, n = 3). The latter consisted of three cases of Grade 4 neutropenia, manageable with granulocyte colony-stimulating factor that did not limit dose escalation, and one Grade 5 intra-abdominal hemorrhage which occurred 12 days post-infusion (DL2), caused by Grade 4 disseminated intravascular coagulation in a patient with multiple comorbidities. One Grade 5 urosepsis was recorded >6 months post-infusion in a patient in ongoing complete response (CR). No Grade ≥3 infections occurred in Ph2.

In Ph1, among 14 efficacy-evaluable patients, 12 responded to treatment (ORR 86%) with 11 achieving a CR (CR rate [CRR] 79%). Four patients progressed after an initial response: two CD19-positive relapses, one CD19−negative relapse, one unconfirmed. ORR and CRR were 86% and 71% at DL1, and 86% and 86% at DL2, respectively. In Ph2, with 7 efficacy-evaluable patients, six responded to treatment (ORR 86%), including four CRs (CRR 57%). PoC manufacturing increased early T-cell phenotype in the final product compared to apheresis starting material, as assessed by exploratory flow cytometry. Lastly, GLPG5101 showed robust expansion in vivo across doses.

Conclusions: The ongoing Atalanta-1 study shows that PoC CAR-T manufacturing with a short vein-to-vein time (median 7 days) is feasible with administration of a fresh product. The safety profile of GLPG5101 is comparable to that of other CD19-targeted CAR-T therapies. CRRs were high in Ph1 and Ph2 in a heavily pretreated patient population.

Clinical Trial Registry: CTIS: 2022-502661-23-00

Disclosure: Marie José Kersten: Research/clinical trial funding from Galapagos and Kite, a Gilead company; honoraria from Adicet Bio, Beigene, BMS/Celgene, Galapagos, Kite, a Gilead company, Miltenyi Biotech, Novartis and Roche

Kirsten Saevels: Research/clinical trial funding from Galapagos

Sophie Servais: Nothing to declare

Yves Beguin: Research/clinical trial funding from Galapagos

Joost S. P. Vermaat: Nothing to declare

Eva Santermans: Employee of, and shareholder in, Galapagos

Stavros Milatos: Employee of Galapagos GmbH, Basel, Switzerland

Maike Spoon: Employee of Galapagos BV, Netherlands; Shareholder in Galapagos NV, Belgium

Marte C. Liefaard: Employee of Galapagos BV, Netherlands

Claire Vennin: Employee of Galapagos BV, Netherlands

Margot J. Pont: Employee of Galapagos BV, Netherlands

Anna D. D. van Muyden: Employee of Galapagos BV, Netherlands

Maria T. Kuipers: Research/clinical trial funding from Galapagos

Sébastien Anguille: Nothing to declare

This study was funded by Galapagos NV (Mechelen, Belgium). Medical writing support was provided by Iain Haslam, PhD, CMPP (Aspire Scientific, Bollington, UK), and funded by Galapagos NV. Publication coordination was provided by John Gonzalez and Slávka Baróniková (Galapagos NV).

4: CAR-based Cellular Therapy – Clinical

Benedetta Rambaldi ¹, Giulia Risca², Stefania Galimberti^2,3, Cristian Meli¹, Giuliana Rizzuto^1,4, Chiara Buracchi^5,3, Alex Moretti^5,6,4, Chiara Francesca Magnani⁷, Marianna Ponzo^3,5, Sarah Tettamanti⁵, Muriel Paganessi¹, Silvia Zaninelli⁸, Elisa Gotti⁸, Chiara Capelli¹, Benedetta Cabiati⁹, Michele Francesco Quaroni⁹, Martino Introna¹, Federico Lussana^1,10, Giuseppe Gritti¹, Silvia Ferrari¹, Anna Grassi¹, Sara Napolitano⁶, Adriana Balduzzi^6,11, Maria Grazia Valsecchi^3,2, Giuseppe Dastoli⁵, Alessandro Rambaldi^1,10, Andrea Biondi^6,11, Giuseppe Gaipa^6,5,9

¹Azienda Socio Sanitaria Territoriale Papa Giovanni XXIII, Bergamo, Italy, ²Fondazione IRCCS San Gerardo dei Tintori Biostatistics and Clinical Epidemiology, Monza, Italy, ³School of Medicine and Surgery, University of Milano-Bicocca, Monza, Italy, ⁴Ph.D. Program in Translational and Molecular Medicine (DIMET), University of Milano-Bicocca, Monza, Italy, ⁵Fondazione IRCCS San Gerardo dei Tintori Tettamanti Center, Monza, Italy, ⁶Fondazione IRCCS San Gerardo dei Tintori Pediatrics, Monza, Italy, ⁷University Hospital Zurich and University of Zurich, Zurich, Switzerland, ⁸SS Centro di Terapia Cellulare “G. Lanzani”, Bergamo, Italy, ⁹Fondazione IRCCS San Gerardo dei Tintori Laboratorio di Terapia Cellulare e Genica Stefano Verri, Monza, Italy, ¹⁰University of Milan, Milan, Italy, ¹¹University of Milano-Bicocca School of Medicine and Surgery, Monza, Italy

Background: Biomarkers associated to treatment response after Chimeric Antigen Receptor (CAR) T cell therapy are needed to understand which patients are likely to benefit from consolidative therapies. We conducted trials in patients with B-cell acute lymphoblastic leukemia (BCP-ALL) relapsed after allogeneic hematopoietic cell transplantation (HCT), receiving donor-derived, CD19-targeted CAR cytokine-induced killer (CARCIK-CD19) cells engineered with the non-viral Sleeping Beauty transposon vector.¹ (Magnani et al. JCI, 2020). The aims of this study were to evaluate the impact of proliferation, differentiation, and expansion of infused CARCIK-CD19 cells on patient’s clinical outcome.

Methods: Data were prospectively collected from consecutive patients enrolled in the FT01CARCIK, phase I/IIb study (NCT03389035) (N = 15), in the compassionate use study (PUC-CARCIK-CD19) (N = 6) and in the FT03CARCIK phase II study (NCT05252403) (N = 12). CARCIK-CD19 cells were counted by flow cytometry (FC) in the peripheral blood (PB) at predefined time points until month 12. We assessed the distribution of T cell maturation subsets in infused CARCIK-CD19 cells, identified as: naïve (CD45RA + CD62L + ), central memory (CM, CD45RA-CD62L + ), effector memory (EM, CD45RA- CD62L-) T cells, and terminal differentiated T cells (TEMRA, CD45RA + CD62L-). We also measured the expression of PD1 as exhaustion marker. CR was defined as absence of leukemia in the BM at day 28. CARCIK-CD19 peak was defined as the maximum amount of CAR⁺ cells/mL measured in PB.

Results: The analysis included 33 patients, 4 of whom were pediatric patients. Circulating CARCIK-CD19 cells peaked in median at day 10 (range, 7-150 days). Median CARCIK-CD19 cell peak was 44.16/μL (range, 1.35-2282.31). In some cases, CARCIK-CD19 cells were still detectable at the last follow up at month 12 with a median of 1.26/μL (range 0-5.86). The majority of CARCIK-CD19 cells in the product bag were CD8⁺ T cells (median 83.86%, range 24.34-93.59) and the same trend persisted in vivo after the CARCIK-CD19 infusion. Responder patients (N = 28) showed lower CD4/CD8 CARCIK-CD19 mean values in time, especially at day 28 after infusion (p = 0.0004, contrast from a longitudinal mixed model) compared to non-responder patients (N = 5), suggesting that an enrichment in CD8⁺ CARCIK-CD19 cells correlates with treatment response at day 28.

In the product bag, the majority of CARCIK-CD19 cells were EM (median 48.5%, range, 11.44-79.02), followed by TEMRA (median 32.34%, range 4.1-79.42). At day 10 of expansion, the majority of CARCIK-CD19 cells were EM (median 46.75%, range 6.6-83.75) and CM (median 40.51%, range 6.3-93.23), followed by TEMRA (median 1.31%, range 0-29.3), and naïve cells (median 4.93%, range 0.16-18.28). The median level of PD1 expression on CARCIK-CD19 cells in the product bag was 7.1%, range 0.55-33.5, and increased soon after infusion (median 76.91%, range 33.17-98.78 at day 10).

Conclusions: CARCIK-CD19 cells efficiently expanded in vivo, reaching the peak at day 10, and persisting in some cases up to 12 months after infusions. The majority of CARCIK-CD19 cells detected in patients were CD8⁺ with a memory phenotype confirming the capability of these cells to persist long-term. A lower CD4/CD8 CARCIK-CD19 ratio at day 28 was associated with early treatment response.

Clinical Trial Registry: Patients included in this study were enrolled in the FT01CARCIK, phase I/IIb study (NCT03389035) (N = 15), in the compassionate use study (PUC-CARCIK-CD19) (N = 6) and in the FT03CARCIK phase II study (NCT05252403)

Disclosure: Nothing to declare:

Benedetta Rambaldi, Giulia Risca, Stefania Galimberti,

Cristian Meli, Giuliana Rizzuto, Chiara Buracchi, Alex Moretti, Chiara Francesca Magnani,

Marianna Ponzo, Muriel Paganessi, Silvia Zaninelli, Elisa Gotti, Chiara Capelli, Benedetta Cabiati, Michele Francesco Quaroni, Martino Introna, Silvia Ferrari, Anna Grassi, Sara Napolitano, Maria Grazia Valsecchi, Giuseppe Dastoli, Giuseppe Gaipa

Sarah Tettamanti, CoImmune (Research Funding)

Federico Lussana Yes: Amgen (Speakers Bureau), Pfizer (Membership on an entity’s Board of Directors or advisory committees and Speakers Bureau), Incyte (Speakers Bureau), Bristol Myers Squibb (Membership on an entity’s Board of Directors or advisory committees and Speakers Bureau), AbbVie (Membership on an entity’s Board of Directors or advisory committees), Clinigen (Membership on an entity’s Board of Directors or advisory committees)

Giuseppe Gritti:

Consultancy: Takeda

Membership on an entity’s Board of Directors or advisory committees: F. Hoffmann-La Roche Ltd, Takeda, Kite-Gilead, Ideogen, Genmab, Italfarmaco

Advisory Board: Roche, Takeda, Kite-Gilead, Italfarmaco, Ideogen, Genmab

Training activity: Takeda, Clinigen, Ideogen, Beigene, Incyte, Novartis

Support for attending meetings: Roche, Sandoz, Beigene, Janssen

Adriana Balduzzi: Novartis, Amgen, Medac, Neovii (Speakers Bureau)

Alessandro Rambaldi

Honoraria: ABBVIE, Celgene-BMS, Janssen, Roche, Incyte, Novartis, Kite-Gilead, Jazz, Astellas, Pfizer, Agmen, Omeros

Andrea Biondi: Colmmune (Membership on an entity’s Board of Directors or advisory committees and Research Funding), Galapagos (Membership on an entity’s Board of Directors or advisory committees), Agmen (Speakers Bureau), Novartis (Speakers Bureau), BMS (Membership on an entity’s Board of Directors or advisory committees)

4: CAR-based Cellular Therapy – Clinical

Lu-Gui Qiu ¹, Chunrui Li², Di Wang², Yongping Song³, He Huang⁴, Jianyong Li⁵, Bing Chen⁶, Jing Liu⁷, Xi Zhang⁸, Yujun Dong⁹, Kai Hu¹⁰, Peng Liu¹¹, Jianqing Mi¹², Kaiyang Ding¹³, Zhenyu Li¹⁴, Qie Dong¹⁵, Peng Guo¹⁵, Fuyuan Zhang¹⁵, Hong-yu Gui¹⁵, Guang Hu¹⁵, Wen Wang¹⁵

¹State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin, China, ²Institute of Hematology; Tongji Hospital,Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, ³The Affiliated Cancer Hospital of Zhengzhou University, Henan Cancer Hospital, Zhengzhou, China, ⁴The First Affiliated Hospital, Zhejiang University, Hangzhou, China, ⁵The First Affiliated Hospital of Nanjing Medical University, Nanjing, China, ⁶Nanjing Drum Tower Hospital,Affiliated Hospital of Nanjing University Medical School, Nanjing, China, ⁷Third Xiangya Hospital of Central South University, Changsha, China, ⁸Medical Center of Hematology, Xinqiao Hospital, State Key Laboratory of Trauma, Burn and Combined Injury, Army Medical University, Chongqing, China, ⁹Peking University First Hospital, Beijing, China, ¹⁰Beijing Boren Hospital, Beijing, China, ¹¹Zhongshan Hospital, Fudan University, Shanghai, China, ¹²State Key Laboratory of Medical Genomics, National Research Center for Translational Medicine at Shanghai, Shanghai Institute of Hematology, Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China, ¹³The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, China, ¹⁴Affiliated Hospital of Xuzhou Medical University, Xuzhou, China, ¹⁵Nanjing IASO Biotechnology Co., Ltd., Nanjing, China

Background: B-cell maturation antigen (BCMA) is a promising therapeutic target for multiple myeloma (MM) and soluble BCMA (sBCMA) is derived from the ectodomain shedding of the BCMA. The serum sBCMA level predicts patients’ survival and can be used to monitor the course of the disease. On other hand, it was reported that sBCMA fragment was capable of inhibiting CAR T-cell function. Sufficient sBCMA can accumulate in the bone marrow of MM patients to inhibit CAR T-cell recognition of tumor cells, and potentially limit efficacy of BCMA-directed T-cell therapy. So during the design and development of equecabtagene autoleucel (eque-cel, CT103A), the competitive inhibition of sBCMA to cell-surface BCMA binding was minimized via selection of proper scFv. The pivotal phase 2 FUMANBA-1 study (NCT05066646) demonstrated deep and durable responses with eque-cel in heavily pretreated patients with RRMM. However, the correlation between baseline sBCMA plasma level and clinical outcome after CAR-T therapy is still elusive. This study aimed to investigate the potential impact of baseline serum sBCMA levels in MM after eque-cel infusion.

Methods: The level of sBCMA was analyzed by enzyme-linked immunosorbent assay (ELISA). CAR transgene copies in the patient PBMCs were monitored by digital droplet polymerase chain reaction (ddPCR). In this subanalysis, univariate and multivariate logistic regression models were used to assess baseline serum sBCMA level and other characteristics that correlate with the likelihood of achieving CR/sCR.

Results: The median baseline sBCMA level in serum of 89 RRMM patients is 225.1 ng/mL. Baseline sBCMA level was not correlated with age, gender, except patients with high tumor burden (defined as ≥50% bone marrow plasma cells) (median 605.9 ng/mL in high tumor burden, median 206.2 ng/mL in low tumor burden; P = 0.0033) and R-ISS III (median 880.7.9 ng/ml in R-ISS III, median 213.2 ng/mL in R-ISS I-II; P = 0.012). The median peak copy number (Cmax) of eque-cel was comparable between patients with high baseline serum sBCMA level (≥225.1 ng/mL) (104237.00 copies /μg DNA) and those with low baseline serum sBCMA level (<225.1 ng/mL) (86591.75copies /μg DNA). Descriptive analysis demonstrated comparable baseline sBCMA in patients with CR/sCR (median 228.06 ng/mL) versus non-CR/sCR (median 185.13 ng/mL). The ORRs and ≥CR rates were 100% and 80% patients with high baseline serum sBCMA level vs. 97.7% and 77.3% patients with low baseline serum sBCMA level. Univariate analysis of CR/sCR by baseline characteristics including gender, age, extramedullary disease, high risk cytogenetics, tumor burden, R-ISS disease stage, received bridging therapy, baseline sBCMA, showed that none of them were significantly associated with CR/sCR. And all the subgroups showed a sharp sBCMA decrease with the response to CT103. sBCMA decreased to LLOQ (lower limit of quantification) at a median of 2 months post-infusion and remained below LLOQ at month 12 in patients remaining in response.

Conclusions: These results demonstrate that eque-cel represents a promising treatment option for RRMM patients, including those with higher baseline serum sBCMA level and higher tumor burden. And the relatively long persistence of eque-cel is another demonstration that eque-cel is not inhibited or exhausted by sBCMA.

Disclosure: Nothing to declare. Professor Lu-Gui Qiu is also the corresponding author of this study.

3: CAR-based Cellular Therapy – Preclinical

Mi Shao ¹, Xiaohui Si¹, Yue Huang¹, Xinyi Teng¹, Pengxu Qian¹, He Huang¹

¹Zhejiang University, Hangzhou, China

Background: In recent years, CAR-T therapy has achieved remarkable success in the treatment of R/R B-cell malignancies. Despite its success, CAR-T therapy faces several obstacles that limit its long-term efficacy and application to broader cancer types. In addition to antigen escape, poor tumor infiltration, and treatment-related adverse effects, CAR-T cell exhaustion and short-term persistence pose major challenges. The in vivo environment employs complicated immunosuppressive mechanisms to suppress the antitumor function of endogenous or adoptively transferred T cells. Metabolic intervention thus represents a promising approach for enhancing CAR-T cell function and improving clinical performance in difficult-to-treat cancers. Mitochondria play crucial roles in the adaptive immune system, mediating the development, metabolism, and activation of T cells, as reported before. In this study, we will investigate the relationship between mitochondrial dynamics and CAR-T cell exhaustion by analyzing the characteristics and transcriptomic changes in an in vitro model of exhausted CAR-T cells.

Methods: CAR-T cells were manufactured and expanded in vitro, followed by tumor antigen exposure. Expression of inhibitory receptors, differentiation markers, and activation markers was detected by flow cytometry. Whole genomic RNA sequencing (RNA-seq) was applied to detect gene expression levels in each group. The mitochondrial morphology of CAR-T cells was observed using cryoelectronic microscopy. The cellular oxygen consumption rate was detected using a cellular energy metabolism analyzer. Mitochondrial profiles, including MitoTracker, TMRM, MPTP, MitoSox, ROS and ATP, were detected by live cell staining and colorimetric methods.

Results: Tumor antigen exposure drove CAR-T cells into an exhaustion state, as evidenced by decreased proliferative capacity, increased apoptosis, decreased cytokine production, increased expression levels of multiple inhibitory receptors, increased mRNA expression levels of exhaustion-related genes, and increased terminally differentiated subsets, along with decreased activation levels. Analysis of RNA-seq revealed the downregulation of mitochondria-related pathways in the exhausted group. The mitochondria in exhausted CAR-T cells showed abnormal morphology, specifically characterized by swollen and enlarged mitochondria with few and short cristae. Exhausted CAR-T cells showed decreased oxidative phosphorylation, decreased mitochondrial mass, decreased membrane potential, increased membrane permeability, increased MitoSox and ROS levels, decreased intracellular ATP content, and decreased mRNA expression of mitochondria-related genes.

Conclusions: Multiple rounds of tumor cell exposure in vitro successfully establish a tumor antigen-induced exhaustion CAR-T cell model. Exhausted CAR-T cells display abnormal mitochondrial morphology and function.

Disclosure: Nothing to declare

5: Cellular Therapies other than CARs

Andrew McIntyre ^1,2, Maria Cuadrado², Joshua Hughes¹, Aideen O’Neill^1,2, William Wilson³, Darren Edwards³, Ka Man Mak³, Jessica Hunt³, Laura Clifton-Hadley³, Adrian Bloor⁴, Andrew Clark⁵, Anne Parker⁵, Maria Gilleece⁶, Richard Lovell⁷, Alexander M. Martin⁸, Ann Hunter⁸, Deborah Richardson⁹, John Snowden¹⁰, Stephen Robinson¹¹, Ben Uttenthal¹², Charles Crawley¹², Mark Lowdell¹³, Karl Peggs², Persephone Borrow¹⁴, Ronjon Chakraverty^1,2

¹Weatherall Institute of Molecular Medicine, University of Oxford, Oxford, United Kingdom, ²UCL Cancer Institute, London, United Kingdom, ³CRUK and UCL Cancer Trials Centre, London, United Kingdom, ⁴Christie Hospital, Manchester, United Kingdom, ⁵Beatson West of Scotland Transplant Centre, Glasgow, United Kingdom, ⁶St. James’s University Hospital, Leeds, United Kingdom, ⁷University Hospitals Birmingham, Birmingham, United Kingdom, ⁸Leicester Royal Infirmary, Leicester, United Kingdom, ⁹University Hospitals Southampton, Southampton, United Kingdom, ¹⁰Royal Hallamshire Hospital, Sheffield, United Kingdom, ¹¹Bristol Royal Hospital for Children, Bristol, United Kingdom, ¹²Addenbrooke’s Hospital, Cambridge, United Kingdom, ¹³Centre for Cell, Gene and Tissue Therapeutics, Royal Free Hospital, London, United Kingdom, ¹⁴Centre for Immuno-Oncology, University of Oxford, Oxford, United Kingdom

Background: Relative to donor CD8⁺ T cells, the recovery of CD4⁺ T cell populations is severely delayed following allo-SCT. We reasoned CD8⁺ T cell differentiation would therefore take place under ‘helpless’ conditions and impair anti-tumour immunity. To test whether accelerating CD4⁺ T cell recovery would improve CD8⁺ T cell fitness and improve anti-tumour immunity, we conducted a randomized phase II, multicentre trial to evaluate the efficacy of prophylactic transfer of CD4⁺ T lymphocytes after alemtuzumab-based, reduced intensity HLA-matched related donor transplant for haematological cancers (PROT4; NCT01240525). The randomised intervention permitted unbiased measurement of the effects of CD4⁺ T cell infusion upon T cell population architecture and function, which we report here.

Methods: 106 patients were registered for the trial and tapered cyclosporine from day 40-day 70. 54 patients without GVHD, disease progression and normal graft function were randomised at a 2:1 ratio to receive 1 x 10⁶ CD4⁺ T cells/kg between day 100-120. Blood samples were taken at day 90 (pre-infusion) and at 12-week intervals over the next 12 months. Immune population architecture and function were analysed using a combination of high-parameter flow cytometry, effector molecule expression, bulk/ single cell (sc) RNAseq and TCRseq.

Results: Compared to controls, CD4⁺ T cell infusion significantly delayed CD8⁺ T cell reconstitution both in terms of frequency and absolute numbers. Although per-cell cytokine generation was similar upon ex vivo stimulation, gene expression profiling showed that CD4⁺ T cell infusion induced signatures of CD8⁺ T cellular ‘fitness’ (increased DNA replication, TCR activation and ATP generation). These qualitative changes following CD4⁺ T cell infusion were matched by a dramatic shift in population architecture towards less differentiated CD8⁺ T cells as evaluated by unbiased clustering of flow cytometry and scRNAseq data. By pairing trajectory analysis and TCRseq, we found that CD4⁺ T cell addback diverted CD8⁺ T cell clonal expansion and differentiation away from terminal effector-like cellular states. In parallel, CD4⁺ T cell addback remodelled the CD4⁺ T cell compartment by enriching for circulating PD-1⁺CXCR5⁺ T follicular helper (Tfh)-like cells with features of central memory cells, but antagonized clonal expansion of cytotoxic CD4⁺ T cells. To test whether Tfh cells could provide direct help to CD8⁺ T cells, we cultured human volunteer CD8⁺ T cells in vitro under conditions of repetitive anti-CD3/CD28 stimulation with or without addition of Tfh cells. Whereas helpless CD8⁺ T cells underwent terminal differentiation, co-incubation with Tfh cells promoted early differentiation-like cellular states and proliferative fitness, a process that required the prototypic Tfh cytokine, IL-21.

Conclusions: CD4⁺ T cell infusion after allo-SCT in human patients shifts CD8⁺ T cell clonal expansion and differentiation trajectory to less differentiated cellular states. Our finding that human Tfh cells use IL-21 to directly promote fitness in human CD8⁺ T cells provides a potential mechanism involved and is consistent with recent data from pre-clinical models demonstrating Tfh-mediated help for anti-tumour and anti-pathogen CD8⁺ T cells. These data provide further impetus to explore how the properties of CD4⁺ T cells can be best exploited in immunotherapeutic targeting of cancer.

Clinical Trial Registry: NCT01240525

Disclosure: Nothing to declare

5: Cellular Therapies other than CARs

Francesca Farina ¹, Bernhard Gentner^2,3, Gaetano Finocchiaro⁴, Marica Eoli⁵, Alessia Capotondo³, Elena Anghileri⁵, Matteo Barcella³, Valentina Brambilla⁶, Matteo Carrabba¹, Valeria Cuccarini⁷, Quintino Giorgio D’Alessandris⁸, Francesco DiMeco⁹, Valeria Ferla¹, Paolo Ferroli⁹, Filippo Gagliardi¹⁰, Federico Legnani⁹, Stefania Mazzoleni⁶, Alessandro Olivi⁸, Roberto Pallini⁸, Marco Saini⁹, Luigi Naldini³, Carlo Russo¹¹, Fabio Ciceri¹

¹Hematology and Bone Marrow Transplant Unit - San Raffaele Hospital – Milano, Milano, Italy, ²Centre Hospitalier Universitaire Vaudois and Université de Lausanne, Lausanne, Switzerland, ³San Raffaele Telethon Institute for Gene Therapy (SR-Tiget), Milano, Italy, ⁴Neuro-Oncology Unit - San Raffaele Hospital, Milano, Italy, ⁵Neuro-Oncology Unit - Istituto Neurologico Carlo Besta, Milano, Italy, ⁶Genenta Science Spa, Milano, Italy, ⁷Neuroradiology Unit - Istituto Neurologico Carlo Besta, Milano, Italy, ⁸Neurosurgery Unit - Policlinico Gemelli, Roma, Italy, ⁹Neurosurgery Unit - Istituto Neurologico Carlo Besta, Milano, Italy, ¹⁰Neurosurgery Unit - San Raffaele Hospital, Milano, Italy, ¹¹Genenta Science Spa, New York, United States

Background: Glioblastoma (GBM) multiforme is an incurable glial tumor affecting the central nervous system. The reported median survival in the unfavorable patients’ subgroup with unmethylated MGMT promoter is 12.7 months with less than the 15% of patient surviving up to 2-years. Despite immunotherapies being able to slow or eradicate numerous tumors, even those metastasized to the brain, none so far have extended survival in GBM.

Methods: We are conducting a Phase 1/2a dose-escalating non-randomized open label study in newly diagnosed GBM multiforme with unmethylated MGMT gene promoter, involving an i.v. injection of autologous CD34+ HSPCs genetically modified with a lentiviral vector encoding for human interferon-a2 (Temferon). Temferon has been designed to deliver IFN-α2 within the TME by Tie-2 expressing macrophages (TEMs). The study aims to evaluate the short-term (up to 90 days) and long-term (up to 2 years) tolerability and safety of five escalating doses of Temferon in up to 27 GBM patients. Autologous CD34 + HSPC are mobilized with lenograstim and plerixafor, collected by apheresis, purified and ex vivo modified with a lentiviral vector. Following a sub-myeloablative conditioning regimen (Thiotepa + BCNU or Thiotepa + Busulfan or Busulfan alone), up to 3 million Temferon cells/kg are administered following a fixed dose of non-manipulated CD34+ supporter cells

Results: As of 27th July 2023, 21 GBM patients in 7 cohorts received incremental doses of Temferon up to 3 million cells/kg and 5 patients were still alive with a median OS of 17 months and a median PFS of 8.3 months following initial surgery. To date, no DLTs have been identified. In all patients, rapid engraftment of modified progenitors and fast hematological recovery have been observed within two weeks from the transplantation. Grade 4 or 5 serious adverse events (n = 7 patients), including graft failure, febrile neutropenia, pneumonia, steroid restart, and disease progression, were mostly attributed to the conditioning chemotherapy, but steroid usage or disease progression were also listed. 1 SUSAR (persistent GGT elevation) occurred. The higher transduced cells proportion in the BM was observed 30 days after Temefron administration with a chimerism for the highest dose tested up to 40%, still detectable in the long-term. Very low concentrations of IFNa were detected in the plasma, indicating a tight regulation of vector expression. Seven patients underwent second surgery and genetically engineered cells (6 out of 7 patients assessed) were detected within the bulk tumor lesion (>3% in 3 patients). Moreover, IFN-responsive genes were up-regulated, suggesting a local IFNα release by TEMs. In one patient, a stable lesion had a higher proportion of T cells & TEMs, an increased IFN-response signature and myeloid re-programming revealed by scRNAseq, as compared to a synchronous, progressing tumor. TCR sequencing of blood and tumor samples showed a post-treatment increase of peripherally expanded clones frequency in the tumor samples, as evidence of a crosstalk between peripheral and tumor repertoires of T cells.

Conclusions: These data corroborate the initial evidence on safety and tolerability of Temferon. They also suggest that Temferon has potential to counteract disease progression in patients affected by unmethylated MGMT GBM.

Clinical Trial Registry: NCT03866109

Disclosure: Nothing to declare

5: Cellular Therapies other than CARs

Ioannis Kyriakou ¹, Zoi Boussiou¹, Georgios Karavalakis¹, Chrysa Pantazi¹, Ioanna Vallianou¹, Maria Giannaki¹, Fani Stavridou¹, Maria Liga¹, Ioulia Mavrikou¹, Kyriakos Koukoulias¹, Apostolia Papalexandri¹, Fotini Kika¹, Eleftheria Sagiadinou², Ioannis Batsis¹, Alexandros Spyridonidis², Ioanna Sakellari¹, Anastasia Papadopoulou¹, Evangelia Yannaki¹

¹George Papanikolaou Hospital, Thessaloniki, Greece, ²University of Patras, Patra, Greece

Background: Increasing use of alternative-donor hematopoietic cell transplantation (HCT) has given prominence to morbidity and mortality that stems from opportunistic infections in this setting. Conventional drug treatment is often suboptimal or/and associated with toxicity and resistance growth, whereas adoptive transfer of pathogen-specific T-cells (pSTs) has emerged as an alternative therapeutic option. We present combined results of two phase I/II studies, evaluating the safety and efficacy of multivalent-specific T-cells targeting CMV, EBV and BKV (trivalent-STs), or in addition, adenovirus-AdV and Aspergillus fumigatus-AF (pentavalent-STs).

Methods: GMP-grade, multivalent-STs were generated by pulsing blood mononuclear cells(PBMCs) from the graft donors with peptides spanning immunogenic antigens of EBV:LMP2/EBNA1/BZLF1, AdV:Hexon/Penton, CMV:pp65/IE1, BKV:Large-T/VP1 and AF:Gel1/SHMT/Crf1 and expanding them for 10days in culture. Specificity was assessed by Elispot in the products and patient PBMCs post-infusion. pSTs (2x10⁷ cells/m²) were administered as single infusion when infection from one or more of the targeted pathogens occurred. Patients with GvHD at infusion, requiring >0.5mg/kg methylprednisolone, were excluded. Participants were followed for up to 6 months post-infusion.

Results: So far, 29 patients post haplo- (n = 27) or volunteer unrelated-donor (n = 2) HCT, received pSTs for CMV, EBV and BKV infections. pST products reached a median 7x(2-31) expansion over baseline and mean 222±30x10⁶cells in total. The products were enriched in CD3⁺cells (93.2 ± 1%) and comprised of both CD4⁺ (50.9 ± 3%) and CD8⁺cells (41.7 ± 3%) while also expressed central- (40.4 ± 4%) and effector-memory markers (43.5 ± 4%). Twenty, six and three patients received pSTs for single (CMV = 8/EBV = 6/BKV = 6), double (BKV + EBV = 3/BKV + CMV = 2/EBV + CMV = 1) or triple infection (CMV + EBV + BKV = 3), at median 70days(32-1289) post-transplant. pST administration was safe, with no infusion-related toxicity or GVHD occurrence/exacerbation. From a total 41 infections at infusion, 21 were viral diseases (BKV cystitis/nephropathy n = 14, CMV gastritis/colitis/pneumonia n = 4, EBV colitis/encephalitis/PTLD n = 3) and 20 viremias/virurias. We observed 35 (85.4%) complete responses (CMV = 13, EBV = 11, BKV = 11), 5 (12.2%) partial responses (BKV = 2, EBV = 2, CMV = 1) and 1 (2.4%) no response (CMV = 1). The latter was not unexpected as the non-responding to CMV patient with a triple infection at infusion, had received a dual specificity (EBV&BKV) product from a CMV-seronegative donor. Clinical and viral load responses occurred at a median 3(1-14) weeks post-infusion and corresponded to circulating pST expansion. Importantly, 9/41 and 7/41 infections (including one EBV-PTLD) resolved without or with short-courses of anti-viral medication respectively, and 3BKV/41 infections, previously refractory to full-course cidofovir treatment, were resolved post-infusion. During follow-up, 17/29 patients experienced 22 new viral reactivations, 11 of which resolved without or with only short antiviral treatment, due to efficient in vivo expansion of specific-T-cells. Importantly, the rest 11 reactivations which required drug treatment, occurred in 9 patients, the majority of whom (7/9), either had relapsed or were under increased immunosuppression at reactivation and 5/9 underwent a 2^nd infusion. 9 of these 11 reactivations, including one EBV-PTLD, were fully resolved, 1/11 responded partially and one CMV reactivation was refractory due to continuing immunosuppression.

Conclusions: We show that pSTs can serve as potent, long-lasting and safe therapeutic alternative for opportunistic infections after allo-HCT, minimizing or even abolishing the need for drug therapy via efficient self-control of current and future reactivations.

Clinical Trial Registry: EudraCT2020-004725-23, NCT05471661

Disclosure: none

5: Cellular Therapies other than CARs

Dominik Heim¹, Christoph Bucher¹, Michael Medinger¹, Udo Holtick ², Gérard Socié³, Christos Stylianou⁴, Simone Dertschnig⁵, Malika Souquieres⁵, Stephan Oehen⁵, Elisabeth Kueenburg⁵, Jakob Passweg¹

¹University Hospital Basel, Basel, Switzerland, ²University Hospital of Cologne, Cologne, Germany, ³APHP Hospital Saint Louis and University of Paris, Paris, France, ⁴ClinBAY, Limassol, Cyprus, ⁵Priothera, St. Louis, France

Background: In allogeneic stem cell transplantation (allo-HCT), relapse is one of the major causes of mortality. There is an unmet need to mitigate this risk and thus increase the overall survival (OS) of patients (pts) undergoing allo-HCT. Mocravimod (MOC), a sphingosine-1-phosphate receptor modulator, has demonstrated a dual effect, enhanced graft-versus-leukemia (GvL) and reduced graft-versus-host disease (GvHD), by sequestering T cells in lymphoid organs. Study CKRP203A2105 (NCT01830010) was a Phase I/II trial investigating MOC in pts with hematologic diseases undergoing allo-HCT.

Methods: In this multicenter, open label trial 23 patients comprising different hematologic diseases were included across three treatment arms to receive a daily oral dose of either 3mg MOC+cyclosporine A (CsA), 1mg MOC+CsA or 3mg MOC+tacrolimus (Tac). Pts were treated for 111 days or until relapse or unacceptable toxicity and then entered an OS-follow-up (FU) period of up to 1 or 2 years. We performed a post-hoc analysis of OS. Survival estimates were compared to matched pts from a real-world cohort, the CIBMTR data base (n = 5191). In addition, OS data of a subgroup of pts receiving allo-HCT for treatment of AML (n = 7) were compared to a matched cohort of CIBMTR (n = 303). Bayesian comparison of OS of MOC-treated pts to controls from the CIBMTR database was performed.

Results: Pts characteristics were similar in both groups with a mean age of 48.9 vs 51 years in the MOC patients and CIBMTR controls, respectively. Hematologic diseases were equally distributed with the majority being treated for either AML or ALL: 30% and 39 % AML, 17% and 21% ALL of MOC pts and CIBMTR controls, respectively. Most pts received stem cells from matched donors (78% vs 85%). Survival estimates of all pts demonstrated OS rates of 83% in MOC treated pts vs 58% in the control group at 1a post allo-HCT. A prolonged OS could be shown at 2a FU time after allo-HCT with 69% in the MOC treated pts and 45% in the control group (Fig.1). The median age was 51 years for both, MOC-treated AML and CIBMTR AML pts. While all CIBMTR AML pts were in CR, only 57.1% of MOC AML pts were in CR at time of transplant. Survival estimates for AML pts demonstrated OS rates of 91% vs 73% and 84% vs 66% for MOC pts vs CIBMTR pts at 1- and 2-years post-transplant, respectively. The probability that MOC treated AML patients have higher survival rates compared to controls was 99% and 94% at 1- and 2-years post-transplant, respectively. Primary disease was main cause of death in both groups.

Conclusions: This post hoc analysis demonstrated an OS benefit for pts with hematologic diseases receiving MOC as add-on treatment to allo-HCT in comparison to a matched real-world cohort undergoing allo-HCT without MOC. These data support the development of MOC as an adjunctive and maintenance treatment to allo-HCT, which is currently evaluated in the Phase III MO-TRANS trial, a worldwide double-blind trial investigating the GvL effect of MOC translating into prolonged relapse-free-survival and OS in AML pts undergoing allo-HCT (NCT05429632).

Clinical Trial Registry: ClinicalTrials.gov: NCT01830010

Disclosure: Simone Dertschnig, Malika Souquieres, Stephan Oehen and Elisabeth Kueenburg are employees of Priothera SAS

29: Chronic Leukaemia and Other Myeloproliferative Disorders

Benjamin McCormick ¹, Anthony Findley¹, James Foran¹, Mohamed Kharfan-Dabaja¹, Madiha Iqbal¹, Lisa Sproat², Jeanne Palmer², William Hogan³, Kebede Begna³, Mark Litzow³, Talha Badar¹

¹Mayo Clinic, Jacksonville, United States, ²Mayo Clinic, Phoenix, United States, ³Mayo Clinic, Rochester, United States

Background: Despite significant improvements in outcomes of chronic myeloid leukemia (CML-CP) following tyrosine kinase inhibitor (TKI) implementation, CML in chronic/advanced phases (CP/AP) with TKI resistance and CML in blast phase (CML-BP) continue to have suboptimal outcomes. Allogeneic hematopoietic stem cell transplantation (alloHCT) remains the most viable treatment option for CML-BP or CML-CP/AP with TKI resistance. We aimed to evaluate clinical outcomes among patients with CML who underwent alloHCT at Mayo Clinic.

Methods: We retrospective analyzed the clinical outcomes of 59 adult patients with CML who underwent alloHCT at Mayo Clinic between 2004-2023 (n = 20 CML-myeloid BP [MBP], n = 24 lymphoid BP [LBP], n = 15 CML-CP/AP). We used statistical methods including Kaplan-Meier (KM) survival analysis, cumulative incidence estimation, and Cox proportional hazards regression to analyze patient and transplant factors associated with post-transplant outcomes.

Results: The baseline characteristics and clinical outcomes are summarized in Table 1. The median age of the patient was 46 years (range, 20.5-70). ABL1 kinase mutations (mainly T315I) were present in 44% patients with 17% receiving a 3rd generation TKI prior to alloHCT (22 [37%] received ≥ 3 TKI prior to allo-HCT). At the time of blast phase progression, 70% MBP and 28.5% LBP exhibited somatic mutations with clonal cytogenetic progression observed in 66% and 18%, respectively.

Ninety five percent of patients received myeloablative conditioning, 54% of patients had acute graft versus host disease (GVHD; ≥ 3 [10%]), and 42% had chronic GVHD (all grades). At the time of review, 39 patients were alive with 3-year and 5-year overall survival of 70% and 50%, respectively. The median OS (mOS) following alloHCT was 46.6 months. In MBP, LBP, and CML-CP/AP, the mOS in months was 46.5, 37.7 and not reached (NR) (p = 0.039) with 5-year survival of 33.3%, 45.5% and 75%, respectively. The median time from diagnosis to transplant was 51.7 and 12.5 months in CML-CP/AP and CML-BP (p < 0.001), respectively. Patients who progressed from CML-CP/AP to MBP had significantly worse mOS (15.3 months) compared to those with de novo MBP (p = 0.025). In multivariate analysis adjusting for MBP at diagnosis, 7/7q- mutation at blast phase portended significantly worse survival with hazard ratio 6.88 (95% CI, 1.1-42.8; p = 0.039). TKI maintenance therapy post-transplant was utilized in 62% patients (74% CML-BP vs 33% CML-CP/AP) without significant survival benefit amongst this cohort, including multivariate analysis accounting for ABL1 mutations.

Conclusions: We evaluated clinical outcomes of 59 patients with CML at various disease stages who underwent alloHCT at Mayo Clinic. MBP was associated with the worst outcomes followed by LBP and CML-CP/AP. MBP outcomes were worse when progressing from chronic to blast phase while on TKI before alloHCT; however, there was no significant difference in LFS or post-transplant relapse between disease groups, suggesting the difference in survival between groups relates to the disease severity prior to alloHCT rather than transplant complications. TKI maintenance therapy following alloHCT did not appear to affect survival in any group irrespective of BCR-ABL1 mutations. Further studies are warranted to explore the utility of TKI maintenance post allo-HCT and timing of allo-HCT in CML-CP/AP with T315I mutations.

Disclosure: Nothing to declare

16: Conditioning Regimens

Michael Scordo ¹, Miguel-Angel Perales¹, Audrey Mauguen², Andrew Lin², Binni Kunvarjee², Jennifer Bieler², Linh K. Nguyen², Maria Paes Pena³, Christina Cho⁴, Boglarka Gyurkocza¹, Andrew C. Harris², Ann A. Jakubowski¹, Richard J. Lin¹, Esperanza B. Papadopoulos¹, Ioannis Politikos¹, Doris M. Ponce¹, Brian C. Shaffer¹, Gunjan L. Shah¹, Barbara Spitzer⁴, Roni Tamari¹, Sergio A. Giralt¹, Jaap Jan Boelens², Kevin J. Curran²

¹Adult Bone Marrow Transplant Service, Memorial Sloan Kettering Cancer Center, New York, United States, ²Memorial Sloan Kettering Cancer Center, New York, United States, ³Sloan Kettering Institute - Pediatrics, Memorial Sloan Kettering Cancer Center, New York, United States, ⁴John Theurer Cancer Center, Hackensack University Medical Center, Hackensack, United States

Background: Ex vivo CD34-selected allogeneic hematopoietic cell transplantation (allo-HCT) is associated with favorable chronic graft-versus-host disease relapse-free survival but limited by delayed immune reconstitution (IR) and in some studies higher non-relapse mortality (NRM) (BMT CTN 1301; Luznik et al., JCO 2023). We recently reported that the use of traditional weight-based ATG dosing in ex vivo CD34-selected allo-HCT led to high post-HCT ATG pharmacokinetic (PK) exposure that was associated with delayed CD4 + T cell IR (CD4 + IR) and increased NRM risk (Lakkaraja et al., Blood Advances 2022).

Methods: Between 5/2021-12/2023, we conducted an investigator-initiated, single-center, phase II study of population PK model-based ATG dosing targeting a low post-HCT exposure with the aim of improving CD4 + IR (NCT04872595). Pediatric and adult patients with hematologic malignancies undergoing their first allo-HCT received model-based ATG to target a post-HCT exposure of <20 AU*d/mL beginning on day -12 followed by one of two myeloablative conditioning regimens (Table 1) and a peripheral blood mobilized, ex vivo CD34-selected allograft (CliniMACS CD34 Reagent System [Miltenyi Biotec, Gladbach, GER]) from a HLA 7-8/8 matched donor.

The primary endpoint was the proportion of patients who reached CD4 + IR, defined as CD4 + T cell values > 50/µL at 2 consecutive timepoints by day +100. The trial size of 56 patients was based on an optimal Simon 2-stage design that had 90% power to detect a 20% increase from a historical control rate of 32% to ≥52%, with 5% 1-sided α. Successful CD4 + IR was required in ≥24 of 56 evaluable patients to reject the null hypothesis. We defined time-to-CD4 + IR as the time from HCT to CD4 + IR. For this endpoint, we censored patients without CD4 + IR at day +100 and considered death before CD4 + IR by day +100 a competing risk. Additional endpoints included the cumulative incidences of NRM and relapse, and rates of relapse-free (RFS) and overall survival (OS).

Results: To date, data were available for 59 patients: 3 were not evaluable and 2 are still within the 100-day period, leaving 54 for the current analysis. Among evaluable patients, the median age was 55 (range, 4-70), 32 (59%) were male, most (N = 30; 56%) were treated for AML, and most (N = 42; 78%) received regimen B. Median estimated pre- and post-ATG exposures were 52 (35.3-73.8) and 9.9 (4.3-16.1) AU*d/mL, respectively (Table 1).

The median follow-up was 17 months (range, 1-28). A total of 37 patients reached CD4 + IR by day +100. The cumulative incidence of CD4 + IR was 69% (95% CI, 56-81%). The 2-year incidences of NRM and relapse were 9% (0-17%) and 13% (3-24%), respectively. The 2-year RFS and OS rates were 78% (66-91%) and 86% (76-97%), respectively.

Table 1. Patient and HCT Characteristics

Conclusions: In patients undergoing ex vivo CD34-selected allo-HCT, the use of model-based ATG dosing to achieve optimal post-HCT exposure led to high rates of CD4 + IR, thereby exceeding our study’s primary objective. These CD4 + IR rates came with low NRM resulting in highly favorable survival, suggesting that the NRM rate observed in trials involving ex vivo CD34-selected allo-HCT such as BMT CTN 1301 were driven by high ATG exposure and may be offset by model-based dosing.

Clinical Trial Registry: NCT04872595

https://clinicaltrials.gov/study/NCT04872595

Disclosure: Michael Scordo: Consultancy - McKinsey & Company, Angiocrine Bioscience, Inc., and Omeros Corporation; Research funding - Angiocrine Bioscience, Inc., Omeros Corporation, Amgen, Inc.; Ad hoc advisory boards for Kite – A Gilead Company; Honoraria - i3Health, Medscape, and CancerNetwork for CME activity.

Miguel-Angel Perales: Honoraria - Adicet, Allogene, Allovir, Caribou Biosciences, Celgene, Bristol-Myers Squibb, Equilium, Exevir, ImmPACT Bio, Incyte, Karyopharm, Kite/Gilead, Merck, Miltenyi Biotec, MorphoSys, Nektar Therapeutics, Novartis, Omeros, OrcaBio, Sanofi, Syncopation, VectivBio AG, and Vor Biopharma. He serves on DSMBs for Cidara Therapeutics and Sellas Life Sciences, and scientific advisory board - NexImmune. Ownership interests: NexImmune, Omeros and OrcaBio. Institutional research support: Allogene, Incyte, Kite/Gilead, Miltenyi Biotec, Nektar Therapeutics, and Novartis.

Boglarka Gyurkocza: Research funding - Actinium Pharmaceuticals, Inc.

Andrew C. Harris: Horizon Pharma PLC: Consultancy; Incyte Corporation: Consultancy, Research Funding; Janssen Scientific Affairs, LLC: Consultancy.

Ioannis Politikos: ExcellThera: Other: Membership on DSMB; Merck: Research Funding.

Doris M. Ponce: Kadmon/Sanofi Pharmaceuticals: Membership on a Board of Directors or advisory committees - Kadmon/Sanofi Pharmaceuticals, Ceramedix, Incyte Corporation, Evive Biotechnology; Research Funding - Incyte Corporation.

Brian C. Shaffer: Gamida Cell: Consultancy, Research Funding; Hansa Biopharma: Consultancy. Attia: Anumana: Patents & Royalties.

Gunjan L. Shah: Amgen: Research Funding; Beyond Spring: Research Funding; Janssen: Research Funding; BMS: Research Funding; ArcellX: Other: DSMB.

Sergio A. Giralt: Amgen, Actinuum, Celgene/BMS, Omeros, Johnson & Johnson, Miltenyi, Takeda: Research Funding; Amgen, Actinuum, Celgene/BMS, Kite Pharma, Janssen, Jazz Pharmaceuticals, Johnson & Johnson, Novartis, Spectrum Pharma, Takeda: Membership on a Board of Directors or advisory committees.

Jaap Jan Boelens: Bluerock: Consultancy, Honoraria; Omeros: Consultancy, Honoraria; Sobi: Consultancy, Honoraria; Sanofi: Consultancy, Honoraria; Immusoft: Consultancy, Honoraria; Advanced Clinical: Honoraria; Bluebird Bio: Honoraria; SmartImmune: Consultancy, Honoraria.

Kevin J. Curran: Kevin J. Curran Consultancy and Research funding – Cellectis, Celgene, Novartis, Atara; Consultancy – Turn Bio.

The remaining authors report no conflicts of interest.

16: Conditioning Regimens

Tim Bognàr ¹, Moises Garcia², Arief Lalmohamed^1,3, A. C. G. Egberts^1,3, Caroline A. Lindemans^4,5, Geoffrey D. E. Cuvelier⁶, Tayfun Güngör⁷, Mathias Hauri-Hohl⁷, Christen L. Ebens⁸, Robbert G.M. Bredius⁹, Hannah Lust¹⁰, Sonali Chaudhury¹⁰, Susan Prockop¹¹, Layne Oram¹¹, Sun-Yung Pai^11,12, Robert Chiesa¹³, Marc Ansari^14,15, Maja Krajinovic¹⁶, Anne-Charlotte Teyssier¹⁶, Yves Théoret¹⁶, Jordan Brooks¹⁷, Rada M. Savic¹⁷, Christopher C. Dvorak¹⁷, Janel R. Long-Boyle¹⁷, Cary Martinez¹⁸, Erin Morales¹⁸, Mary Slatter¹⁹, Maria P. Cicalese^20,21,22, Robert F. Wynn²³, Mary Coussons²³, Christa E. Nath²⁴, Peter J. Shaw²⁴, Steven J. Keogh²⁴, Imke H. Bartelink²⁵, Jaap-Jan Boelens²

¹University Medical Center Utrecht/Wilhelmina Children’s Hospital, Utrecht, Netherlands, ²Cell Transplantation and Cellular Therapies, Memorial Sloan-Kettering Cancer Center, New York, United States, ³Division of Pharmacoepidemiology and Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, Netherlands, ⁴Princess Máxima Center Utrecht, Utrecht, Netherlands, ⁵UMC Utrecht, Utrecht, Netherlands, ⁶Pediatric Blood and Marrow Transplantation, CancerCare Manitoba, University of Manitoba, Manitoba, Canada, ⁷Division of Stem Cell Transplantation and Children’s Research Center, University Children’s Hospital Zurich, University of Zürich, Zürich, Switzerland, ⁸Division of Pediatric Blood and Marrow Transplant & Cellular Therapy, MHealth Fairview Masonic Children’s Hospital, University of Minnesota, Minneapolis, United States, ⁹Willem Alexander Children’s Hospital, Leiden University Medical Center, Leiden, Netherlands, ¹⁰Stem Cell Transplant Program, Ann & Robert Lurie Children’s Hospital/Northwestern University, Chicago, United States, ¹¹Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, Boston, United States, ¹²Center for Cancer Research, National Cancer Institute (NCI), National Institutes of Health (NIH), United States, ¹³Great Ormond Street, Hospital for Children & Stem Cell Program, London, United Kingdom, ¹⁴Cansearch Research Platform for Pediatric Oncology and Hematology, Faculty of Medicine, University of Geneva, Geneva, Switzerland, ¹⁵University Geneva Hospitals, Geneva, Switzerland, ¹⁶Centre de Cancérologie Charles-Bruneau Centre de recherche - Hôspital Sainte-Justine Montréal, Québec, Canada, ¹⁷University of California San Francisco (UCSF), San Francisco, United States, ¹⁸Center for Cell and Gene Therapy, Baylor College of Medicine, Texas Children’s Hospital, Texas, United States, ¹⁹Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle, United Kingdom, ²⁰San Raffaele Telethon Institute for Gene Therapy (SR-Tiget), IRCCS San Raffaele Scientific Institute, Milan, Italy, ²¹Faculty of Medicine and Surgery, Vita-Salute S. Raffaele University, 20132, Milan, Italy, ²²Pediatric Immunohematology Unit and BMT Program, IRCCS San Raffaele Scientific Institute, 20132, Milan, Italy, ²³Royal Manchester Children’s Hospital, Manchester, United Kingdom, ²⁴Children’s Hospital at Westmead, Sydney, Australia, ²⁵Amsterdam University Medical Center, Location VUmc, Amsterdam, Netherlands

Background: Previously, a clear relation between cumulative busulfan exposure (AUC) and clinical outcomes after pediatric and young adult allogeneic hematopoietic cell transplantation (HCT) was shown. As these studies involved only a small number of primary immune deficiency (PID) patients, the optimal target for this group remains unclear. The objective of this study was to assess the optimal busulfan exposure in pediatric and young adult PID patients.

Methods: Patients who received a busulfan-based conditioning regimen between 2000 and 2023 from 17 centers in the Netherlands, USA, Canada, Switzerland, UK, Italy, Germany, and Australia, were included. Main outcome of interest was event-free survival (EFS); events considered were graft-failure (GF) and mortality. Other outcomes of interest were the most recent (myeloid or whole blood) donor chimerism, acute and chronic graft-versus-host disease, veno-occlusive disease and transplant-related mortality (TRM). Patients were categorized into 4 PID subgroups: combined immunodeficiency (CID), severe combined immunodeficiency (SCID), neutrophil disorders and hemophagocytic lymphohistiocytosis (HLH)-related disorders (including primary immune regulatory disorders; PIRD). Busulfan exposure was calculated by individual centers (AUC_CENTER) and was re-estimated using all raw concentration-time profiles with nonlinear mixed-effects modeling (AUC_NONMEM) by applying an externally validated busulfan pharmacokinetic (PK) model (Bartelink, 2012). To assess the validity of the AUC prediction among centers, we compared the AUC_CENTER with AUC_NONMEM. To evaluate the busulfan AUC_NONMEM in relation with the outcomes of interest, we used propensity score adjusted Weibull survival functions and Fine-Gray competing risk regression.

Results: Overall, 562 patients were included: 154 (27.4%) SCID, 173 (30.8%) CID, 101 (18.0%) HLH (14 PIRD) and 134 (23.8%) neutrophil disorders. Median age was 1.7 years (range 0.08-27.0). Median busulfan AUC_NONMEM was 69.0 mg×h/L and correlated poorly with AUC_CENTER (r² = 0.51). CID disease subtype was an effect modifier (p = 0.03), therefore, patients with SCID, HLH-related, and neutrophil disorders were analyzed together (n = 389). Estimated 2-year EFS was 79.6%. In patients with the found optimal busulfan AUC_NONMEM of 70-90 mg×h/L, 2-year EFS was 87.9% (95% CI 80.3-92.6%), superior to <70 mg×h/L (adj-HR 1.97, 95% CI 1.11-3.49, p = 0.02), and >90 mg×h/L (adj-HR 5.05, 95% CI 2.43-10.49, p < 0.0001). In patients with AUC_NONMEM < 70 mg×h/L, events were primarily due to GF, whereas those with AUC_NONMEM > 90 mg×h/L predominantly experienced TRM. In patients with AUC_NONMEM < 50 mg×h/L, the incidence of GF is considerably increased. Donor chimerism increased with higher busulfan AUC_NONMEM, plateauing at 90 mg×h/L. For CID patients, optimal AUC_NONMEM for donor chimerism was found to be >70 mg×h/L, while no optimal AUC_NONMEM for EFS > 50 mg×h/L was found.

Conclusions: Improved EFS and higher donor chimerism may be achieved by targeting a cumulative busulfan AUC_NONMEM of 80 mg×h/L (range 70-90). The data stresses the importance to uniformly use a validated population PK-model to estimate the AUC_NONMEM. For patients with a higher co-morbidity index, targeting a busulfan AUC_NONMEM of 60 mg×h/L (range 50-70) may reduce toxicity, at the risk of increased mixed chimerism and GF.

Clinical Trial Registry: N/A

Disclosure: Declaration of Interest: Christopher Dvorak has disclosed affiliations with Jazz Pharmaceuticals and Alexion Pharmaceuticals, Inc. The other authors report no relevant conflicts of interest

16: Conditioning Regimens

Moniek de Witte ^1,2, Klaartje Nijssen¹, Tim Bognar¹, Anna van Rhenen¹, Lotte van der Wagen¹, Laura Daenen¹, Anke Janssen¹, Anniek Stuut², Peter van de Ven³, Dirk-Jan Moes⁴, Peter van Balen⁴, Stijn Halkes⁴, Kim van der Elst¹, Jurgen Kuball^1,2

¹University Medical Center, Utrecht, Netherlands, ²Center of Translational Immunology, University Medical Center, Utrecht, Netherlands, ³UMC Utrecht, Julius Centrum, Utrecht, Netherlands, ⁴Leiden University Medical Center, Leiden, Netherlands

Background: We have shown in a retrospective study that in adult patients receiving fludarabine (FLU) during conditioning for allo-HCT, 57% of patients were overdosed and 1% were underdosed, and that a high exposure of FLU increased non-relapse mortality (NRM), whereas low exposure was associated with graft failure. We therefore proposed that optimizing FLU exposure by therapeutic drug monitoring (TDM) would decrease complications. Here we present the primary analyses of a prospective randomized clinical trial, comparing TDM-guided FLU dosing with conventional dosing in adults with hematological malignancies who received a TCRalphabeta/CD19 depleted allograft.

Methods: The TARGET study (NL64877.041.18) enrolled adult patients with hematological malignancies. Patients with a creatinine clearance < 40 ml/min were excluded. Allografts were derived from peripheral blood derived stem cells of matched related donors or matched unrelated donors (10/10 or 9/10). Conditioning consisted of Thymoglobulin®, fludarabine and busulfan (AUC 90 mg*h/L). TCRalphabeta/CD19-depletion was performed as previously described. Patients received FLU from day -5 to -2. Conventional FLU (FLU-C) dosing was 40 mg/m²/day and FLU-TDM targeted a cumulative area under the curve (AUC) of 20 mg*h/L (range 15-25 mg*h/L). Primary endpoint was the cumulative incidence of CTC-AE grade 3 viral infections at day 100. Secondary endpoints were NRM, acute graft-versus-host disease (GVHD), chronic GVHD, engraftment, overall survival (OS), event free survival (EFS), GVHD-relapse free survival (GRFS), immune reconstitution of CD8, CD4, γδT, NK and B cells, and FLU exposure.

Results: 48 patients were randomized to the FLU-C arm and 50 patients to the FLU-TDM arm. Patients were well balanced for age, creatinine clearance, disease and donor type. Median follow up was 18 (IQR 3-32) months. Primary endpoint analyses showed that the incidence at day 100 of severe viral infections was not different when FLU-C was compared to FLU-TDM, with a cumulative incidence of CMV in CMV seropositive patients of 67% (95% CI 50-89) vs 60% (95% CI 41-87) and of 2% (95% CI 0.3-18) vs 6% (95% CI 2-18) for EBV. There was no difference in immune reconstitution of CD8, CD4, γδT, NK and B cells between both arms. To assess the lack of difference between both study arms, we measured the FLU AUC for all patients. In the FLU-TDM arm all patients achieved the target dose (median 19.6 mg*h/L, 18.1–24.3) and in the FLU-C arm, 75.5% reached the target range and only 22.5% was overexposed. OS and EFS at 2Y were 62% (95% CI 52-73) and 54% (95% CI 44-65) for the complete cohort and no significant differences where observed between FLU-C and FLU-TDM in multivariable analyses adjusted for potential disbalanced confounders. We also compared the primary and secondary endpoints as well as immune reconstitution between patients with target exposure (n = 86) and those with overexposure (n = 12) and found no difference in outcomes.

Conclusions: This is the first prospective study comparing TDM of FLU to conventional dosing of FLU within the context of allo-HCT. Within this cohort of patients with a creatinine clearance of >40 ml/min, only a fraction of patients are overexposed and TDM does not have substantial addition.

Clinical Trial Registry: NL64877.041.18

www.clinicaltrialsregister.eu/ctr-search/trial/2018-000356-18/NL

Disclosure: J.K. reports grants Novartis, and Miltenyi Biotech and is inventor on patents dealing with γδ T-cell–related aspects as well as cofounder and shareholder of Gadeta. The remaining authors declare no competing financial interests.

16: Conditioning Regimens

Monzr Al Malki ¹, Susanta Hui¹, Dongyun Yang¹, Sally Mokhtari¹, Kay Chanea¹, Firoozeh Sahebi¹, Hoda Pourhassan¹, Matthew Mei¹, Amanda Blackmon¹, Brian Ball¹, Shukaib Arslan¹, Salman Otoukesh¹, Haris Ali¹, Guido Marcucci¹, Ryotaro Nakamura¹, Stephen Forman¹, Anthony Stein¹, Jeffrey Wong¹

¹City of Hope, Duarte, United States

Background: TMLI-based conditioning, developed at City of Hope, has shown to be safe and effective at 20 Gy dose in younger patients with active disease undergoing HCT. Here, we are reporting outcomes of our phase 1 clinical trial of TMLI and Fludarabine-Melphalan for matched donor HCT in older patients with refractory and relapsed (r/r) leukemia and MDS. (NCT03494569)

Methods: Patients ≥55-years-old with r/r acute leukemia or MDS (morphologically or by MRD flow) who had a matched donor available were enrolled. Conditioning regimen included TMLI (days -8 to -5), fludarabine (30 mg/m²/day; days -4 to -2), and melphalan (100 mg/m²; day -2). Radiation dose was escalated in increments of 2 Gy (12 to 18 Gy). All patients received peripheral blood stem cells followed by Tacrolimus and Sirolimus for GvHD prophylaxis. Toxicity was recorded by Bearman and CTCEA-v5.0.

Results: We accrued a total of 30 patients: 24 patients for dose finding portion with 4 dose-levels (DLs) and 6 patients per dose level and 6 more patients at the recommended phase-2 dose (P2RD). Table 1 summarizes patient and HCT characteristics. Median off-target organ dose for lungs, upper GI, heart, testes, parotids, and thyroid was 6-7 Gy. Lower GI, liver, bladder, and brain received 7-8 Gy dose. Oral cavity dose was kept at 3.5 Gy. Almost all patients had one or more Grade-3 GI toxicities by CTCEA, including abdominal pain, diarrhea, nausea, ileus, and colitis (DL1 n = 5; DL2 n = 6; DL3 n = 10; DL4 n = 3), and only one had Grade-4 diarrhea at DL3. Mucositis including mouth, throat or esophagitis was common at lower grades (DL1 n = 1; DL2 = 3; DL4 n = 8; DL4 n = 5). Other common Grade-3 non-hematological AEs were fatigue (DL3 n = 2; DL4 n = 1) and bone pain (DL2 n = 1; DL3 n = 5; and DL4 n = 2). Grade-3 febrile neutropenia was common (DL1 n = 3; DL2 = 3; DL3 n = 9; DL4 n = 6). Three patients had dose limiting toxicities (DLTs) mostly renal, GI, and pulmonary (DL3 n = 2; and DL4 n = 1), all deemed not to be related to TMLI.

All patients achieved engraftment with full donor chimerism, with a median time to neutrophil engraftment of 14 days (range: 13-17) and platelet engraftment of 18 days (range: 16-29). All evaluable patients achieved complete remission with negative MRD by multicolor flowcytometry on day 30. With a median follow-up duration of 23.3 months, 2-year overall survival and progression-free survival were 64% (95% CI: 0.41-0.80) and 62% (95% CI: 0.40-0.78), respectively. One-year GVHD-free Relapse-free survival was at 60% (95% CI: 0.40-0.76). Two-year cumulative incidence (CI) of relapse and non-relapse mortality (NRM) were 23% (95% CI: 0.09-0.41) and 15% (95% CI: 0.05-0.32), respectively. Day-100 CI of grade 2-4 and 3-4 acute-GvHD were 27% (95% CI: 0.12-0.48),13% (95% CI: 0.04-0.28), respectively. CI of chronic-GvHD at 2-years was 36% (95% CI: 0.16-0.56).

Conclusions: TMLI combined with FM was feasible and safe in older patients with high-risk r/r acute leukemia and MDS with acceptable NRM. Disease control was promising in this advanced and high-risk population. Due to increased infections, mucositis, and GI toxicities seen at 18 Gy (DL4), 16 Gy (DL3) was deemed as P2RD.

Table 1. Patient and HCT characteristics

Clinical Trial Registry: NCT03494569

Disclosure: Authors have no COI to disclose.

16: Conditioning Regimens

Tommy Alfaro Moya ^1,2, Mats Remberger³, Carol Chen¹, Armin Gerbitz^1,2, Dennis Dong Hwan Kim^1,2, Rajat Kumar^1,2, Wilson Lam^1,2, Arjun Datt Law^1,2, Jeffrey H. Lipton^1,2, Fotios V. Michelis^1,2, Igor Novitzky Basso^1,2, Auro Viswabandya^1,2, Jonas Mattsson^1,2, Ivan Pasic^1,2

¹Princess Margaret Cancer Centre, Toronto, Canada, ²University of Toronto, Toronto, Canada, ³Uppsala University Hospital, Uppsala, Sweden

Background: Treosulfan has shown promise in allogeneic hematopoietic cell transplantation (HCT) for its myeloablative properties and low toxicity. However, existing retrospective studies lack data from well-matched cohorts whereas phase III data failed to define optimal treosulfan dosing. In this single-centre retrospective propensity score-matched cohort study we compared treosulfan- and busulfan-based conditioning in allogeneic HCT for myeloablative conditioning-ineligible patients with myelodysplastic syndrome (MDS).

Methods: This study included 138 adults who underwent allogeneic HCT for MDS or chronic myelomonocytic leukemia (CMML) at Princess Margaret Hospital, Toronto 2015-2022. Using propensity score matching, we compared transplant outcomes between two well-matched cohorts who received conditioning with either fludarabine-treosulfan (FT) (n = 46) or fludarabine-busulfan with total body irradiation (FBT200) (n = 92).

Results: Patient characteristics are shown in Table 1. A scoring system based on patient age, Karnofsky performance score and hematopoietic cell transplant comorbidity index was used to assign patients based on fitness to low-dose (30 g/m²) or high-dose (42 g/m²) treosulfan: 32 (69.6%) received high-dose treosulfan. Primary outcomes were analyzed at a median follow-up of 747 days.

Patients who received FT had a superior 2-y overall survival (OS) compared to those who received FBT200: 66.9% (95% confidence interval (CI): 46.1-81.2) vs. 44.5% (95% CI: 34-54.4), hazards ratio (HR): 0.43, 95% CI: 0.22-0.84 (P = 0.013). In multivariate analysis (MVA), only the use of fresh grafts (P = 0.02) and FT (P = 0.01) were associated with improved OS.

FT was associated with superior 2-y relapse-free survival (RFS) compared FBT200: 63.1% (95% CI: 42.6-77.9) vs. 39.1% (95% CI: 29.1-49.1), HR: 0.44 (95% CI: 0.24-0.81), P = 0.008. In MVA, the use of fresh grafts (P = 0.03) and FT (P = 0.009) were associated with improved RFS.

Recipients of FT demonstrated superior 2-y graft versus host disease relapse-free survival (GRFS) compared to those who received FBT200: 57.4% (95% CI: 37.8-72.8) vs. 35.1% (95% CI: 25.5-45). In MVA, only FT was associated with superior GRFS (P = 0.02).

FT was associated with lower 1-y non-relapse mortality (NRM) compared to FBT200 in univariate analysis (9.9% (95% CI: 3.0-21.8) vs. 29.7% (95% CI: 20.6-39.3), HR: 0.41 (95% CI: 0.17-0.96), P = 0.04) and MVA (P = 0.04).

FT recipients exhibited markedly superior 1-y event-free survival (EFS) compared to recipients of FBT200 in univariate analysis (40.3% (95% CI: 25.9-54.2) vs. 9.2% (95% CI: 4.4-16.3), HR: 0.47 (95%CI: 0.30-0.72), P < 0.001) and MVA (P = 0.004).

Conclusions: This study is the first to utilize propensity score matching in assessing the role of treosulfan in stem cell transplantation of MAC ineligible patients with MDS and contributes to the increasing body of evidence supporting the antileukemic properties of the drug while underscoring its safety, even at the dose of 42 g/m². Randomized controlled trials comparing treosulfan given at 30 g/m² and 42 g/m² are warranted.

Disclosure: Nothing to declare

16: Conditioning Regimens

Hemalatha G. Rangarajan ¹, Antonio Pérez Martínez², Jacques-Emmanuel Galimard³, Arnaud Dalissier⁴, Rose-Marie Hamladji⁵, Oana Mirci-Danicar⁶, Franco Locatelli⁷, Monika Mielcarek-Siedziuk⁸, Adriana Balduzzi⁹, Jan Styczynski¹⁰, Franca Fagioli¹¹, Cécile Renard¹², Jean Hugues Dalle¹³, Mouhab Ayas¹⁴, Jochen Buechner¹⁵, Marianne Ifversen¹⁶, Petr Sedlacek¹⁷, Kanchan Rao¹⁸, Robert Wynn¹⁹, Marc Ansari²⁰, Katharina Kleinschmidt²¹, Krzysztof Kalwak⁸

¹Nationwide Children’s Hospital, Columbus, United States, ²University Hospital La Paz, Madrid, Spain, ³EBMT Statistical Unit, Paris, France, ⁴EBMT Paris Study Unit, Paris, France, ⁵Centre Pierre et Marie Curie, Alger, Algeria, Algeria, ⁶Bristol Royal Hospital for Children, Pediatric BMT, Bristol, United Kingdom, ⁷IRRCS Ospedale Pediatrico Bambino Gesù, Catholic University of the Sacred Heart, Rome, Italy, ⁸Wroclaw Medical University, Wroclaw, Poland, ⁹Centro Trapianti di Midollo Osseo Clinica Pediatrica Università di Milano Bicocca Ospedale San Gerardo, Monza, Italy, ¹⁰University Hospital, Collegium Medicum UMK Pediatric Hematology and Oncology Bydgoszcz, Bydgoszcz, Poland, ¹¹Onco-Ematologia Pediatrica Centro Trapianti Cellule Staminali, Torino, Italy, ¹²Institut dOncologie Pediatrique, Lyon, France, ¹³Hôpital Robert Debre, Paris, France, ¹⁴King Faisal Specialist Hospital & Research Centre (Paeds), Riyadh, Saudi Arabia, ¹⁵Oslo University Hospital, Olso, Norway, ¹⁶Copenhagen University Hospital,Rigshospitalet, Copenhagen, Denmark, ¹⁷University Hospital Motol, Prague, Czech Republic, ¹⁸Great Ormond Street Hospital for Children NHS Foundation Trust, London, United Kingdom, ¹⁹Royal Manchester Children’s Hospital, Machester, United Kingdom, ²⁰University Hospital of Geneva, University of Geneva, Geneva, Switzerland, ²¹University of Regensburg, Regensburg, Germany

Background: Treosulfan, with its favorable toxicity profile, is increasingly being considered for conditioning in pediatric patients with acute myeloid leukemia (AML) in need of an allogeneic hematopoietic cell transplantation (HCT). Published studies comparing outcomes following Treosulfan vs. Busulfan-based conditioning mainly include recipients of ≥ 2nd HCT. Therefore, we conducted a study comparing outcomes of pediatric AML patients undergoing first allogeneic HCT following either a Treosulfan or Busulfan-based conditioning.

Methods: This retrospective EBMT study included all children ( < 18 years) with de novo AML in CR1 or CR2 who underwent first allogeneic HCT between 2014-2022 and received any of the following myeloablative regimens: Busulfan/Cyclophosphamide (BuCy), Busulfan/Cyclophosphamide/Melphalan (BuCyMel) and Treosulfan/Fludarabine/Thiotepa (TreoFluThio). We excluded recipients of cord blood, haploidentical and < 9/10 mismatched HCT due to low number of patients within at least one of the regimens of interest. We compared the following outcomes between the three groups: leukemia-free survival (LFS), overall survival (OS), non-relapse mortality (NRM), incidence of acute and chronic graft versus host disease (aGVHD, cGVHD), relapse incidence (RI), and GVHD relapse-free survival (GRFS).

Results: 988 (575 males) patients with a median age of 9.3 years [IQR: 3.8-13.7] met the study inclusion criteria, of whom 492 received BuCy, 371 BuCyMel, and 125 TreoFluThio. Baseline characteristics of the three groups (BuCy vs. BuCyMel vs TreoFluThio) were comparable except that a significant proportion of BuCyMel recipients were younger [median age in years: 10.2 vs 7.8 vs 9.5; p < 0.001], received more often bone marrow graft (64.2% vs 74.7% vs 57.6%; p < 0.001), more TreoFluThio recipients had a performance score (PS) < 90 (10.2% vs. 15.1% vs. 28.8%; p < 0.001) and BuCy recipients less often received in-vivo T cell depletion (37.6% vs 66.7% vs 61.3%; p < 0.0001) and more often sibling donors (57.9% vs. 32.9% vs 33.6%; p < 0.001). At a median follow-up of 3.1 years, the 2-year LFS, OS, GRFS, RI, and NRM of the entire cohort were 72.1% (95% CI: 68.8-75), 79.3% (95% CI: 76.4-82), 61.9% (58.4-65.1%), 21.8% (95% CI: 19-24.7) and 6.2% (95% 4.7-7.9) respectively. The cumulative incidence of grade II-IV aGVHD by day+100 was 26.4% (95% CI: 23.6-29.2), and cGVHD at 2 years was 15.1% (95% CI: 12.7-17.6). In multivariable analysis, as shown in Table 1 below, TreoFluThio was not significantly different from BuCy or BuCyMel in terms of LFS, OS, NRM, and GRFS. The incidence of relapse was, however, greater following TreoFluThio compared to BuCyMel but not to BuCy. There were no differences in the incidence of aGVHD and cGVHD. Increasing age (5-year increments) was associated with higher NRM (HR 1.66; p < 0.001), cGVHD (HR 1.35; p = 0.002), and PS of ≥ 90 with better LFS (HR: 0.66; p = 0.03) and OS (HR:0.67; p = 0.048).

Conclusions: In our study, RI was significantly lower following BuCyMel-based conditioning, with other outcomes being not significantly different among the three groups. Data collection on veno-occlusive disease post-HCT is ongoing. The prospective randomized trial comparing BuCyMel vs TreoFluThio in pediatric AML patients will provide more definitive answers on the differences in outcomes between these two regimens.

Clinical Trial Registry: Not Applicable, Retrospective Registry Based Study

Disclosure: Jan Styczynski: Medac (Congress Fee)

All other authors have no conflicts of interest relevant to this abstract.

16: Conditioning Regimens

Ryszard Swoboda ¹, Myriam Labopin², Nicolaus Kröger³, Matthias Stelljes⁴, Thomas Schroeder⁵, Peter Dreger⁶, Gesine Bug⁷, Andrew McDonald⁸, Tülay Özçelik⁹, Montserrat Rovira¹⁰, Igor Wolfgang Blau¹¹, Desiree Kunadt¹², Gerard Socié¹³, Werner Rabitsch¹⁴, Matthias Eder¹⁵, Marie Therese Rubio¹⁶, Alexandros Spyridonidis¹⁷, Bipin Savani¹⁸, Zinaida Peric¹⁹, Sebastian Giebel¹, Mohamad Mohty², Fabio Ciceri²⁰

¹Maria Sklodowska-Curie National Research Institute of Oncology, Gliwice Branch, Gliwice, Poland, ²EBMT Unit, Sorbonne University, Saint-Antoine Hospital, AP-HP, INSERM UMRs 938, Paris, France, ³University Hospital Eppendorf, Hamburg, Germany, ⁴University of Muenster, Muenster, Germany, ⁵University Hospital, Essen, Germany, ⁶University of Heidelberg, Heidelberg, Germany, ⁷Goethe-Universitaet, Frankfurt Main, Germany, ⁸Alberts Cellular Therapy, Pretoria, South Africa, ⁹Demiroglu Bilim University Istanbul Florence Nightingale Hospital, Istanbul, Turkey, ¹⁰Institute of Haematology and Oncology, IDIBAPS, Hospital clinic, University of Barcelona. Josep Carreras Leukaemia Research Foundation, Barcelona, Spain, ¹¹Medizinische Klinik m. S. Hämatologie, Onkologie und Tumorimmunologie, Berlin, Germany, ¹²University Hospital TU Dresden, Dresden, Germany, ¹³Saint-Louis Hospital, Paris, France, ¹⁴Medizinische Universitaet Wien, Vienna, Austria, ¹⁵Hannover Medical School, Hannover, Germany, ¹⁶CHRU NANCY, Vandoeuvre les Nancy, Nancy, France, ¹⁷University of Patras, Patras, Greece, ¹⁸Vanderbilt University Medical Center, Nashville, United States, ¹⁹University of Zagreb and University Hospital Centre Zagreb, Zagreb, Croatia, ²⁰IRCCS Ospedale San Raffaele s.r.l., Haematology and BMT, Milano, Italy

Background: Total body irradiation (TBI)-based conditioning is considered as a standard therapy before allogeneic hematopoietic stem cell transplantation (alloHSCT) in patients with acute lymphoblastic leukemia (ALL). Although treosulfan (Treo) is a frequently used component of TBI-free conditioning in patients with acute myeloid leukemia, there are very limited data on Treo-based conditioning before alloHSCT in ALL patients. As Treo is commonly used in combination with fludarabine (Flu), we aimed to compare Flu/Treo with Flu/TBI as a conditioning therapy before alloHSCT in ALL patients.

Methods: Based on EBMT registry we retrospectively analyzed data of 2055 adult patients with ALL who met inclusion criteria: 1) first alloHSCT in first or second complete remission (CR1, CR2) between years 2006 and 2022; 2) matched sibling (MS), unrelated (UD) or haploidentical donor; 3) Flu/Treo (30-42g/m2) or Flu/TBI (8-12 Gy) as a conditioning therapy. 153 patients in Flu/Treo group and 431 patients in Flu/TBI group were included in a final matched-pair analysis.

Results: The median patients age was 54.1 and 56.5 years in Flu/TBI and Flu/Treo group, respectively. Male to female proportion was 52%/48% in both groups. Philadelphia negative (Ph-), Ph positive (Ph + ) B-ALL and T-ALL were diagnosed in 61 (39.9%), 42 (27.5%), 22 (14.4%) patients in Flu/Treo group and in 168 (39%), 111 (25.8%), 70 (16.2%) patients in Flu/TBI group. CR1/CR2 proportion was 75%/25% in both groups. The allograft was derived from UD (54.2% in Flu/Treo and 52% Flu/TBI group), MS (30.7% and 32.5%) and haploidentical donor (15% and 15.5%).

Graft failure was diagnosed in 5 (3.5%) and 7 (1.7%) patients in Flu/Treo and Flu/TBI group, respectively. The incidence of acute graft versus host disease (GVHD) grade II-IV and III-IV was 26.6% and 13.7% in Flu/Treo and 30% and 9.3% in Flu/TBI group (HR 0.87 (95% CI 0.6-1.24), p = 0.43, and HR 1.5 (0.89-2.54), p = 0.13, respectively). The 2-year incidence of overall and extensive chronic GVHD was 33.5% and 18.8% in Flu/Treo while 40.7% and 20.2% in Flu/TBI group (HR 0.81 (0.57-1.14), p = 0.23, and HR 1.04 (0.65-1.68), p = 0.87). The 2-year incidence of relapse and non-relapse mortality was 28.1% and 25.6% in Flu/Treo and 25.1% and 20.4% in Flu/TBI group (HR 1.31 (0.91-1.88), p = 0.15 and HR 1.19 (0.8-1.77), p = 0.38, respectively). Leukemia-free survival and overall survival was 46.2% and 55% in Flu/Treo and 54.5 and 61.3 in Flu/TBI group (HR 1.25 (0.96-1.63), p = 0.1 and HR 1.19 (0.9-1.59), p = 0.22, respectively). GVHD and relapse-free survival was 33.7% and 40.8%, respectively in Flu/Treo and Flu/TBI group (HR 1.25 (0.99-1.57), p = 0.06).

Conclusions: The combination of Flu with Treo as a conditioning therapy before alloHSCT in ALL patients results in comparable transplant outcomes as Flu combined with TBI. Therefore Flu/Treo appears a valuable alternative for ALL patients with contraindications for irradiation. Our findings require confirmation in prospective trials.

Disclosure: Gesine Bug has received honoraria from Novartis, Jazz, BMS and Gilead and travel grants from Jazz, Gilead and Neovii.

Other authors have no conflict of interest to declare.

16: Conditioning Regimens

Margaret MacMillan ¹, Qing Cao¹, David McKenna¹, Meera Srikanthan¹, John Wagner¹

¹University of Minnesota, Minneapolis, United States

Background: We hypothesized that TCRalpha/beta depletion would sufficiently deplete graft-versus-host disease (GVHD) causing a/b T cells to eliminate the need for prolonged immunosuppression (IS) after related or unrelated donor peripheral blood stem cell (PBSC) transplantation in patients with Fanconi anemia (FA).

Methods: Patients were conditioned with fludarabine, cyclophosphamide and methylprednisolone alone if the patient had marrow failure only and an HLA matched sibling donor (n = 1), or in combination with busulfan (n = 14) or total body irradiation (n = 12) if there was clonal disease, or the donor was a haploidentical related (n = 2) or HLA matched or mismatched unrelated donor (n = 22). All but the first patient received one dose of rituximab the day before transplant. PBSCs collected after Neupogen mobilization were depleted of TCRa/b cells prior to transplant. Mycophenolate mofetil was given after transplant if the TCRa/b dose ≥2 x 10⁵ TCR α/β T cells/kg recipient weight (n = 4).

Results: To date, 27 patients with FA (14 FANCA, 3 FANCC, 6 BRCA2, and one each with FANCB, FANCF, FANCG and FANCJ) received TCRa/b depleted PBSCs for marrow failure (n = 13), early MDS (n = 7), late MDS (n = 2), relapsed AML (n = 1), ALL (n = 1) or pre-emptively for BRCA2 (n = 3), with 2 patients having had a prior transplant. Median age was 9.4 years (6 > 18 years). Median CD34+ cells/kg was 1.11 x 107/kg and median TCRa/b dose was 7.0 x 104/kg. Neutrophil engraftment occurred in all patients at a median of 9 days. Platelet recovery ≥20,000/uL occurred in 96% at a median of 15 days. Complete donor myeloid engraftment was achieved by day 21 in all patients with sustained neutrophil engraftment who did not relapse. Donor lymphoid engraftment was slower with complete CD3+ chimerism achieved in only 15 patients at 1 year after PBSCT. To date, 3 patients developed grade II-IV acute GVHD. No patient has developed chronic GVHD requiring systemic therapy.With a median follow-up of 22 months, probability of survival at 2 years is 93%, including 5 of 6 BRCA2 patients and all 6 adult patients (Figure). Secondary graft failure occurred in two patients; aplasia after a febrile illness in one FANCF patient; and autologous recovery in one BRCA2 patient. Both patients were successfully retransplanted. Relapse was observed in 3 patients: two BRCA2 patients with AML or MDS, and one FANCC patient with relapsed MDS after a prior transplant. The two BRCA2 patients were successfully retransplanted and remain in remission. 2 patients developed PTLD, treated successfully with rituximab. 5 patients reactivated viral infections (3 CMV, 1 adenovirus,1 BK virus).

Conclusions: TCRa/b depleted PBSC without CNI results in excellent primary engraftment, minimal GVHD and excellent survival in FA patients including BRCA2 and older patients. Patients with biallelic BRCA2 require a TBI based regimen to achieve sustained engraftment using this approach. The absence of IS after HCT provides an opportunity for future cell therapy interventions to optimize recovery after HCT.

Clinical Trial Registry: Clinical Trials.gov NCT03579875

Disclosure: Nothing to declare

6: Experimental Transplantation and Gene Therapy

Maria Escolar ¹, Michele D. Poe¹, Juan Ruiz¹, Mark Vander Lugt², Raymond Wang³, David Buchbinder⁴, Paul Szabolcs⁵

¹Forge Biologics, Grove City, United States, ²University of Michigan, Ann Harbor, United States, ³Children’s Hopital of Orange County, Orange, United States, ⁴Children’s Hospital of Orange County, Orange, United States, ⁵University of Pittsburgh, Pittsburgh, United States

Background: REKLAIM is a novel intravenous AAVrh10.GALC gene therapy administered during the period of myeloablation or immune suppression after umbilical cord blood transplantation (UCBT) for the treatment of Infantile and Late Infantile Krabbe disease (IKD, LIKD). We report the results of the first 5 subjects with IKD treated with a low dose intravenous FBX-101 (1.6 x10¹³ gc/kg). IKD is a fatal neurodegenerative disorder due to galactocerebrosidase (GALC) deficiency that results in psychosine toxicity to myelinating cells in the brain and peripheral nervous system. If untreated death occurs in average by 2 years. Currently, pre-symptomatic neonates are treated with UCBT in the first 30-45 days of life, halting brain disease. However, follow up studies demonstrate that motor function slowly declines due to progressive peripheral neuropathy. One of the most challenging problems when administering AAV intravenously is the production of antibodies against the transgene and the capsid. UCBT is immunological naive and the healthy graft should not recognize the over expression of GALC as an antibody. Therefore, we hypothesized that FBX-101 administered during myeloablation or later during immunesuppression, will override the antibody response to the vector’s capsid and transgene, improving transduction and increasing the levels of GALC.

Methods: REKLAIM is a Phase 1b dose-escalating intravenous gene therapy to evaluate safety and efficacy of FBX-101 administered intravenously >21 days after UCBT for the treatment of Infantile and Late Infantile Krabbe disease while the subject is myeloablated, or later when the subject is immune-suppressed. Each subject at the time of gene therapy administration may be at different levels of immune suppression. Therefore, an independent committee reviews the immune status and suggest adjustments as needed in order to minimize an immune response to AAV. A Rituximab, Serolimus and Prednisolone based regime is used to adjust immune suppression. After gene therapy administration the subjects are followed for 2 years and enrolled in 3 additional years of long term follow up.

Results: FBX-101 was well tolerated, with no treatment-related serious adverse events with follow up ranging from 4 to 24 months. In the two subjects treated during myeloablation, there were no antibodies to AAV. Plasma and CSF GALC increased to supraphysiological levels with psychosine dropping below detection levels. The white mater integrity of the corticospinal tracts that rapidly myelinate in the first year, followed a normal trajectory and subjects achieved normal gross motor skills above those seen after UCBT only. The three subjects treated during immune suppression developed minimal total antibodies to AAV with no signs of humoral or cellular toxicity and improved in gross motor skills.

Conclusions: In summary, FBX-101 after UCBT leverages the myeloablation and immune suppression after UCBT, resulting in efficient AAV transduction providing increased GALC enzyme that supports normal brain myelination and gross motor development.

Clinical Trial Registry: NCT05739643

Disclosure: Escolar is an inventor on the patent application describing FBX-101(WO2018/136710)licensed to Forge Biologics by the University of Pittsburgh where she is an Adjunct Professor of Pediatrics.

6: Experimental Transplantation and Gene Therapy

Maddalena Migliavacca ^1,2, Federica Barzaghi^2,1, Claudia Fossati¹, Paola M.V. Rancoita³, Michela Gabaldo⁴, Francesca Dionisio¹, Stefania Giannelli¹, Federica Andrea Salerio¹, Francesca Ferrua^2,1, Francesca Tucci^1,2, Calbi Valeria^2,1, Vera Gallo^2,1, Salvatore Recupero^2,1, Giulia Consiglieri^2,1, Roberta Pajno^2,1, Vittoria Garella⁵, Ilaria Monti¹, Paolo Silvani⁶, Silvia Darin¹, Miriam Casiraghi¹, Ambra Corti¹, Stefano Zancan⁴, Margherita Levi⁴, Daniela Cesana¹, Anna Pituch-Noworolska⁷, Dalia AbdElaziz⁸, Ulrich Baumann⁹, Andrea Finocchi^10,11, Caterina Cancrini^10,11, Saverio Ladogana¹², Andrea Meinhardt¹³, Isabelle Meyts¹⁴, Davide Montin¹⁵, Lucia Dora Notarangelo¹⁶, Fulvio Porta¹⁷, Marlène Pasquet¹⁸, Carsten Speckmann^19,20, Polina Stepensky²¹, Alberto Tommasini²², Marco Rabusin²², Zeynep Karakas²³, Miguel Galicchio²⁴, Lucia Leonardi²⁵, Marzia Duse²⁵, Sukru Nail Guner²⁶, Clelia Di Serio^3,27, Fabio Ciceri^5,28,1, Maria Ester Bernardo^5,2,1, Alessandro Aiuti^5,2,1, Maria Pia Cicalese^5,2,1

¹San Raffaele Telethon Institute for Gene Therapy (SR-Tiget), IRCCS San Raffaele Scientific Institute, Milano, Italy, ²Pediatric Immunohematology and Bone Marrow Transplantation Unit, IRCCS San Raffaele Scientific Institute, Milan, Italy, ³University Centre for Statistics in the Biomedical Sciences (CUSSB), Vita-Salute San Raffaele University, Milan, Italy, ⁴Fondazione Telethon, Milano, Italy, ⁵Università Vita-Salute San Raffaele, Milano, Italy, ⁶IRCCS San Raffaele Scientific Institute, Milano, Italy, ⁷University Children’s Hospital of Cracow, Cracow, Poland, ⁸Cairo University, Cairo, Egypt, Cairo, Egypt, ⁹Hannover Medical School, Hannover, Germany, ¹⁰Bambino Gesù Children’s Hospital, Scientific Institute for Research and Healthcare (IRCCS), Rome, Italy, ¹¹University of Rome Tor Vergata, Rome, Italy, ¹²Paediatric Onco-haematology Unit, “Casa Sollievo della Sofferenza” Hospital, IRCCS, San Giovanni Rotondo, Italy, ¹³Medical Center, University Hospital Giessen, Giessen, Germany, ¹⁴Immunology and Transplantation, KU Leuven; Childhood Immunology, UZ Leuven, Leuven, Belgium, ¹⁵University of Torino, Regina Margherita Children’s Hospital, AOU Città della Salute e della Scienza di Torino, Turin, Italy, ¹⁶Medical Direction, Children’s Hospital, ASST-Spedali Civili, Brescia, Italy, ¹⁷Pediatric Oncology-Hematology and BMT Unit, Spedali Civili di Brescia, Brescia, Italy, ¹⁸Pediatric Hematology and Immunology, Children’s Hospital, Toulouse, France, ¹⁹Institute for Immunodeficiency, Center for Chronic Immunodeficiency (CCI), Faculty of Medicine, Medical Center-University of Freiburg, Freiburg, Germany, ²⁰University Medical Center Freiburg, University of Freiburg, Freiburg, Germany, ²¹Hadassah-Hebrew University Medical Center, Jerusalem, Israel, ²²IRCCS Burlo Garofolo, Trieste, Italy, ²³Istanbul University, Istanbul School of Medicine, Istambul, Turkey, ²⁴Allergy and Immnunology Service, Hospital de Niños VJ Vilela, Rosario, Argentina, ²⁵La Sapienza University of Rome, Rome, Italy, ²⁶Division of Pediatric Allergy and Immunology, Faculty of Medicine, Necmettin Erbakan University, Konya, Turkey, ²⁷Faculty of Biomedical Sciences, Università della Svizzera Italiana, Lugano, Switzerland, ²⁸Hematology and Bone Marrow Transplantation Unit, IRCCS San Raffaele Scientific Institute, Milan, Italy

Background: Autosomal recessive adenosine-deaminase (ADA) deficiency leading to severe combined immunodeficiency (SCID) is a life‐threatening immunodeficiency, characterized by lymphopenia in B/T/NK subpopulations, failure to thrive, severe infections and autoimmunity. Allogeneic hematopoietic stem cell transplant (HSCT) and Strimvelis, an ex-vivo retroviral hematopoietic stem cell gene therapy (GT) approved in the EU in 2016, are standard of care treatments for ADA-SCID.

Methods: Here we describe the post-marketing experience of 19 patients (STRIM cohort) up to 5 years of follow up, the largest cohort of patients treated worldwide with an approved product based on ex vivo transduced CD34+ cells. We also provide an extended follow-up data on 22 subjects treated in the clinical development/named patient program (CDP + NPP cohort) and report on two patients treated with mobilized peripheral blood CD34+ cells under hospital exemption (mPB-HE). Age at diagnosis and at GT, CD34+ cell dose and vector copy number (VCN) per genome in the drug product were similar in the CDP + NPP and STRIM cohorts.

Results: Nineteen patients initially referred for Strimvelis did not receive treatment due to lack of funding, other treatment choice or ineligibility due to different reasons. At data cut-off (March 2023), all 43 patients were alive, with a median follow-up of 5 years (interquartile range 2.4-15.4) and 2 year intervention-free survival (no need for long-term ERT or rescue HSCT) of 88% (95% CI: 78.7%-98.4%). Long-term persistence of multilineage gene-corrected cells was associated with immune-reconstitution. At the last available observation, 16/17 CDP + NPP patients and all STRIM patients who reached 5 years post GT, discontinued IVIg supplementation and responded to most vaccinations. Response to anti aCD3i and PHA were normal at last follow up in all STRIM and CDP + NPP patients who did not fail GT. Native pathogens’ infections had a regular evolution and resolved without sequelae. Estimated mixed-effects models showed that higher dose of CD34+ cells infused affected positively the plateau of CD15+ and CD3+ gene corrected cells. In addition, younger age at GT affected positively the plateau of CD3+ gene corrected T cells as well as lymphocytes, CD3 and CD4 + CD45RA+ naïve T cell counts. We also observed a significantly higher plateau of transduced CD15+ and CD3+ cells and better dAXP detoxification in STRIM vs CDP + NPP, but no differences in the lymphocyte or T cell reconstitution.

Most adverse events/reactions were related to disease background, busulfan conditioning or immune reconstitution. A lymphoid T cell leukemia with retroviral insertion near LMO2 emerged 4.7 years post-GT in a patient from the CDP + NPP group, who is currently in clinical remission 2.5 years after haploidentical HSCT.

Conclusions: In the post-marketing population, the safety profile was in line with the clinical development program and we did not find evidence of clonal proliferation, nor new treatment-related events. Due to the risk of insertional oncogenesis, long-term safety monitoring remains important.

Following disinvestment from the industrial holder, the marketing authorization of the retroviral GT for ADA-SCID was transferred to Fondazione Telethon which established a nonprofit model of distributing for ultrarare-diseases, allowing this gene therapy still available to eligible patients.

Clinical Trial Registry: NCT00598481[https://clinicaltrials.gov/study/NCT00598481];NCT03478670[https://clinicaltrials.gov/study/NCT03478670].

Disclosure: Gene therapy for ADA-SCID was developed at SR-Tiget and licensed to GlaxoSmithKline (GSK) in 2010. Strimvelis Marketing Authorization in Europe occurred in 2016 (under GSK holding) and then transferred to Orchard Therapeutics (Netherlands) B.V. in 2018, which divested the program and transferred the authorization to Fondazione Telethon that became the holder in July 2023.

A. Aiuti was the PI of pilot, pivotal and long-term follow up study of SR-Tiget clinical trial of gene therapy for ADA SCID. M.P. Cicalese and M. Migliavacca are PI and deputy PI, respectively, of the Strimvelis Registry, RIS and RMMs studies. All other authors declare no competing interests.

The work was partially supported by Fondazione Telethon and grants from the European Commission (ERARE-3-JTC 2015 EUROCID, A.A.), Ministero della Salute, Ricerca Finalizzata NET-2011-02350069 (to AA, CC). AA is the recipient of Else Kröner-Fresenius-Stiftung (EKFS) prize.

6: Experimental Transplantation and Gene Therapy

Giorgio Orofino ¹, Pier Edoardo Rovatti¹, Marco Punta¹, Matteo Maria Naldini¹, Matteo Barcella¹, Laura Zito¹, Matteo Carrabba¹, Massimo Bernardi¹, Jacopo Peccatori¹, Bernhard Gentner¹, Fabio Ciceri^1,2, Luca Vago^1,2, Cristina Toffalori¹

¹IRCCS Ospedale San Raffaele Scientific Institute, Milan, Italy, ²San Raffaele Vita-Salute University, Milan, Italy

Background: Allogeneic Hematopoietic Cell Transplantation (allo-HCT) represents a potentially curative treatment for high-risk Acute Myeloid Leukemia (AML) patients. However, leukemia relapses are frequent and characterized by poor prognosis. Three major alternative mechanisms by which AML cells modify their features and escape immune control have been described: genetic loss of the mismatched HLA haplotype (HLA loss), downregulation of HLA-II molecules and upregulation of inhibitory ligands, all negatively impacting on T cell-mediated leukemia recognition. Increasing evidence suggests that leukemic cells weave a broader net of interactions, involving several immune cell types. Leveraging on single-cell RNA-sequencing (scRNA-seq), we profiled NK cells in the bone marrow (BM) microenvironment to understand whether and how they are affected by the emergence of immune-escape variants and why they are unable to prevent clinical relapse.

Methods: We performed scRNA-seq throught the 10x Genomics platform on BM aspirates from 25 acute AML patients at the time of relapse after allo-HCT (relapse modalities included genomic HLA loss, downregulation of MHC-II molecules, upregulation of inhibitory ligands), from 5 post-transplant patients in complete remission (CR), each tested at two timepoints ( + 90 and +365 days) and from 6 healthy controls (HC).

Results: Based on the expression of canonical NK cell transcripts we identified a total of 26,264 NK cells from the original dataset. We annotated three major subsets (CD56^bright, CD56^dim and transitional NK cells) and a small cluster of CD56^bright NK cells, characterized by high expression of IFN-related genes and of NK exhaustion markers, identified by previous studies as “inflamed-NK”. Interestingly, inflamed-NK were more represented in patients with upregulation of T-cell inhibitory ligands compared to other relapsed patients (5.97% vs 0.91%, respectively, p.value = 0.03), suggesting a shared mechanism of dysfunction between T and NK cells in this relapse modality. Next, we specifically focused on NK cells in HLA loss relapses. In this relapse modality, the loss of some HLA alleles might in fact mean losing ligands for inhibitory KIRs, and thus trigger NK-cell alloreactivity. Using recently published gene signatures of NK cell activation, we generated a module score for CD56^dim NK cells. As expected, HC exhibited the lower score, while CR samples showed a higher activation profile, similar to that of upregulation on inhibitory ligands relapses, suggesting that in most patients upon transplant NK cell immunosurveillance is activated. Interestingly, HLA loss relapses showed a very low activation score, regardless of potential alloreactivity, indicating that they might be functionally impaired by a mechanism currently under investigation.

Conclusions: Our observations support the hypothesis that, as T cells, also NK cells are differentially affected by the features of relapse, and inflammation within BM plays a central role on NK cell transcriptional features. Understanding the drivers of NK cell dysfunction in relapses may have important implications for rewiring their cytotoxic functions and developing innovative salvage strategies.

Disclosure: No disclosures

6: Experimental Transplantation and Gene Therapy

Anne Charlotte Adriane Mudde ¹, Soraia Silva¹, Emily Hu¹, Giandomenico Turchiano¹, Nathan White¹, Claire Booth¹

¹UCL, London, United Kingdom

Background: X-linked inhibitor of apoptosis (XIAP) deficiency is a rare, life-threatening inborn error of immunity, caused by mutations in the XIAP/BIRC4 gene. It is characterised by immune dysregulation that is associated with a spectrum of manifestations, including haemophagocytic lymphohistiocytosis (HLH) and severe inflammatory bowel disease. Elevated serum IL-18 levels are typical of the disease, irrespective of clinical status. Allogeneic haematopoietic stem cell (HSC) transplantation can be curative, but outcome post-HSCT is worse compared to other forms of HLH and graft versus host disease is associated with high mortality rates in these patients. An autologous HSC gene editing approach may offer a suitable alternative treatment, removing any risk of alloreactivity and permitting the use of less toxic preparative chemotherapy regimens. We previously demonstrated proof of concept for a lentiviral gene addition approach in a murine model and patient cells. However, we are now investigating a gene editing approach, in which the endogenous regulation of protein expression is maintained which may offer safety and efficacy advantages.

Methods: A CRISPR/Cas9 based gene editing strategy was used to achieve precise correction at the XIAP locus in human CD34+ cells, targeting early in the gene to allow correction of most described mutations. After initial experiments in T-lymphocytes, healthy donor CD34+ cells were electroporated with ribonucleoprotein (RNP) using a guide RNA targeting the first intron, followed by AAV6 transduction of a donor cassette containing codon optimised XIAP cDNA and GFP reporter. Cells were differentiated into monocytes and stimulated with MDP or LPS to assess XIAP dependent function. In vitro colony forming unit (CFU) assay and an in vivo engraftment study in NSG mice were used to assess differentiation and engraftment potential of edited CD34+ cells.

Results: As a first step, successful editing in T-lymphocytes was demonstrated with targeted integration rates of 25%. Following optimisation for HSCs, editing rates of up to 60% were achieved as measured by GFP expression, without impacting cell viability. Correction levels of up to 44% were achieved in XIAP patient HSCs derived from peripheral blood, resulting in restoration of XIAP protein expression and partial reconstitution of TNFα secretion after differentiation into monocytes and stimulation with MDP. Edited HSCs from healthy donors generated similar number of colonies compared to untreated cells in CFU assay, indicating preservation of clonogenic and differentiation potential in vitro. Engraftment potential is being further investigated in vivo; edited healthy donor CD34+ cells have been successfully engrafted into an immunodeficient NSG mouse model and analysis of engraftment specifically of gene edited cells is underway. Analyses to assess off-target endonuclease activity are currently being undertaken by sequencing of the in silico predicted top 10 off-target sites and karyotype analysis.

Conclusions: This study confirms proof-of-principle for an HSC gene editing approach as a potential durable treatment for XIAP deficiency. High editing rates in CD34+ cells were achieved at the XIAP locus, with good viability and in vitro differentiation potential. High levels of targeted integration were also observed in XIAP patient HSCs resulting in partial restoration of XIAP protein dependent immune function in differentiated, corrected cells.

Disclosure: Nothing to declare

6: Experimental Transplantation and Gene Therapy

Francesca Tucci ¹, Maria Luz Uria Oficialdegui², Giulia Consiglieri¹, Matilde Cossutta¹, Chiara Filisetti¹, Luca Galimberti¹, Mariana Priviero Spiridon³, Maria Pia Manitto¹, Carla Butera¹, Roberto Santangelo¹, Giulia Danè¹, Chiara Camesasca¹, Silvia Darin¹, Mireia Del Toro², Cristina Diaz de Hereida Rubio², Rossella Parini⁴, Bernhard Gentner⁴, Alessandro Aiuti¹, Maria Ester Bernardo¹

¹IRCCS San Raffaele Scientific Institute, Milano, Italy, ²Hospital Universitari Vall d’Hebron, Barcelona, Spain, ³Vita-Salute San Raffaele University, Milan, Italy, ⁴San Raffaele Telethon Institute for Gene Therapy, Milano, Italy

Background: Preliminary results of a phase I/II study (NCT03488394) of ex-vivo haematopoietic stem and progenitor cells gene therapy (HSC-GT, also known as OTL-203) in 8 Mucopolysaccharidosis type I Hurler syndrome (MPSIH) patients have demonstrated superior metabolic correction compared to allogeneic haematopoietic stem cell transplantation (HSCT). Nevertheless, HSPC-GT effects on non-neurological, non-skeletal manifestations such as corneal clouding (CC), hearing loss (HL), carpal tunnel syndrome (CTS) and cardiac involvement have not been extensively reported. Here we analysed for the first time these latter outcomes after HSPC-GT and performed a retrospective comparison with an external cohort of MPSIH patients undergoing HSCT.

Methods: MPSIH patients undergoing HSPC-GT (N. 8; NCT03488394) were retrospectively compared with MPSIH patients receiving HSCT (N. 9) in Vall d’Hebron Hospital, Barcelona. Patient and transplant characteristics of the HSC-GT and HSCT cohorts are reported in Table 1. Two of nine patients who received HSCT experienced primary graft failure and required a second HSCT.

CC was assessed using clinical grading scores (Fahnehjelm et al. Acta Opthalmol 2012 for HSC-GT patients), HL was evaluated using brainstem auditory evoked responses using dBnHL threshold. Cardiac function was studied clinically and by echocardiography. CTS was evaluated by electroneurography of the median nerve. Baseline results were compared with those obtained after treatment. HSC-GT patients were followed for up to a median of 4 years (range: 4-5) after treatment, while HSCT patients were followed-up for a median of 15 years (range: 4-21). However, to enable appropriate comparison between the two cohorts, this was performed within the same time window (4-5 years after treatment).

Table 1

Results: At baseline, all subjects had evidence of mild to moderate CC. Of the HSC-GT patients, 3/8 achieved CC resolution and 5/8 had mild CC at last follow-up. In the HSCT cohort, no patient experienced CC resolution with all patients showing moderate CC at 5 years after HSCT.

In the HSPC-GT cohort at baseline, 50% of the patients had mild to severe HL. An improvement in most of the patients was observed with 4/7 showing normal hearing function at last follow-up. In the HSCT cohort, 7/9 had mild or moderate HL at baseline with an improvement at the last follow-up as 4/9 showed normal hearing function. ENT surgery incidence and timing after treatment was comparable in the 2 cohorts.

In the HSC-GT-cohort none of the patients developed severe cardiomyopathy or valvular disease, while in the HSCT-cohort 2/9 patients experienced progression of valvular insufficiency but did not require valve replacement.

One HSC-GT patient underwent CT surgery before treatment, while another one 6 months after HSC-GT due to pre-existing severe damage before treatment. The remaining patients did not develop CTS. In the HSCT-cohort 7/9 patients developed CTS, requiring surgery at a median of 3.7 years post-HSCT.

Conclusions: Our results indicate stability or improvement of CC, HL, cardiac manifestations and CTS in MPSIH patients after HSC-GT up to 4 years after treatment. A longer follow-up and a prospective comparison with HSCT patients are required to draw definitive conclusions on the efficacy of HSC-GT in addressing these disease burdens.

Clinical Trial Registry: NCT03488394 (https://www.clinicaltrials.gov/study/NCT03488394)

Disclosure: The study is sponsored by IRCCS San Raffaele Scientific Institute. Fondazione Telethon was the initial financial sponsor; since May 2019 Orchard Therapeutics Ltd (OTL) is the financial sponsor when MPSIH gene therapy was licensed to OTL. Fondazione Telethon and OSR are entitled to receive milestone payments and royalties for such a therapy from OTL. AA was the study PI between May 2018 and July 2020, MEB is the PI since July 2020. BG, RP and MEB have acted as ad hoc consultant for an Orchard Therapeutic advisory board in 2020.

6: Experimental Transplantation and Gene Therapy

Giulia Consiglieri ¹, Francesca Fumagalli¹, Francesca Tucci¹, Chiara Filisetti¹, Matilde Cossutta¹, Stefano Scarparo¹, Marina Sarzana¹, Cristina Puricelli¹, Francesca Ciotti¹, Maurizio Barbera¹, Silvia Darin¹, Rossella Parini¹, Cristina Baldoli¹, Bernhard Gentner², Alessandro Aiuti¹, Maria Ester Bernardo¹

¹IRCCS San Raffaele Scientific Institute, Milano, Italy, ²San Raffaele Telethon Institute for Gene Therapy, Milano, Italy

Background: Early treatment with allogeneic haematopoietic stem cell transplantation (HSCT) was demonstrated in many cases to prevent deterioration of cognitive functions in Mucopolysaccharidosis type I Hurler (MPSIH) patients. Nevertheless, HSCT is not able to fully address central nervous system (CNS) manifestations, leaving patients with residual disease burden. The neurological outcomes of 8 MPSIH patients enrolled in a phase I/II study (NCT03488394) of autologous ex-vivo haematopoietic stem cell gene therapy (HSC-GT, also known as OTL-203) were evaluated up to 4 years after treatment.

Methods: IDUA activity, dermatan (DS) and heparan sulphate (HS) levels were measured in cerebral spinal fluid (CSF) at baseline and at different timepoints after HSC-GT. Ages of achievement of motor and language milestones were compared with the window of achievement defined by the World Health Organization (WHO) in healthy children. Gross and fine motor skills were analyzed using the Peabody Developmental Motor Scales, while measurement of cognitive functions was evaluated by the Bayley Scales of Infant and Toddler Development and the Wechsler Preschool and Primary Scale of Intelligence. Typical MRI brain abnormalities were assessed by a specific neuroradiologic scoring system previously developed for MPSIH patients treated with HSCT.

Results: From 3 months after GT, IDUA activity became detectable in CSF and persisted over time, accompanied by local reduction of both DS and HS with undetectable DS in 7/8 patients at +4 year-follow-up. Motor and language milestones were accomplished in all patients at the latest follow up, either at a normal or delayed time for age, except for one patient who suffers from cognitive and language delay accompanied by a behavioural disturbance, determined to be unrelated to treatment. Regarding motor abilities, we observed a stability of normality or a gain of normality in 5/8 patients in terms of total motor quotient (TMQ) and gross motor quotient (GMQ) at last follow-up, while fine motor quotients (FMQ) remained or became normal after treatment in 7/8 patients. After HSC-GT 7/8 patients continued to gain cognitive skills in line with typically developing children. All patients experienced an improvement or stabilization of the brain MRI scores with a significant reduction of the typical brain abnormalities. Compared to baseline, brain MRI images at last follow-up showed a meaningful reduction of multiple enlarged perivascular space and of subarachnoid space enlargement, regression of faint white matter alterations and normal progression of myelination.

Conclusions: This preliminary neurological analysis suggests biochemical detoxification in the CSF and initial clinical response in terms of motor and cognitive abilities up to 4 years after treatment, except for one patient with cognitive delay and behavioural disturbance determined to be unrelated to treatment. CSF detoxification suggests migration of gene-modified cells across the blood brain barrier and local engraftment in the CNS. Neuroradiological modifications correspond well with biochemical and clinical outcomes. A longer follow-up and a prospective comparison with HSCT are needed to draw definitive conclusions on the effect of HSC-GT on the typical neurological manifestations of MPSIH.

Clinical Trial Registry: NCT03488394 (https://www.clinicaltrials.gov/study/NCT03488394)

Disclosure: The study is sponsored by IRCCS San Raffaele Scientific Institute. Fondazione Telethon was the initial financial sponsor; since May 2019 Orchard Therapeutics Ltd (OTL) is the financial sponsor when MPSIH gene therapy was licensed to OTL. Fondazione Telethon and OSR are entitled to receive milestone payments and royalties for such a therapy from OTL. AA was the study PI between May 2018 and July 2020, MEB is the PI since July 2020. BG, RP and MEB have acted as ad hoc consultant for an Orchard Therapeutic advisory board in 2020.

6: Experimental Transplantation and Gene Therapy

Francesca Tucci¹, Giulia Consiglieri¹, Maurizio De Pellegrin², Matilde Cossutta¹, Chiara Filisetti¹, Marina Sarzana¹, Stefano Scarparo¹, Cristina Baldoli¹, Renata Mellone¹, Ludovica Santi², Stefania Crippa², Rossella Parini², Bernhard Gentner², Alessandro Aiuti¹, Maria Ester Bernardo ¹

¹IRCCS San Raffaele Scientific Institute, Milano, Italy, ²San Raffaele Telethon Institute for Gene Therapy, Milano, Italy

Background: Skeletal dysplasia is a prominent feature of Mucopolysaccharidosis type I Hurler syndrome (MPSIH) which is not fully addressed by allogeneic haematopoietic stem cell transplantation (HSCT). Autologous ex vivo haematopoietic stem cell gene therapy (HSC-GT) is a promising therapeutic strategy for MPSIH since it is capable of producing supraphysiologic IDUA levels, leading to extensive metabolic correction. Here we report the interim results from a Phase I/II clinical trial (NCT03488394) on the skeletal outcome of 8 MPSIH patients treated with HSC-GT (OTL-203) up to a median follow-up of 4 years. As there is limited knowledge of the pathogenesis of skeletal defects and skeletal correction mechanisms in MPSIH after HSC-GT, we have characterized in vitro the key players of skeletal remodelling before and after HSC-GT in patients and mice.

Methods: Skeletal outcome of the 8 MPSIH patients was assessed at baseline and specific timepoints after treatment by clinical (growth), functional (motor performance and joint range of motion) and radiological parameters (X-Rays and MRIs of the hip and spine) up to a median of 4 years after HSC-GT. Specific skeletal measures were retrospectively compared with an external cohort of HSCT patients transplanted in Monza. Patient-derived osteoclasts (OCs), mesenchymal stromal cells (MSCs) and osteoblasts (OBs) were isolated and characterized before and at several timepoints after HSC-GT.

Results: IDUA activity reached supraphysiologic levels in all patients after HSC-GT resulting in extensive substrate depletion, maintained at last follow-up. All patients exhibited longitudinal growth within expected reference ranges according to age and gender, with a median height gain greater than the one observed in Monza HSCT patients. All children progressively acquired motor skills (stability of Total Motor Quotient score over time) and experienced a complete and earlier normalization of joint mobility as compared with HSCT patients. Both hip X-ray and hip MRI parameters improved after HSC-GT, representing an amelioration of hip dysplasia up to the latest follow-up. A specific spine MRI score showed stabilization of typical spine MRI features (i.e. dorsolumbar kyphosis, foramen magnum, spinal stenosis) in all patients at last follow-up.

In the HSC-GT scenario, we hypothesized that OCs, deriving from genetically corrected haematopoietic precursors, could deliver the missing enzyme in the skeletal microenvironment as mediators of cross-correction of non-haematopoietic cells. We demonstrated that patient-derived OCs after HSC-GT and their supernatant show supraphysiologic IDUA activity which can be up-taken by patient MSCs and OBs, thus rescuing GAGs engulfment in MPSIH-OBs. Osteomedullary biopsies of MPSHI patients obtained at baseline showed foamy appearance of cytoplasm of the soft tissue cells adjacent to bone cells (i.e., entheses) which fully resolved after HSC-GT. Moreover, HSPC-GT ameliorated the skeletal disease of the growth plate in terms of organization and morphology in MPSI mice.

Conclusions: Clinical, functional and radiological parameters indicate an early beneficial effect of HSC-GT on MPSIH-typical skeletal features. A longer follow-up and direct comparison with HSCT are needed to draw definitive conclusions on HSC-GT impact on skeletal dysplasia. Biological studies aimed at unveiling skeletal correction mechanisms after HSC-GT will be key to advance the treatment of MPSIH patients.

Clinical Trial Registry: NCT03488394 (https://www.clinicaltrials.gov/study/NCT03488394)

Disclosure: The study is sponsored by IRCCS San Raffaele Scientific Institute. Fondazione Telethon was the initial financial sponsor; since May 2019 Orchard Therapeutics Ltd (OTL) is the financial sponsor when MPSIH gene therapy was licensed to OTL. Fondazione Telethon and OSR are entitled to receive milestone payments and royalties for such a therapy from OTL. AA was the study PI between May 2018 and July 2020, MEB is the PI since July 2020. BG, RP and MEB have acted as ad hoc consultant for an Orchard Therapeutic advisory board in 2020.

12: Graft-versus-host Disease – Clinical

Jose A. Perez-Simon ¹, Zachariah DeFilipp², Stephanie J. Lee³, Wendy Ingram⁴, Jennifer White⁵, Helene Schoemans^6,7, Laetitia Souchet⁸, Daniel Wolff⁹, Carrie L. Kitko¹⁰, Simona Sica^11,12, Avichai Shimoni¹³, Britnie Thomas¹⁴, Vedran Radojcic¹⁵, Chuan Tian¹⁴, Amandeep Salhotra¹⁶, Corey Cutler¹⁷

¹Hospital Universitario Virgen del Rocío Instituto de Biomedicina de Sevilla (IBIS), CSIC, University of Sevilla, Seville, Spain, ²Massachusetts General Hospital, Boston, United States, ³Fred Hutchinson Cancer Center, Seattle, United States, ⁴University Hospital of Wales, Cardiff, United Kingdom, ⁵The University of British Columbia, Vancouver General Hospital, Vancouver, Canada, ⁶University Hospitals Leuven, Leuven, Belgium, ⁷KU Leuven - University of Leuven, Leuven, Belgium, ⁸Clinical Hematology Unit, Groupe Hospitalier Pitié-Salpêtrière, AP-HP Sorbonne Université, Paris, France, ⁹University Hospital of Regensburg, Regensburg, Germany, ¹⁰Vanderbilt University Medical Center, Nashville, United States, ¹¹Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy, ¹²Università Cattolica Sacro Cuore, Rome, Italy, ¹³Chaim Sheba Medical Center, Tel-Hashomer and Sackler Medical School Tel-Aviv University, Tel-Aviv, Israel, ¹⁴Incyte Corporation, Wilmington, United States, ¹⁵Syndax Pharmaceuticals, Inc, Waltham, United States, ¹⁶City of Hope Medical Center, Duarte, United States, ¹⁷Dana-Farber Cancer Institute and Harvard Medical School, Boston, United States

Background: Chronic graft-versus-host disease (cGVHD) is a major cause of morbidity after allogeneic hematopoietic stem cell transplantation. Colony-stimulating factor 1 receptor (CSF-1R)-dependent monocytes and macrophages potentiate inflammation and fibrosis, which are key processes leading to multiorgan damage in cGVHD. Axatilimab, an investigational, high-affinity anti–CSF-1R monoclonal antibody, was evaluated in a pivotal, phase 2, open-label, randomized, multicenter study (AGAVE-201; NCT04710576), and primary results have been reported (Wolff D, et al. Blood. 2023;142(suppl_1):1). This analysis focuses on fibrosis-dominant organ-specific responses and related changes in patient-reported symptom burden. Safety is also examined, with an emphasis on infections.

Methods: Study design details have been reported previously. Organ-specific responses were measured using the 2014 National Institutes of Health cGVHD consensus guidelines. Modified Lee Symptom Scale organ subdomain analyses were performed to identify changes in symptom burden associated with fibrosis.

Results: Across all cohorts (n = 241), 193 patients (80%) had skin involvement (93% of whom had sclerotic skin), 162 (67%) had joints/fascia involvement, 108 (45%) had lung involvement, and 61 (25%) had esophageal involvement. Responses were seen in all involved organs, including complete responses (CRs). Notable efficacy was documented in the fibrosis-dominant organs across all doses with highest responses in the 0.3 mg/kg every 2 weeks (Q2W) cohort (Table). Median (range) time to response for fibrosis-dominant organs was concordant with the previously reported time to overall response, except for skin (0.3-mg/kg cohort, 3.7 [1.0–8.4] months) and lungs (0.3-mg/kg cohort, 2.9 [1.0–7.2] months), where a slight prolongation was observed; median (range) time to response for joints/fascia and esophagus in the 0.3-mg/kg cohort were 1.9 (1.0–9.5) and 1.9 (0.9–10.5) months, respectively. Lung responses (47% in the 0.3-mg/kg cohort), including CR, were based on both forced expiratory volume in 1 second and symptom score improvement criteria. Skin responses were accompanied by reductions in the percentage of sclerotic body surface area in almost half of patients (46% overall), and clinician-reported skin and joint tightening improvement was documented in the majority of patients (61% overall). A ≥ 2-point improvement in photographic range of motion score was recorded in all cohorts and was highest with the 0.3-mg/kg dose (55%), consistent with overall response results. The majority of patients with symptoms related to thickened skin reported improvement (72% overall). Adverse events in the overall population were mostly low grade, reversible, and increased with higher doses, with no unexpected safety signals. No cytomegalovirus or Epstein–Barr virus infections (including reactivations) and no invasive fungal infections occurred in the 0.3-mg/kg cohort; these events were infrequent in the higher-dose cohorts (1–4 patients in each cohort). COVID-19 infections occurred in 44 patients overall (18%); none were fatal.

Table. Responses to Axatilimab in Fibrosis-Dominant Organs.

Conclusions: In AGAVE-201, clinical activity in fibrosis-dominant organs is supported by clinician-reported changes in the majority of patients and patient-reported reductions in organ-specific symptom burden. Axatilimab had a generally well-tolerated safety profile; opportunistic infections, which are typically common in patients with cGVHD under heavy immunosuppression, were infrequent with axatilimab.

Clinical Trial Registry: NCT04710576

Disclosure: JAP-S served as a consultant for Incyte Corporation and Sanofi.

ZD served as a consultant for Incyte Corporation, Inhibrx, MorphoSys, Ono Pharmaceuticals, PharmaBiome AG, and Sanofi; received honoraria from Incyte Corporation and MorphoSys; and received research funding from Incyte Corporation, Regimmune, and Taiho Oncology.

SJL received research funding from Amgen, AstraZeneca, Incyte Corporation, Kadmon, Pfizer, and Syndax Pharmaceuticals; served as a consultant for Equillium, Incyte Corporation, Kadmon, Mallinckrodt, and Novartis; received study medication from Janssen; and served as a steering committee member for Novartis and Sanofi.

WI received honoraria from Medac Pharma and served as an advisory board member for Gilead, Novartis, Roche, Sobi, and Synairgen.

JW received honoraria from Novartis and Sanofi.

HS received personal fees from Incyte Corporation, Janssen, Novartis, Sanofi, and the Belgian Hematological Society (BHS) and research grants from Novartis and BHS, all paid to her institution. She also received nonfinancial support (travel grants) from Gilead, Pfizer, EBMT, and CIBMTR.

LS, SS, and A. Shimoni have no disclosures to report.

DW received honoraria from Behring, Gilead, Incyte Corporation, Mallickrodt, Novartis, Sanofi, and Takeda and received research funding from Novartis.

CLK served as an advisory committee member for Horizon Therapeutics and Incyte Corporation.

BT and CT are employees and shareholders of Incyte Corporation.

VR is an employee and shareholder of Syndax Pharmaceuticals.

A. Salhotra received research funding from Bristol Myers Squibb, Gilead, Jazz Pharmaceuticals, Kura Oncology, Orca Bio, and Rigel Pharmaceuticals; served as a speakers bureau member for Sanofi; and served as a board of directors/advisory committee member for Swedish Orphan Biovitrum.

CC served as a consultant for Astellas, Incyte Corporation, Inhibrx, Rigel Pharmaceuticals, Sanofi, and Syndax Pharmaceuticals; served as a data safety monitoring board member for AlloVir and Pluristem Therapeutics; and served as a board of directors/advisory committee member for Cimeio Therapeutics and Oxford Immune Algorithmics.

12: Graft-versus-host Disease – Clinical

Florent Malard ¹, Michael Loschi², Thomas Cluzeau², Faezeh Legrand³, Anne Huynh⁴, Sarah Guenounou⁴, Cécile Borel⁴, Jean-Baptiste Méar⁵, Faustine Lhomme⁵, Déborah Desmier⁶, Niels Moya⁶, Amandine Charbonnier⁷, Delphine Lebon⁷, Hélène Labussière-Wallet⁸, Martin Carré⁹, Jérôme Cornillon¹⁰, Vincent Camus¹¹, Patrice Ceballos¹², Francesco Saraceni¹³, Corentin Orvain¹⁴, Sylvain Chantepie¹⁵, Jakob Rudzki¹⁶, Marie-Anne Couturier¹⁷, Patrice Chevallier¹⁸, Clémence Mediavilla¹⁹, David Beauvais²⁰, Etienne Daguindau²¹, Karin Bilger²², Stefan Klein²³, Jaime Sanz²⁴, Sarah Altmeyer²⁵, Francesca Patriarca²⁶, Marion Bruelle²⁷, Emilie Plantamura²⁷, Mohamad Mohty¹

¹Saint-Antoine Hospital, Paris, France, ²CHU Nice, Nice, France, ³Institut Paoli Calmettes, Marseille, France, ⁴IUCT Oncopole, Toulouse, France, ⁵University Hospital of Rennes, Rennes, France, ⁶University Hospital of Poitiers, Poitiers, France, ⁷CHU Amiens-Picardie, Amiens, France, ⁸Hôpital Lyon Sud, Hospices Civils de Lyon, Pierre-Bénite, France, ⁹CHU Grenoble Alpes, Grenoble, France, ¹⁰CHU de St-Etienne, Saint Priest en Jarrez, France, ¹¹Centre Henri Becquerel, Rouen, France, ¹²Hôpital Saint Eloi, Montpellier, France, ¹³Ospedali Riuniti Torrette di Ancona, Ancona, Italy, ¹⁴CHU Angers, Angers, France, ¹⁵CHU Caen Normandie, Caen, France, ¹⁶Medizinische Universität Innsbruck, Innsbruck, Austria, ¹⁷CHRU Brest, Brest, France, ¹⁸Nantes University Hospital, Nantes, France, ¹⁹CHU Bordeaux, Bordeaux, France, ²⁰CHRU Lille, Lille, France, ²¹Hôpital Jean Minjoz, Besançon, France, ²²Institut de Cancérologie de Strasbourg, Strasbourg, France, ²³Medizinische Klinik Hämatologie und Onkologie Universitätsmedizin, Mannheim, Germany, ²⁴Hospital Universitario y Politécnico La Fe, Valencia, Spain, ²⁵Universitätsklinikum des Saarlandes, Homburg, Germany, ²⁶Azienda Sanitaria Universitaria Friuli Centrale, Udine, Italy, ²⁷MaaT Pharma, Lyon, France

Background: Fecal microbiotherapy has shown promising results in several pilot studies in aGvHD patients. Here we report clinical outcomes from 140 steroid-refractory (SR) or steroid-dependent (SD) GI-aGvHD patients treated with the pooled allogeneic microbiotherapy MaaT013 as part of the Early Access Program (EAP) in Europe.

Methods: 140 patients (including 2 pediatric patients aged 12 and 15) with SR/SD GI-aGvHD were treated with MaaT013 therapy as part of the EAP in Europe. These patients had previously failed 1 to 6 systemic lines. For each patient, a total of 3 MaaT013 administrations were planned every 7+/-2 days. Each dose is composed of 30 g of feces in 150 mL suspension from 4 to 8 healthy donors, administered by enema (except 1).Treatment response was calculated among all treated patients based on aGvHD staging and grading at D28 at the time of the EAP request.

Results: D28 GI-ORR was 52%: 39 complete response (CR,28%), 26 very good partial response (VGPR,19%), 8 partial response (PR,6%).

Administration of MaaT013 in the two pediatric patients was well tolerated. For one, treatment with MaaT013 led to CR of GI and skin symptoms for up to M12. Liver symptoms improved from M6 without additional therapy. For the other, VGPR was achieved for GI symptoms.

OS was 54% at M6, 47% at M12 and 42% at M18. OS was significantly higher in patients achieving at least GI-PR at D28 (Responder, R; n = 73) compared to patients in treatment failure (Non-responder, NR; n = 67): 74% versus 33% at M6, 68% versus 24% at M12, 58% versus 24% at M18 (p < 0.0001).

In 49 patients previously treated with ruxolitinib (2^nd line) and MaaT013 (3^rd line), D28 GI-ORR was 63%, with 49% CR. OS was 52% (M6), 49% (M12) and 42% (M18). OS was significantly higher in R, when compared to NR (76% versus 11% at M6,76% versus 6% at M12 and 64% versus 6% at M18 for R and NR respectively, p < 0.0001).

MaaT013 displays a good safety profile in the EAP population: 35 pharmacovigilance cases were reported in 33 patients, including 22 cases that could possibly be considered related to MaaT013 either by the physician or the company: sepsis in 6, bacteremia in 10, rectal bleeding/anorectal disorder in 3, C. difficile colitis in 1, E. coli osteoarthritis in 1, G. silvicola in stools in 1. No pathogen transmission was reported. In 2 patients, non-pathogenic commensal bacteria isolated following infectious events were detected in the administered MaaT013. Causality could not be formally excluded in these cases.

70 deaths were reported: GvHD in 28, severe infection in 24, relapse in 11, hemorrhage in 2, neurological complications post allo-HCT in 1, respiratory distress in 1, cardiac arrest in 2 and unknown for 1 patient. No causality link with MaaT013 administration has been identified.

Conclusions: Clinical data showed that MaaT013 was safe and effective for the treatment of SR/SD-GI-aGvHD, especially in patients having failed ruxolitinib. GI-response correlates with increased OS, suggesting a favorable benefit-risk profile for MaaT013. A Phase 3 trial is currently ongoing to confirm these results in ruxolitinib-refractory patients (NCT04769895).

Disclosure: MBR and EPL are MaaT Pharma’s employees.

12: Graft-versus-host Disease – Clinical

Ivan Moiseev ¹, Mikhail Drokov², Yulia Kitaeva³, Tatyana Rudakova¹, Maria Barabanshchikova¹, Ksenia Afanasyeva¹, Sergei Bondarenko¹, Larisa Kuzmina², Elena Morozova¹, Natalia Popova², Vera Vasilyeva², Yulia Vlasova¹, Dmitrii Zhogolev¹, Tatyana Konstantinova³, Elena Parovichnikova², Alexander Kulagin¹

¹RM Gorbacheva Research Institute, Pavlov University, Saint Petersburg, Russian Federation, ²National Research Center for Hematology, Moscow, Russian Federation, ³Regional Hospital #1, Ekaterinburg, Russian Federation

Background: Post-transplant cyclophosphamide (PTCy) is a widely used graft-versus-host disease (GVHD) prophylaxis in haploidentical and unrelated hematopoietic cell transplantation (HCT). Based on the results of a phase I/II study in myelofibrosis (Morozova et al., Acta Hematologica, 2021), it is hypothesized that the combination of PTCy and ruxolitinib may have comparable efficacy in controlling GVHD and better safety profile.

Methods: This multicenter randomized trial (PTCyRuxo, CINC424ARU01T, clinicaltrials.gov NCT04669210) enrolled 128 adult HCT recipients from Nov 2020 to Feb 2023. Inclusion and exclusion criteria defined study group as acute leukemia patients in complete remission without major co-morbidities or complications. Patients were randomized 1:1 with type of donor and Disease Risk Index as strata. Study procedures included FB2-FB4 conditioning according to centers’ standard operating procedures and GVHD prophylaxis with PTCY 50 mg/kg on days +3, + 4. Control group received tacrolimus from day 5 to 100, MMF 30 mg/kg/day from day 5 to 35 and PTCyRuxo group received ruxolitinib 15 mg/day on days 5-21, and 10 mg/day on days 22-150 (PTCyRuxo). Groups were balanced by major patient and HCT characteristics (Table).

Results: Median follow up was 16 months. There was no differences in the incidence of engraftment: 89% vs 98%, median day of engraftment - 19 in both groups (р=0.56). The study met its primary endpoint of acute GVHD grade II-IV non-inferiority: acute GVHD incidence 11.4% (95%CI 5.0-20.8%) in PTCyRuxo vs 19.7% (95%CI 11.1-30.1%) in Control group, non-inferiority p = 0.0031, superiority p = 0.1932. Also non-inferiority was demonstrated for PTCyRuxo prophylaxis of moderate and severe chronic GVHD: 24.8% (95%CI 14.7-36.3%) vs 42.0% (95%CI 28.6-54.9%), non-inferiority p = 0.0020, superiority p = 0.1220. Cumulative incidence of non-relapse mortality in PTCyRuxo and Control groups was 8.2% (95%CI 3.0-16.7%) vs 23.5% (95%CI 11.6-37.7%), р=0.0964, cumulative incidence of relapse – 18.3% (95%CI 9.3-29.8%) vs 21.5% (95%CI 6.7-41.8%), р=0.3633, respectively. Overall survival was 91.8% (95%CI 81.4-96.5%) vs 72.4% (95%CI 55.6-83.7%), р=0.0451, event-free survival - 73.5% (95%CI 59.8-83.2%) vs 62.8% (95%CI 44.9-76.4%), р=0.6922, GVHD-relapse-free survival – 63.4% (95%CI 49.2-74.6%) vs 46.1% (95%CI 29.1-61.6%), p = 0.2142, in PTCyRuxo and Control groups respectively. Reduced incidence of cytomegalovirus reactivation (45% vs 63%, р= 0.0439), grade ≥2 kidney toxicity (0% vs 8%, р= 0.0296) and reduced incidence of transplant-associated microangiopathy (0% vs 8%, р=0.0319) was observed in PTCyRuxo arm, while the incidence of other complications was comparable. The incidence of severe adverse events was 27% vs 42%, р= 0.0756. The pattern of immunologic recovery was different in PTCyRuxo group with more pronounced naïve T-cell depletion and higher prevalence of memory and effector CD8+ cells.

Conclusions: PTCyRuxo trial has met its primary endpoint of comparable GVHD control and secondary endpoint of improved safety profile. The final analysis of survival outcomes is planned after the end of follow up.

Clinical Trial Registry: NCT04669210, clinicaltrials.gov

Disclosure: The study was supported through Investigator Initiated Trial program by Novartis. The exploratory laboratory studies were supported by RSF grant №23-15-00327. COI: I.M.: recieved honoraria from Novartis, Sanofi, Takeda, J&J.

12: Graft-versus-host Disease – Clinical

Alexandros Spyridonidis ¹, Myriam Labopin², Bipin Savani³, Alexander Kulagin⁴, Didier Blaise⁵, Jurjen Versluis⁶, Simona Sica⁷, Anna Maria Raiola⁸, Jan Vydra⁹, Goda Choi¹⁰, Montserrat Rovira¹¹, Mi Kwon¹¹, Jaime Sanz¹², Maija Itäla-Remes¹³, Peter von dem Borne¹⁴, Rodrigo Martino¹⁵, Yener Koc¹⁶, Eolia Brissot¹⁷, Arnon Nagler¹⁸, Mohamad Mohty¹⁷, Fabio Ciceri¹⁹

¹University of Patras, Patras, Greece, ²Sorbonne University, Saint-Antoine Hospital, Paris, France, ³Vanderbilt University Medical Center, Nashville, United States, ⁴RM Gorbacheva Research Institute, Pavlov University, St. Petersburg, Russian Federation, ⁵Programme de Transplantation & Therapie Cellulaire, Marseille, France, ⁶Erasmus MC Cancer Institute, Rotterdam, Netherlands, ⁷Universita Cattolica S. Cuore, Rome, Italy, ⁸IRCCS Ospedale Policlinico San Martino, Genova, Italy, ⁹Institute of Hematology and Blood Transfusion, Prague, Czech Republic, ¹⁰University Medical Center Groningen, Groningen, Netherlands, ¹¹Hospital General Universitario Gregorio Marañon, Madrid, Spain, ¹²IRCCS Ospedale Policlinico San Martino, Valencia, Spain, ¹³Turku University Hospital, Turku, Finland, ¹⁴Leiden University Hospital, Leiden, Netherlands, ¹⁵Hospital Santa Creu i Sant Pau, Barcelona, Spain, ¹⁶Medicana International Hospital Istanbul, Istanbul, Turkey, ¹⁷Hospital Saint Antoine, Paris, France, ¹⁸Chaim Sheba Medical Center, Tel-Hashomer, Israel, ¹⁹IRCCS Ospedale San Raffaele s.r.l., Milano, Italy

Background: Post-transplant cyclophosphamide (PTCY) is increasingly used as an effective GvHD prophylaxis in allogenic HCT. However, PTCY is associated with specific organ toxicities. Whether patients with specific comorbidities are more vulnerable to cyclophosphamide-induced toxicity is unclear.

Methods: We evaluated the impact of individual comorbidities for NRM risk in an ALWP-EBMT registry cohort of adult AML patients who underwent between 2010 and 2023 an allogenic HCT (cord blood excluded) with PTCY (in vivo and ex vivo TCD excluded). Comorbidities were defined per standard definitions in the HCT-CI, whereas cardiac comorbidity included also valvular disease and arrhythmia. Outcomes were analyzed using Cox-proportional hazard models.

Results: A total of 5883 AML patients (CR-1 64%, CR2 + 17%, advanced 20%) median age of 58 years (interquartile range, 43-64) received PTCY as GVHD prophylaxis after haploidentical/UD/MSD (61%/28%/11%), RIC/MAC (49%/51%), PBSC/BM (81%/19%) transplantation. 75% patients (n = 4390) had no comorbidities, while 21% (n = 1229) of them had 1 and 4.5% (n = 269) had 2 comorbidities. The most frequently comorbidities were moderate or severe pulmonary dysfunction (15%, n = 870 and 10% n = 578), cardiac disease (9.4%, n = 555), active infection (9%, n = 519), prior malignancy (8%, n = 466) and mild or severe hepatic dysfunction (4.7%, n = 275 and 1.4%, n = 80). NRM incidence was 18.2% (17.2-19.3) and the most frequently NRM causes were multiorgan failure (28%, n = 308), infection (27%, n = 294) and GvHD (11%, n = 122). Cardiac toxicity was identified as primary cause of NRM in 17 cases (1.5%). In a multivariate analysis which included 7 individual comorbidities selected in a stepwise model using the lower Akaike criteria, 5 of them (cardiac, kidney, severe hepatic, severe pulmonary, infection) were independently associated with adverse transplant outcomes (NRM and OS) without influencing the REL rate. Besides individual organ dysfunction, NRM was also significantly influenced by incremental age (per 10years), KPS < = 80%, active disease or infection at transplant and use of an haploidentical donor. Regarding specific organs, cardiac disease was associated with an increased NRM risk in univariate and multivariate analysis (hazard ratio [HR], 1.47; 95% confidence interval [CI], 1.22-1.77, p < 0.0001) and reduced OS (HR, 1.3; 1.3-1.5, p = 0.0002). Severe hepatic or severe pulmonary were also independently associated with NRM (HR 1.64; 1.02-2.62 and 1.22; 1-1.49, respectively) and OS (HR 1.6; 95% CI, 1.13-2.26 and 1.22; 95% CI, 1.06-1.4, respectively). Renal disease, albeit rare (0.9%, n = 54), remained significant for NRM (HR, 2.27; 95% CI, 1.45-3.56, p = 0.0004) and OS (HR, 1.69; 95% CI, 1.16-2.46, p = 0.006). Diabetes and obesity, though prevalent (5% and 4.5%, respectively), were not associated with increased NRM. A simplified model using the absence (n = 4390), presence of 1 (n = 1229) or presence of 2 or 3 (n = 269) of the 5 comorbidities which were determined individually to contribute to NRM, could stratify patients into 3 NRM risk (16.2%, 21.6%, 36%, respectively, p = 0.001) and OS categories (64%, 56%, 36.4%. respectively, p = 0.001).

Conclusions: Individual dysfunction of key organs (cardiac, pulmonary, kidney) confers significant additional NRM risk in PTCY treated patients. In contrast, diabetes or obesity should not preclude for using PTCY. Programs of PTCY dose modulation should be promoted in patients with selected comorbidities.

Clinical Trial Registry: EBMT Registry study, approved by the ALWP of EBMT

Disclosure: Nothing to declare related to this study. Mi Kwon decalred following COIs: speaker for Gilead, Jazz, Pfizer.

12: Graft-versus-host Disease – Clinical

Arnon Nagler ¹, Myriam Labopin^2,3, Thomas Schroeder⁴, Rose-Marie Hamladji⁵, Laimonas Griskevicius⁶, Urpu Salmenniemi⁷, Alessandro Rambaldi⁸, Stephan Mielke⁹, Sergey Bondarenko¹⁰, Jakob Passweg¹¹, Peter Dreger¹², Tobias Gedde-Dahl¹³, Edouard Forcade¹⁴, Grzegorz Helbig¹⁵, Matthias Stelljes¹⁶, Cristina Castilla-Llorente17¹⁷, Alexandros Spyridonidis¹⁸, Bipin Savani¹⁹, Eolia Brissot³, Fabio Ciceri²⁰, Mohamad Mohty^2,3

¹Sheba Medical Center, Division of Hematology, Tel Hashomer, Israel, ²EBMT Paris study office; Saint Antoine Hospital, Paris, France, ³Sorbonne University, Saint-Antoine Hospital, AP-HP, INSERM UMRs 938, Paris, France, ⁴University Hospital of Essen, Essen, Germany, ⁵Centre Pierre et Marie Curie, Alger, Algeria, ⁶Vilnius University, Hospital Santaros Klinikos, Vilnius, Lithuania, ⁷HUCH Comprehensive Cancer Center, Helsinki, Finland, ⁸University of Milan and Azienda Socio-Sanitaria Territoriale, Papa Giovanni XXIII, Bergamo, Italy, ⁹Karolinska University Hospital, Stockholm, Sweden, ¹⁰RM Gorbacheva Research Institute, Pavlov University, Petersburg, Russian Federation, ¹¹University Hospital of Basel, Basel, Switzerland, ¹²University of Heidelberg, Heidelberg, Germany, ¹³Oslo University Hospital, Rikshospitalet,, Norway, ¹⁴CHU Bordeaux, Hôpital Haut-Leveque, Pessac, France, ¹⁵Silesian Medical Academy, Katowice, Poland, ¹⁶University of Muenster, Muenster, Germany, ¹⁷Gustave Roussy Cancer Campus, Villejuif, France, ¹⁸University Hospital of Patras, Patras, Greece, ¹⁹Vanderbilt University, Medical Center, TN, United States, ²⁰IRCCS Osspedale San Raffaele, Vita-Salute San Raffaele University Hematology and BMT, Milano, Italy

Background: The BMT CTN 1703 phase III randomized study compared post-transplant cyclophosphamide with tacrolimus with mycophenolate mofetil (PTCy/TAC/MMF) to TAC /methotrexate (MTX) as GVHD prophylaxis post allogeneic transplantation (HSCT) from matched siblings (MSD) or 7-8/8 unrelated donors (MUD), demonstrating a lower incidence of severe acute (a) and chronic (c) GVHD and better GRFS. However, the control arm did not contain ATG, and CSA was not included.

Methods: The study aim was to compare PTCy with TAC or cyclosporine A (CSA) and MMF (PTCy/TAC or CSA & MMF) vs. ATG with TAC or CSA and MTX (ATG/TAC or CSA & MTX) in AML patients (pts) undergoing MSD or 9-10/10 MUD HSCT in CR1. Statistical tests included a multivariate analysis (MVA) using a Cox proportional-hazards regression model for main outcomes. A propensity score matching was also performed as a sensibility analysis.

Results: 6050 pts met the inclusion criteria, 402 received PTCy/TAC or CSA & MMF and 5648 received ATG/TAC or CSA & MTX. Follow-up was 23.4 (IQR, 20.3-24.9) and 41.8 (IQR, 39.6-43.3) months (p < 0.0001). Pts in the PTCy/TAC or CSA & MMF group were younger, with a median age of 48.7 (range 18-5.6) vs. 51.5 (8-77.8) years (y) (p = 0.024). The cytogenetic risk category did not differ between pts in the PTCy/TAC or CSA & MMF and ATG/TAC or CSA & MTX groups, respectively (p = 0.35). More pts in the PTCy/TAC or CSA & MMF group received RIC 51.5% vs. 41.1% in the ATG/TAC or CSA & MTX group, respectively (p < 0001). Neutrophil engraftment (ANC > 0.5 x 10⁹/L) was 98.7% vs. 98.6%, respectively (p = 0.84). Day 180 incidence of aGVHD grade II-IV and III-IV was 21.2% vs. 20.4% (p = 0.92) and 8.1% vs. 6% (p = 0.1), in pts receiving PTCy/TAC or CSA & MMF vs. the ATG/TAC or CSA & MTX, respectively. The 2- y total and extensive cGVHD were 33.7% vs. 30% (p = 0.09) and 10.7 % vs. 11.2%, respectively (p = 0.81). GVHD was the cause of death in 11.6% vs. 13.9% of pts who died. In the MVA, both aGVHD grade II-IV or III-IV) and cGVHD (total or extensive) did not differ between the groups with hazard ratios for PTCy/TAC vs CSA & MMF (HRs) = 0.95 (95% CI 0.71-1.27, p = 0.71), HR = 1.38 (95% CI 0.88-2.16, p = 0.16), HR = 1.2 (95% CI 0.91-1.56, p = 0. 19 and HR = 0.66 (95% CI 0.41-1.04, p = 0.074). Data were confirmed by the pair match analysis. The HRs for NRM and RI were 0.70 (95% CI 0.47-1.07, p = 0. 010) and 1.10 (95% CI 0.83-1.46, p = 0. 50). The HRs for LFS, OS, and GRFS were HR = 0.96 (95% CI 0.76-1.21, p = 0.71), HR = 0.93 (95% CI 0.72-1.21, p = 0.59) and HR = 1.03 (95% CI 0.84-1.26, p = 0.76).

Conclusions: In this retrospective analysis, comparing PTCy in combination with TAC or CSA and MMF to ATG in combination with TAC or CSA and MTX as GVHD prophylaxis, we observed a similar incidence and severity of both aGVHD and cGVHD. Likewise, NRM, RI, LFS, OS, and GRFS did not differ between the two study groups.

Disclosure: Nothing to declare

12: Graft-versus-host Disease – Clinical

Anne B. Verbeek ¹, Erik G J. von Asmuth¹, Erik B. van den Akker¹, Anja M. Jansen-Hoogendijk¹, Marco W. Schilham¹, Arjan C. Lankester¹, Gertjan Lugthart¹, Alexander B. Mohseny¹, Emilie P. Buddingh¹

¹Leiden University Medical Center, Leiden, Netherlands

Background: In this explorative study we aimed to identify inflammatory serum proteins measured before allogeneic hematopoietic stem cell transplantation (HSCT) that are associated with acute Graft-versus-Host Disease (aGvHD) and mortality in pediatric HSCT recipients.

Methods: In this single-center retrospective cohort study, we measured 28 inflammatory serum proteins using a Luminex immunoassay in 384 pediatric patients (2010-2022). A sample before start of the conditioning (TP1) was included, as well as a sample on the day of HSCT (TP2). For patients who developed aGvHD we also included a sample at the time of diagnosis, before initiation of systemic treatment (TP3). Associations with aGvHD, steroid-refractory aGvHD, non-relapse mortality (NRM) and overall survival (OS) were analyzed using robust (cause-specific) Cox models. A sensitivity analysis for underlying disease subgroups was performed. We evaluated the dynamics in serum protein levels before and after conditioning. All reported p-values were significant (p ≤ 0.05) after multiple testing correction (Benjamini-Hochberg).

Results: We included 384 children (age range 0.2-18.6 years) with malignant (30%) and non-malignant (70%) indications for allogeneic HSCT. Patient and transplant characteristics are shown in Table 1. At 1-year post-HSCT OS was 86%, NRM was 10% and the incidence of aGvHD was 18%.

At TP1, TNFR1 (cause-specific hazard ratio (csHR) 1.61, p < 0.001), sIL2-Rα (csHR 1.54, p < 0.001) and TNFR2 (csHR 1.53, p < 0.001) were highly associated with the development of aGvHD. CXCL16, TNFR1 and SCF at TP1 were highly associated with NRM and OS. The associations of serum proteins with outcomes were not driven by one specific underlying disease subgroup.

At the onset of aGvHD (TP3) ST2 levels were significantly associated with the development of steroid-refractory aGvHD (csHR 2.21, p < 0.001, AUC 83.6), as previously reported in biomarker studies in predominantly adult HSCT populations.

Conclusions: Pre-conditioning inflammatory serum protein levels are associated with adverse outcomes after HSCT in children. To our knowledge, this is the largest aGvHD biomarker study in a pediatric HSCT cohort. TNFR1, sIL2-Rα and TNFR2 before conditioning had the strongest association with aGvHD and CXCL16, TNFR1 and SCF were most-strongly associated with NRM and OS. After confirmation of these findings by future studies, the potential prognostic biomarkers reported here might be used to select pediatric patients pre-HSCT for alternative prophylactic aGvHD strategies. ST2 levels at aGvHD diagnosis is a validated predictive biomarker for steroid-refractory aGvHD in children.

Table 1: Patient and Transplant characteristics

Disclosure: Nothing to declare

12: Graft-versus-host Disease – Clinical

Javier Delgado Serrano ^1,2,3, Lucía López Corral^4,5, David Valcárcel^6,7, Jorge Sierra⁸, Christelle Ferra-Coll⁹, Teresa Caballero Velázquez^1,2,3, Francisco Martín Domínguez^1,2,3, Reyes Jiménez León^1,2,3, Juan Luis Reguera Ortega^1,2,3, Cristina Blázquez Goñi^1,2,3, Estefanía Pérez López^4,5, Almudena Cabero Martínez^4,5, Irene García Cadenas⁸, José Antonio Pérez Simón^1,2,3

¹Hospital Universitario Virgen del Rocío, Sevilla, Spain, ²Instituto de Biomedicina de Sevilla, Sevilla, Spain, ³Universidad de Sevilla, Sevilla, Spain, ⁴Hospital Universitario de Salamanca, Salamanca, Spain, ⁵IBSAL, Salamanca, Spain, ⁶Vall Hebron Institute of Oncology, Barcelona, Spain, ⁷Vall Hebron University Hospital, Barcelona, Spain, ⁸Hospital Santa Creu i Sant Pau, Barcelona, Spain, ⁹Hospital Universitari Germans Trias i Pujol, Badalona, Spain

Background: Most strategies of graft-versus host disease prophylaxis (GvHD) are based on early postrasplant interventions; nevertheless, the pathophysiology of chronic GvHD involves different cell subpopulations that remain abnormal at the long term and depend on nuclear factor kappa-beta (NF-kb) for their activation and, therefore, can be targeted using proteosome inhibitors. The aim of this study is to demonstrate the safety and efficacy of ixazomib administered from day +100 in the prevention of moderate or severe cGVHD.

Methods: We performed a prospective, randomised, multicentre trial including 73 patients undergoing allogeneic hematopoietic stem cell transplantation (allo-HSCT) between 2019-22. Patients in the treatment arm received 4 mg of oral Ixazomib (Ixa) on days +1, +8 and +15 from +d100 every 28 days up to a maximum of 15 cycles. Immunological characterisation by flow cytometry was performed on peripheral blood at days +100, +180, +270, +365, +540 and +720.

Results: Thirty-nine patients were randomized to the treatment arm and thirty-four to the control arm. There were no significant differences between the two groups in terms of baseline characteristics, conditioning, acute GvHD (aGvHD) prophylaxis or history of previous aGVHD.At a median follow-up of 24 months, significant differences were observed in the 2-year cumulative incidence of moderate-severe cGvHD between the Ixa and control groups (13% vs 43% respectively, HR = 0.23, 95% CI = 0.07-0.071, p = 0.001), as well as overall cGvHD (44% vs 63% respectively, HR = 0.52, 95% CI = 0.27-1.01, p = 0.048). There were no significant differences in the risk of relapse, although there was a trend in favour of Ixa (14% vs 27%, p = 0.22). Event and GvHD-free survival (EGFS) was significantly better in the Ixa group (81% vs 51%, HR = 0.30, 95% CI = 0.12-0.72, p = 0.007). The main toxicity associated with ixazomib was digestive, observed in 36% vs 3% in the first 3 cycles (p = 0.001) and 25% vs 3% in cycles 4-9 (p = 0.013).On day +180 median fluorescence intensity (MFI) for Stat3 phosphorylation (pStat3) in IL-6 stimulated CD4 + T cells (HR = 6.78, p = 0.015), MFI for pStat3 in stimulated CD8 + T cells (HR = 5.4, p = 0.029), the proportion of B cells that have undergone class switching (SMBC) (HR = 12.8, p = 0.016) and the presence of circulating plasma cells (HR = 4.98, p = 0.042) were significantly related to the risk of developing moderate/severe cGvHD. In addition, the increase of expression of BAFF receptor (ΔBAFFR) (MFI on day 180 respect day 100) in B cells was significantly higher in patients with moderate/severe cGvHD (HR = 8.81, p = 0.042). The proportion of SMBC (16.94 vs 27.35, p = 0.012), ΔBAFFR (-1334.96 vs 1230.27, p = 0.009) and the concentration of soluble BAFF (ng/ ml) (0.03 vs 1.21, p = 0.013) were significantly different between Ixa and control group.

Table 1: Predictive biomarkers of moderate-severe chronic GVHD on day + 180 after allogeneic transplant

Conclusions: The administration of Ixazomib from day +100 post-allogeneic transplant prevents the development of cGvHD, especially in its moderate or severe form, improving EGFS and cGvHD. Several biomarkers related to T and B cells activation were associated with an increased risk of developing cGvHD.

Clinical Trial Registry: NCT03225417

https://ichgcp.net/es/clinical-trials-registry/NCT03225417

Disclosure: No conflict of interest

12: Graft-versus-host Disease – Clinical

Virginia Escamilla Gómez ¹, Beatriz Astibia Mahillo², Patricia Alcalde Mellado¹, Francisco Manuel Martín Domínguez¹, Marta González Vicent³, Pedro Asensi Cantó⁴, Juan Montoro Gómez⁴, Inés Gómez Seguí⁴, María Isabel Benítez Carabante⁵, Cristina Díaz de Heredia Rubio⁵, Luisa Sisinni⁶, Antonio Pérez Martínez⁶, José María Pérez Hurtado¹, José Antonio Pérez Simón¹, Valentín García Gutiérrez²

¹Hospital Universitario Virgen del Rocío, Sevilla, Spain, ²Hospital Universitario Ramón y Cajal, Madrid, Spain, ³Hospital Niño Jesús, Madrid, Spain, ⁴Hospital Universitario y Politécnico La Fe, Valencia, Spain, ⁵Hospital Universitari Vall d´Hebron, Vall d’Hebron Institut de Recerca (VHIR), Barcelona, Spain, ⁶Hospital La Paz, Madrid, Spain

Background: The JAK1/2 inhibitor ruxolitinib has been approved for the treatment of adults and youngers above ≥12 years for acute or chronic graft-versus-receptor disease (GvHD) based on the REACH-2 and 3 studies. Recently, the REACH-4 trial evaluates its effectiveness in the pediatric population; however, experience in this population is scarce.

Methods: Analysis of 42 pediatric patients ( < 14 years) with acute or chronic GvHD refractory to corticosteroids through a retrospective, descriptive and multicenter study from 17 Spanish centers belonging to the Spanish Group of Transplantation and Cellular Therapy (GETH-TC) between October/2015 and September/2022. The severity of acute GvHD was evaluated using the Harris scale; for chronic GvHD, NIH consensus scale was used.

Results: The baseline characteristics and decrease in immunosuppression are shown in table 1 and figure 1.

Acute GvHD group. 29 patients were analyzed. The overall response (OR) was 82.7 % (51.7% complete remissions, CR). The median time to achieve maximum response was 4 weeks (range: 1-16). The responses by organ were: skin 88.8% (CR: 66.6%), gastrointestinal tract 69.5% (CR: 56.5%) and liver 66.6% (CR: 33.3%). No differences were observed in terms of response depending on the previous lines of treatment. 11 patients died due to the following causes: GVHD progression 3, relapse 2, infection 3, others 3. Overall, 3 patients relapsed from their underlying disease.

Among responders, 66.6% and 75% were able to withdraw corticosteroids and ruxolitinib, respectively. The 6-month OS between responders vs non-responders was 77.4% (95% CI: 53.6-90) vs 20% (95% CI: 0-58.2), p = 0.00.

Chronic GvHD group. 13 patients were analyzed. The OR and CR were 100% and 23%, respectively. By organs, all patients with gastrointestinal involvement achieved CR, and all patients with lung involvement and scleroderma achieved partial response. No differences were observed in terms of response depending on the previous lines of treatment. A single patient died due to relapse of his underlying hematological disease. In total, 2 patients relapsed in this group.

Among responders, 30.7% were able to withdraw corticosteroids and 30.7% were able to withdraw ruxolitinib. The 2-year OS was 100%.

Table 1. Patients characteristics

Table 1. Baseline characteristics of the study population. AA: aplastic anemia; ALL: acute lymphoblastic leukemia; AML: Acute myeloblastic leukemia; CR: complete remission; MDS: myelodysplastic syndrome; PR: partial response; SD stable disease.

Figure 1. Evolution of Immunosuppression among responding patients. The figure shows the total number of responding patients taking each drug at the beginning of treatment with ruxolitinib and at the end of treatment or at the last visit.

Conclusions:

• Real-life ruxolitinib data show very favorable results in the pediatric population.

• By organs, the results in acute GvHD are very favorable in the skin and gastrointestinal tract. In chronic GvHD all patients responded, achieving CR rates of 100% at the gastrointestinal level.

• These results seem not to be influenced by lines of treatment (<2 vs ≥2; <3 vs ≥3).

• In the group of responders, a reduction in the number of patients taking immunosuppressants was achieved at the end of treatment.

Disclosure: *This project has been carried out through the GETH-TC (Spanish Group for Transplantation and Cellular Therapy - Grupo Español de Trasplante Hematopoyético y Terapia Celular) and funded by Novartis.

12: Graft-versus-host Disease – Clinical

Carlos Echecopar Parente¹, David Bueno Sánchez¹, Blanca Molina Angulo², María Isabel Benitez Carabante³, Iván López Torija⁴, María Trabazo⁵, Carolina Fuentes⁶, Mercedes Plaza⁷, Antonia Pascual Martínez⁸, Andrea Urtasun⁹, Susana Riesco¹⁰, Soledad González Muñíz¹¹, Sara Vinagre², Cristina Díaz de Heredia³, Julia Marsal ⁵, Marta González-Vicent², Antonio Pérez-Martínez^1,12,13

¹La Paz University Hospital, Madrid, Spain, ²Niño Jesús University Hospital, Madrid, Spain, ³Vall d´Hebron University Hospital, Barcelona, Spain, ⁴Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, ⁵Sant Joan de Deu University Hospital, Barcelona, Spain, ⁶La Fe University Hospital, Valencia, Spain, ⁷Hospital Clínico Universitario Virgen de la Arrixaca, Murcia, Spain, ⁸Hospital Regional Universitario de Málaga, Málaga, Spain, ⁹Clínica Universidad de Navarra, Pamplona, Spain, ¹⁰Complejo Asistencial Universitario de Salamanca, Salamanca, Spain, ¹¹Hospital Universitario Central de Asturias, Oviedo, Spain, ¹²University Hospital La Paz Research Institute (IdiPAZ), Madrid, Spain, ¹³Universidad Autónoma de Madrid, Madrid, Spain

Background: Graft-versus-host disease (GVHD) contributes to the elevated morbidity and mortality rates in hematopoietic stem cell transplantation. Mesenchymal stromal cells (MSCs) is a cell therapy used in several immune-mediated diseases, such as GVHD. Our objective was to evaluate the efficacy of MSCs in pediatric GVHD and identify factors associated with treatment response.

Methods: This is a retrosprospective, multicenter study that includes pediatric patients (0-18 years old) who have received MSCs for the treatment of GVHD (both acute and chronic) in spanish centers between June 2014 and July 2023.

Results: A total of 102 patients from 11 different centers were included in the study, 78 patients (76.5%) had acute GVHD, and 24 (23.5%) chronic GVHD.The median follow-up duration was 11 months. Overall response rate was 76.5% at day 28 [30.4% complete response (CR) and 46.1% partial response (PR)] without statistically significant differences between the type of GVHD. Response was associated with the dosage used in the infusions (p = 0.007), showing a CR rate of 37.3% in patients receiving ≥ 1x10e⁶ cells/kg vs a CR rate of 11.1% in patients receiving < 1x10e⁶ cells/kg (Table 1). Other treatment-related variable associated with response was MSCs source. There were no progressions in the group of patients who received MSCs extracted from Wharton’s jelly, whereas in patients who received MSCs from bone marrow and adipose tissue the progresion rate was 16.4% and 25%, respectively (p = 0.017) (Table 1). Finally, the total number of MSC infusions demonstrated a statistically significant association with the response level (p = 0.011). Patients receiving less than four infusions showed a CR rate of 14.3%, compared to 38.9% for those receiving a total of four infusions and 25.9% for those receiving more than four. The progression rate was 23.8% in the group with fewer than four infusions, 3.7% in the group with four, and 18.5% in the group with more than four infusions (p = 0.011) (Table 1). In acute GVHD, the organ that showed better results was the skin, with a complete response rate of 71.9% and no progressions. Response to MSCs, infusion dosage, and total number of infusions were correlated with three-year overall survival and transplant-related mortality.

TABLE 1. Response level at day 28 based on treatment-related variables

Conclusions: MSCs represent a potentially effective treatment for pediatric GVHD, with promising results in this retrospective analysis. Treatment-related factors like MSCs dosage, source and total of infusions might be related to response to this therapy. The results of this study needs to be confirmed and validated through multicenter prospective randomized studies.

Disclosure: Nothing to declare

12: Graft-versus-host Disease – Clinical

Margarita Perminova¹, Larisa Shelikhova¹, Maria Klimentova¹, Maria Dunaykina¹, Anna Vasilieva¹, Sergey Blagov¹, Svetlana Kozlovskaya¹, Dmitriy Balashov¹, Alexander Popov¹, Daria Kobyzeva¹, Natalya Miakova¹, Yakov Muzalevsky¹, Galina Novichkova¹, Alexey Maschan¹, Michael Maschan¹, Alexandra Laberko ¹

¹Dmitry Rogachev Federal Research Centre of Pediatric Hematology, Oncology and Immunology, Moscow, Russian Federation

Background: GvHD remains a crucial complication associated with significant morbidity and mortality after allogeneic HSCT. High-dose PTCy targets alloreactive donor T cells proliferating early after HSCT, promotes regulatory T cells, and prevents severe GvHD. The effectiveness of PTCy-based regimens varies depending on the graft source and intensity of the preparative regimen. Improvement of GvHD control with novel targeted agents, added to the backbone of PTCy, could further increase the safety of HSCT and reduce the morbidity and complications of standard regimens of pharmacological immune suppression. In this pilot trial, we evaluated the safety and efficacy of abatacept and vedolizumab in unmanipulated haploidentical transplantation among children with leukemia.

Methods: A total of 58 pts with acute leukemia (AML- 20, ALL-38, 26 females, 30 males, the median age at HSCT - 7,4 years), underwent unmanipulated allogeneic BM (n = 56) or PBSC (n = 2) transplantation between March 2022 and May 2023. Fifty-four pts received haploidentical graft, 4 a graft from MUD. Disease status at transplant was CR1 in 27 pts, CR > 2 in 31 pts. For all patients it was the 1-st transplantation. All patients received myeloablative conditioning, either TBI-based (n = 31) or treosulfan-based (n = 27). Prophylaxis of GvHD consisted of CsA since day-1 to day 180, 50 mg/kg of PTCy on days +3 and +4, abatacept at 10 mg/kg on days +5, +14, +28, +45, +60 and vedolizumab at 10 mg/kg (capped at 300 mg) on days -1, +14, +28. The median dose of CD34+ cells was 4,79 x106/kg, and the CD3 dose was 42 x106/kg.

Results: Primary engraftment was achieved in 57 (98%) pts; the median time to neutrophil and platelet recovery was 20 and 23 days, respectively. One patient failed to engraft, relapsed, and required rescue with a second HSCT from an alternative donor. The regimen was generally well tolerated. Acute toxicity consisted of grade 3-4 mucositis (n = 10); hemorrhagic cystitis grade II-III (n = 4); severe VOD (n = 2) (all cases of VOD developed after pre-HSCT use of inotozumab); bloodstream infections (n = 8).

The median follow-up time for survivors was 12,5 months. TRM was 3,8% (95% CI:1-15): two patients died (one due to consequences of severe VOD and infection on day + 60, one due to the infection and GvHD on day +268). 19 patients had aGvHD, the CI of aGvHD grade II-IV was 33% (95% CI, 23 - 47), and grade 3-4 was 9% (95% CI, 4 - 20). The CI of gut aGvHD grade II-IV was 10% (95% CI, 4 - 24), all pts had norovirus infection. The CI of ChGvHD was 8% (95% CI, 3 - 21). The CI of relapse was 12% (95%CI:6-25) in the entire cohort. EFS and OS were 84% (95%CI: 74-94) and 94% (95%CI: 87-100), respectively. The GRFS (survival without aGVHD3-4, extensive chGvHD and relapse) was 73% (95%CI:62-85).

Conclusions: Our data suggest that adding vedolizumab and abatacept to the PTCy-based GVHD prophylaxis backbone is well tolerated and ensures high engraftment rate and low transplant-related mortality. This approach can be further developed to reduce exposure to calcineurin inhibitors and deploy additional anti-leukemia effectors after HSCT.

Clinical Trial Registry: NCT05515029

Disclosure: M. Maschan - Miltenyi Biotec Honoraria

D. Balashov - Octafarm - Lecturers fee

12: Graft-versus-host Disease – Clinical

Antoine Capes ¹, Jarl Mooyaart², Didier Blaise³, Stefania Bramanti⁴, Mi Kwon⁵, Mohamad Mohty¹, Patrice Chevallier⁶, Jan Vydra⁷, Péter Reményi⁸, Edouard Forcade⁹, Lucía López Corral¹⁰, Maija Itäla-Remes¹¹, Ali Bazarbachi¹², Enrico Derenzini¹³, Jorinde D. Hoogenboom¹⁴, Jürgen Kuball¹⁵, Giorgia Battipaglia¹⁶, Florent Malard¹, Annalisa Ruggeri¹⁷

¹Hôpital Saint-Antoine, APHP, Paris, France, ²EBMT Statistical Unit, Leiden, Netherlands, ³Programme de Transplantation & Therapie Cellulaire, Marseille, France, ⁴IRCCS Humanitas Research Hospital, Milano, Italy, ⁵Hospital Gregorio Marañón, Madrid, Spain, ⁶CHU Nantes, Nantes, France, ⁷Institute of Hematology and Blood Transfusion, Prague, Czech Republic, ⁸Dél-pesti Centrumkórház, Budapest, Hungary, ⁹CHU Bordeaux, Hopital Haut-Leveque, Bordeaux, France, ¹⁰Hospital Clínico, Salamanca, Spain, ¹¹Turku University Hospital, Turku, Finland, ¹²American University of Beirut Medical Center, Beirut, Lebanon, ¹³European Institute of Oncology, Milano, Italy, ¹⁴EBMT Leiden Study Unit, Leiden, Netherlands, ¹⁵University Medical Center Utrecht, Utrecht, Netherlands, ¹⁶Federico II University of Naples, Naples, Italy, ¹⁷San Raffaele Scientific Institute, Hematology and Bone Marrow Transplantation Unit, Milan, Italy

Background: The use of haploidentical hematopoietic cell transplantation (haplo-HCT) with peripheral blood stem cells (PBSCs) to treat high risk hematological malignancies is increasing worldwide. In this setting, prevention of graft-versus-host disease (GVHD) with post-transplant cyclophosphamide (PTCy) has emerged as a gold standard. Some centers add anti-thymocyte globulin (ATG) to PTCy aiming to reduce the risk of GVHD. In a few retrospective studies this approach decreased the GVHD incidence with no impact on overall survival (OS) or disease-free survival (DFS). However, those studies were limited to acute myelogenous leukemias in first or second remission and included only a small number of patients in the PTCy+ATG group. Here, we sought to compare patients’ outcomes after PBSC haplo-HCT using PTCy alone versus PTCy+ATG, on behalf of the Cellular Therapy and Immunobiology Working Party of the European Society of Blood and Marrow Transplantation (EBMT).

Methods: We compared the outcomes of 5326 adult patients receiving a haplo-HCT with PBSCs for a hematological malignancy, with either PTCy alone (n = 4639) or PTCy+ATG (n = 687) for GVHD prophylaxis, reported to the EBMT registry between January 2014 and December 2021. In addition, all patients received a calcineurin inhibitor and mycophenolate mofetil. All hematological malignancies except plasma cell dyscrasia, were included.

Results: The median follow-up was 29 months among surviving patients. The two groups were similar in terms of disease and transplant characteristics. Transplant indications were acute leukemias followed by myelodysplastic or myeloproliferative neoplasms then lymphomas. Two-year GVHD-relapse-free survival (GRFS) was similar between the PTCy+ATG and PTCy groups, being 36.1%, versus 39.5%, respectively (p = 0.074). This was confirmed in the multivariable analysis (HR 1.08 [0.97, 1.21], p = 0.02). The 2-year OS was significantly lower in the PTCy+ATG group (52.5%), versus the PTCy group (56.7%, p = 0.015), with no significant difference in the multivariable analysis (HR 1.13 [1.00, 1.28], p = 0.059). DFS was also lower in the PTCy+ATG group (46.9%), versus the PTCy group (52.5%, p = 0.006), and it was confirmed in the multivariable analysis, (HR 1.16 [1.03, 1.31], p = 0.0015). There was no difference in the cumulative incidence of relapse in the PTCy+ATG versus PTCy group (HR 1.14 [0.96, 1.36], p = 0.14), while the non-relapse mortality (NRM) was significantly higher with PTCy+ATG (HR 1.18 [1.00, 1.40], p = 0.048). There was no difference in the day-180 cumulative incidence of grade II-IV and III-IV acute GVHD between the PTCy+ATG and the PTCy groups (HR 0.93 [0.80-1.08], p = 0.4 and HR 1.21 [0.96-1.53], p = 0.11). The 2-year cumulative incidence of chronic GVHD (cGVHD) and extensive cGVHD was significantly lower in the PTCy+ATG group compared to the PTCy group, being 21.2% versus 31.4% (p < 0.001), and 10.4% versus 13.5% (p = 0.031), respectively. The lower incidence in the PTCy+ATG group was confirmed in the multivariable analysis for overall cGVHD (HR 0.66 [0.55-0.80], p < 0.001) and extensive cGVHD (HR 0.75 [0.57-0.99], p = 0.042).

Conclusions: Overall, while the addition of ATG to PTCy was associated with a lower DFS and a higher NRM after haplo-HCT with PBSCs compared to PTCy alone, GRFS and OS did not significantly differ and the CI of cGVHD, including extensive cGVHD were significantly lower in patients that received ATG+PTCy.

Disclosure: The authors declare having no conflict of interest on that topic

12: Graft-versus-host Disease – Clinical

Attilio Olivieri ¹, Giorgia Mancini¹, Francesco Saraceni¹, Ilaria Scortechini¹, Irene Federici¹, Alessandro Busca², Luisa Giaccone², Alessandro Rambaldi³, Maria Caterina Micò³, Alessandra Algarotti³, Giuseppe Milone⁴, Salvatore Leotta⁴, Jacopo Peccatori⁵, Maria Teresa Lupo Stanghellini⁵, Francesca Bonifazi⁶, Mario Arpinati⁶, Maura Faraci⁷, Fabio Benedetti⁸, Simone Cesaro⁸, Franca Fagioli⁹, Francesco Saglio⁹, Irene Maria Cavattoni¹⁰, Vincenzo Pavone¹¹, Anna Mele¹¹, Riccardo Saccardi¹², Chiara Nozzoli¹², Domenico Russo¹³, Michele Malagola¹³, Carlo Borghero¹⁴, Daniele Vallisa¹⁵, Adriana Balduzzi¹⁶, Francesca Patriarca¹⁷, Antonella Giromin¹⁷, Edoardo Benedetti¹⁸, Nicola Mordini¹⁹, Francesco Merli²⁰, Fabio Ciceri⁵, Massimo Martino²¹

¹AOU delle Marche, Marche University Hospital, Ancona, Italy, ²S.S.C.V.D Trapianto di Cellule Staminali, A.O.U Citta della Salute e della Scienza di Torino, Torino, Italy, ³Azienda Socio Sanitaria Territoriale Papa Giovanni XXIII, University of Milan, Bergamo, Italy, ⁴Azienda Policlinico Ferrarotto, Catania, Italy, ⁵San Raffaele Scientific Institute, Milano, Italy, ⁶University Hospital S. Orsola-Malpighi, Bologna University, Bologna, Italy, ⁷IRCCS Istituto Giannina Gaslini, Genova, Italy, ⁸Azienda Ospedaliera Universitaria Integrata, Verona, Italy, ⁹Regina Margherita Hospital, Torino, Italy, ¹⁰Ospedale Centrale di Bolzano, Bolzano, Italy, ¹¹Hospital Card. G. Panico, Tricase, Italy, ¹²Careggi University Hospital, Firenze, Italy, ¹³ASST-Spedali Civili di Brescia, Brescia, Italy, ¹⁴San Bortolo Hospital, Vicenza, Italy, ¹⁵Ospedale di Piacenza, Piacenza, Italy, ¹⁶Clinica Pediatrica Università degli Studi di Milano Bicocca, Monza, Italy, ¹⁷S. Maria della Misericordia Hospital, Azienda Sanitaria Universitaria Friuli Centrale, Udine, Italy, ¹⁸Pisa University Hospital, Pisa, Italy, ¹⁹A.O. S. Croce e Carle Hospital, Cuneo, Italy, ²⁰Azienda Unità Sanitaria Locale-IRCCS, Reggio Emilia, Italy, ²¹BMM Hospital, Reggio Calabria, Italy

Background: We conducted a prospective study in 266 consecutive patients receiving HSCT and developing newly diagnosed cGVHD from 2017 to 2022, enrolled by 20 Italian GITMO centers by using a previously validated software which automatically calculates both the cGVHD NIH score and the kind of response to treatment according to the 2014 NIH consensus criteria.

Methods: The primary Endpoint was failure free survival (FFS) at 1 year, defined as the percentage of patients alive without cGVHD progression or not needing new immunosuppressive treatment, or experiencing malignancy relapse or severe (grade 3-4) toxicity. Secondary end points were overall response rate (ORR) at 6 and 12 months after systemic therapy (TX) (including both first&second line TX); other included OS and validation of our software in a multicenter prospective setting; this in order to collect the main standardized clinical informations of these patients, who received the current best available therapy (BAT) in 20 GITMO centers.

Results: Among the 266 patients enrolled, 41 (17%) had mild cGVHD; 149 (61%) moderate and 56 (23%) severe cGVHD respectively. Most patients had multiorgan involvement with skin (66%), mouth (60%), eyes (35%), and gastrointestinal tract (25%) being the most frequently involved districts; less frequently involved were liver (23%), joints (16%), genitalia (13%) and lung (10%); the median reported Lee symptom scale score was 18 (0-66). The median follow-up was 40 months (4-77).

Among the 231 evaluable patients, 32 did not receive any treatment, 16 received only topical treatment, 173 required an immediate first line TX which was represented by steroids alone in 142 (82%) and the remaining patients received association therapy including ECP (8) or Ruxolitinib (6) or other drugs (17). The ORR, calculated by the physician (with the help of the software) at 6 months, was 67% (16% CR), with 33% patients experiencing NR/progressive cGVHD.

Overall, during the observational interval, 112 patients required a second line TX represented by: ECP in 33 patients (29.5%), Ruxolitinib in 17 (15%), Imatinib in 12 (5.4%), Ibrutinib in 3 (3%) and by other TX in the remaining 41 (32%), in most cases represented by a second steroid challenge; 27 patients never received any Tx while 14 received only topical treatment.

185 patients were evaluable at 12 months for response assessment, showing an ORR of 50% (16% of CR). Among the 222 patients evaluable for long term follow-up, the median FFS was 32 months (60% and 43% at one and two years respectively), while the 2 years OS was 84%.

Conclusions: This prospective observational study shows that a standardized cGVHD evaluation and the response evaluation according the 2014 NIH criteria, are feasible in a multicenter setting, when supported by a standard tool shared by all the investigators. Furthermore this study allows to evaluate the realistic ORR and the outcome of patients receiving the BAT in a real life scenario. Both the ORR and the hard outcomes should be considered as a reliable current benchmark reference for future studies aimed to evaluate the efficacy of new drugs when randomized clinical trials are not feasible.

Disclosure: Nothing to declare

12: Graft-versus-host Disease – Clinical

XiaoYu Zhang ¹, Yi He¹, Sizhou Feng¹, Mingzhe Han¹, Erlie Jiang¹

¹Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Tianjin, China

Background: Allogeneic hematopoietic stem cell transplantation (allo-HSCT) offers rapid hematopoietic reconstitution and long-term wellness for aplastic anemia (AA). However, the high incidence of acute graft versus host disease (aGVHD) contributed to high treatment-related mortality (TRM). Since AA is a nonmalignant disorder, avoiding aGVHD remains a clinical concern. Our study aimed to determine the prophylactic value of ruxolitinib for aGVHD in HSCT recipients of AA.

Methods: Between August 2021 and January 2023, 35 consecutively aplastic anemia patients who applied for allo-HSCT were retrospectively enrolled. Despite the standard GVHD prophylaxis including cyclosporine A or tacrolimus, short-term methotrexate, ruxolitinib was initiated at the beginning of the conditioning regimen and continued to 3 months after transplantation.

Results: This study included 35 AA patients (median age: 32, ranging from 12 to 57). With the administration of ruxolitinib as prophylaxis, neutrophil and platelet engraftment took a median time of 14 (10-24) days and 23 (8-112) days, respectively, without primary graft failures (GF) cases observed. Ruxolitinib for GVHD prevention enhanced Tregs development after transplantation. Compared to the historical-matched control, the ruxolitinib group exhibited a significant advantage in decreasing the incidence of moderate (grade II ~ IV: 17.1% vs 48.6%, P = 0.033) and severe (grade III – IV: 8.6% vs 20%, P = 16) aGVHD. The ruxolitinib group exhibits a lower incidence of cGVHD as compared to the historical control (26.2 ± 9.5% vs 38.3 ± 8.7%, P = 0.24). The peri-transplant ruxolitinib didn’t increase infection incidence rate. After a median follow-up of 417 days (range, 112-725), only one patient in the HID-HSCT group died of TB-related treatment complications on +day 112. The 1-year overall survival (OS), failure-free-survival (FFS), and GFFS rate at 97.1 ± 2.82 (95%CI 91.8 ~ 100), 91.4 ± 4.73, (95%CI 82.6 ~ 100), and 85.7 ± 5.91, (95%CI 74.9 ~ 98.1), respectively. The treatment-related mortality (TRM) was 2.9 ± 2.82 (95%CI 0 ~ 8.2). Though nosignificant differences were documented, we noted more than 10% advantage of GVHD free-failure survival in the ruxolitinib group compared to the control group (85.7% vs 68.6%, p = 0.18).

Conclusions: This is the first study investigating the prophylactic application of ruxolitinib for aplastic anemia patients after allo-HSCT. Ruxolitinib would be a promising and safe choice in preventing aGVHD, but this awaits further study in larger cohorts.

Disclosure: The authors declare that they have no competing interests.

11: Graft-versus-host Disease – Preclinical and Animal Models

Marie Czech ¹, Robert Zeiser²

¹University Medical Center Freiburg, Freiburg, Germany, ²University Medical Center, Freiburg, Germany

Background: The success of allogeneic hematopoietic cell transplantations (allo-HCT) remains limited by acute graft-versus-host disease (aGVHD). We previously demonstrated that neutrophil granulocytes increase early aGVHD by releasing reactive oxygen species and promoting tissue damage. Additionally, they can migrate to mesenteric lymph nodes, where they present antigen and thereby activate allogeneic T cells. Here, we identified a neutrophil population in the gastrointestinal tract of allo-HCT recipients that was characterized by expression of Lipocalin-2 (LCN2). LCN2 is an antimicrobial peptide with protective functions against inflammatory bowel disease, making it an interesting target to study in the context of aGVHD.

Methods: To characterize neutrophils infiltrating the gastrointestinal tract, we implemented single-cell RNA sequencing, leading to the definition of a new neutrophil subset. We confirmed these initial findings by using major mismatch models of allo-HCT, where mice received treatment regiments or adoptive transfer of neutrophils or primary macrophages. For mechanistic analyses, we performed 16S rRNA gene sequencing, microarray and kinase array, the latter of which we confirmed with western blot and qPCR. Finally, we used biopsies, serum and stool samples from aGVHD patients to translate murine findings to the human system.

Results: Single-cell RNA sequencing identified a donor-derived neutrophil population, characterized by expression of the antimicrobial peptide LCN2. We confirmed the presence of LCN2⁺ neutrophils during aGVHD, which coincided with increased expression of Lcn2 in the small intestine and increased secretion of LCN2 into the intestinal lumen. Loss of hematopoietic LCN2 led to aggravated disease, as well as reduction of protective bacteria in the stool and increased infiltration of immunogenic immune cells into intestinal tissues. Transfer of LCN2-overexpressing neutrophils protected from aGVHD. Treatment with recombinant LCN2 also protected from aGVHD, while modifying the immune cell composition of the GI tract, where less immunogenic and more tolerogenic cell types were detected. To decipher the mode of action, we screened for expression of a LCN2 receptor (Slc22a17), and found high expression in macrophages. Treatment with LCN2 decreased the expression of MHC class II in macrophages and adoptive transfer of LCN2-primed tolerogenic macrophages reduced disease activity. Additionally, we found increased expression of Il10 in LCN2-treated macrophages. IL10^-/- macrophages did not respond to LCN2 treatment with decreased MHC-II expression, strengthening our hypothesis of LCN2 being IL10-dependent in a treatment setting. Indeed, genetic or systemic loss of IL10 abrogated the protective effect of LCN2 in an in vivo model of aGVHD. To quantify the effect of LCN2 on macrophages, we performed a kinase array and found insulin-like growth factor receptor 1 (IGF1R) to be activated upon LCN2 treatment. Additionally, we found that loss of epithelial LCN2 in the gastrointestinal tract aggravates aGVHD, strengthening its position as protective disease modulator. Finally, we used samples from four different patient cohorts and could demonstrate increased expression of LCN2 in allo-HCT patients that developed GVHD, which correlated with expression of IL10 and serum LCN2.

Conclusions: We found a protective LCN2-expressing neutrophil population present in aGVHD, which led us to the identification of LCN2 as a potential treatment option for patients suffering from aGVHD.

Disclosure: RZ received honoraria from Novartis, INcyte, MNK and Sanofi outside of this study. This study was supported by the Deutsche Forschungsgemeinschaft, Germany, SFB-1479 – Project ID: 441891347 (P01 to RZ, S1 to MB, P03 to NK), Project-ID 259373024 – TRR 167 (Z01 to M.B.), Germany’s Excellence Strategy – (CIBSS - EXC 2189 Project ID: 390939984), DFG individual grant 872/4-2 to R.Z., SFB1160 (Project ID 256073931 - TP B09 to R.Z. and Z02 to M.B.), CRC1453 (Project ID 431984000-S1, M.B.), the European Union: GVHDCure (ERC consolidator grant to R.Z.), the Deutsche Krebshilfe (grant number 70113473), the Jose-Carreras Leukemia Foundation (grant number DJCLS 01R/2019) (R.Z.), by the Max Planck Society (D.G.), and by the German Research Foundation (DFG) (322977937/GRK2344 MeInBio) (D.G.). We also acknowledge funding from the German Federal Ministry of Education and Research (BMBF) within the Medical Informatics Funding Scheme - MIRACUM-FKZ 01ZZ1801B (M.B.) and EkoEstMed–FKZ 01ZZ2015 (G.A.).

11: Graft-versus-host Disease – Preclinical and Animal Models

Suze Anne Jansen ^1,2, Alessandro Cutilli¹, Coco de Koning^2,1, Marliek van Hoesel^2,1, Leire Saiz Sierra¹, Stefan Nierkens^2,1, Michal Mokry¹, Edward Nieuwenhuis^1,3, Alan Hanash⁴, Enric Mocholi¹, Paul Coffer¹, Caroline A. Lindemans^2,1

¹University Medical Center Utrecht, Utrecht, Netherlands, ²Princess Máxima Center of Pediatric Oncology, Utrecht, Netherlands, ³Utrecht University, Utrecht, Netherlands, ⁴Memorial Sloan Kettering Cancer Center, New York City, United States

Background: Chemotherapy (chemo) used in the conditioning regimen before allogeneic hematopoietic stem cell transplantation (HSCT) is thought to contribute to the development of acute graft-versus-host disease (GVHD). Whether damage to the intestinal epithelium induced by chemotherapy directly influences allogeneic (allo) T cell behavior is not exactly known.

Methods: Using a variety of 3D human intestinal organoid-based models we studied the direct effects of chemo-induced damage of intestinal epithelial cells (IEC) on allo T cell responses.

Results: Treatment of human intestinal epithelial organoids with clinically relevant chemo busulfan (Bu), fludarabine (Flu) and clofarabine (Clo) induced morphological changes, stalled DNA replication, reduced proliferation, induced oxidative stress and apoptosis. Functionally, chemo-treatment impaired single organoid cells to form new organoids, thus impairing regenerative capacity. On transcriptional level, chemo treatment induced relatively subtle changes with 66, 106 and 118 differentially expressed genes (DEGs) respectively. Organoids treated with the nucleoside analogs Flu and Clo shared more DEGs, than each did with alkylating agent Bu, reflecting common mechanisms of action. Interestingly, gene set enrichment analysis showed a transcriptional change related to an inflammatory response.

Using a transwell system we evaluated the effect of chemo-treatment of organoids in the lower compartment on functional migration of peripheral blood T cells placed in the upper compartment. Both resting and polyclonally preactivated (using anti-CD3/CD28) CD4+ and CD8 + T cells demonstrated increased migration to chemo-treated organoids. Next, the effect of chemo-damaged intestinal epithelium on concurrent polyclonal activation of T cells was evaluated in a co-culture system. T cell proliferation and activation markers were assessed by flow cytometry after 4 days of co-culture. Co-culture with chemo-treated organoids led to increased proliferation and IFNγ-production in CD4+ and CD8 + T cells. To gain insight in the mechanism underlying increased T cell migration and activation, we performed proteomic analysis of culture-medium (CM) exposed to chemo-treated organoids using the Olink platform. We found that galectin-9 (Gal-9) levels, a beta-galactoside-binding lectin with pleiotropic and divergent roles in inflammation, were increased in CM predominantly from Flu- and Clo-treated organoids. Interestingly, Gal-9 levels were also increased in plasma of pediatric patients that eventually developed gut GVHD after HSCT. To determine whether Gal-9 indeed played a role in the migration and activation of T cells we used Gal-9 blocking antibodies and organoids in which Gal-9 was genetically ablated using CRISPR/Cas9 technology (Gal-9-KO). Indeed, blocking of Gal-9 reduced migration and activation of T cells, and co-culture with Gal-9-KO organoids abrogated the increased activation of T cells that was found previously with Clo-treatment of organoids.

Conclusions: T cells demonstrate increased migration to chemotherapy-treated organoids, and upon exposure to them, proliferated more and expressed higher levels of IFNγ. Damaged organoid-derived Gal-9 appeared to be at least partially responsible for these effects. As such, Gal-9 serves as a potential biomarker for intestinal damage and inflammation, and a probable target for future prevention and/or treatment of GVHD and other immune-mediated disease.

Disclosure: Nothing to declare

11: Graft-versus-host Disease – Preclinical and Animal Models

Edmund K. Waller ¹, Yi-Wen Li¹, Shuhua Wang¹, Hanwen Zhang¹, Yutong Liu¹, Po-Chih Hsu¹, Pankoj Das¹, Passang Lim¹, Tamas Nagy², Zhaohui Qin¹, Cynthia Giver¹

¹Winship Cancer Institure of Emory University, Atlanta, United States, ²University of Georgia College of Vetrinary Medicine, Athens, United States

Background: Graft-versus-host disease (GvHD) in allo-transplant recipients is characterized by donor T cell-mediated damage to recipient organs and changes in the gut microbiome. Vasoactive intestinal peptide (VIP), an immunosuppressive neuropeptide, directs T cell responses towards a Th2 phenotype while inhibiting proinflammatory cytokine production. Furthermore, VIP-producing enteric neurons contribute to the mucosal barrier and maintain gut microbiome homeostasis. We previously described the role of VIP-producing donor plasmacytoid dendritic cells in limiting GvHD. While diet and antibiotic treatment have been shown to affect the gut microbiota, the impact of genes expressed in the gut on GvHD and the composition of the microbiota needs to be clarified. We hypothesized that the absence of VIP in allogeneic bone marrow transplant (allo-BMT) recipients would increase GvHD compared to wild-type mice and that local VIP synthesis in the gut would regulate the microbiota.

Methods: We examined the effect of endogenous VIP synthesis in modulating GvHD in multiple murine MHC mismatched allo-BMT models. VIP KO mice and radiation chimeras in which either the hematopoietic or somatic tissue compartments were derived from VIP KO mice were used as BMT recipients. Measurements of serum cytokines and intracellular staining for cytokines and transcription factors characterized the immune polarization of donor T cells. Cell transplantation of luciferase+ donor T cells tracked tissue homing an in vivo expansion after allo-BMT. Gut microbiota was classified after shotgun DNA sequencing of stool pellets, characterizing bacterial species in wild-type and VIP KO mice before and after co-housing and following allo-BMT and treatment with a VIP-receptor antagonist.

Results: Lethally irradiated VIP knock-out (VIP-KO) mice rejected grafts composed of MHC mismatched T cell-depleted bone marrow (BM) without added donor T cells, consistent with heightened host cell allorreactivity. Transplanting donor splenocytes or purified T cells with T cell-depleted BM led to rapid death from GvHD, with histology of GVHD-target organs showing the predominance of GVHD in the gut. Heightened activation and expansion of Th1- and Th17-polarized T cells occurred in VIP-KO recipients. Transplanting radiation chimeric mice demonstrated that VIP production by non-hematopoietic tissues limited the GvHD activity of donor T cells. Immunofluorescence staining demonstrated VIP expression in enteric neurons in allo-BMT recipients, consistent with local effects of VIP on T cells and the gut microbiome. VIP-KO mice had less microbiome diversity, lower stool pH, and higher frequencies of the Bacillota phylum, including Lactobacillus murinus and Ruminococcus gnavus species, post-transplantation. Co-housing with wild-type mice did not mitigate alterations in the gut microbiome or the GvHD-susceptibility of VIP-KO mice undergoing allo-BMT. Daily treatment with a VIP-receptor antagonist recapitulated the increase in Lactobacillus murinus seen in VIP-KO mice, demonstrating that VIP-signaling on the gut epithelial cells regulates the gut microbiome.

Conclusions: These data support the crucial role of VIP in maintaining the diversity of the gut microbiome with beneficial commensal species and in modulating the alloreactivity of donor T cells after allo-BMT. Ongoing experiments on the impact of VIP receptor signaling on the gut microbiome composition and its relationship with anti-cancer activity will provide insights into the broader therapeutic applications of VIP and its antagonists.

Clinical Trial Registry: Not applicable

Disclosure: EKW and Y-WL have an intellectual property related to using VIP agonists to limit GvHD. EKW founded and has equity in Cambium Oncology LLC, which has licensed intellectual property from Emory University covering medicinal uses of VIP agonists and antagonists.

11: Graft-versus-host Disease – Preclinical and Animal Models

Yuxi Xu^1,2, Guancui Yang¹, Qing Xu¹, Ya Zhou¹, Rui Wang¹, Xiaoqi Wang¹, Shijie Yang¹, Qingxiao Song ^1,2, Xi Zhang^1,2

¹Medical Center of Hematology. Xinqiao Hospital of Army Medical University (Third Military Medical University), Chongqing, China, ²Jinfeng Laboratory, Chongqing, China

Background: Chronic graft‐versus‐host disease (cGVHD) is a leading cause of morbidity and mortality following allogeneic hematopoietic stem cell transplantation (allo-HSCT). The expansion of pathogenic T helper cells (Th) and reduction of regulatory T cells (Tregs) result in cGVHD. Phosphoinositide-3 kinase (PI3K), a lipid kinase, plays a pivotal role in the proliferation, differentiation, survival, and metabolism of activated T cells. As a result, this study aimed to evaluate the efficacy of PI3K inhibitors in the treatment of cGVHD.

Methods: BALB/c recipients were conditioned with a total body irradiation of 7.5Gy. For the B10.D2 → BALB/c scleroderma model, recipients were transplanted with 10×10⁶ bone marrow cells and 15×10⁶ splenocytes from B10.D2 donors. For the C57BL/6 → BALB/c systemic GVHD model, recipients were transplanted with 5x10⁶ bone marrow cells and 1.25x10⁶ splenocytes from C57BL/6 donors. Treatment groups received PI3Kδ inhibitor (Parsaclisib (5mg/kg)) or solvent by daily gavage starting from day 10 post-HSCT. Mice were monitored for body weight change, clinical and cutaneous signs of cGVHD, and survival. HE staining was used to examine pathological changes in the target organs (skin, lung, liver, small intestine, and colon). Skin collagen content was detected by Masson staining. Lymphocytes from the target organs of the recipients are collected for flow cytometry analysis.

Results: Here, in a mouse model of minor histocompatibility mismatch model (B10.D2 → BALB/c) that presents with a cutaneous cGVHD and systemic inflammatory response. Treatment with (Parsaclisib) substantially reduced skin GVHD scores and extended overall survival among recipients. On day 30 post-HSCT, compared to the non-GVHD (syngeneic HCT) group, the recipients given solvent showed inflammatory cell infiltration, collagen deposition, and serious fat atrophy in the skin. In contrast, recipients given Parsaclisib significantly reduced cell infiltration and collagen deposition in the skin. Parsaclisib treatment significantly reduced CD4⁺ and CD8⁺ T cells infiltrated in the skin while inhibiting the transcription factor Hobit expression to suppress the CD69⁺CD103⁺ tissue-resident CD8⁺ T cells in the lung and liver. More importantly, Parsaclisib treatment upregulates Foxo1 and Foxo3α to promote Foxp3 expression and expands a population of Tregs that is stable and highly suppressive in the skin, which could contribute to the suppression of cutaneous cGVHD. The frequency and number of CD19⁺B220⁺ B cells and the CD11b⁺F4/80⁺ macrophages in GVHD target organs (skin, liver, and lung) were markedly reduced in the recipients treated with Parsaclisib compared to those given solvent. In addition, the therapeutic effect of Parsaclisib was validated in a major MHC mismatch model (C57BL/6 → BALB/c). Similarly, Parsaclisib effectively treated mice with active GVHD in the MHC mismatched model.

Conclusions: PI3Kδ inhibitor (Parsaclisib) modulates the tissue microenvironment by enriched Tregs while suppressing Trm cells, B cells, and macrophages in the tissue, effectively alleviating cGVHD. These data provide the scientific rationale for PI3Kδ inhibition as a therapeutic approach for patients with cGVHD.

Clinical Trial Registry: N/A

Disclosure: Nothing to declare

1: Haematopoietic Stem Cells

Qiaomei He ¹, Xi Sun¹, Lin Chen², Jun Yang¹, Huixian Wu¹, Shixuan Tang¹, Liping Wan¹, Yan Zhang¹, Bo O. Zhou², Fang Zhang³, Xianmin Song¹

¹Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, ²State Key Laboratory of Cell Biology, Shanghai Institute of Biochemistry and Cell Biology, CAS Center for Excellence in Molecular Cell Science, Chinese Academy of Sciences, Shanghai, China, ³Zhongshan School of Medicine, Sun Yat-Sen University, Guangzhou, China

Background: Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is hinged on the successful engraftment of donor-derived hematopoietic stem cells (HSCs) and the achievement of full-donor chimerism (FDC) for the reconstitution of the recipient’s hematopoietic and immune system. Moreover, early cell immune reconstitution after allo-HSCT is pivotal to prevent pathogenic microorganism infection and original disease relapse, which are the two most important causes of transplantation failure. Therefore, it is of great clinical significance to effectively promote donor HSC implantation and accelerate immune reconstitution after allo-HSCT. However, little is known about how to promote implantation and early immune reconstitution of HSCs through regulating its own functions.

Methods: Single-cell transcriptome sequencing analysis on bone marrow cells from clinical samples and mouse models in two states, from mixed donor chimerism (MDC) to FDC, was performed to explore the key regulatory target affecting HSC implantation after allo-HSCT. The effects of the agonists that modulate this target on HSC implantation after transplantation were investigated in a novel MDC mouse model. The effects of the agonist regulating this target on HSC function were investigated by competitive transplantation mouse model and serial umbilical cord blood (UCB) xenograft transplantation model. The mechanisms of the agonist were further explored by single-cell sequencing.

Results: We identified very-late antigen (VLA-4) as a pivotal target in regulating HSC implantation. Critically, we screened out a VLA-4 agonist, which could promote the implantation and long-term hematopoietic reconstitution of donor HSCs in a MDC mouse model and serial xenotransplantation model of human CD34+ cells from UCB. Competitive transplantation mouse model and limited dilution assay for UCB CD34+ cells revealed that VLA-4 agonist could enhance the self-renewal capacity of HSCs and increase the number of severe combined immunodeficient (SCID)-repopulating cells (SRCs) to five-fold in vivo. Furthermore, the VLA-4 agonist could facilitate early post-transplant cell immune reconstitution by enhancing T cell output of HSCs. VLA-4 agonist regulates HSC functions and lymphoid progenitor differentiation through affecting the ERK1/2 phosphorylation process, whereas may not increase the risk of leukemia transformation.

Conclusions: Our research underscores the significant potential of the VLA-4 agonist in addressing two of the foremost challenges in allo-HSCT: enhancing engraftment efficiency and facilitating early immune reconstitution of HSCs. These findings pave the way for further clinical exploration and application of the VLA-4 agonist in HSCT.

Disclosure: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.This research was funded by National Natural Science Foundation of China (82270226 for Xianmin Song), National Natural Science Foundation of China (32270838 for Fang Zhang), Shanghai Shen Kang Hospital Development Center (SHDC2020CR1012B for Xianmin Song).

1: Haematopoietic Stem Cells

Martin Maiers ¹, Sapir Israeli², Yoram Louzoun², Stephen R. Spellman¹, Heather E. Stefanski¹, Jeffery J. Auletta¹, Steven M. Devine¹

¹Center for International Blood and Marrow Transplant Research, National Marrow Donor Program, Minneapolis, United States, ²Bar-Ilan University, Ramat Gan, Israel

Background: We have devised a machine learning algorithm for forecasting hematopoietic cell transplantation (HCT) outcomes. This algorithm addresses the decision-making process when selecting among available donor sources based on the patient’s relatives and unrelated donors (URD) identified in a patient’s registry search. The four donor types considered were matched related donor (MRD), mismatched related donor (MMRD), matched unrelated donor (MUD), and mismatched unrelated donor (MMUD).

Methods: We analyzed a cohort of 31,515 U.S. transplants reported to the Center for International Blood and Marrow Transplant Research (CIBMTR) between 2016 and 2019 (Table 1). Using this dataset, we trained and validated multiple prediction algorithms for the following clinical endpoints at 1-year post-HCT: overall survival, disease relapse, graft rejection, chronic graft-versus-host disease (cGvHD) and the composite endpoint of event-free survival (EFS), representing survival in the absence of these events. We conducted searches in the U.S. URD registry for each recipient and applied predictions to each potential donor. Analysis included the best URD 1-year EFS risk predicted by the model compared to that of the donor actually selected for the transplant. Notably, GvHD prophylaxis strategy for the potential donor aligned with that actually used in the recipient. Matching criteria involved up to 1 mismatch out of 10 alleles for the loci HLA-A, -B, -C, -DRB1, and -DQB1.

Results: Key factors impacting predictive outcomes include donor age, sex, donor type, HLA match, and post-transplant cyclophosphamide (PTCy) use. Comparing HLA-10/10 to 9/10 URD, outcomes generally favor matched transplants although some mismatches yielded better predictions. MRDs are predicted to provide lower rates of GvHD and mortality. However, in specific cases, URDs surpass MRDs in survival predictions. URDs consistently show lower rejection risk than MRDs. Considering MMRD, predominantly HLA haploidentical relatives, URDs demonstrate better predicted EFS risk in the majority of cases. The EFS risk differences are driven by survival and rejection, influenced by donor age and higher HLA mismatch levels, respectively.

Conclusions: Post-transplant cyclophosphamide (PTCy) is increasingly recognized as a standard GvHD prophylaxis in recipients undergoing allogeneic HCT. This model is consistent with previous reports that PTCy use has reduced the outcome differential between MUD and MMUD transplants. A consequence of this substantial growth in alternative donor options has resulted in more scenarios wherein an URD presents a better predicted outcome than a relative. The analytical approach in this study employs a straightforward linear model, facilitating its implementation through a user-friendly nomogram. By scoring all donor options (both related and unrelated) based on a limited set of accessible variables, clinicians can effectively assess and compare potential outcomes for patients.

Disclosure: Nothing to disclose

Table 1 Characteristics of patients in the U.S. receiving adult donor HCT (2016-2019)