Hematopoietic stem cell transplantation for relapsed classical Hodgkin lymphoma: A benchmark Study on 19,498 patients from the EBMT lymphoma working party

Abstract Introduction: Autologous hematopoietic stem cell transplantation (auto-HSCT) remains a standard of care for patients with chemosensitive relapsed/refractory Hodgkin lymphoma (HL) and allogeneic transplant (allo-HSCT) for patients who failed a prior auto-HSCT. Brentuximab vedotin (BV) and checkpoint inhibitors (CPI) have changed the landscape of HL treatment. Methods: Using the European Society for Blood and Marrow Transplantation (EBMT) registry, we assessed changes over time in 15,648 patients who received an auto-HSCT and 3850 an allo-HSCT between 2010 and 2022. Results: For auto-SCT, when comparing three periods (2010-2014 N = 6,896; 2015-2018 N = 4,739; and 2019-2022 N = 4,013), patients transplanted in recent years were older, more likely to have received BV and/or CPI pretransplant, and more frequently transplanted in complete remission (CR) and PET scan negativity. The 2-year progression free survival (PFS) increased from 63% to 69% and 73% over the three time periods and the 2-year overall survival (OS) increased from 85% to 91% and 93% respectively. Improvement over time was noted in patients transplanted in partial response (PR) (2-year PFS 56%, 57% and 65%) whereas that of patients autografted in CR remained stable at 77%. In multivariate analysis (MVA), transplantation in recent years significantly improved PFS (HR 0.9, p=0.046 for 2015-2018; and 0.79, p<0.001 for 2019-2022, both compared to 2010-2014) and OS (HR 0.6, p<0.001 and 0.55, p<0.001). PFS and OS were positively affected by female gender (HR 0.89 and 0.83 respectively, p=0.009 for both), time from diagnosis to auto-HSCT > 24 months (HR 0.71, p<0.001; and 0.82, p=0.005) and Karnofsky score > 90 (HR 0.85, p=0.003; and 0.75, p<0.001). PFS and OS were negatively affected by transplantation in PR (HR 1.62 and 1.81 respectively, p<0.001 for both) or in refractory status (HR 2.45 and 2.56 respectively, p<0.001 for both) as well as HSCT comorbidity index > 3 (HR 1.57 and 1.33, p<0.001 for both). Older age negatively affected OS (HR for 5-years increase 1.15 p<0.001). For allo-SCT, when comparing the three periods (2010-2014 N = 1673; 2015-2018 N = 1229; and 2019-2022 N = 948), patients transplanted in recent years were older, had a longer time from diagnosis to transplant, more likely to have received BV and/or CPI pretransplant, more frequently transplanted in CR and PET negativity, more likely to have received transplant from mismatched relative (predominantly haplo-identical) donor (MMRD) and post-transplant cyclophosphamide (PTCy). The 2-year PFS increased from 44% to 58% and 62% over the three time periods and the 2-year OS went from 66% to 73% and 72% respectively. Importantly, the 2-year GVHD and relapse free survival (GRFS) increased from 32% to 43% and 50% respectively. In MVA, transplantation in recent years significantly improved PFS (HR 0.78, p<0.001 for 2015-2018; and 0.7 p<0.001; both compared to 2010-2014), OS (HR 0.8, p=0.001 and 0.84, p=0.049) and GRFS (HR 0.88, p=0.012 and 0.75, p<0.001). PFS and OS were negatively affected by older age (HR for 5-years increase 1.08 and 1.14, p<0.001 for both), transplantation in PR (HR 1.58 and 1.34, p<0.001 for both) or refractory disease (HR 2.28 and 2.1, p<0.001 for both). OS was also negatively affected by myeloablative conditioning (HR 1.22, p=0.002). Compared to matched related donor (MRD) without PTCY, OS was negatively affected by the use of MMRD without PTCy (HR 1.38, p=0.019), unrelated donor (UD) without PTCy (HR 1.29, p<0.001), or MMRD with PTCy (HR 1.21, p=0.031), but not significantly affected by other combinations, whereas PFS was positively affected by the use of UD with PTCy (HR 0.64, p=0.005) but not significantly affected by other combinations. In separate MVA starting from 2015, CPI use before allo-HSCT significantly improved PFS (HR 0.79, p=0.025) whereas BV use before allo-HSCT did not significantly affect PFS and OS.Conclusion: outcomes after auto-HSCT and allo-HSCT continue to improve over time. For auto-HSCT, transplantation in CR, younger age, female gender, Karnofsky > 90, low comorbidity index and a longer interval from diagnosis to HSCT significantly improve survival. For allo-HSCT, transplantation in CR, younger age, and reduced intensity conditioning significantly improve survival. CPI use before allo-HSCT significantly improved PFS. MRD is the preferred donor option in the absence of PTCy, whereas the combination of UD and PTCy yield the best outcomes.