Reduced Intensity Allogeneic Stem Cell Transplantation for Follicular Lymphoma Results in An Improved Progression Free Survival When Compared to Autologous Stem Cell Transplantation. An Analysis from the Lymphoma Working Party of the EBMT

Both autologous stem cell transplantation (autoSCT) and reduced intensity allogeneic stem cell transplantation (RIST) may be considered as treatment options in patients with relapsed follicular lymphoma (FL). It is currently unknown which transplant strategy is optimal in patients with FL. We have therefore conducted a retrospective comparison of patients who have undergone an autoSCT or a RIST. Adult patients undergoing a 1^st transplant procedure and reported to the EBMT between Jan 1998 and Dec 2005 were included. 1394 patients underwent an autoSCT and 110 a RIST with median ages of 51 (20–73) and 50 (32–65) respectively. Patients undergoing a RIST had significantly more advanced disease (stage IV 74% vs 60% p=0.01) and B symptoms (39% vs 27% p=0.05) at diagnosis but there was no difference in the size of the largest mass or the LDH. The median time from diagnosis to transplant was 27 months for the whole group but was significantly longer for patients undergoing a RIST (34 months vs 26 months p=0.005). Patients undergoing a RIST had received more lines of prior therapy (61% 3 or more lines vs 34% for those undergoing autoSCT p<0.001) but there was no difference in the percentage of patients receiving rituximab prior to transplantation. 49% of the autoSCT were performed prior to 2002 compared to 26% of the RIST (p<0.001). At transplant the disease status for RIST and autoSCT patients was: CR 34% vs 43% ; PR 52% vs 56%; refractory disease 10% vs 5% respectively (p=0.02). Patients undergoing a RIST received stem cells from matched sibling donors in 94 cases and matched unrelated donors (MUD) in 16. All patients undergoing an autoSCT received unpurged bone marrow stem cells. With a median follow-up post transplant of 26 months (range 6–112) 84 patients are alive in the RIST group and 1154 patients in the autoSCT group. In multivariate analysis non-relapse mortality (NRM) was significantly worse for patients undergoing a RIST (relative risk (RR) 3.5, p<0.001) and for those older than 45 (RR 2.3, p=0.01) or with refractory disease at transplant (RR 3.3, p<0.001). The cumulative incidence of NRM at 100 days and 1 year was 5% and 15% respectively for those undergoing a RIST compared to 2% and 3% for those undergoing an autoSCT. Disease relapse was significantly worse for those receiving an autoSCT (RR 2.8, p<0.001), more lines of prior therapy (RR 1.3, p=0.02), stage IV disease at diagnosis (RR 1.4, p=0.004) and for those with refractory disease (RR 1.6, p<0.001) or poor performance status at transplant (RR=3.0, p=0.005). Patients undergoing a RIST had a 5 year relapse rate of 19% compared to 47% for those undergoing an autoSCT. For patients with follow-up beyond 30 months (RIST 26 patients, autoSCT 449 patients) there were no relapses beyond this time point in the RIST group and 87 relapses in the autoSCT group. Progression free survival was significantly worse for those with chemorefractory disease (RR 3.1, p<0.001), more lines of prior therapy (RR 1.3, p=0.004) or with a poor PS (RR 2.5, p=0.03). The PFS at 3 years and 5 years was 62% for patients receiving a RIST compared to 58% and 48% respectively for those receiving an autoSCT. The comparison of PFS between autoSCT and RIST was non-proportional over time with a significant advantage to RIST after 2 years (RR 9.0, p=0.002). In conclusion RIST is associated with a higher NRM but a lower relapse rate when compared to autoSCT resulting in an improved PFS following a RIST.