Abstract
The choice of allogeneic or autologous haematopoietic stem cell transplantation (HSCT) in follicular lymphoma (FL) remains controversial due to the lack of randomised studies, however the increasing evidence of a graft versus lymphoma (GvL) effect favours the use of allo-HSCT. We analysed 22 patients with (FL) who underwent reduced intensity conditioned (RIC) allo-HSCT (19 BEAM-C, 3 FMC) and compared the outcome of patients receiving BEAM-C to 21 patients who underwent BEAM autologous transplantation (ASCT). BEAM-C (BCNU, cytarabine, etoposide, melphalan, alemtuzumab) and FMC (Fludarabine, melphalan, alemtuzumab). Patients receiving RIC allo-HSCT and ASCT were similar in age, time from diagnosis to transplant, histological grade and stage but differed in that fewer patients in the allogeneic group were in complete remission (CR) at time of transplant and fewer patients in the ASCT group received rituxumab. Sibling donor was used in 12/22 RIC allo-HSCT, volunteer unrelated donor (VUD) in 10/22 (3 HLA mismatch). The median follow-up in the RIC group was 626days (range 22–1901days) with a transplant related mortality (TRM) at day+100 of 23% (siblings 8%, VUD 30%). The higher TRM in the VUD group occurred in patients who were heavily pre-treated with high-grade transformation of their disease, 2 also received HLA mismatched transplants. Acute graft versus host disease (GvHD) occurred in 5 cases (1 grade III). The cumulative incidence of chronic GvHD post day+100 was 25% (1 grade III). 5 patients received donor leukocyte infusions (DLI) for mixed chimerism, 2 also had evidence of disease relapse. 4/5 reverted to full donor chimerism with CR induced in 1/2 patients. For all patients receiving RIC allo-HSCT the 1year OS was 77% (92% siblings, 70% VUD) and actuarial DFS 73% (83% siblings, 70% VUD). Multivariate analysis identified the number of previous therapies (p=0.03) as having significant effect on OS and the duration from diagnosis to transplant (p=0.02) having significant effect on DFS.
As a result we compared the outcome from time of diagnosis to follow-up in the BEAM-C versus BEAM groups which showed a trend for improved OS (p=0.08) and a statistically significant difference in DFS (p=0.02) favouring RIC allo-HSCT, with a 5year DFS of 84% versus 64% and 8year DFS of 84% versus 36% respectively. The 1year and 3year OS from transplantation in the BEAM-C versus BEAM was 79%, 59% versus 86%, 61% respectively and 1year and 3year DFS was 74%, 74% versus 65%, 45% respectively. In summary, the longer the duration from diagnosis to RIC allo-HSCT and the more lines of therapy administered adversely affected outcome, therefore supporting the use of RIC allo-HSCT in second remission. Superior DFS in the allogeneic group is thought to be as a result of the GvL effect, which can be enhanced through the use of DLI. Longer follow-up is however required in order to determine whether the results will translate into an improved OS. A randomised study of RIC allo-HSCT versus ASCT is warranted.
Author notes
Corresponding author
