Allogeneic Peripheral Blood Stem Cell vs. Bone Marrow Transplantation for Patients with Chronic Myeloid Leukemia: Results of an Individual Patient Data Meta-Analysis.

Abstract

The Stem Cell Trialist’s Collaborative Group. An individual-patient data meta-analysis using data from 9 randomized trials enrolling 1,111 patients (pts.) to compare allogeneic peripheral blood stem cell transplantation (PBSCT) with bone marrow transplantation (BMT) has recently been completed (Hematology J2004;5:s89). Here, we report the results of this analysis for patients suffering from CML in 1st chronic phase (CP1) (n=422) or accelerated phase/blast crisis (AP/BC) (n=50). Overall survival at five years (70%) was not significantly different in patients who were transplanted with peripheral blood (PB) as compared to bone marrow (BM) cells if all pts. [absolute difference (AD) 3.3 %, p=0.16] or pts. in CP1 [AD 1.1 %, p=0.72] were considered. However, pts. transplanted for AP or BC survived significantly longer following PBSCT than BMT [AP 40 %, p=0.015]. The difference between two subgroups was significant (test of interaction chi²=4.3;p=0.04). Disease-free survival (DFS) was significantly better after PBSCT for all pts. [AD 15.2 %, p=0.004], for pts. in CP1 [AD 11.5 %, p=0.069], and AP/BC [AD 30.3 %, p=0.009]. Test of interaction between two subgroups was chi²=4.3;p=0.07. Relapse rates for all pts. [AD 18.6 %, p=0.00005] and pts. in CP1 [AP 17.0 %, p=0.0002] were significantly lower after PBSCT; no statistical difference was found for pts. in AP/BC most likely due to small patient numbers. Extensive chronic GVHD occurred more often after PBSCT [AD 20%, p<0.00001 ) than after BMT.Treatment-related mortality (TRM) at 5 yrs was 27% for all pts and 26% for patients in CP phase with no significant differences between pts. transplanted with PB instead of BM.TRM was higher in the BMT arm (62% vs. 22%, p=0.02) for pts. in AP/BC. The difference between two subgroups was statistically significant (chi²=5.5;p=0.02). Pts. with CML in AP/BC should be offered a transplant with G-CSF mobilized PB. For pts. in CP1 the decision to use PB or BM needs to value the higher risk of attracting extensive cGVHD against the lower risk of relapse after PBSCT. Further follow up will be necessary to see if the difference in relapse rates after BMT as compared to PBSCT continues to increase over time.