Use of Bone Marrow or Peripheral Blood Stem Cell Grafts in Non T Cell Depleted Haploidentical Transplants Using Post-Transplant Cyclophosphamide, an ALWP-EBMT Analysis

The use of mobilized peripheral blood (PB) as stem cell source in allogeneic stem cell transplantation (HSCT) from sibling or unrelated donors is associated with higher incidence of chronic graft versus host disease (GVHD) as compared to HSCT with bone marrow (BM). The reported incidence of GVHD in the haploidentical transplants with PT-Cy with BM grafts is low, while GVHD incidence using PB grafts varied between studies between 30% to 40% in single center reports.

With the aim to analyze the effect of stem cell (SC) source in the setting of non T cell depleted haploidentical transplant using post-transplant Cyclophosphamide (PT-Cy), we analyzed patients (pts) transplanted for AML or ALL in first or second complete remission (CR) and reported to the EBMT from 2010 to 2014.

Four hundred fifty one pts were analyzed, the majority of the pts in both groups were transplanted for AML (73%) in CR1 (67%). BM was the source of SC for 260 pts and PB for 191. Median follow up was longer for pts receiving BM (22.8 vs 18.3 months) and those pts were more likely transplanted with a MAC (61% vs 49%, p=0.008). All pts received PT-Cy as GVHD prophylaxis and prophylaxis was mainly in combination to calcineurin inhibitor and MMF, according to transplant center policy. ATG use was not different for the 2 groups (5% vs 7%, p=0.41).

Overall OS, LFS and GRF at 2 year were 55%, 51% and 43%. CI of grade II-IV acute GVHD, chronic GVHD, NRM and Relapse at 2 year were 18%, 35%, 23% and 24%, respectively.

Pts transplanted with BM had a longer time to neutrophil engraftment (18 vs 17 days, p= <0.001). In the univariate analysis pts transplanted with BM had a lower incidence of grade II-IV and grade III-IV acute GVHD (12% vs 27%, p=<0.01; and 3%% vs 8%, p=<0.01, respectively). No difference in chronic GVHD (36% vs 32%, p=0.28) was reported according to the SC source (PB vs BM), nor in relapse (26% vs 22%, p=0.38) and NRM (23% vs 23%, p=0.60).

At 2 year LFS was 49% vs 54%, p=074 and GRFS was 44% and 42%, p=0.59, for pts receiving BM and PB, respectively. LFS was 53% and 47% (p=0.31) for pts transplanted for AML and ALL, and 56% and 46% (p=0.004) for MAC vs RIC.

In the multivariate analysis adjusted for age, diagnosis, disease status, CMV serostatus, conditioning regimen, center effect and SC source, the use of PB was the sole factor associated with an increased risk of grade II-IV acute GVHD (HR 2.2, 95%CI 1.27-3.9, p=0.005). The type of SC (PB vs BM) was not significant associated with grade III-IV aGVHD (HR 2.5, 95%CI 0.9-7.3, p=0.07), cGVHD (HR 1.0, 95%CI 0.57-1.9, p=0.88), relapse (HR 0.7, 95%CI 0.51-1.15, p=0.21), NRM (HR 0.80, 95%CI 0.49-1.32, p=0.4), GRFS (HR 0.90, 95%CI 0.65-1.25, p=0.56), LFS (HR 0.73, 95%CI 0.52-1.04, p=0.08) and OS (HR 0.79, 95%CI 0.54-1.15, p=0.23).

For LFS and OS, the use of RIC regimen was the only factor associated with a higher risk of treatment failure, (LFS: HR 1.39, 95%CI 1.01-1.93, p=0.04; OS: HR 1.5, 95%CI 1.06-2.12, p=0.02) and higher risk of relapse (HR 1.61, 95%CI 1.06-2.44, p=0.02).

Finally, a center effect was found to be significant for NRM, LFS, GRFS, OS and cGVHD , and entered a frailty variable in the multivariate model.

In conclusion, our study indicates that in patients with acute leukemia in first or second CR receiving haploidentical transplant with PT-Cy, the outcomes according to the use of PB or BM are not different. The ultimate choice depends on transplant center experience and policy.