Favorable Outcome of Sequential HLA-Haploidentical Transplantation Using Ptcy As GvHD Prophylaxis in High Risk AML and MDS of the Elderly

Sequential conditioning regimens, comprising cytoreductive chemotherapy shortly applied prior to reduced intensity conditioning are successfully used for high-risk (HR) AML/MDS in matched related and unrelated donor hematopoietic stem cell transplantation. However, few data are available for sequential conditioning in the context of HLA-haploidentical transplantation (haplo-HSCT), especially in the elderly. To investigate the relative merits of sequential haplo-HSCT in the elderly we retrospectively analyzed the outcome of thirty-five patients (pts) with advanced AML/MDS (>=50 years old).

Thirty-three pts suffering from HR AML, defined by refractory, relapsed or secondary leukemia, or complete remission with adverse-risk genetics according to ELN criteria and two pts with HR MDS according to IPSS-R, who underwent T-cell-replete haplo-HSCT at our institution between January 2009 and November 2016 were included. Disease was active in 29 pts while 6 had achieved CR. Pre-transplantation risk factors were scored using the hematopoietic cell transplantation-specific comorbidity index (HCT-CI) which was ≥3 in 13 pts (median HCT-CI:2, range 0-8). A sequential therapeutic concept using either FLAMSA (n=26) or clofarabine (n=9) for cytoreduction was used prior to RIC in all pts. Bone marrow (54%) and peripheral blood stem cells (46%) were both used as graft source. Post-grafting immunosuppression consisted of high-dose cyclophosphamide, tacrolimus and MMF in all pts.

Median age was 60 years (50-70). One graft rejection occurred. Three pts died early in aplasia. Neutrophil and platelet engraftment was achieved in 95% and 77% of evaluable pts, respectively at a median of 16,5 (13-89) and 31,5 (11-103) days.Acute GvHD grade I-III occurred in 25/32 of the pts (grade III n=2); no patient developed grade IV aGvHD. Chronic GvHD was observed in 13/29 pts and was most frequently assessed as mild (n=6) or moderate (n=5) while 2 pts developed severe cGvHD. No GvHD related death was observed. CI of NRM at day 100, 1-year and 3-years was 11%, 23% and 23%, respectively. CI of relapse at 1- and 3-years was 15% and 27%, with a median time to relapse of 152 days (20-413). At a median follow up of 27 months (4-74), estimated one- and three-year overall survival (OS) was 62% and 52% respectively. One- and three-year leukemia-free survival (LFS) was 59% and 52%.

Our results suggest that using a sequential therapeutic concept in PTCY-based haplo-HSCT is safe and properly tolerated while it provides a favorable disease control when treating elderly HR MDS/AML pts. Thus, sequential haplo-HSCT seems to be a valuable alternative in pts who lack a conventional donor or are in urgent need for prompt transplantation.