Haploidentical, G-CSF Primed, Unmanipulated Bone Marrow Transplantation with Anti-CD25 Monoclonal Antibody As Part of the Gvhd Prophylaxis for Patients with High-Risk Acute Myeloid Leukemia

Background. During the last years, hematopoietic stem cell transplant from haploidentical donor is increasing by using the unmanipulated graft. We report the results of allogeneic bone marrow transplantation (BMT) from haploidentical donor primed with G-CSF in patients (pts) with high-risk acute myeloid leukemia (AML), who received an uniform TBF (Thiotepa, Busulfan, Fludarabine) conditioning regimen including antitymocyte globulin and an identical GVHD prophylaxis combining MTX+CSA, MMF and an anti-CD25 monoclonal antibody (Basiliximab).

Patients and Methods. Since August 2005 to December 2014, 165 pts with high-risk hematological malignancy underwent unmanipulated BMT from fully haploidentical donor primed with 4µg/kg/day of G-CSF over 7 days. Of these 165 pts, 71 (median age 49 yrs, (range5-66) were affected by high-risk AML and were in early (CR1 and CR2: n=58) or advanced (>CR2 or active disease: n=13) phase at time of transplant. All pts received the same myeloablative (MAC: n=47) or reduced intensity (RIC: n=24) TBF conditioning regimen including antithymocyte globuline for 4 days (days-4 through -1) and an identical graft-versus-host disease (GVHD) prophylaxis consisting of the classical association of MTX+CSA combined with MMF from day +7 to day +100 and the administration of 20 mg of Basiliximab at days 0 and +4. The donor were represented by: son (n=25); brother (n=18), mother (n=11), daughter (n=8), sister (n=6) and father (n=3). Anti-infective management, supportive care and transfusion policy were identical for all pts. The experience was conducted at 2 Institutions: Stem Cell Transplant Unit at University Tor Vergata of Rome (n=45) and at Civil Hospital of Pescara (n=26), Italy.

Results. All pts engraftedwith full donor chimerism: median time 20 days (range, 11-35) and CI 94±3%. With a median follow-up (FU) of 3.9 yrs (range, 0.7-7.4), the 7-yrs probability of overall survival (OS) and disease free survival (DFS) for all pts was 45±6% and 43±6%, respectively. OS calculated by disease status at transplant was 53±7% at 7 yrs for 58 pts transplanted in early phase and 0.0% at 4 yrs for 13 transplanted in advanced phase (p=0.0027), DFS was 49±7% and 15±10% at 7 and 2 yrs (p=0.007), respectively.

TBF-MAC Regimen: 47 pts (median age 43 yrs, range 5-63), in early (n=38) or advanced (n=9) phase at transplant received a TBF-MAC regimen. For all these pts the CI of engraftment was 94±4% at a median of 20 days (range, 12-35). The 100 day CI of acute GVHD (AGVHD) grade II-IV and III-IV was 32±7% and 6±4%, respectively. The CI of extensive chronic-GVHD (CGVHD) was 19±7%. In the patients transplanted in early phase the CI of transplant related mortality (TRM) at 6 months and 7 yrs was 21±7 and 27±7%, respectively. The CI of relapse at 1 and 7 yrs was 16±6% and 24±7%, respectively. With a median FU of 3.9 yrs (range, 1.3-7.4), the 7-yr probability of OS and DFS was respectively 60±8% and 52±8% with a plateau of the curve from 2 yrs after transplant. None of the 15 variables considered in multivariate analysis (Cox and Fine and Gray models) influenced significantly the outcomes.

TBF-RIC Regimen: 24 pts (median age 61.5 yrs, range 41-66) were conditioned with TBF-RIC. All pts engrafted with full donor chimerism (CI 96±6%). The 100 day CI of AGVHD grade II-IV and III-IV was 33±10% and 13±7%, respectively. The CI of extensive CGVHD was 8±8%.Four pts were in advanced and 20 in early phase at transplant. For these last pts, the CI of TRM at 6 months and 6 yrs was 15±8% and 25±10%, respectively. The CI of relapse at 1 and 6 yrs was 15±8% and 27±11%, respectively. With a median FU of 4.1 yrs (range, 0.8-6), the 6-yr probability of OS and DFS was respectively 42±12% and 41±12% with a plateau of the curve from 2 yrs after transplant. In multivariate analysis, no significant factor was identified.

Conclusions. Although the number of pts does not allow an extensive analysis of the results, the homogeneity of our series in terms of diagnosis, patient selection, conditioning regimen and GVHD prophylaxis supports the conclusion that haploidentical, G-CSF primed, unmanipulated BMT is a valid alternative for high-risk AML pts lacking an HLA identical sibling. The regimen of GVHD prophylaxis is highly effective and can be proposed in alternative to the use of post-transplant cyclophosphamide, proposed by the Baltimore group. Finally, TBF-RIC regimen can be confidently offered to unfit or older AML pts candidate to an haploidentical transplant.