Cyclophosphamide-Driven Efficacy in GVHD Unaffected by Matched Unrelated Donor Status

Cyclophosphamide administered alone or in combination with other agents is an effective prophylactic treatment for graft-versus-host disease (GVHD), regardless of stem cell source, conditioning intensity, or patient age in 10/10 and 9/10 human leukocyte antigen (HLA) matched unrelated donor (MUD) hematopoietic stem cell transplantation (HSCT) recipients, according to findings presented at the 2019 European Society for Blood and Marrow Transplantation Annual Meeting.

Improved non-relapse mortality (NRM) rates were observed in both 10/10 and 9/10 MUD patients with cyclophosphamide, demonstrating similar incidence rates of acute and chronic GVHD.

“PTCy is successfully used as a GVHD-prophylaxis backbone in haplo-identical transplantation; these encouraging results prompted us to assess PTCy in the unrelated donor setting, said Francesca Lorentino, MD, of the San Raffaele Scientific Institute, Hematology and Bone Marrow Transplantation Unit in Milan, Italy, who presented the results on behalf of the Acute Leukemia Working Party (ALWP) of the EBMT.

The ALWP assessed the impact of HLA-allele matching on patient outcomes following 10/10 or 9/10 unrelated HSCT when post-transplant cyclophosphamide was used for GVHD prophylaxis. The study included adult patients aged ≥18 years with acute leukemia who underwent HSCT from 2010 to 2017. Of these, 431 patients had HSCT from 10/10 and 234 had HSCT from 9/10 HLA-MUDs. All patients received high-dose, post-transplant cyclophosphamide-based GVHD prophylaxis alone or combined with other immunosuppressant agents.

GVHD-free survival and relapse-free survival were the primary endpoints, and secondary endpoints included leukemia-free survival, overall survival, acute and chronic GVHD, as well as relapse and transplant-related mortality.

Characteristics were similar in the 10/10 and 9/10 groups; median follow-up was 18 months (range, 5-34) versus 19 months (range, 11-31), median age was 43 years (range, 31-57) versus 48 years (range, 32-59), and 25% versus 28% of patients had Karnofsky performance score >90%, respectively. The disease diagnosis was acute myeloid leukemia (AML) in 73% and acute lymphoblastic leukemia (ALL) in 27% of both groups. Also in the 10/10 and 9/10 groups, 65% and 57% of patients had CR1 disease status at transplant, 19% and 24% had >CR1, and 16% and 19% had advanced disease, respectively. Mismatched HLA loci were determined in 37% of patients in HLA-A, 20% in HLA-B, 24% in HLA-C, 5% in HLA-DRB1, and 14% in HLA-DQB1.

Regarding intensity of the conditioning regimen, 55% of patients received myeloablative conditioning (MAC) and 45% received reduced-intensity conditioning (RIC) in the 10/10 arm, while 58% and 42% of patients in the 9/10 arm received MAC and RIC regimens, respectively. In the respective arms, the donor stem cell source was peripheral blood for 90% and 88% of patients, with the remainder receiving stem cells from bone marrow.

Post-transplant cyclophosphamide was administered as monotherapy for GVHD to 7% and 5% of patients in the respective groups, and in combination with a calcineurin inhibitor (CI) with/without mycophenolate mophetil (MMF, 66%; 76%), with methotrexate +/- a calcineurin inhibitor (19%; 14%), and with sirolimus plus MMF (4% in both).

Patient cumulative incidence (CI) outcomes were similar between the two groups across all time points. The 100-day CI (standard deviation) of grades 2/3 acute GVHD was 31% (±5%) and 28% (±6%) in the 10/10 and 9/10 groups (P = .4) and the CI for grades 3/4 was 10±3% and 9±3% (P = .5) respectively.

At 2 years, the CI of chronic GVHD in the respective groups was 32% (±5%) and 38% (±7%; P = .2) and the CI of extensive GVHD 17% (±4%) and 15% (±5%; P = .4), respectively. The 2-year rates of NRM were 19% (±4) in 10/10 MUD compared with 17% (±5; P = .4).

“No interaction was found between donor and the use of in vivo T-cell depletion,” she commented.

By multivariate analysis, no effect of donor HLA-matching was found to affect patient outcome. The covariates assessed included disease (ALL vs AML), disease status (CR1 versus CR>1 versus advanced), Karnofsky performance status (<90 versus ≥90), RIC versus MAC, in vivo T-cell depletion (yes versus no), and stem cell source (peripheral blood versus bone marrow). Only disease status of advanced versus CR1 (HR, 1.5; P = .03) and Karnofsky performance status ≥90 (HR 0.6; P = .001) showed an association with acute GVHD grades ≥2. Karnofsky performance score ≥90 also showed an association with chronic GVHD (HR ,0.6; P = .03).

Regarding NRM, several variables associated significantly, including increased patient age (HR, 1.3; P <.001); disease status with CR>1 versus CR1 (HR, 1.7; P = .06), advanced versus CR1 disease status (HR, 2.5; P <.001), and Karnofsky performance score ≥90 (HR, 0.5; P = .001).

"The potential to limit immunosuppression after HSCT makes post-transplant cyclophosphamide a promising platform for post-graft strategies to prevent relapse; if prospectively confirmed, these data could lead to expansion of a suitable donor pool,” Lorentino concluded.

Lorentino F, Labopin M, Ciceri F, et al. Comparable outcome after 10/10 and 9/10 HLA-matched unrelated donor stem cell transplantation (HSCT) using post-transplant cyclophosphamide (PTCY) for GVHD prophylaxis, on behalf of the ALWP-EBMT. Presented at: 2019 European Society for Blood and Marrow Transplantation Annual Meeting; March 24 to 27, 2019; Frankfurt, Germany. Abstract OS9-1.