Comparative Safety and Efficacy of Posttransplant Cyclophosphamide Versus Antithymocyte Globulin in Patients with Cardiovascular Comorbidity Undergoing Allogeneic Hematopoietic Stem Cell Transplantation in First Complete Remission from Unrelated Donors: A Retrospective Pair-Match Analysis of EBMT ALWP

Background

Posttransplant cyclophosphamide (ptCy) or antithymocyte globulin (ATG) is commonly added to a calcineurin inhibitor and an antimetabolite for the prophylaxis of graft-versus-host disease (GVHD) after allogeneic hematopoietic stem cell transplantation (alloHSCT). High-dose cyclophosphamide is associated with a risk of cardiac toxicity, which is not an issue with ATG. We retrospectively analyzed data from the EBMT registry on the outcomes of transplants using ptCy and ATG-based prophylaxis in patients with pretransplant comorbidities associated with cardiovascular risk, as identified on the pretransplant Hematopoietic Cell Transplant Comorbidity Index (HCT-CI) form. Our aim was to evaluate the safety of ptCy in these patients.

Patients and Methods

Patients who underwent their first alloHSCT for AML in CR1 in EBMT centers between 2010 and 2022 from 10/10 or 9/10 HLA antigen-(mis)matched unrelated donors using ptCy or ATG-based GVHD prophylaxis were included in the analysis. Patients had to have a positive history of cardiac disease, arrhythmia, heart valve disease, or cerebrovascular disease on the HCT-CI form and have received a conditioning regimen with a known Transplant Conditioning Intensity (TCI) score without cyclophosphamide or TBI.

To eliminate heterogeneity in patient characteristics between ptCy and ATG groups, a pair-match analysis using a propensity score was performed based on age, year of transplant, donor type, Karnofsky score, TCI, female donor to male patient, and AML cytogenetic classification. The standardized mean difference of variable between groups after pair matching was < 0.1. Kaplan-Meier estimates were calculated for overall survival (OS) and leukemia-free survival (LFS), and the cumulative incidence function in the presence of competing risks was calculated for non-relapse mortality (NRM), relapse, and GVHD. Univariate Cox analysis was used to compare transplant outcomes in ATG vs. ptCy cohorts after matching.

Results

A total of 1256 patients were identified: 1127 received ATG and 129 ptCy. After propensity score pair-matching, 432 patients were included in the analysis: 313 in the ATG group and 119 in the ptCy group.

After matching, the age of patients was 59.9 (IQR 51.8, 66.0) and 60.9 (IQR 52.2, 65.8) years, with 68% females in both groups, 53% and 42% had a HCT-CI <=2, 75% and 78% had a Karnofsky score ≥ 90, TCI was 1-2 in 29% and 29% of patients, 2.5-3.5 in 55% and 52%, and 4-6 in 16% and 18% in the ATG and ptCy groups, respectively. 14% received TBF-based regimens, 25% TreoFlu-based regimens, 44% BuFlu-based regimens, and 15% FluMel-based regimens. The distribution of HCT-CI comorbidities did not differ between groups: 26% had diabetes, 38% cardiac comorbidity, 17% arrhythmia, 9.8% cerebrovascular disease, 13% heart valve disease, and 23% obesity.

No transplant outcome differed significantly between the groups: OS (2 years) was 67.5% (61 - 73.2) and 68.6% (56.7 - 77.8); LFS (2 years) was 60.4% (53.8 - 66.4) and 62.6% (50.4 - 72.6); RI (2 years) was 22.1% (17 - 27.7) and 23.2% (14.3 - 33.4); NRM (2 years) was 17.5% (13.1 - 22.4) and 14.1% (7.5 - 22.8); aGVHD (Grade II or higher at 180 days) was 23.7% (19 - 28.8) and 29.7% (21.7 - 38.1); aGVHD (Grade III or higher at 180 days) was 8.9% (6 - 12.6) and 11.9% (6.8 - 18.5); cGVHD (2 years) was 30.4% (24.5 - 36.5) and 28.3% (18.5 - 38.9); ext-cGVHD (2 years) was 12.1% (8.2 - 16.7) and 13.6% (7 - 22.5) for ATG and ptCy, respectively.

Conclusions

No significant differences in transplant outcomes were observed between ptCy and ATG-based GVHD prophylaxis in patients with comorbidities associated with cardiovascular risk identified on the pretransplant HCT-CI.