Decreased Chronic GvHD with Post-Transplant Cyclophosphamide (PTCY) Compared to ATG in Patients with Myelofibrosis Allografted with Unrelated Donor: A Study from the Chronic Malignancies Working Party of the EBMT

Background

Prospective randomized trials have reported a benefit for anti-thymoglobuline (ATG)-based graft-versus-host disease (GvHD) prophylaxis in the setting of allogeneic hematopoietic stem cell transplantation (alloHSCT) using unrelated donors (UD). However, the optimal GvHD prophylaxis has been recently challenged by emerging post-transplant cyclophosphamide (PTCY) data. We here present the outcomes of PTCY vs ATG as GvHD prophylaxis in myelofibrosis (MF) patients transplanted with UD from the EBMT registry.

Method

A total of 2607 MF patients transplanted from 2012 to 2022 with unrelated donors across 234 participating centers, receiving either PTCY or ATG, were included. Primary endpoints were relapse-free survival (RFS) and the cumulative incidence of aGvHD grade II-IV. Secondary endpoints included overall survival (OS) and the cumulative incidences of engraftment, relapse, non-relapse mortality (NRM), aGvHD grade III-IV and cGvHD. Cox (cause-specific) proportional hazards models were used to obtain adjusted estimates of the effect of PTCY compared to ATG on all endpoints, with a predefined set of adjustment factors (age, DIPSS, donor HLA matching, prior ruxolitinib, comorbidity score, MF subtype (PMF vs sMF)) and a frailty term for transplant center.

Results

Overall, 192 patients received PTCY and 2415 received ATG. Patients in the ATG group were older (median 61 years [IQR 55-65] vs 60 years [IQR 53-64]). Among patients with available data for DIPSS classification at alloHSCT (n=1726), there were more patients in the intermediate-II and high risk in the ATG group (59% vs 48%) than in the PTCY one. Patients in the PTCY cohort had less CMV negative patients-CMV negative donors (17% vs 34%), more patients with Karnofsky ≥90 (73% vs 64%), were more frequently transplanted from an HLA mismatched 9/10 donor (40% vs 21%) and more often transplanted in later/more recent years (42% vs 20%), PBSC as stem cell source was 96% for both.

Engraftment by day 28 was significantly better with ATG (88% vs 74.5%, p<0.001). With a median follow-up of 36.6 (95% CI 35.4 – 38.7) months, 4-year OS was 61% (53-69) with PTCY, 55.1% (53-57%) with ATG, p=0.6. 4-year RFS was 57.5% (49-66) with PTCY vs 47.3% (45-50) with ATG, p=0.2 The 4-year cumulative incidence of relapse (CIR) was 13.5% (8-19) with PTCY vs 22.9% (21-25) with ATG, p=0.07, whereas 4-year NRM was 29% (21-37%) vs 29.7% (28-32) in PTCY and ATG respectively, p=0.9. Grade II-IV aGvHD was 26.7% (20-33%) with PTCY vs 31.8% (30-34%) with ATG, p=0.15 and grade III-IV aGvHD was 14.4% (9-19%) with PTCY vs 15.3% (14-17%) with ATG, p=0.7 while there was a lower incidence at 4 years of cGvHD with PTCY, 28.4% (20-37%) vs 43.8% (42-46) with ATG, p<0.001 as well as for extensive cGvHD with 13.2% (7-20%) vs 22.1% (20-24%), p=0.02.

At multivariable analysis, there were no difference between GvHD prophylaxis groups for the primary endpoints with, for RFS, a HR with PTCY of 0.80 (0.59-1.08), p=0.15 and for aGvHD grade II-IV of 0.83 (0.58-1.19), p=0.31. However, the multivariable analysis confirmed the benefit of PTCY for cGvHD with a HR of 0.57 (0.38-0,84), p=0.005, with a tendency for extensive cGvHD with a HR of 0.58 (0.31-1.07), p=0.08 and showed a tendency of decrease relapse incidence for PTCY with a HR of 0.63 (0.38-1.05), p=0.08 as well as for GRFS with a HR of 0.80 (0.61-1.04), p=0.097.

Conclusion:

This study suggests that MF patients undergoing UD alloHSCT had a decreased incidence of cGvHD using PTCY compared to ATG as GvHD prophylaxis. PTCY is a valid option and considering availability and costs of ATG this may offer a good choice/alternative for regions which do have difficulties to have access to ATG. However, engraftment remains an issue in this setting.