The Impact of Post-Transplant Cyclophosphamide (PTCy) with or without MMF in the Prevention of De Novo and Progressive / Relapsing Chronic Graft-Versus-Host Disease after HLA-Matched Donor Transplant

Background: We previously showed that PTCy/Tac ± MMF is associated with a significant reduction in the rate of chronic graft-versus-host disease (cGvHD) as compared to Tac/MTX after HLA-matched sibling (MSD) but not after HLA-matched unrelated (MUD) donor stem cell transplant (SCT). Notably, ATG was used with Tac/MTX in MUD but not in MSD. [Transplant Cell Ther. 2022;28(10):695] We aimed to further elucidate the epidemiology of cGvHD with the use of PTCy/Tac by comparing the impact of Tac/MTX vs PTCy/Tac ± MMF on de novo and progressive/relapsing cGvHD.

Methods: Recipients of a MSD (Tac/MTX no ATG, or PTCy-based) or MUD (Tac/MTX with ATG, or PTCy-based) peripheral blood SCT between 2015-21 at our institution were eligible. To evaluate the impact of prophylaxis on the incidence of de novo and progressive/relapsing cGvHD, we performed two separate landmark analyses starting 3 months (mo) post-SCT among patients who had not and those who had been diagnosed with acute GvHD (aGvHD) within 3 mo post-SCT, respectively. Patients who died or experienced relapse of malignancy (n=135) or were diagnosed with cGvHD within 3 mo post-SCT (n=13) were excluded from the landmark analyses. There was no difference in the rate of early death/relapse or cGvHD by prophylaxis.

Results: 1040 patients met the eligibility criteria. Median follow-up in survivors was 28 mo (IQRT: 18, 50). The cumulative incidence (CumInc) of cGvHD in the Tac/MTX, PTCy/Tac without MMF, and PTC/Tac with MMF groups was 39% (34-45, reference), 20% (14-29, p<0.001) and 20% (13-30, p=0.003) respectively in MSD; and 23% (18-30, reference), 17% (12-26; p=0.2), and 19% (14-26; p=0.6) respectively in MUD. Consistent with our previous findings, within the PTCy/Tac group, MMF was not associated with the overall rate of cGvHD in MSD (HR=1.2, 95% CI 0.6-2.1, p=0.6) or MUD (HR=1.3, 95% CI 0.7-2.2, p=0.4). However, in the current study, stratified analyses performed to evaluate the impact of prophylaxis on the risk of de novo vs progressive/relapsing cGvHD revealed a potential role for MMF in the development of de novo cGvHD. Subsequent results are presented separately for the de novo and progressive/relapsing cGvHD risk cohorts (Table).

The de novo cGvHD risk cohort included patients (N=442) who had not been diagnosed with aGvHD within 3 mo post-SCT. AML / MDS (60%) and CML / MPD (16%) were the most common diagnoses. 124 cases of cGvHD were diagnosed 3 -36 mo post-SCT with a CumInc of 32% (27-37). In multivariate analysis (MVA), compared to Tac/MTX, the use of PTCy/Tac without MMF (HR=0.3, 95% CI 0.2-0.6, p<0.001) was associated with a significant reduction in the incidence of cGvHD. Such reduction was not observed (Figure) with PTCy/Tac with MMF (HR vs Tac/MTX=1.0, 95% CI 0.6-1.5, p=0.9). Additional independent predictors were myeloablative conditioning (HR=1.9, 95% CI 1.3-2.9, p=0.001) and MSD (HR=1.5, 95% CI 1.2-2.6, p=0.003). Notably, progression-free survival (PFS) was similar in PTCy/Tac/MMF (univariate HR: 0.8, 95% CI 0.5-1.4, p=0.4) vs PTCy/Tac.

The progressive/relapsing cGvHD risk cohort included patients (N=450) who had been diagnosed with grade 1 (27%), 2 (59%) or 3-4 (14%) aGvHD within 3 mo post-SCT. AML / MDS (64%) and CML / MPD (14%) were the most common diagnoses. 109 cases of cGvHD were diagnosed 3 -36 mo post-SCT with a CumInc of 28% (24-33). In MVA analysis, the impact of prophylaxis differed by donor type. As compared with Tac/MTX, PTCy/Tac ± MMF was associated with significantly lower rate of progressive/relapsing cGVHD in MSD (HR=0.2, 95% CI 0.1-0.4, p<0.001) but not in MUD (HR=0.6, 95% CI 0.4-1.1, p=0.09). This may be related to the use of ATG with Tac/MTX in MUD which was not used in MSD. HCT-CI ≥3 was the only other significant predictor of cGvHD (HR=0.6, 0.4-0.9, p=0.03). PFS was similar in PTCy/Tac ± MMF vs Tac/MTX in MSD (univariate HR: 0.8, 95% CI 0.5-1.3, p=0.3) or MUD (univariate HR: 0.8, 95% CI 0.5-1.2, p=0.3).

Conclusions: PTCy/Tac ± MMF prophylaxis slows the development of progressive/relapsing cGvHD in MSD vs (Tac/MTX no ATG), but not in MUD (vs Tac/MTX with ATG). The addition of MMF to PTCy/Tac (vs PTCy/Tac) has no significant impact on the risk of progressive/relapsing cGVHD in MSD or MUD; but is associated with a higher risk of de novo cGvHD in MSD and MUD. Our data suggest that multi-center studies are warranted to determine whether MMF should be excluded from cGvHD prevention regimens in the HLA-matched donor setting with PTCy- based GvHD prophylaxis.