Abstract
Introduction. Results of our previous studies have suggested that the use of mycophenolate mofetil (MMF) for graft-versus-host disease (GvHD) prophylaxis may not confer clinical benefit and may adversely affect outcomes in certain hematopoietic cell transplantation (HCT) contexts. In HLA-matched HCT utilizing post-transplant cyclophosphamide (PTCy) and tacrolimus, MMF was associated with a higher incidence of moderate acute GvHD, delayed immune reconstitution, and increased bacterial infections, with a modest reduction in relapse rates (Mehta et al, Transplant Cell Ther, 2022). These observations prompted further investigation in a larger, single-center cohort, the findings of which are presented below.
Methods. AML/MDS patients who underwent their first HLA-matched unrelated (MUD) or related (MRD) HCT with peripheral blood stem cells between 2018-2024 at our institution and received PTCy-based (± MMF) were eligible for this study. The impact of MMF on HCT outcomes was evaluated in univariate and multivariate analyses, accounting for competing risks as applicable.
Results. A total of 846 consecutive patients underwent their first MUD (N=547) or MRD (N=299) HCT during the study period. Median (range) recipient age and Hematopoietic Cell Transplantation-specific Comorbidity Index (HCT-CI) were 63 (18-78) years and 3 (0-10), respectively, and 52% (N=442) of patients had HCT-CI ≥3. Seventy-two percent (N=612) of patients had AML and 46% (N=390) had high-risk (defined by ELN17 for AML and IPSS-R for MDS) disease. Conditioning was myeloablative in 71% (N=599) of transplants, and MMF was included in the GvHD prophylaxis in 70% (n=595) of transplants. The median (range) follow-up was 30 (2-86) months. At 2 years post-HCT, overall survival (OS) was 62% (95% Confidence Interval [CI]: 59-66), non-relapse mortality (NRM) was 16% (95% CI: 13-18), and disease progression was 26% (95% CI: 23-29). In multivariate analysis, the use of MMF was associated with significantly inferior OS (53% vs 65%, HR=1.6, 95% CI: 1.3-2.1, P<0.001) and higher NRM (20% vs 13%, HR=1.6, 95% CI: 1.1-2.3, P=0.007) in patients with HCT-CI ≥3. MMF did not have a significant impact on OS or NRM in patients with HCT-CI<3. The relapse rate did not differ in the MMF vs no MMF subsets (HR=0.97, 95% CI: 0.7-1.3, P=0.8) in multivariate analysis. The cumulative incidence of grade III-IV acute GvHD was 8% (95% CI: 6-10) at 6-months post-HCT. The rate of grade III-IV acute GvHD did not differ by MMF (6-months incidence of 8% vs 7% in the MMF vs no MMF subsets, HR=1.1, 95% CI: 0.6-1.9, P=0.8) or by HCT-CI (6-months incidence of 8% vs 7% in HCT-CI ≥ vs <3, HR=1.1, 95% CI: 0.7-1.8, P=0.7). Multivariate analysis was not indicated for grade III-IV acute GvHD because no significant predictors were identified in univariate analysis. The cumulative incidence of chronic GvHD was 17% (95% CI:15-20) at 2 years. The rate of chronic GvHD did not differ by MMF (2-years incidence of 18% vs 16% in the MMF vs no MMF subsets, HR=1.05, 95% CI: 0.7-1.5, P=0.8) or by HCT-CI (2-years incidence of 15% vs 20% in HCT-CI ≥ vs <3, HR=0.8, 95% CI: 0.5-1.1, P=0.1), in univariate analysis. There was also no significant (HR=1.3, 95% CI: 0.8-1.8, P=0.2) association between MMF and chronic GVHD in multivariate analysis. Bacterial (14% vs 5%, p=0.3) and fungal (11% vs 0%, p=0.2) infections were more common causes of NRM in the MMF vs no MMF subsets in patients with HCT-CI ≥3. Acute GvHD accounted for 11% and 16% (p=0.4) of NRM in these groups, respectively.
Conclusions: Among patients undergoing matched unrelated or related donor HCT, the addition of MMF prophylaxis was associated with significantly inferior OS and increased NRM, which was likely driven by infections, among patients with higher comorbidity burden. There was no discernable effect of MMF on severe acute or chronic GVHD.
