Triple-Drug Graft-versus-Host Disease Prophylaxis after HLA-Matched Unrelated Donor Nonmyeloablative Allogenic Hematopoietic Stem Cell Transplantation

Abstract

Adding sirolimus to graft-versus-host disease (GVHD) prophylaxis with cyclosporin and mycophenolate mofetil (MMF) reduced the risk of grade II-IV acute GVHD after nonmyeloablative (NMA) allogenic hematopoietic stem cell transplantation (HSCT) with an HLA-matched unrelated donor in a randomized clinical trial. We analyzed real-life data to investigate the impact of implementing the triple-drug regimen with cyclosporin, MMF and sirolimus as standard GVHD prophylaxis after NMA HSCT with an HLA-matched unrelated donor at our institution. We studied all adult patients (age ≥18 years) who underwent NMA HSCT with an HLA-matched unrelated donor at Rigshospitalet, Copenhagen University Hospital, Denmark between 2018 and 2021 and received GVHD prophylaxis with cyclosporin, MMF and sirolimus (triple-drug group [TDG]). Comparisons were made with a historical cohort who received tacrolimus and MMF as GVHD prophylaxis after HLA-matched unrelated donor NMA HSCT between 2014 and 2017 (control group [CG]). Outcomes were grade II-IV and grade III-IV acute GVHD, chronic GVHD, relapse, nonrelapse mortality (NRM) andoverall survival (OS). A total of 264 patients were included (TDG, n = 137; CG, n = 127). Median age was 66 years (interquartile range [IQR], 58 to 69 years) in the TDG and 63 years (IQR, 57 to 68 years) in the CG. Acute myeloid leukemia and myelodysplastic syndrome were the most frequent indications for HSCT in both groups (TDG, 33% and 23%, respectively; CG, 36% and 22%, respectively). The cumulative incidence at day +110 of grade II-IV GVHD was 17% (95% confidence interval [CI] 11% to 23%) in the TDG versus 29% (95% CI, 21% to 37%) in the CG (P = .02, Gray's test) and that of grade III-IV acute GVHD was 3% (95% CI, 0 to 6%) versus 5% (95% CI, 1% to 8%), respectively (P = .4, Gray's test). In a Cox regression model adjusted for age, donor age and female donor to male recipient the risk of grade II-IV acute GVHD was lower in the TDG compared to the CG (hazard ratio [HR], .51; 95% CI .30 to .86; P = .01). The 2-year OS was 77% (95% CI, 70% to 84%) in the TDG and 69% (95% CI, 61% to 77%) in the CG (P = .04), and this difference remained significant after adjustment for age and Karnofsky Performance Status (HR, .65; 95% CI, .42 to .99; P = .04). The 2-year cumulative incidences of chronic GVHD, relapse and NRM were 60% (95% CI, 51% to 69%), 21% (95% CI, 13% to 28%), and 12% (95% CI, 6% to 17%), respectively, in the TDG and 62% (95% CI, 54% to 71%), 27% (95% CI, 19% to 35%) and 14% (95% CI, 8% to 20%), respectively, in the CG. Multivariable analyses revealed no difference in the risk of chronic GVHD (HR, .91; 95% CI, .65 to 1.26; P = .56), relapse (HR, .70; 95% CI, .42 to 1.15; P = .16) or NRM (HR, .56; 95% CI, .31 to 1.05; P = .07). After changing the standard GVHD prophylaxis in patients undergoing NMA HSCT with an HLA-matched unrelated donor from tacrolimus and MMF to cyclosporin, MMF and sirolimus, we observed a reduction in the incidence of grade II-IV acute GVHD and improved 2-year OS.