A randomized 3-arm Phase II trial involving 208 older or medically infirm patients (pts) with hematological malignancies given unrelated HCT after minimal intensity conditioning demonstrated that adding sirolimus to tacrolimus and MMF resulted in less grades II-IV GVHD, less steroid use, and less CMV reactivation (Hematologica 2014; 99(10); 1624). Based on this trial we designed a Phase III multi-site trial comparing triple therapy with sirolimus/MMF/CSP (Arm2) to the standard immunosuppressive regimen of MMF/CSP (Arm1). The primary objective of the trial was to compare the respective incidences of grades II-IV acute GVHD. Secondary objectives included comparing non-relapse mortality, survival, and progression-free survival. Pts in both Arms received CSP 5 mg/kg bid starting on day -3 through day 96 with a taper through day 150. In the first 28 days after HCT CSP was targeted to 400 and 350ng/ml in Arm1 and Arm2, respectively and 120-360ng/ml after day 28 in both arms. Arm 1: MMF was given daily at 15 mg/kg Q8 hours until day +30, reduced to 15 mg/kg Q12 hours until day 150, then tapered through day 180. Arm 2: MMF was given daily at 15 mg/kg Q8 hours until day +30, reduced to 15 mg/kg Q12 hours until day 40, then discontinued (no taper). In addition to the MMF/CSP, sirolimus was administered starting on day -3 at 2.0 mg/day through day 150 with a target level of 3-12ng/ml, with tapering off by day 180. The target enrollment was 300 pts with a built-in interim analysis for futility. At the time of the interim analysis, 158 pts ineligible for high-dose conditioning had been enrolled (Nov. 2010 to Jan. 2016: Arm1 n=74, Arm2 n=84) Their median age was 62 (range 18-79) yrs. The median HCT comorbidity index (HCT-CI) was 3 (range 0-10). Five pts had 6 previous allogeneic HCT and 32 pts (20%) had 36 previous autologous HCT. All pts were matched for HLA-DRB1 and -DQB1 at the allele level: 8 had single allele mismatches at HLA-A, -B or -C and the remainder (n=150) were fully HLA-matched. Diagnoses included AML (n=64), MDS/MPD (n=30), NHL (n=23), MM (n=13), CLL (n=13), ALL (n=11), HL (n=2), and CML (n=2). Randomization was stratified by transplant center. Unmodified PBSC grafts contained a median of 8.0 ×106 CD34 and 2.9 × 108 CD3 cells/kg. Conditioning consisted of fludarabine 90mg/m2 and 2-3 Gy TBI. Sustained donor engraftment occurred in 99% of pts. The median follow-up of surviving pts was 24 (range, 1-65) months. Table 1 and Figures 1 and 2 summarize results. The day-100 cumulative incidences of grades II-IV acute GVHD were in Arm1: 53%, and Arm2: 25% (p=0.0001) and the grades III-IV acute GVHD were in Arm1: 8%, Arm2: 2% (p=0.04). The 1-year cumulative incidence of chronic extensive GVHD was similar between Arm1: 49%, and Arm2: 48% (p=0.94). The 1-year cumulative incidence of non-relapse mortality was lower in Arm2 (Arm1: 15% and Arm2: 5%; p=0.007), while relapse/progression was similar at 1 year for Arm1: 21%, and Arm2: 19% (p=0.86). The 1-year overall and progression-free survivals for Arm1 vs. Arm2 were: 72% vs. 85% (p=0.03), and 65% vs. 77% (p=0.08); respectively. T-cell (CD3) donor chimerism was lower in Arm 2 on day 28 (median, Arm1 85%; Arm2 80%; p=0.05) with no differences seen in granulocyte (CD33: Arm1 98%, Arm2 95%) or NK cell (CD56: Arm1 96%, Arm2 97%) donor chimerisms.

In summary, the interim analysis showed that adding sirolimus to MMF and CSP not only reduced the risks of grades II-IV but also of grades III-IV acute GVHD and of non-relapse mortality without increasing the risk of relapse or progressive malignancy. Based on these findings and the significantly improved survival, the DSMB recommended the trial be closed. This triple immunosuppressive regimen should, therefore, be considered in the future as the standard of care in pts given unrelated donor grafts after minimal intensity conditioning.


Pulsipher:Novartis: Consultancy, Other: Study Steering Committee; Chimerix: Consultancy; Jazz Pharmaceutical: Consultancy; Medac: Other: Housing support for conference.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution