Results of a Two-Arm Phase II Clinical Trial Using Post-Transplantation Cyclophosphamide for Prevention of Graft-Versus-Host Disease in Haploidentical and Mismatched Unrelated Donors Hematopoietic Stem-Cell Transplantation

Allogeneic stem cell transplantation offers curative therapy for many patients (pts) with high-risk hematologic malignancies. Donor availability remains a major limitation for many pts. The introduction of high-dose post-transplant cyclophosphamide (PTCy) has significantly improved the outcomes of pts undergoing haploidentical (HAPLO) stem cell transplants. The choice between a HAPLO or a one-antigen HLA mismatched unrelated donor (9/10 MUD) for pts lacking an HLA-matched donor remains unclear.

Methods:

We conducted a prospective non-randomized phase 2 clinical trial with two parallel arms, HAPLO (n=60) and 9/10 MUD (n=46) transplants, for pts with advanced hematologic malignancies or aplastic anemia who lacked an HLA-matched unrelated donor type at 10 loci (HLA-A, -B, -C, -DRB1, and -DQB1) using a MEL-based reduced-intensity conditioning regimen. The regimen included a single intravenous dose of MEL 140 mg/m² (day -7), thiotepa 5 mg/kg (day -6), and four daily IV doses of fludarabine 40 mg/m² (day -5 to day -2) (FM140). Thiotepa was intermittently available and was replaced by total body irradiation at a dose of 2 Gy on day -1. Pts >55 years (yr) old or with significant comorbidities received a lower MEL dose (100 mg/m²) (FM100). All pts with CD20-positive lymphoma received rituximab (375 mg/m²) on days -13, -6, +1 and +8. GVHD prophylaxis consisted of PTCy 50 mg/kg on day +3 and +4, and tacrolimus and mycophenolate for 6 and 3 months (mo), respectively. The stem cell source was unmodified bone marrow for both arms.

Results:

Patient characteristics are shown in Table 1. The median follow-up duration was 24 mo in the HAPLO arm and 29 mo in the 9/10 MUD arm. The cumulative incidence (CI) of neutrophil (ANC) recovery at day 45 was 95% and 98% in the HAPLO and 9/10 MUD arm, respectively. The median time to ANC recovery was 18 days in both arms; the median time to platelet recovery was 25 days in the HAPLO arm and 28 days in the 9/10 MUD arm. Primary graft failure developed in two pts in the HAPLO arm (one due to anti-donor HLA antibodies) and one patient in the 9/10 MUD arm. One pt in both arms developed mixed donor chimerism at day 100; otherwise, all pts in both arms achieved full (>95%) donor chimerism. Bone marrow was the graft source in all pts except 2 in the HAPLO arm and 8 in the 9/10 MUD arm who received a peripheral blood graft. The 1-yr overall and progression free survival were 70% and 60%, respectively, in the HAPLO arm (Fig. 1A) and 60% and 47%, respectively, in the 9/10 MUD arm (Fig. 1B). Day 100 CI of grade II-IV aGVHD and III-IV aGVHD were 28% and 3%, respectively, in the HAPLO arm versus 33% and 13%, respectively, in the 9/10 MUD arm; the 2-yr CI of chronic extensive GVHD was 13% and 14% in the two groups, respectively. The 1-yr CI of non-relapse mortality was 21% in the HAPLO arm and 31% in the 9/10 MUD arm, while the 1-yr relapse rate was 19% and 25% in the two groups, respectively.

Conclusions:

This study establishes PTCy, tacrolimus, and mycophenolate as an effective regimen for GVHD prevention in mismatched transplantation using both haploidentical and mismatched unrelated donor sources. Melphalan-based reduced-intensity conditioning is an effective regimen for a broad range of hematologic malignancies. Prospective randomized studies comparing haploidentical and unrelated donor sources are needed.

Figure 1.

View largeDownload slide

Figure 1.

View largeDownload slide

Close modal