Post-Transplant Cyclophosphamide Is Associated with Improved Clinical Outcomes in HLA-Mismatched Unrelated Donor Hematopoietic Cell Transplantation

The current standard of care to prevent graft versus host disease (GVHD) after HLA mismatched unrelated donor (URD) allogeneic hematopoietic cell transplantation (HCT) is tacrolimus, methotrexate (MTX) and anti-thymocyte globulin (ATG). Recently, an approach based on administration of post-transplant cyclophosphamide (PT-Cy) demonstrated promise in a prospective trial for HLA mismatched URD HCT. Here, we compared tacrolimus/MTX/ATG (ATG group) treatment, with that of PT-Cy, mycophenolate mofetil, and tacrolimus or sirolimus (PT-Cy group) after HLA mismatched URD HCT, at two academic centers with similar peri-transplant supportive care practices.

Subjects that underwent HCT from a URD mismatched at one or more loci among HLA-A, -B, -C, and -DRB1 from 2010-2020 were included. The primary endpoint of our study was one-year GVHD-free, relapse-free survival (GRFS). Of 128 total subjects included (ATG: 46; PT-Cy: 82), the median age was 54.7 (range: 21.0 - 72.0) in the ATG group and 60.3 (21.0 - 75.2) in the PT-Cy group. Notably, 74 subjects (57.8%) belonged to a racial/ethnic minority, while 53 (41.4%) received a bone marrow (BM) graft, and 26 (20.3%) received a highly mismatched (i.e., <6/8 HLA) graft. A higher proportion of PT-Cy subjects received BM (50% vs. 26.1%, P= 0.01) and highly mismatched grafts (30.5% vs. 2.2%, P=0.001). The groups were well matched for key demographic data, including indication for HCT, high-risk disease, and HCT comorbidity index.

The rate of primary neutrophil engraftment at day 28 post HCT was similar between the two groups: 91% (83%-99%), in the ATG group, and 90% (84%-97%), in the PT-Cy group. The median day of neutrophil engraftment was +17 in the ATG, and +20 in the PT-Cy groups. Platelet engraftment at day 100 was 84.8% (74.4% - 95.2%) and 86.6% (79.2% - 94%) in the ATG and PT-Cy groups, respectively. Among subjects that survived 3-months post HCT and had donor chimerism testing, full donor chimerism (≥95% donor) was achieved in 24/26 (92.3%) patients in the ATG group, and 61/68 (89.7%) subjects in the PT-Cy group.

Overall, the one-year GRFS was 16% (95% confidence interval: 8% - 31%) vs. 54% (44% - 66%, p < 0.001) in the ATG and PT-Cy groups, respectively. The adjusted, multivariate hazard ratio (HR) for GRFS was 0.34 (0.21 - 0.55, p < 0.001), in association with the use of PT-Cy. The one-year overall survival (OS) in the ATG group was 45% (32% - 62%) vs. 75% (66% - 85%, p < 0.001) in the PT-Cy group. There was no difference in 2-year relapse incidence (27%, 14-39% vs. 19%, 10% - 28%, p = 0.2), whereas 1-year non-relapse mortality (NRM) was increased with ATG-based prophylaxis: (38%, 23% - 52% vs. 16%, 9% - 25%, p = <0.001). The use of PT-Cy was associated with improved GRFS for patients treated with BM (HR = 0.27, 0.11 - 0.62; p = 0.002) and G-CSF mobilized, peripheral blood derived (PB) grafts (HR = 0.39, 0.21 - 0.71; p = 0.002). BM grafts were not an independent predictor of GRFS (HR: 1.2, 0.8 - 1.9; p = 0.5) in a multivariate Cox regression model.

The use of PT-Cy, compared to ATG, was associated with lower rates of grade 3-4 acute GVHD in the entire cohort (ATG: 22%, 18% - 45% vs. PT-Cy: 15%, 8% - 23%; p = 0.03), and in the sub-cohorts of patients transplanted with PB (ATG: 33%, 17% - 49% vs. PT-Cy: 20%, 9% - 33%) and BM grafts (ATG: 25%, 1% - 50% vs. PT-Cy: 10%, 1% - 19%). Chronic GVHD requiring systemic immune suppression was more frequent in the ATG/PB group (26%, 13% - 42%) than in the ATG/BM (8%, 0% - 24%), the PT-Cy/PB (8%, 2% - 19%), and the PT-Cy/BM (10%, 1% - 19%) groups.

Despite greater HLA-mismatching, PT-Cy following mismatched URD HCT resulted in superior GRFS, OS and lower NRM and GVHD rates compared to ATG-based GVHD prophylaxis. These results indicate that PT-Cy is the preferred method to prevent GVHD after mismatched URD HCT and confers a benefit in patients treated with either PB or BM-derived allografts. Combinatorial approaches, using the T-cell co-stimulatory blockading agent abatacept, in addition to tacrolimus/methotrexate, may also prevent GVHD in this population, and can be compared to a PT-Cy-based approach in a prospective clinical trial.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal