In this prospective study, allogeneic HCT using PTCy in patients with higher-risk MDS demonstrated a considerably low incidence of grade II-IV and III-IV acute GVHD, at 9.2 and 2.6%, respectively. The cumulative incidence of moderate-to-severe chronic GVHD was 22.7% at 2-years. These results align with previous studies investigating allogeneic HCT with GVHD prophylaxis utilizing PTCy. After Luznik et al. first introduced non-myeloablative conditioning followed by PTCy in HLA-haploidentical HCT,4 Bacigalupo et al. reported outcomes of a retrospective study involving 148 patients with hematologic malignancies undergoing myeloablative conditioning HLA-haploidentical BM transplantation using PTCy.18 In the study, grade II-IV and III-IV acute GVHD incidence was 24 and 10%, respectively, and that of extensive chronic GVHD was 12%. In a prospective study of allogeneic HCT in patients with acute leukemia from unrelated donors using PTCy, the incidences of grades II-IV and III-IV acute and chronic GVHD were 19, 4, and 16%, respectively.7 Another phase 2 randomized prospective study reported that PTCy-based GVHD prophylaxis showed incidences of 27% for grade II-IV acute GVHD, 2% for grade III-IV acute GVHD, and 22% for chronic GVHD requiring immunosuppressive therapy.9
When compared with ATG, our study demonstrated that PTCy-based GVHD prophylaxis exhibited a significantly lower incidence of grades II-IV and III-IV acute GVHD, whereas that of moderate-to-severe chronic GVHD was similar between the two groups. The benefit was more pronounced in HCT from HLA-matched donors compared to haploidentical familial donors. Controversy exists regarding the differences in GVHD incidence between PTCy and ATG. In a prospective study by Moiseev et al., the PTCy group showed significantly lower incidences of grades II-IV and III-IV acute and chronic GVHD than the historical ATG group, resulting in lower NRM and longer OS and EFS.7 However, the registry data from the European Society for Blood and Marrow Transplantation in the setting of HLA-mismatched unrelated donor HCT for patients with acute myeloid leukemia (AML) exhibited similar incidences of grade II-IV acute, chronic, and extensive chronic GVHD compared to ATG, except for lower incidence of grade III-IV acute GVHD in the PTCy group.19 Brissot et al. conducted a retrospective study including 174 patients with AML receiving HCT from HLA-mated unrelated donors and observing no significant difference between PTCy and ATG regarding acute and chronic GVHD, NRM, and CIR.11 In a meta-analysis, PTCy was associated with a lower incidence of grades II-IV and III-IV acute GVHD and NRM, without a difference in chronic GVHD.20 However, these inconsistent results should be interpreted cautiously owing to imbalances in major factors affecting GVHD, including disease types and status, donor types, and conditioning regimens across the studies.
We found a similar recovery pattern between the two groups regarding immune cell reconstitution. The number of NK cells remained consistently higher in the ATG group compared to the PTCy group, and an earlier recovery of CD8 + T cells was observed in the ATG group. Leserer et al. analyzed the reconstitution of T cell subsets in the context of matched unrelated donor HCT comparing ATG and PTCy. They found that absolute counts of the CD4 + T cell were consistently higher in the PTCy group, and the CD8 + T cell counts were similar between the two groups.21 Additionally, the absolute number of regulatory T cells was significantly higher in the PTCy group for the first 6 months than the ATG group, whereas the absolute counts of NK and B cells showed no significant difference between the two groups. Given the variability in patient populations and transplantation procedures between these studies, the results should be interpreted cautiously.
In this study, we observed significantly worse OS and EFS in patients receiving HCT from HLA-haploidentical donors than other donors. These inferior outcomes of haploidentical HCT are associated with a higher incidence of NRM, including GVHD, rather than disease relapse. Since we applied the same intensity of immunosuppression across all donor types and observed different outcomes after adjusting for potential confounders, different approaches may be required according to donor types. For instance, in a phase 2 study of GVHD prophylaxis with sole PTCy in 49 patients with AML/MDS receiving reduced-intensity conditioning HCT from matched donors, the PTCy cohort showed higher grade II-IV acute GVHD and NRM and shorter OS compared with a tacrolimus/MTX cohort, suggesting PTCy as a sole GVHD may not be sufficient in this transplantation setting.22 Pedraza et al. conducted a retrospective study of PTCy and tacrolimus without MMF in HCT from mismatched or matched unrelated donors.23 They found no differences between mismatched and matched unrelated donor HCT regarding GVHD, CIR, and NRM, suggesting that reduced immunosuppression could be an acceptable strategy. These results suggest that the degree of immunosuppression should be decided based on conditioning intensity, donor types, and disease status, and further prospective studies are needed to support these findings.
We observed that older patient age, higher IPSS-R pre-HCT, and HLA-haploidentical familial donors were associated with worse survival outcomes, and IPSS-R pre-HCT was strongly correlated with CIR and OS. Reducing disease burden before transplantation is recognized to improve survival after HCT for MDS, especially in patients receiving reduced-intensity conditioning.24 While patients with higher-risk MDS have traditionally been recommended for HCT consideration at diagnosis,25 pre-transplant treatment to lower disease burden with active treatment including novel agents, should be carefully evaluated.
Our study had some limitations. First, data from the comparative ATG group were collected retrospectively, introducing potential selection bias. Second, owing to the study’s retrospective nature, patient characteristics were not fully balanced between the two groups. Considering that patients in the ATG group underwent HCT much earlier than those in the PTCy group, supportive care may have differed between the two groups. Despite these limitations, our study effectively demonstrates the efficacy of allogeneic HCT with PTCy in patients with higher-risk MDS, employing a uniform strategy encompassing conditioning and immunosuppression and highlighting divergent outcomes based on donor types.
In summary, our study uncovered notable activity of PTCy in preventing acute GVHD, resulting in a low NRM incidence without an associated increase in disease relapse for patients with higher-risk MDS compared with ATG. However, patients undergoing HCT from HLA-haploidentical donors still exhibited inferior outcomes compared to a matched sibling or unrelated donors, highlighting the need for a more nuanced approach in these patients with augmenting GVHD prevention without increasing the risk of disease relapse. Further prospective studies comparing the two GVHD prophylactic strategies are warranted to delineate optimal patient population and transplantation procedures.