Highlight
– Replacing mycophenolate mofetil (MMF) with post‑transplant cyclophosphamide (PTCy) in combination with sirolimus (SIR) and cyclosporine (CSP) after nonmyeloablative or reduced‑intensity unrelated donor peripheral blood stem cell (PBSC) transplantation substantially improved 1‑year chronic GVHD‑free relapse‑free survival (CRFS: 73% vs 48%; HR 0.46, P = .005).
– PTCy markedly reduced moderate‑to‑severe chronic GVHD (3% vs 33%) without increasing relapse (15% vs 15%) or overall survival at 1 year (86% vs 86%).
– However, patients receiving PTCy experienced a higher rate of grade ≥3 infections (HR 2.65; P = .003), highlighting a trade‑off that requires vigilance in infection prophylaxis and monitoring.
Background
Allogeneic hematopoietic cell transplantation (HCT) remains a potentially curative therapy for many hematologic malignancies, but graft‑versus‑host disease (GVHD) — particularly chronic GVHD (cGVHD) — is a major cause of long‑term morbidity and impaired quality of life. Reducing cGVHD while preserving graft‑versus‑tumor effects and avoiding excess treatment‑related mortality is a central challenge in transplant practice. Post‑transplant cyclophosphamide (PTCy) has emerged over the last decade as an effective strategy to reduce GVHD in haploidentical and matched donor settings, principally through selective elimination of highly proliferative alloreactive T cells and promotion of regulatory T‑cell recovery. Sirolimus, an mTOR inhibitor, and calcineurin inhibitors (like cyclosporine) are commonly used as GVHD prophylaxis because sirolimus favors regulatory T‑cell expansion and dampens effector T‑cell proliferation.
The randomized Phase II trial by Ueda Oshima et al. (J Clin Oncol. 2025) examined whether combining PTCy with SIR and CSP after nonmyeloablative or reduced‑intensity unrelated donor PBSC transplantation could reduce cGVHD without worsening relapse or survival outcomes compared with a regimen containing MMF instead of PTCy.
Study design
Design, patients, and setting
This Phase II randomized trial (ClinicalTrials.gov identifier: NCT03246906) enrolled 145 adults with hematologic malignancies who were ineligible for myeloablative conditioning and were receiving PBSC grafts from HLA‑matched or mismatched unrelated donors. Patients were randomized 1:1 to two GVHD prophylaxis strategies after nonmyeloablative or reduced‑intensity conditioning:
– Intervention arm: sirolimus + cyclosporine + PTCy (50 mg/kg once daily on days +3 and +4).
– Control arm: sirolimus + cyclosporine + mycophenolate mofetil (MMF).
The primary endpoint was 1‑year chronic GVHD‑free relapse‑free survival (CRFS), a composite capturing absence of moderate‑to‑severe cGVHD, relapse, and death. Secondary endpoints included rates of acute GVHD (aGVHD), moderate‑to‑severe cGVHD, relapse, progression‑free survival (PFS), overall survival (OS), nonrelapse mortality (NRM), and severe infections. Median follow‑up among survivors was 3.0 years (range 0.6–7.0 years).
Key results
Primary endpoint: CRFS
At 1 year, estimated CRFS favored the PTCy arm: 73% (95% CI, 61%–82%) versus 48% (95% CI, 36%–59%) in the MMF arm. This corresponded to a hazard ratio (HR) for CRFS failure of 0.46 (95% CI, 0.26–0.79; P = .005) in the PTCy group — a clinically meaningful and statistically significant reduction in composite failure events at 1 year.
GVHD outcomes
– Acute GVHD: Rates of grade II–IV aGVHD were similar between groups (40% PTCy vs 42% MMF). Grade III–IV aGVHD was numerically lower in the PTCy group (6% vs 10%).
– Chronic GVHD: The effect on cGVHD was striking. The 1‑year incidence of moderate‑to‑severe cGVHD was 3% (95% CI, 1%–9%) in the PTCy arm versus 33% (95% CI, 22%–44%) in the MMF arm.
These data indicate that the CRFS benefit was driven primarily by prevention of clinically significant cGVHD rather than differences in early acute GVHD.
Relapse, survival, and nonrelapse mortality
– Relapse: Identical 1‑year relapse rates of 15% in both arms, suggesting that replacing MMF with PTCy did not reduce graft‑versus‑tumor activity.
– Progression‑free survival: 75% (PTCy) versus 78% (MMF) at 1 year.
– Overall survival: Identical 1‑year OS estimates of 86% in both arms.
– Nonrelapse mortality: 10% (PTCy) versus 7% (MMF) at 1 year — not a statistically significant difference in reported data.
Taken together, these secondary endpoints suggest that the substantial reduction in cGVHD with PTCy did not come at the cost of increased relapse or early mortality.
Infections and safety
Serious safety considerations emerged. The hazard ratio for grade ≥3 infections in the PTCy group compared with the MMF group was 2.65 (95% CI, 1.41–4.97; P = .003). This indicates a significantly higher risk of severe infections with the PTCy SIR/CSP combination. Clinically, this requires careful attention to peri‑transplant and post‑transplant antimicrobial prophylaxis, immune monitoring, and possibly tailored vaccination and preemptive strategies.
Expert commentary and interpretation
The trial provides randomized evidence that PTCy, when used in combination with SIR and CSP after unrelated donor PBSC transplant with reduced‑intensity or nonmyeloablative conditioning, substantially reduces clinically important cGVHD without increasing relapse. The large absolute reduction in moderate‑to‑severe cGVHD (33% down to 3%) is notable and, if replicated in larger phase III settings, could shift prophylaxis paradigms for unrelated donor transplants.
Mechanistic plausibility supports these findings. PTCy, given early after infusion, preferentially depletes proliferating alloreactive T cells while sparing quiescent and regulatory T cells, promoting tolerance. Sirolimus adds a complementary mechanism by inhibiting mTOR‑dependent effector T‑cell proliferation and supporting regulatory T‑cell expansion. Cyclosporine reduces early T‑cell activation through calcineurin inhibition. The combined immunologic effect plausibly explains the powerful protection against chronic, immune‑mediated tissue injury that defines cGVHD.
However, the increased incidence of severe infections is an important counterbalance. Potential contributors include intensified early immune suppression from PTCy plus dual maintenance immunosuppression (sirolimus + cyclosporine), delayed lymphocyte reconstitution, and toxicity from cyclophosphamide. The net clinical benefit will depend on whether infection risk can be mitigated (through enhanced prophylaxis, preemptive therapy, and monitoring) while retaining the cGVHD reduction. Importantly, relapse rates were unchanged, alleviating a major concern that deeper immune suppression might impair anti‑tumor effects.
Limitations and generalizability
– Phase II design and sample size: While randomized, this was a phase II trial, and the sample size may limit power for less frequent events and long‑term outcomes.
– Population: The trial enrolled adults deemed ineligible for myeloablative conditioning; results may not generalize to younger, fitter patients receiving myeloablative regimens, nor to bone marrow grafts rather than PBSCs.
– Donor types: Both HLA‑matched and mismatched unrelated donors were included; subgroup analyses (not detailed here) are needed to understand whether the benefit is consistent across donor match levels.
– Infection signal: The marked increase in grade ≥3 infections requires detailed breakdown (pathogens, timing, attributable mortality), which will be crucial to inform implementation.
– Follow‑up duration: Median follow‑up among survivors was 3.0 years; longer follow‑up will clarify durability of cGVHD protection and late complications.
Clinical implications and practical considerations
– For centers performing reduced‑intensity or nonmyeloablative unrelated donor PBSC transplants, the SIR/CSP/PTCy combination merits consideration as a strategy to lower cGVHD incidence and improve CRFS, provided local capacity for infection prevention and monitoring.
– Before broad adoption, institutions should plan standardized infection prophylaxis and monitoring pathways, and consider enrolling patients in confirmatory phase III trials or registries to accumulate safety and long‑term outcome data.
– The unchanged relapse risk is reassuring, suggesting preservation of graft‑versus‑tumor effects, but disease‑specific analyses (e.g., myeloid vs lymphoid malignancies, MRD status) will be informative to tailor use of PTCy.
Research priorities and next steps
– Conduct a definitive phase III randomized trial comparing SIR/CSP/PTCy versus current standard prophylaxis in unrelated donor PBSC transplantation with sufficient power for survival and infection outcomes.
– Detailed immunologic studies to characterize T‑cell subset reconstitution, regulatory T‑cell dynamics, and pathogen‑specific immunity after the combined regimen to identify predictors of infection and GVHD protection.
– Optimize infection prophylaxis and reconstitution strategies (antiviral, antifungal, antibacterial prophylaxis; immunoglobulin replacement; preemptive therapy thresholds) for patients receiving PTCy plus dual maintenance immunosuppression.
– Evaluate the regimen’s efficacy across donor types (matched vs mismatched), graft sources (bone marrow vs PBSC), and conditioning intensities.
Conclusion
This randomized Phase II trial demonstrates that substituting MMF with PTCy in a sirolimus/cyclosporine backbone for GVHD prophylaxis after unrelated donor PBSC transplantation dramatically reduces clinically important chronic GVHD and substantially improves 1‑year CRFS without increasing relapse or early mortality. The benefit is biologically plausible and clinically meaningful. Nevertheless, the elevated risk of severe infections identifies an important safety concern that must be addressed before widespread adoption. Confirmation in larger trials, accompanied by strategies to mitigate infection risk and mechanistic studies of immune reconstitution, should be prioritized.
Funding and clinicaltrials.gov
ClinicalTrials.gov identifier: NCT03246906.
Funding: See original publication (Ueda Oshima et al., J Clin Oncol. 2025) for detailed funding and disclosures.
References
Ueda Oshima M, Vo PT, Boeckh M, Bouvier ME, Carpenter PA, Mielcarek M, Petersdorf EW, Storb R, Gooley T, Sandmaier BM. Sirolimus and Cyclosporine With Post‑Transplant Cyclophosphamide or Mycophenolate Mofetil as Graft‑Versus‑Host Disease Prophylaxis in Unrelated Donor Hematopoietic Cell Transplantation. J Clin Oncol. 2025 Nov 20;43(33):3600-3609. doi: 10.1200/JCO-25-01238. Epub 2025 Oct 3. PMID: 41043099; PMCID: PMC12614434.
ClinicalTrials.gov. Study of Sirolimus and Cyclosporine With Post‑Transplant Cyclophosphamide or Mycophenolate Mofetil in Unrelated Donor Hematopoietic Cell Transplantation. NCT03246906. Available at: https://clinicaltrials.gov/study/NCT03246906
