Abstract
Background: Haploidentical allogeneic stem cell transplantation (haplo-SCT) has been a transformative option for patients lacking matched donors. Post-transplant cyclophosphamide (PTCy) at 100 mg/kg total dose (50 mg/kg on Day +3 and +4) is a standard for graft-versus-host diseases (GvHD) prophylaxis. Efforts to decrease toxicity while maintaining efficacy have informed pursuits of dose de-escalation. However, data regarding the efficacy and safety of reduced-dose PTCy (<100 mg/kg) remain scattered and variably reported.
Methods: We performed a systematic review registered on PROSPERO (CRD420250650291) and conducted per PRISMA guidelines. Six databases (Medline, Embase, Web of Science, Scopus, Cinahl, Cochrane) were searched initially up to May 2024. An updated search was conducted to include studies till June 2025. Eligible studies included prospective or retrospective cohorts or randomized trials evaluating low-dose PTCy in haplo-SCT recipients reporting on acute/chronic GvHD, overall survival (OS), relapse-free survival (RFS), immune reconstitution, or complications such as hemorrhagic cystitis or infections. Non-comparative studies, reviews, and abstracts were excluded. Data was extracted independently by two reviewers, and risk-of-bias (ROB) was assessed using ROBINS-I and Cochrane RoB-2 tools.
Results: Thirty-three studies, including 31 observational studies and 2 randomized controlled trials (RCTs), encompassing over 2500 patients, met our eligibility criteria. Most studies originated from Asia, with smaller contributions from Europe, the United States, India and Latin America. Low-dose PTCy regimens ranged from 14.5 to 80 mg/kg, commonly administered as 50 mg/kg ×1 on day +3 or 40 mg/kg ×2 on days +3 and +4. Comparators included standard-dose PTCy (100 mg/kg), ATG-based prophylaxis, or other regimens. Median follow-up ranged from 4 months to over 5 years.
Acute GvHD (grades II–IV) rates in low-dose PTCy arms ranged from 9% to 34%. Most observational data showed comparable or reduced incidence of aGvHD compared to controls. Chronic GvHD outcomes varied from 8-46% and showed no clear benefit of dose reduction in most studies.Two-year OS ranged widely from 47% to 100%, with some observational studies favoring low-dose PTCy but others showing no survival advantage. RFS was similarly variable. Hemorrhagic cystitis occurred in 3–33% of patients, with no clear dose-dependent pattern. Infections were inconsistently reported, but, when data were available, viral infections, including BK virus, CMV, and EBV reactivation, were common, with no consistent dose-dependent trend. Median time to immune reconstitution was slightly shorter in some cohorts, but clinical relevance remains uncertain. Two studies reported significantly reduced incidence of cardiac complications with lower PTCy exposure. One randomized controlled trial comparing 80 mg/kg PTCy + ATG with the Beijing Protocol reported significantly reduced aGvHD (11.5% vs. 39.3%, p=0.001) and better OS (75.4% vs. 54.1%, p = 0.021) and RFS ((72.7% vs. 55.0%, p = 0.044)) in the experimental arm. In contrast, the only other randomized study used ultra-low-dose PTCy (14.5 mg/kg ×2) vs. ATG and found no significant difference in acute GvHD (18.2% vs. 18.2%, p=0.996) or survival and was stopped early because of futility. Both RCTs showed comparable rates of cGvHD, toxicity, and immune reconstitution.Importantly, 87% of observational studies had a serious or critical ROB. Both RCTs had some concerns of bias due to inadequate allocation concealment and open-label design but were otherwise at low ROB.
Conclusions: Low-dose PTCy for GvHD prophylaxis in haplo-SCT carries comparable or lower rates of acute GvHD, similar chronic GvHD incidence, and non-inferior survival outcomes in many studies. Marked heterogeneity in regimens, patient populations, high ROB, and selective reporting preclude definitive conclusions. Randomized trials must focus on identifying subgroups benefiting from de-escalated PTCy. Until then, low-dose PTCy remains a promising but investigational alternative in haploidentical transplantation.
