Comparison of different dose of anti-T lymphocyte globulin (ATLG) in haploidentical HSCT for gvhd prophylaxis

Abstract Background Allogeneic hematopoietic stem cell transplantation (allo-HSCT) remains a curative option for various hematologic malignancies. Anti-T lymphocyte globulin (ATLG) is widely used for graft-versus-host disease (GVHD) prophylaxis, particularly in haploidentical transplantation. While a dose of 30 mg/kg is commonly recommended in Europe, a lower dose of 20 mg/kg is frequently adopted in China. However, direct comparative data between the two doses in the context of haplo-HSCT remain limited. Methods This multicenter, prospective, randomized, parallel-group clinical trial enrolled patients aged 16–65 years with AML, ALL, or MDS undergoing haploidentical HSCT. Participants were randomly assigned to receive either 20 mg/kg or 30 mg/kg of ATLG, in combination with uniform myeloablative conditioning and standard GVHD prophylaxis. The primary endpoint was the cumulative incidence of grade II–IV acute GVHD (aGVHD) within 100 days post-transplant. Secondary endpoints included chronic GVHD (cGVHD), viral infections, adverse events (AEs), and immune reconstitution. This trial is registered at ClinicalTrials.gov (NCT06525519). Results As of July 2025, a total of 54 patients had been enrolled in the study, including 29 patients in the 20 mg/kg ATLG group and 25 in the 30 mg/kg group. Baseline characteristics, including age, sex, disease type, and ECOG performance status, were well balanced between the two groups, with no statistically significant differences observed. By day 100 post-transplant, the cumulative incidence of acute GVHD (aGVHD) was 13% overall, with 14% in the 20 mg/kg group and 12% in the 30 mg/kg group (P > 0.9). For grade II–IV aGVHD, the overall incidence was 9.3%, occurring in 10% of patients in the 20 mg/kg group and 8% in the 30 mg/kg group (P > 0.9). Grade III–IV aGVHD was observed in 10% of the 20 mg/kg group but was not reported in the 30 mg/kg group, with the difference not reaching statistical significance (P = 0.2). No cases of moderate-to-severe chronic GVHD were documented in either group. Regarding infections, EBV viremia within 100 days occurred in 6.9% of patients receiving 20 mg/kg and 12% of those receiving 30 mg/kg of ATLG (P = 0.7). CMV viremia was observed only in the 30 mg/kg group (4%), while no cases were reported in the 20 mg/kg group (P = 0.5). High-grade fever, defined as a body temperature ≥39°C, occurred at a similarly low frequency in both groups (3.4% vs. 4.0%). Immune reconstitution was assessed on day 21 post-transplant. Patients in the 20 mg/kg group demonstrated significantly better immune recovery, with higher proportions of CD3⁺, CD4⁺, CD8⁺, activated T cells, and regulatory T cells compared to those in the 30 mg/kg group. In terms of safety, adverse events (AEs) occurred in 83.3% of patients overall, with a slightly lower incidence in the 20 mg/kg group (79.3%) compared to the 30 mg/kg group (88.0%). Severe AEs (grade ≥3) were reported in 6.9% of patients in the 20 mg/kg group, while no such events occurred in the 30 mg/kg group. The most common AEs included fever, diarrhea, hypoalbuminemia, and hypomagnesemia. Conclusion This interim analysis suggests that 20 mg/kg ATLG is non-inferior to 30 mg/kg for GVHD prophylaxis in haploidentical HSCT, with a trend toward lower infection and AE rates, and better early immune reconstitution. Final conclusions await the completion of full patient enrollment and follow-up.