Optimal Dose of Antithymocyte Globulin in Conditioning Regimens for Unmanipulated Haploidentical Haematopoietic Stem Cell Transplantation: Long-Term Outcomes of a Prospective Randomised Trial

Background: Antithymocyte globulin (ATG) is an important component of conditioning regimens to prevent severe GVHD in patients undergoing unmanipulated haploidentical stem cell transplantation (haplo-SCT). However, the optimal dose of ATG is unknown.

Methods: In this open-label extension of a prospective, randomised trial, we compared the long-term outcomes of two ATG doses used in myeloablative conditioning before unmanipulated haplo-HSCT. Patients were randomly assigned (1:1) to 10 mg/kg (ATG-10) or 6 mg/kg (ATG-6) ATG. Patients and individuals assessing outcomes were masked to treatment allocation. Analysis of disease-free survival (DFS), GVHD-free/relapse-free survival (GRFS), relapse, non-relapse mortality (NRM), and chronic GVHD (cGVHD) included the entire population. Late effects were assessed in disease-free patients who had survived for at least 6 months and had received regular follow-up evaluations.

Results: Between December 2010 and May 2012, 224 patients were recruited. The median follow-up period was 1614 days (28-1929 days). The 5-year cumulative incidence was comparable between the ATG-6 and ATG-10 groups for relapse (12·8% vs. 13·4%, p=0·832) and NRM (11·6% vs. 17·0%, p=0·263). The 5-year probability of DFS was comparable between groups (75·6% vs. 69·6%, p=0·283). The 5-year cumulative incidence was higher with ATG-6 for cGVHD (75·0% vs. 56·3%, p=0·007; moderate-to-severe cGVHD: 56·3% vs. 30·4%, p<0·0001) and late effects (71·2% vs. 56·9%, p=0·043). The 5-year probability of GRFS was higher with ATG-10 (41·0% vs. 26·8%, p=0·008). In the multivariate analysis, ATG-10 was associated with lower cGVHD risk and improved GRFS.

Conclusions: ATG 10 mg/kg may be the optimal dose for conditioning regimens before unmanipulated haplo-SCT.