Addition of anti-thymocyte globulin to standard graft-versus-host disease prophylaxis versus standard treatment alone in patients with haematological malignancies undergoing transplantation from unrelated donors: final analysis of a randomised, open-label, multicentre, phase 3 trial

Background: Previous trials testing prevention strategies for chronic graft versus host disease (GVHD) have measured its cumulative incidence. In this trial of anti-thymocyte globulin, we measured treatment-independence at a long-term timepoint as the primary endpoint. Methods: This was a randomised, open-label, multicentre, phase 3 trial done at ten centres in Canada and one in Australia. Eligible patients had a haematological malignancy (leukaemia, myelodysplastic syndrome, or lymphoma), were between 16 and 70 years of age, eligible for transplantation with a Karnofsky score of at least 60, and received an unrelated donor (fully matched or one-locus mismatched at HLA-A, HLA-B, HLA-C, or DRB1 loci) graft following myeloablative or non-myeloablative–reduced intensity conditioning. Patients were randomly assigned to receive anti-thymocyte globulin 4·5 mg/kg plus standard GVHD prophylaxis (cyclosporine or tacrolimus plus methotrexate or mycophenolate) or standard GVHD prophylaxis alone. The primary endpoint, freedom from immunosuppressive therapy without resumption at 12 months, was previously reported. Here we report on the prespecified 24-month analysis. Analyses were per-protocol, excluding those patients who did not proceed to transplantation. This trial is registered as ISRCTN 29899028 and NCT01217723, status completed. Findings: Between June 9, 2010, and July 8, 2013, we recruited and randomly assigned 203 eligible patients to receive anti-thymocyte globulin (n=101) or no additional treatment (n=102) along with standard GVHD prophylaxis. 7 (3%) patients did not receive a transplant and were excluded from the analysis. 38 (38%) of 99 evaluable patients in the anti-thymocyte globulin plus GVHD prophylaxis group were free from immunosuppressive therapy at 24 months compared with 18 (19%) of 97 patients in the standard GVHD prophylaxis group (adjusted odds ratio [OR] 3·49 [95% CI 1·60–7·60]; p=0·0016. At 24 months, the cumulative incidence of relapse was 16·3% (95% CI 8·9–23·7) in the anti-thymocyte globulin plus GVHD prophylaxis group compared with 17·5 (9·9–25·1) in the standard GVHD prophylaxis group (p=0·73) and non-relapse mortality was 21·2% (95% CI 13·2–29·2) versus 31·3% (21·9–40·7; p=0·15). The cumulative incidence of chronic GVHD at 24 months was 26·3% (95% CI 17·5–35·1) in the anti-thymocyte globulin group and 41·3% (31·3–51·3) in the standard GVHD prophylaxis group (p=0·032). Overall survival at 24 months was 53·3% (95% CI 42·8–62·7) in the anti-thymocyte globulin plus GVHD prophylaxis group compared with 70.6% (95% CI 60·6–78·6) in the standard GVHD prophylaxis group (adjusted hazard ratio [HR] 0·56, 95% CI [0·35–0·90]; p=0·017. Symptoms of chronic GVHD by the Lee Scale were more prevalent in the standard GVHD prophylaxis group, with scores of 13·27 (SD 10·94) in the anti-thymocyte globulin plus GVHD prophylaxis group and 20·38 (SD 14·68) in the standard GVHD prophylaxis group (p=0·040). Depressive symptoms were more prominent in the standard GVHD prophylaxis group, the mean Center for Epidemiological Studies Depression scale (CES-D) scores were 10·40 (SD 9·88) in the anti-thymocyte globulin group and 14·62 (SD 12·26) in the standard GVHD prophylaxis group (p=0·034). Serious adverse events (CTCAE grade 4 or 5) occurred in 38 (38%) patients in the anti-thymocyte globulin group and in 49 (51%) in the standard GVHD prophylaxis group, the most common being infection and GVHD. One patient died of Epstein-Barr virus hepatitis. Interpretation: The results of this prespecified 24-month analysis suggest that pretreatment with anti-thymocyte globulin provides clinically meaningful benefits when added to standard GVHD prophylaxis in patients undergoing unrelated donor transplantation, including a decrease in use of immunosuppressive therapy, chronic GVHD and its symptoms, depressive symptoms, and improved overall survival. Anti-thymocyte globulin could be included in the preparative regimens of patients with haematological malignancies selected for unrelated donor transplantation. Funding: Canadian Institutes of Health Research and Sanofi.