Randomized Trial of Two Dosages of Rabbit Antithymocyte Globulin in Patients with Aplastic Anemia

Introduction: In patients with severe aplastic anemia (SAA) who are not candidates for stem cell transplantation (SCT), immunosuppressive therapy (IST) with antithymocyte globulin (ATG) and cyclosporine A (CyA) is the treatment of choice. A randomized controlled trial demonstrated the inferiority of rabbit ATG compared with horse ATG for first-line treatment of SAA (Scheinberg et al. N Engl J Med 365:430-438, 2011). However, horse ATG remains unavailable in many countries outside the USA. In Asian countries, the recommended dose of rabbit ATG ranges from 2.5 to 3.5 mg/kg/day. Here we report the first prospective randomized multicenter study comparing two dosages of rabbit ATG in patients with SAA (www.umin.ac.jp UMIN000011134).

Methods: Patients with SAA who required initial IST in Japan (n = 89), China (n = 85), and Korea (n= 48) were enrolled between May 2012 and October 2017. Randomization was 1:1 in blocks between the two dosages of rabbit ATGs. The IST regimen comprised rabbit ATG (thymoglobulin^®, 2.5 or 3.5 mg/kg/day for 5 days), CyA (6 mg/kg/day for minimum 6 months), and methylprednisolone (2 mg/kg/day for 5 days) with subsequent tapering of the dose for 28 days. The dose of CyA was adjusted to maintain whole blood trough concentrations between 150 and 250 ng/mL. The primary endpoint was hematologic response at day 180. Secondary endpoints included frequency of Epstein-Barr virus-associated lymphoproliferative disorder (EBV-LPD), hematologic response at day 360, relapse, and overall survival (OS).

Results: In total, 222 patients (age 0.5-71.0 years) were randomized, with 112 patients receiving 2.5 mg/kg and 110 receiving 3.5 mg/kg of rabbit ATG. Patient characteristics were well matched between the two groups. Median follow-up duration for all patients was 742 days from first ATG infusion (range, 16-2165) and 887 days for those surviving till the end of the study (range, 32-1757).

After 3 months, in the 2.5mg/kg of rabbit ATG group, 4 (4%) patients achieved a complete response (CR) and 35 (31%) a partial response (PR), for an overall response rate (ORR) of 35%. In the 3.5mg/kg of rabbit ATG group, 1 (1%) patient achieved a CR and 43 (39%) a PR, for an ORR of 40%. After 6 months, in the 2.5mg/kg of rabbit ATG group, 55 patients (49%) achieved an ORR including 7 (6%) who achieved a CR and 48 (43%) PR. In the 3.5mg/kg of rabbit ATG group, 53 patients (48%) achieved a response including CR in 9 patients (8%) and PR in 44 (40%). The response rates did not differ significantly between the two groups at 3 (P = 0.488) or 6 (P = 0.894) months. Multivariate logistic regression analysis identified disease severity [odds ratio (OR), 2.09; 95% confidence interval (CI), 1.18−3.67; P = 0·011] and interval between diagnosis and treatment (OR, 1.86; 95% CI, 1.07−3.22; P = 0.027) as independent predictors of response to IST at 6 months.

Despite relatively short follow-up, relapse and clonal evolution occurred with similar frequencies between the two groups, with 4 relapse (among 55 responders) and 1 clonal evolution in the 2.5 mg/kg group and 4 relapse (among 53 responders) and no evolutions in the 3.5 mg/kg group. One patient in the 3.5 mg/kg group developed EBV-LPD at day 109 after IST. During follow-up, 22 (10%) patients died, all of whom showed no response to IST. The causes of death were infections (n = 13), hemorrhage (n = 4), SCT-related complications (n = 2), clonal evolution (n = 1), and unknown (n = 2).

Transplantation free survival (TFS), failure free survival (FFS), and OS at three-year were similar between the two groups [TFS, 73% (95% CI, 63%−81%) vs. 71% (95% CI, 60%−80%); P = 0.991; FFS, 45% (95% CI, 35%−54%) vs. 45% (95% CI, 61%−84%); P = 0.909; OS, 85% (95% CI, 76%−91%) vs. 91% (95% CI, 82%−96%); P = 0.107]. Multivariate logistic regression analysis revealed that lack of response to IST was an independent predictor of OS (OR, 6.95; 95% CI, 1.76%−27.46%; P = 0.006).

Conclusions: The current study revealed no differences in the efficacy and safety between 2.5 mg/kg and 3.5 mg/kg dose of rabbit ATG for patients with SAA. However, as the follow-up period was relatively short, longer surveillance is warranted to compare long-term complications and survival rates between the two groups in the follow up study.