Background: Severe steroid-refractory acute graft-versus-host disease (SR-aGvHD) following allogeneic stem cell transplantation (HSCT) is associated with a high mortality rate. Mesenchymal stromal cell (MSC) therapy has emerged as a promising treatment option for SR-aGvHD, yet its clinical efficacy has never been conclusively demonstrated. Aims: The randomized double-blind placebo-controlled HOVON-113 trial was designed as an international effort to evaluate whether addition of MSC to standardized second line treatment of SR-aGVHD improves clinical outcome. Patients developing aGvHD grade II to IV after alloSCT or donor lymphocyte infusions (DLI) with gut and/or liver involvement and stable or progressive disease or mixed response after 5 to 10 days of 2 mg/kg steroids or equivalent, were eligible for the study. Methods: Patients were randomized to receive 2 intravenous infusions at a 7-day interval of either MSC 2 x 106/kg body weight or placebo. In addition, all patients received mycophenolate mofetil (MMF) as standardized second line treatment. Patients who developed aGvHD while on prophylactic MMF were ineligible for inclusion. The trial’s primary endpoint was complete or partial response at 28 days following the 1st MSC or placebo infusion, reported as overall response (OR). Secondary endpoints included overall survival (OS), cumulative incidence of chronic GvHD, and adverse events. A sample size of 150 patients was calculated to detect an increase in the aGvHD response rate from 40% to 65% with a power of 0.80 and a 2-sided significance level of 0.05. Results: In the course of 5 years, 41 eligible patients were included in the Netherlands, Germany, Spain, Italy, and Belgium. 7 of these were < 18 years (age range 1-16). 20 patients were randomized to the MSC arm and 21 to the placebo arm. At diagnosis, 12% of patients had grade II aGvHD, 68% grade III, and 20% grade IV. In 63% of cases, aGvHD had been elicited by alloSCT and in 37% of cases by DLI. These variables were distributed evenly between the treatment arms. Of the 41 patients, 36 (85%) received both infusions. 3 did not receive any infusions and 2 received only 1 infusion due to refusal (1) or swift clinical deterioration and death (4). No infusion toxicity was observed. OR at day 28 was reached by 8 (38%) of patients in the placebo arm and by 12 (60%) in the MSC arm (odds ratio = 2.48, 95% confidence interval (CI) 0.65-9.51, p = 0.19). 7 patients died before day 28 and could not be evaluated, 6 (29%) in the placebo arm and 1 (5%) in the MSC arm, and were considered as non-responders in the primary analysis. OS at 1 year after randomization was 33% (95% CI 15-53) in the placebo arm and 45% (95% CI 23-65) in the MSC arm (HR = 0.81, 95% CI 0.37-1.77, p = 0.60). Infectious complications were comparable between treatments arms; placebo arm: 25% grade 3, 45% grade 4; MSC arm: 22% grade 3, 39% grade 4. The main causes of death were GvHD and infections. The median follow-up of the 15 patients still alive was 24.0 months (range 9.4 – 73.8 months). Summary/Conclusion: Due to regulatory issues and insufficient accrual, the trial could only include 41 patients and did not have sufficient power to detect the anticipated difference between the two treatment arms. No statistically significant effect of MSC infusions on aGvHD response on day 28, overall survival, or safety could be demonstrated. An extensive biomarker analysis, immunological monitoring, quality of life analysis, health technology assessment, and ethical examination are still ongoing.
Read full abstract