Phase 3 trial of defibrotide for the treatment of severe veno-occlusive disease and multi-organ failure
The primary report of Study 2005-01: 38 percent of transplant patients with severe veno-occlusive disease and organ failure were alive at day 100 on defibrotide, against 25 percent of carefully matched historical controls.
Overview
Phase 3 study of defibrotide in patients with established hepatic veno-occlusive disease (sinusoidal obstruction syndrome) and advanced multi-organ failure after haematopoietic stem cell transplantation, a condition with more than 80 percent mortality untreated. Patients (n = 102) given defibrotide 25 mg per kilogram per day were compared with 32 historical controls identified from 6867 medical charts of transplant patients by blinded independent reviewers; baseline characteristics were well balanced.
The primary endpoint was survival at day +100 post-transplant: observed rates were 38.2 percent in the defibrotide group and 25 percent in the controls (23 percent estimated difference; 95.1% CI 5.2 to 40.8; P = .0109, propensity-adjusted analysis). Observed day +100 complete response rates were 25.5 percent for defibrotide and 12.5 percent for controls (19 percent difference; 95.1% CI 3.5 to 34.6; P = .0160). Related adverse events included haemorrhage or hypotension; common haemorrhagic events (pulmonary alveolar 11.8 and 15.6 percent, gastrointestinal 7.8 and 9.4 percent) were similar between groups.
- Day +100 survival 38.2% with defibrotide vs 25% in historical controls; estimated difference 23% (95.1% CI 5.2 to 40.8), P = .0109.
- Day +100 complete response 25.5% vs 12.5%; difference 19% (95.1% CI 3.5 to 34.6), P = .0160.
- Haemorrhagic adverse events similar between groups: pulmonary alveolar bleeding 11.8% vs 15.6%, gastrointestinal bleeding 7.8% vs 9.4%.
This is the trial behind defibrotide's March 2016 US approval, the only approved treatment for hepatic veno-occlusive disease with organ failure after transplant. The historical-control design was accepted because a randomised trial in a condition this lethal and rare was judged unfeasible; the gain is meaningful but most patients still died.
- Historically controlled, not randomised; the 32 controls were matched from chart review.
- 95.1% confidence intervals adjust for an interim analysis.
- Absolute survival remained low (38% at day 100).