RESUMO
In 2012 we reported promising results from a phase 2 clinical trial of HP802-247, a novel spray-applied investigational treatment for chronic venous leg ulcers consisting of human, allogeneic fibroblasts and keratinocytes. We now describe phase 3 clinical testing of HP802-247, its failure to detect efficacy, and subsequent investigation into the root causes of the failure. Two randomized, controlled trials enrolled a total of 673 adult outpatients at 96 centers in North America and Europe. The primary endpoint was the proportion of ulcers with confirmed closure at the end of 12 weeks of treatment. An investigation into the root cause for the failure of HP802-247 to show efficacy in these two phase 3 trials was initiated immediately following the initial review of the North American trial results. Four hundred twenty-one patients were enrolled in the North American (HP802-247, 211; Vehicle 210) and 252 in the European (HP802-247, 131; Vehicle 121) trials. No difference in proportion of closed ulcers at week 12 was observed between treatment groups for either the North American (HP802-247, 61.1%; Vehicle 60.0%; p = 0.5896) or the European (HP802-247, 47.0%; Vehicle 50.0%; p = 0.5348) trials. Thorough investigation found no likelihood that design or execution of the trials contributed to the failure. Variability over time during the trials in the clinical response implicated the quality of the cells comprising HP802-247. Concordance between the two separate, randomized, controlled trials with distinct, nonoverlapping investigative sites and independent monitoring teams renders the possibility of a Type II error vanishingly small and provides strong credibility for the unexpected lack of efficacy observed. The most likely causative factors for the efficacy failure in phase 3 was phenotypic change in the cells (primarily keratinocytes) leading to batch to batch variability due to the age of the cell banks.