Hyperforin plasma level as a marker of treatment adherence in the national institutes of health hypericum depression trial

Background: A previously reported clinical trial of Hypericum perforatum (St John's wort) in depression did not demonstrate efficacy. We assessed treatment adherence by measuring plasma hyperforin and evaluated the possible impact of adherence on study results. Methods: Outpatients with major depression (N = 340) were randomized to an 8-week trial of H. perforatum (900-1500 mg/d), sertraline (50-100 mg/d) as active comparator, or placebo. Plasma was available from 292 patients (86% of randomized). Samples from the placebo and H. perforatum groups were assayed for hyperforin, and samples from the sertraline group for sertraline/N-desmethyl-sertraline. Results: Of the 104 patients randomized to placebo, 18 (17%) had detectable plasma hyperforin. Of the 97 patients randomized to H. perforatum, 17 (17%) had no detectable plasma hyperforin. All the assayed sertraline patients (N = 91) had plasma sertraline/N-desmethyl-sertraline. The clinical trial conclusions remained unchanged when only patients with plasma assay consistent with random assignment were included in the analyses. Conclusions: One of every 6 patients assigned to placebo had plasma hyperforin, and 1 of every 6 patients assigned to H. perforatum had no detectable plasma hyperforin. The finding underscores the difficulty of enforcing treatment adherence in clinical trials of preparations that are readily available in the community.