Problems in the design and reporting of trials of antifungal agents encountered during meta-analysis

Meta-analyses may become biased if the reported data in the individual tria ls are biased and if overlap among trials cannot be identified. We describe the unanticipated problems we encountered in collecting data for a metaana lysis comparing a new antifungal agent, fluconazole, with amphotericin B in patients with cancer complicated by neutropenia. In 3 large trials that co mprised 43% of the patients identified for the meta-analysis, results for a mphotericin B were combined with results for nystatin in a "polyene" group. Because nystatin is recognized as an ineffective drug in these circumstanc es, this approach creates a bias in favor of fluconazole. Furthermore, 79% of the patients were randomized to receive oral amphotericin B, which is po orly absorbed and not an established treatment, in contrast to intravenous amphotericin B, which was administered in 4 of 5 placebo-controlled trials, or 86% of patients. It was unclear whether there was overlap among the "po lyene" trials, and it is possible that results from single-center trials we re included in multicenter trial reports. We were unable to obtain informat ion to clarify these issues from the trial authors or the manufacturer of f luconazole. Two of 11 responding authors replied that the data were with th e drug manufacturer and two indicated that they did not have access to thei r data because of change of affiliation. In the meta-analyses, fluconazole and amphotericin B (mostly given orally) had similar effects (13 trials), w hereas nystatin was no better than placebo (3 trials). Since individual tri als are rarely conclusive, investigators, institutions, and pharmaceutical companies should provide essential details about their work to ensure that meta-analyses can accurately reflect: the studies conducted and that patien ts will realize maximum benefits from treatments. We recommend that: invest igators keep copies of their trial data to help facilitate accurate and unb iased meta-analyses.