Forum report: Issues in the evaluation of diagnostic tests, use of historical controls, and merits of the current multicenter collaborative groups

This forum report contains conclusions about 3 different issues relevant to conducting clinical trials in deep mycoses. (1) Trials of diagnostic tests for deep mycoses must define the population appropriate for testing and the clinical question being asked. The unanswered question for the serum Aspergillus galactomannan assay is whether knowledge of results can change use of empirical therapy to treat febrile patients at high risk of invasive aspergillosis. (2) Use of historical controls is suboptimal but offers a pragmatic solution for studying rare mycoses; use of contemporaneous controls, matched for critical variables and evaluated by a blinded data review committee using detailed criteria, appears optimal. (3) Established groups of independent investigators, such as the European Organization for Research on Treatment of Cancer's Invasive Fungal Infections Group and National Institute of Allergy and Infectious Diseases's Bacteriology and Mycology Study Group, provide a pool of experienced investigators, defined operating rules, impartiality, and specialized expertise. Considering the enormous investment required for adequately powered efficacy trials of antifungal agents and the importance of these trials to guide clinical practice, use of collaborative groups outweighs the extra administrative time that is sometimes required.