Abstract

HIV chemoprevention trials (oral PrEP and topical microbicides) will benefit greatly from an accurate assessment of medication adherence during the trial to assist in the interpretation of prevention success or failure in the trial. HIV infection during the trial may be a result of failure to achieve preventive drug concentrations, which itself may result from either (1) a prescribed regimen, to which the subject fully adhered, that is still insufficient to prevent infection or (2) a result of poor adherence to the prescribed regimen also resulting in insufficient drug concentrations. In addition, an accurate adherence assessment in the midst of a trial could trigger adherence interventions. Objective evidence of adherence in several HIV chemoprevention trials suggests that subjective measures greatly overestimate adherence. Drug concentration has been proposed as a more objective adherence measure. Blood, hair, or other samples are sampled at specified times and the resultant “observed” drug concentration is compared to the “expected” drug concentration. The proportional difference between the expected and observed drug concentrations may be used as an estimate of the proportion of doses for which the subject was adherent to the prescribed regimen. This method attempts to provide more accurate and quantitative measures than those currently employed, but obstacles to their application and feasibility in estimating adherence remain to be demonstrated. Problems of white-coat effect, inter- and intra-individual variability, dose-proportionality, and backward looking temporal frame of reference all need to be addressed as part of the validation and interpretation of drug concentration as an adherence measure.