Our group has pioneered the use of small hair samples to monitor adherence to antiretroviral (ARV) therapy in HIV infection. We have demonstrated that hair concentrations of ARVs, which monitor long-term exposure, are stronger predictors of treatment success than self-reported adherence or plasma ARV levels in large cohorts of HIV-infected patients. Data from recent HIV pre-exposure prophylaxis (PrEP) trials indicate that adherence to prophylactic ARVs is not always concordant with self report, so the incorporation of drug exposure measures as biomarkers of adherence has been crucial to trial interpretation. In iPrEx, mean self-reported adherence was 95%, but drug was detected in plasma from only 8% of seroconverters compared with 54% of matched active-arm controls who remained uninfected. Adherence via self-report was similarly high in two large PrEP trials in African women (FEM-PrEP and VOICE), but adequate drug concentrations in plasma were observed in fewer than one-third of participants. Since efficacy results can be substantially dampened or masked by low adherence to study product, the incorporation of pharmacologic measures to assess adherence in HIV prevention trials has become increasingly routine. The use of drug levels to monitor adherence in HIV treatment trials, despite the fact that self-reported adherence may be particularly inaccurate in all clinical trials, has lagged far behind the prevention trial setting. Single plasma levels of ARVs, like single glucose measurements, are limited in their ability to predict long-term treatment outcomes, because they reflect only a short duration of exposure, demonstrating significant day-to- day variation and """"""""white-coat"""""""" effects. In a manner analogous to glycosylated hemoglobin A1C (HbA1C) providing information on average glucose levels over prolonged periods, the concentration of medications in hair reflects drug uptake from the systemic circulation over weeks to months. This proposal seeks to extend our approach of using hair ARV levels to monitor adherence from the cohort and HIV prevention setting to the HIV treatment trial setting, partnering with the AIDS Clinical Trials Group (ACTG) to further test our methodology in a cost-effective manner. The ACTG A5257 study is a phase III randomized comparative study of three non-nucleoside reverse transcriptase inhibitor (NNRTI)-sparing combination regimens for treatment-naive HIV-1-infected patients. The trial was opened to enrollment in May 2009;hair sampling was initiated in August 2010 and continued to the completion of the study in June 2013. This proposal seeks funds to analyze the relevant ARV(s) in the collected hair samples from A5257 and to fund further data analyses using the hair measures. Incorporating hair measures into A5257, which randomized participants to fixed treatment regimens and performed frequent viral load monitoring, will provide additional power over the cohort setting for hair levels to predict impending virologic failure. This proposal's aim may therefore provide useful algorithms for adherence interventions, both early on and later in the treatment course, in the real-world setting.
Aim 1 is descriptive and will assess the relationship between hair concentrations of the target ARV in each of the 3 arms of A5257 and self-reported adherence as quantified using standardized questionnaires. This comparison between hair levels of ARVs and self-reported adherence measures will be performed for the first time in a clinical trial via this proposal.
Aim 2 will characterize relationships between antiretroviral concentrations in hair, as well as self-reported measures of adherence, and treatment efficacy and tolerability in each of the 3 arms of A5257.
We aim to evaluate how concentrations of ARVs in hair predict and explain treatment outcomes in A5257, including virologic failure, failure due to toxicity, rates of virologic suppression, CD4+ T-cell count trajectories, the development of drug resistance, adverse effects, and parameters associated with the metabolic syndrome in a longitudinal fashion to assess their predictive and explanatory potential. At study completion, we expect to have validated ARV concentrations in hair as a novel and practical biomarker of adherence and a useful predictor of treatment outcomes in a large HIV clinical trial. These findings will pave the way for an implementation project designed to assess the feasibility of incorporating hair concentration measures into clinical practice as a tool to enhance adherence interventions and outcomes.
HIV medications (antiretrovirals) taken by HIV-positive individuals can greatly improve the course of HIV infection and extend life. However, it may be difficult for HIV-infected individuals to take medications every day for the rest of their lives. Recent clinical trials in HIV prevention have shown that trial participants don't always tell investigators when they don't take antiretroviral medications as prescribed, so ongoing prevention trials have used more objective ways to measure adherence, such as measuring drug levels. We have proven that measuring antiretroviral levels in small cut pieces of scalp hair are excellent measures of drug adherence in large cohorts of HIV-infected people. However, hair measures to evaluate antiretroviral adherence have never been validated in an HIV clinical trial, where participants may not always reveal problems with adherence. This project will analyze hair measures in a large Adult Clinical Trials Group (ACTG)-funded HIV treatment trial called A5257 (looking at 3 different treatment regimens for HIV) to see how well hair levels predict treatment outcomes. We may be able to come up with algorithms from this proposal to predict who will do well on therapy in the real-world setting and who may have problems with toxicities or adhering to daily therapy. Hair samples are easy to collect, store and analyze;cutting small hair samples is painless and avoids drawing blood. Our innovative way of monitoring medication levels in hair has public health relevance not only for HIV, but for many chronic disease states.
