Hair levels of antiretrovirals independently predicted virologic failure in an AIDS Clinical Trials Group (ACTG) study that randomized treatment-naive adults to atazanavir/ritonavir, darunavir/ritonavir, or raltegravir, all with tenofovir/emtricitabine . Hair levels correlated weakly with self-reported adherence.
Critical to the success of antiretroviral therapy and preexposure prophylaxis (PrEP), adherence can be measured by self-report, pill counts, pharmacy refills, ingested sensor devices, or other subjective or objective methods. The ACTG team noted that measuring antiretroviral concentrations in hair has some advantages over other methods, including easy sampling, room-temperature storage, ability to measure adherence over the long term in one sample, and ability to measure levels of multiple antiretrovirals in one sample.
Researchers assessed the predictive power of hair antiretroviral concentrations in ACTG A5257, which randomized adults in an open-label design to the integrase inhibitor raltegravir or one of two protease inhibitors (PIs), boosted atazanavir or boosted darunavir . Potential utility of drug concentrations in hair had been assessed in cohort studies but not in a randomized trial.
Because the investigators added hair collection to the protocol after the trial began, about 40% of participants gave samples at weeks 4, 8, and 16, and then quarterly. Participants self-reported adherence on a visual analog scale. The researchers measured levels of atazanavir, darunavir, and raltegravir in hair. They used proportional hazards regression models to estimate possible associations between antiretroviral levels in hair and virologic failure.
Of the 599 participants who gave hair samples, 32% were women, 33% black, and 17% Hispanic. Median age stood at 38 years and median follow-up at 124 weeks. After 2 years virologic failure rates were 26% in the lowest antiretroviral hair level tertile, 6% in the middle tertile, and 3% in the highest tertile.
People with antiretroviral concentrations in the lowest tertile had almost a 7-fold higher risk of virologic failure than those in the highest tertile (hazard ratio [HR] 6.79, 95% confidence interval [CI] 2.65 to 23.0, P = 0.004). Sex-stratified results proved similar for men and women. Every 2-fold lower antiretroviral hair level independently boosted the risk of virologic failure (HR 2.43, 95% CI 1.96 to 3.13, P < 0.0001). Hair level proved the strongest predictor of failure in this study, and the association was consistent for all three antiretrovirals.
Black race raised the failure risk 4.5 times (HR 4.5, 95% CI 1.6 to 12.8, P = 0.006), while less than a high-school education tripled the risk (HR 2.9, 95% CI 1.3 to 6.3, P = 0.009). Pretreatment viral load above 100,000 copies doubled the risk of failure, but that association fell short of statistical significance (HR 2.1, 95% CI 0.97 to 4.6, P = 0.06). Female sex did not affect risk of failure.
Self-reported adherence correlated weakly with antiretroviral levels, confirming weak associations between self-report and objectively measured antiretroviral levels in previous studies (Pearson's r 0.18 for atazanavir, 0.15 for darunavir, and -0.04 for raltegravir).
The investigators see a need for further study on whether measuring antiretroviral levels in hair followed by targeted adherence measures can cut virologic failure rates.
1. Gandhi M, Ofotokun I, Bacchetti P, et al. Hair antiretroviral levels strongly predict virologic outcomes in ACTG's A5257 trial. 25th Conference on Retroviruses and Opportunistic Infections (CROI). March 4-7, 2018. Boston. Abstract 24.
2. Lennox JL, Landovitz RJ, Ribaudo HJ, et al. Efficacy and tolerability of 3 nonnucleoside reverse transcriptase inhibitor-sparing antiretroviral regimens for treatment-naive volunteers infected with HIV-1: a randomized, controlled equivalence trial. Ann Intern Med. 2014;161:461-471. https://www.ncbi.nlm.nih.gov/pubmed/25285539