Background
Only 68% of U.S. men-who-have-sex-with-men (MSM) living with HIV have achieved virologic suppression (VS), sufficiently low to compromise Ending the HIV Epidemic goals. Objective monitoring of antiretroviral therapy (ART) adherence via drug-level testing can inform interventions. A rapid, low-cost urine point-of-care (POC) tenofovir (TFV) assay has been developed by our group as an objective adherence metric. However, self-administration of the urine TFV assay among people with HIV (PWH) outside of the clinical setting has not been previously studied.
Methods
We assessed the association between a self-administered POC urine TFV rapid assay and VS in MSM with HIV on TFV-based ART, all of whom reported adherence challenges or virologic non-suppression. Participants, oversampled for stimulant users, were recruited using the Grindr app, received mailed urine tests, and provided plasma at a certified laboratory for viral load testing. Logistic regression examined the effect of TFV detection in urine on the likelihood of achieving VS. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the urine assay compared to VS were calculated.
Results
Of 727 MSM living with HIV, 88% completed and uploaded a photo of the urine test to the study. Among the analytic sample (n=596), the average age was 42.7 years (SD ± 11.02). 50% identified as White, 27% Black, 17% multiracial/other, 3% Asian, 2% Native American; 22% Hispanic. Nearly all (99%) were on ART with TFV alafenamide (TAF), and 61% reported stimulant use in the past 3 months. Of the 86% of participants that achieved VS, 10% of those had low adherence measured via the urine test. TFV detection was significantly associated with VS (OR=9.39, 95% CI5.62-15.84, p< .001). Among participants with no detectable TFV in urine, 13% of those without VS and 61% with VS reported adherence >80% in the past month.
Conclusions
These results are the first to show the feasibility of self-monitoring of adherence using a POC urine test. TFV presence in urine strongly predicted VS with a PPV of 92%, opening the potential for self-testing as an adherence check in-between viral load testing in the clinic. Non-suppression despite TFV detection suggests suboptimal adherence or drug resistance, and low TFV with VS may herald future treatment failure; both situations should prompt further monitoring. POC TFV testing offers a real-time strategy to identify adherence challenges and timely interventions on TFV-based ART.