Urine tenofovir-monitoring predicts HIV viremia in patients treated with high genetic-barrier regimens

Abstract

Objective: Access to viral load measurements is constrained in resource-limited settings. A lateral flow urine tenofovir (TFV) rapid assay (UTRA) for patients whose regimens include TFV offers an affordable approach to frequent adherence monitoring.

Design: We conducted a cross-sectional study of patients to assess the utility of UTRA to predict virologic failure, defined as a viral load greater than 400 copies/ml.

Methods: We assessed urine TFV among 113 participants at increased risk of viral failure (who had previous viral failure on this regimen or had previously been ≥30 days out of care), comparing low genetic-barrier efavirenz (EFV) regimens (n = 60) to dolutegravir (DTG)-boosted or ritonavir-boosted protease inhibitor (PI/r)-based high genetic-barrier regimens (n = 53). Dried blood spots (DBS) for TFV-diphosphate and plasma for TFV concentrations were collected, with drug resistance assessed if viral failure present.

Results: Among 113 participants, 17 of 53 received DTG or PI/r had viral failure at the cross-sectional visit, with 11 (64.7%) demonstrating an undetectable urine TFV; the negative-predictive value (NPV) of undetectable UTRA for viral failure was 85% (34/40); none of the 16 sequenced had dual class drug resistance. In those treated with EFV regimens the sensitivity was lower, as only 1 (4.8%) of 21 with viral failure had an undetectable UTRA (P < 0.001).

Conclusions: Urine tenofovir-testing had a high negative-predictive value for viral failure in patients treated with DTG or ritonavir-boosted protease inhibitor regimens, where viral failure was largely explained by poor drug adherence. Frequent monitoring with inexpensive lateral flow urine TFV testing should be investigated prospectively in between viral load visits to improve viral load suppression on DTG-based first-line therapy in resource-limited settings.

Figure 1:. The association of UTRA results with long-term and short-term TFV exposure in participants treated with a) EFV- or b) PI/r- or DTG-based regimens.

DBS TFV- DP is plotted on X-axis (log scale) and total TFV in blood plasma on the Y-axis. Solid figures represent VF (VL > 400 copies/mL) and open figures represents VL suppression (VL ≤ 400 copies/mL) while round figures represent UTRA TN undetected and triangles UTRA TFV detected. Among EFV-treated participants, there was a weak positive Spearman’s rank order correlation (95% CI): 0.34 (0.07–0.57); the one participant with undetectable TFV in urine had no TFV-plasma result, due to a sample labelling error, and is therefore not shown. Among participants receiving PI/r or DTG-based regimens (panel b), those with VF and undetectable urine TFV (solid circles) most often had both low short term (TFV-plasma) and long-term (TFV-DP), whereas participants with suppressed VLs and detectable urine TFV (open triangles) had mostly higher long- and short-term TFV exposure, except for one outlying participant with undetectable urine TFV (open circle) despite high plasma TFV and DBS TFV-DP. There was a relatively good correlation between TFV-DP and plasma TFV; Spearman rank order correlation (95% CI) 0.69 (0.52–0.81).