Background
The relevance of plasma (trough) concentrations of cabotegravir (CAB) and rilpivirine (RPV) with regard to sustained virologic suppression in people with HIV (PWH) on long-acting treatment (LA) is an ongoing matter of debate. We aimed to add to the currently available data by analyzing the association between CAB and RPV trough concentrations and viral suppression in a real-world setting.
Methods
Retrospective single-center analysis of clinical routine data from PWH receiving CAB and RPV LA intramuscular every eight weeks. Concentrations of CAB and RPV measured within a week prior to (the planned) application of the next injection were considered for analysis; multiple measurements were allowed per person. People with laboratory signs of intercurrent infections were excluded from the analysis. Lower limits of normal for CAB and RPV were 800 ng/mL and 0.038 µg/mL, respectively. The quantification limit of the essay (HIV-1 RNA <20 cps/mL) was used to define viral suppression. Medians and interquartile ranges were used for descriptive statistics. Mann-Whitney and Fisher’s exact test were used for comparisons of continuous and binary variables between two groups, respectively. α=0.05 was used as level of significance. The use of monocentric clinical routine data did not require ethics approval.
Results
Overall, 350 samples from 165 PWH were included. Concentrations for CAB and RPV were 1,460 (1,120; 1,940) ng/mL or 1,160 (879; 1,560) ng/mL (p=0.009) and 0.073 (0.051; 0.101) µg/mL or 0.064 (0.037; 0.075) µg/mL (p=0.027) for PWH with HIV-1 RNA <20 cps/mL or ≥20 cps/mL, respectively. Proportions of samples with HIV-1 RNA ≥20 cps/mL were 8.2% and 21.9% (p=0.027) among samples with normal and low plasma concentrations of CAB, and 7.9% and 18.8% (p=0.035) among samples with normal and low plasma concentrations of RPV, respectively. Odds ratios for HIV ≥20 copies/mL were 3.1 (1.0; 8.4) and 2.7 (1.0; 6.5) for samples with low concentrations of CAB and RPV, respectively.
Conclusions
Our data indicate that low concentrations of CAB and/or RPV are associated with a higher odds of quantifiable viral load (≥ 20 copies/mL) and that, on a population level, lower levels of both drugs are found in people with HIV-1 RNA ≥20 cps/mL. It is unclear if this is also predictive of virologic failure. Based on our findings, however, it warrants further investigation if dosing can be indivualized taking trough concentrations into account.
Table or Figure
