Abstract
BACKGROUND: Lamivudine + dolutegravir maintenance dual therapy (DT) could be less effective than 3-drug therapy (TT) in the context of resistance-associated mutations to nucleoside reverse transcriptase inhibitors (NRTIs). The ARCA database was queried to test this hypothesis with a trial emulation strategy. METHODS: People with HIV taking 2 NRTIs plus a protease inhibitor or a non-NRTI who switched to DT or dolutegravir-based TT were followed up from the first HIV RNA <50 copies/mL (baseline) to virologic failure (VF; ie, 2 consecutive HIV RNA ≥50 copies/mL or 1 HIV RNA ≥200 copies/mL). Those switching to DT within 6 months were assigned to the treatment arm and all other patients to the control arm. Each participant was also cloned, assigned to the opposite strategy, and censored at the time of deviation from that strategy. Using inverse probability of censoring weight Cox regression models, we calculated hazard ratios of VF for DT vs TT stratified for the presence of resistance-associated mutations. RESULTS: Overall 626 people were analyzed: 204 with DT and 422 with TT (73% men; mean age, 44 years). Ten and 31 VFs occurred with DT and TT, respectively, over a median 5.8 years. When compared with a fully active TT, the DT had similar efficacy (adjusted hazard ratio, 0.88; 95% CI, .29-2.61; P = .812) when full susceptibility was confirmed at historical genotype. When previous M184V/I was present in both groups, the risk of VF was higher for DT vs TT but was not statistically significant (adjusted hazard ratio, 3.06; 95% CI, .45-20.84; P = .252). CONCLUSIONS: DT was not associated with a significantly higher risk of VF than dolutegravir-based TT.