Background

International trials have shown that CD4⁺ T-cell-guided structured treatment interruptions (STI) of antiretroviral therapy (ART) lead to worse outcomes than continuous treatment. We simulated continuous ART and STI strategies with higher CD4⁺ T-cell interruption/reintroduction thresholds than those assessed in actual trials.

Methods

Using a model of HIV, we simulated cohorts of African adults with different baseline CD4⁺ T-cell counts (≤200; 201-350; and 351-500 cells/μl). We varied ART initiation criteria (immediate; CD4⁺ T-cell count <350 cells/μl or ≥350 cells/μl with severe HIV-related disease; and CD4⁺ T-cell count <200 cells/μl or ≥200 cells/μl with severe HIV- related disease), and ART interruption/reintroduction thresholds (350/250; 500/350; and 700/500 cells/μl). First-line therapy was non- nucleoside reverse transcriptase inhibitor (NNRTI)-based and second-line therapy was protease inhibitor (PI)-based.

Results

STI generally reduced life expectancy compared with continuous ART. Life expectancy increased with earlier ART initiation and higher interruption/reintroduction thresholds. STI reduced life expectancy by 48–69 and 11–30 months compared with continuous ART when interruption/reintroduction thresholds were 350/250 and 500/350 cells/μl, depending on ART initiation criteria. When patients interrupted/reintroduced ART at 700/500 cells/μl, life expectancies ranged from 2 months lower to 1 month higher than continuous ART. STI-related life expectancy increased with decreased risk of virological resistance after ART interruptions.

Conclusions

STI with NNRTI-based regimens was almost always less effective than continuous treatment, regardless of interruption/reintroduction thresholds. The risks associated with STI decrease only if patients start ART earlier, interrupt/reintroduce treatment at very high CD4⁺ T-cell thresholds (700/500 cells/μl) and use first-line medications with higher resistance barriers, such as PIs.