Background: Ideally patients who fail first line antiretroviral therapy (ART) are switched to second line quickly, yet logistical issues, clinician decisions and patient preferences make some delay in switching to second line ART common. Delays in switching to second line may be associated with poor outcomes on second line if resistance mutations develop or CD4 count declines substantially. This study explores the impact of delaying second line ART after first line failure on rates of virologic failure over multiple years on second line ART. Methods: Observational cohort study using medical records from 9 clinics across South Africa, including patients with documented virologic failure on first line ART (2 consecutive viral load levels >1000 copies/mL) and at least one year of potential follow-‐up time after failure. Cox proportional hazards models analyzed the association between time to switch to second line (categorized) and virologic failure on second line ART. To account for patients who did not switch to second line, all patients were included in marginal structural models for death following first line failure. All models were stratified by peak CD4 count prior to failure to identify groups of patients for which delaying second line ART had the biggest impact on second line outcomes. Results: 5,895 patients who failed first line ART were included in analysis. 37% never switched to second line. Among patients who switched, median time to switch after failure was 3.4 months (IQR: 1.1-‐8.7 months). Median follow-‐up time on second line was 1.4 years (IQR: 0.71-‐2.41 years). Among patients who did not switch, at one year after failure 48% had been lost from care, 5.7% died, and of those remaining in care 53% had evidence of re-‐suppression. In Cox models, delaying switch was associated with increased rates of virologic failure on second line for patients who had peak CD4 count prior to first line failure ≤ 100 cells/mm3 (adjusted HR for switch in 3-‐6 months vs. 0-‐1.5 months = 2.13 (95% CI: 1.01, 4.47)). Marginal structural models, which adjusted for survivor bias, showed increased risk of death was associated with delaying switch to second line, especially among patients with low CD4 count prior to first line failure (adjusted HR for switch in 6-‐12 months vs. 0-‐1.5 months = 1.49 (95% CI: 0.91, 2.42)). Conclusions: Even small delays in switch to second line ART was associated with increased virologic failure and increased death on second line among patients with low CD4 counts on first line.
Conference: CROI, 2015, Seattle, Washington, USA