Objective: To investigate the influence of genotypic resistance-guided HIV-treatment decisions on long-term clinical and virological outcomes in patients failing antiretroviral therapy by a prospective 30-month observation following a 6-month randomized study (Argenta trial).
Methods: Patients (n=174) with virological failure on highly active antiretroviral therapy (HAART) were initially randomized (1:1) to receive empirical therapy or guidance by genotypic resistance results. After month 6, all patients with HIV RNA >1,000 copies/ml received genotypic resistance tests and expert advice. Predictors of virological and clinical outcomes were analysed by logistic regression and Cox's regression models.
Results: There was a gradual increase in the proportion of patients with HIV RNA <400 copies/ml with 29.3% at 36 months (intent-to-treat) without differences between initial randomization arms. Independent predictors of 36-month virological response were the use of a salvage therapy with less daily doses and a more pronounced 3-month viral load drop. At 36 months, 84% survived without new AIDS events/death. Independent predictors of new AIDS events/death were previous AIDS events, higher baseline viral load, less pronounced 3-month viral load drop and, in a separate model, baseline protease substitutions K20M/R and 184V.
Conclusions: The virological benefit of genotype-guided treatment decisions was continuously appreciable over time. Short-term virological response and viral cross-resistance were independent predictors of long-term outcomes.