It is necessary to identify simple biomarkers that can efficiently predict the efficacy of long-term antiretroviral therapy (ART) against human immunodeficiency virus (HIV), especially in underdeveloped countries. We characterized the dynamic changes in plasma interleukin-18 (IL-18) and assessed its performance as a predictor of long-term virological response. This was a retrospective cohort study of HIV-1-infected patients enrolled in a randomized controlled trial with a follow-up of 144 weeks of ART. Enzyme-linked immunosorbent assay was performed to evaluate plasma IL-18. Long-term virological response was defined as HIV-1 RNA <20 copies/mL at week 144. Among the 173 enrolled patients, the long-term virological response rate was 93.1%. Patients with a long-term virological response had significantly lower levels of week 24 IL-18 than non-responders. We defined 64 pg./mL, with a maximum sum of sensitivity and specificity, as the optimal cutoff value of week 24 IL-18 level to predict long-term virological response. After adjusting for age, gender, baseline CD4+ T-cell count, baseline CD4/CD8 ratio, baseline HIV-1 RNA level, HIV-1 genotype and treatment strategy, we found that lower week 24 IL-18 level (≤64 vs. >64 pg./mL, a OR 19.10, 95% CI: 2.36-154.80) was the only independent predictor of long-term virological response. Early on-treatment plasma IL-18 could act as a promising indicator for long-term virological response in patients with HIV-1 infection. Chronic immune activation and inflammation may represent a potential mechanism; further validation is necessary.
Read full abstract