BACKGROUND: The use of surrogate markers for targeting viral load (VL) testing could be an alternative to universal VL testing during antiretroviral treatment (ART) and would allow for more effective resource allocation. We investigated the correlation between levels of HIV RNA and interferon-γ-inducible protein 10 (IP-10) in Ethiopian adults at 12 months after ART initiation. In addition, we specifically investigated differences in IP-10 levels between patients with and without virological suppression.SETTING: Cohort of HIV-positive adults receiving ART at Ethiopian health centers.METHODS: Using a nested case-control design, individuals without virological suppression (HIV RNA ≥ 150 copies/mL) at 12 months after ART initiation were gender-matched with virologically suppressed controls (1:2 ratio). IP-10 levels were correlated with HIV RNA, and the distribution of IP-10 was compared for 3 VL strata: <150 copies/mL (VL < 150), 150-999 copies/mL (VL150-999), and ≥1000 copies/mL (VL ≥ 1000).RESULTS: At 12 months after ART initiation, the following VL distribution was found among 192 individuals (50% women): VL < 150, 122/192 (63.5%); VL150-999, 23/192 (12.0%); and VL ≥ 1000 47/192 (24.5%). IP-10 and HIV RNA levels were positively correlated (r = 0.481; P < 0.0001). Median IP-10 levels for the VL strata were VL < 150: 159 pg/mL [interquartile range (IQR) 121-246], VL150-999: 174 pg/mL (IQR 131-276), and VL ≥ 1000: 343 pg/mL (IQR 190-529), respectively. These differences were statistically significant for VL ≥ 1000 versus VL < 150 (adjusted P < 0.001) and VL150-999 (adjusted P = 0.004), respectively.CONCLUSIONS: IP-10 and HIV RNA levels during ART showed significant correlations, with significantly higher IP-10 concentration in ART recipients with VL ≥ 1000 copies/mL compared to those with suppressed or undetectable VL.
