Background:  Evidence from randomized trials show that ART interruption increases the risk of non-AIDs clinical (including liver-related) outcomes. We hypothesized that liver disease progression in ART treated co-infected patients may be due, in part, to the consequences of repeated treatment interruptions. Therefore, we examined the impact of ART interruption on fibrosis progression, using AST-to-platelet ratio index (APRI) >1.5 as a surrogate marker of liver fibrosis.

Methods:  Data were analyzed from a Canadian, multi-site prospective cohort of HIV-infected adults receiving ART (n = 701) with virologic evidence of HCV infection (n = 617) between 2003 and 2009. Patients with significant fibrosis (n = 136), defined as having an APRI score >1.5 at baseline, were excluded. The exposure of interest, ART interruption, was defined as the cessation of all antiretrovirals for at least 14 days after being on ART for at least 30 days, as a time-updated variable. Univariate and multivariate Cox regression models were used to evaluate the association of baseline characteristics and time-varying covariates such as, recent intravenous drug use (IDU), CD4 cell count and HIV viral load, with developing significant fibrosis.

Results:  In the study, 481 subjects were followed for a median of 13 months (6 to 19); 72% were male, 3% were HBsAg positive, 36% reported IDU during follow-up and 11% interrupted ART. The median age was 45 years; HCV duration, 17 years, baseline APRI score, 0.54 and nadir CD4 cell count, 161 cells/mL. Univariate analyses revealed that baseline APRI and time-updated CD4 cell count, HIV viral load and ART interruption were associated with fibrosis progression, while nadir CD4 cell count, age, duration of HCV and HIV infections, and recent IDU were not. In the multivariate model, baseline lnAPRI (HR 1.61, 95%CI 1.36 to 1.90, 0.25 increase) and HIV RNA (HR = 1.05, 0.95 to 1.17) were associated with fibrosis progression, while a higher current CD4 cell count was protective (HR = 0.94, 0.88 to 1.00/50 cells/mL), After adjustment, the effect of ART interruption was somewhat attenuated (HR = 1.80, 0.73 to 4.40).

Conclusions:  ART interruption appears to be associated with fibrosis progression in HIV-HCV co-infection and was only partially accounted for by HIV viral load and CD4 cell counts. Our findings suggest that some of the liver disease progression observed in ART treated co-infected patients may in fact be due negative consequences of treatment interruption.