Liver fibrosis progression and clinical outcomes are intertwined : role of CD4+ T-cell count and NRTI exposure from a large cohort of HIV/HCVcoinfected patients with detectable HCV-RNA A MASTER cohort study

Introduction: Patients coinfected with human immunodeficiency virus (HIV) and hepatitis C virus (HCV) suffer from faster progression of liver fibrosis (LF) and have greater risk of worse clinical outcomes. We evaluated predictors and incidence of these events in a large multicentre cohort. Methods: We selected all HIV-infected patients starting a first-line combination antiretroviral therapy (cART), with detectable HCVRNA, without exposure to interferon/ribavirin, with ≥2 fibrosis-4 index (FIB-4) classifications before cART. KaplanCMeier analysis was used to estimate incidence of clinical events (AIDS, non-AIDS related, deaths) and LF progression (via transitions: from FIB-4 class 1 to 2 or 3, from class 2 to class 3, and worsening by 0.5 point). Multivariate Cox regression was used to assess predictors, baseline, or time updated. Results: One thousand four hundred thirty-three patients were selected. Overall, 745 clinical events occurred, with an incidence of 7.6% over 9811 person-year of follow-up (PYFU) and a median survival time of 9.36 years. Incidence of LF progression from FIB-4 class 1 to 2 or 3 was 12.4%, and from FIB-4 class 2 to 3 was 7% with a median survival time of 5.67 and 10.35 years, respectively. At multivariate analyses, intravenous drug use and time-updated gamma-glutamyl transferase (γGT) were negative predictors for any outcomes, either clinical or FIB-4 progression. Higher CD4+ T-cell protected from clinical events, and lower HIV-RNA and higher CD4 + T-cell appeared to protect from FIB-4 transitions. Moreover, independently from the viro-immunological status, current FIB-4 class 3 predicted clinical events. Occurrence of AIDS and cardiovascular/kidney events were significant predictors of 0.5 point worsening and transitions of FIB-4, respectively. Prolonged exposure to nucleos(t)ide reverse transcriptase inhibitors (NRTI) was a negative predictor for any outcomes. Conclusion: Both clinical and LF progression in HIV/HCV-coinfected patients depend strongly on immune status. Intravenous drug users and patients with high γGT (a possible proxy for alcohol abuse) are most-at-risk for both outcomes, as well those who had prolonged exposures to the NRTI class. Therefore, these patients should be prioritized for the access to anti-HCV therapy and a test-and-treat strategy should be implemented for early initiation of cART. Possible benefits of NRTI sparing regimens in HIV/HCVcoinfected patients should be investigated