Background:  In the ESPRIT study that tested interleukin-2 (IL-2) safety and efficacy in HIV patients, all-cause mortality for patients co-infected with hepatitis B virus (HBV) or hepatitis C virus (HCV) was higher than in HIV-mono-infected patients. The aim of this investigation was to assess whether stage of liver fibrosis as measured by 3 liver fibrosis markers influenced all-cause mortality in the co-infected population.

Methods:  For HCV RNA⁺ or hepatitis B surface antigen (HBsAg)⁺ patients stored plasma collected at entry to ESPRIT was tested for the fibrosis marker hyaluronic acid. Fibrosis indices APRI (aspartate aminotransferase [AST] to platelet ratio index) and FIB-4 [age x AST/(platelets x ALT^1/2) were calculated based on locally collected data. Biomarkers were categorized according to levels of fibrosis using previously reported cut-points for significant fibrosis. Adjusted and unadjusted hazard ratios (HR) were estimated using Cox models. Adjustment was for age, gender, race, injecting drug use, AIDS, and nadir and baseline CD4⁺ count. To facilitate comparisons among biomarkers HR per 1 standard deviation (SD) change in the log₁₀-transformed marker were estimated. Associations were studied in the IL-2 and control groups separately.

Results:  Compared to HIV-mono-infected persons, the HR for all-cause mortality among those with co-infection were 1.8 (95%CI 1.1 TO 2.8; 99 deaths) and 2.8 (95%CI 1.9 TO 4.2; 104 deaths), for the IL-2 and control groups, respectively. Data were available for all 3 fibrosis markers at baseline for 463 co-infected patients (HBV, n = 140, 30.3%; HCV, n = 316, 68.3%; HBV/HCV, n = 5, 1.1%). Overall, 33 of these patients died during 2972 person-years of follow-up (4 in the IL-2 group and 5 in the control group) due to liver-related causes. In the control group, but not in the IL-2 group, each biomarker was significantly associated with all-cause mortality (see the table). Of note, the rate of death was comparable between control and IL-2 group for those with normal biomarker levels. For FIB-4 (p = 0.02) and APRI (p = 0.02), but not hyaluronic acid (p = 0.11), the biomarker associations varied by treatment group. Among controls, the association with mortality was stronger for hyaluronic acid than FIB-4 or APRI.

Conclusions:  Liver fibrosis markers predicted mortality in HIV/HBV- or HIV/HCV-co-infected persons randomized to ART alone (control group), but not in the group randomized to receive adjunctive IL-2 therapy. This suggests that IL-2 may dampen the clinical consequences of viral hepatitis-induced liver damage.