Study sample
The present cross-sectional study was conducted in Dehong Prefecture of Yunnan Province at China’s southwest border, with injection drug use (IDU) as the predominant mode of HIV transmission through the early 2000s and continues to be an important mode of HIV and HCV infection[15]. By March 2016, 1,017 HIV/HCV coinfected patients with 170 (16.7%) deaths had been reported to and registered in the Comprehensive Response Information Management System (CRIMS) for HIV/AIDS in China[16]. Out of the remaining 847 patients, 463 (54.7%) were identifiable during the study period from April to October in 2016. Of them, 390 (84.2%) were receiving cART and gave informed consent to participate in the present study. Forty-seven patients were further excluded from the study due to missing data on biochemical measures for FIB-4 calculation including aspartate aminotransferase (AST), alanine aminotransferase (ALT), and platelet count (PLT). Thus, a total of 343 HIV/HCV coinfected patients were included in the final analysis. These participants showed no significant differences with the 120 identified yet excluded patients in the distribution of age, sex, marital status, HIV transmission route, HBsAg serostatus, baseline CD4 cells count, years on cART, current HCV RNA and HIV RNA levels, while showed difference in ethnicity (Table S1). The study was approved by the Institutional Review Board of Fudan University School of Public Health, Shanghai, China. All subjects provided informed consent prior to enrollment.
Data Extraction
Demographical and clinical epidemiological data were extracted from the CRIMS. The data included age, sex, marital status, ethnicity, HIV transmission route, date of cART initiation, cART regimen, CD4+ T cell counts at cART initiation and follow-up visits.
Blood testing
Biochemical tests for liver fibrosis assessment
Liver biopsy is an invasive technique with complications and costly, though it is a gold reference standard for determining liver fibrosis. In the light of these limitations , we assessed liver fibrosis by FIB-4 score, calculated as (AST [IU/L] × age [years]) / (PLT [109/L] × ALT[IU/L]1/2) using Sterling’s formula [17], which is an internationally recognized well-established noninvasive indicator for liver fibrosis[18]. Biochemical measurements of AST, ALT and PLT were performed using an automatic biochemistry analyzer (Beckman Coultner, USA), according to the manufacturer’s protocol.
FIB-4 is generally divided into three categories: no or mild liver fibrosis with FIB-4 <1.45, intermediate liver fibrosis with 1.45 ≤ FIB-4 ≤ 3.25, and advanced liver Fibrosis with FIB-4 >3.25. In this study, we considered advanced liver fibrosis (FIB-4 >3.25) as the major outcome.
HCV RNA Quantification and genotyping
Plasma HCV viral RNA was extracted (Roche diagnostic products (Shanghai) Co., Ltd., China), and quantified by a real-time polymerase chain reaction (RT-PCR) technique using commercially available kits for the quantification of HCV RNA (PCR-Fluorescent Probing, PG Biotech Ltd., Shenzhen, China). The limit of detection was 500 copies/ml, and the linear range of HCV RNA quantification was from 1.0 × 103 to 5.0 × 107 copies/ml.
Amplification was completed by a nested PCR with E1- or NS5B‐specific primers. Splicing, proofreading, and aligning sample sequences were performed using ChromasPro 1.5 and BioEdit7.0.9.0 software. HCV genotype reference sequences were retrieved from the HCV database (http://hcv.lanl.gov/content/sequence/HCV/ToolsOutline.html). The HCV gene subtype was determined according to the phylogenetic tree,which established by the Neighbor-joining method using MEGA 7.0 software[19].
Multiplex cytokine bead assay and sCD14
Twenty-seven cytokines, chemokines and growth factors cytokines in plasma specimens were quantified by a multiplex analysis performed on the BioPlex®200 Multiplex System platform using the Bio-Plex Human 27-plex panel of cytokines/chemokines/growth factors (Bio-Rad, Hercules, CA, USA), according to the manufacturer’s instructions. These biomarkers included 1) proinflammatory biomarkers such as interleukin (IL)-1β, IL-2, IL-6, IL-7, IL-8, IL-9, IL-12, IL-15, IL-17, granulocyte colony stimulating factor (G-CSF), granulocyte macrophage colony stimulating factor (GM-CSF), Interferon-γ (IFN-γ), and tumor necrosis factor (TNF-a); 2) anti-inflammatory biomarkers such as IL-1ra, IL-4, IL-5, IL-10, and IL-13; 3) chemokines such as IFN-γ-inducible protein (IP-10), monocyte chemoattractant protein-1 (MCP-1), macrophage inflammatory proteins-1alpha (MIP-1a), MIP-1b, and regulated upon activation normal T-cell expressed and secreted (RANTES); and 4) growth factors such as Eotaxin, fibroblast growth factor 2 (FGF-basic), platelet-derived growth factor (PDGF-BB), and vascular endothelial growth factor (VEGF). The detection limit for each molecule was determined by the recovery of the corresponding standard, and the lowest values with more than 70% recovery were set as the lowest detection limits. Measurements less than the lower limit of quantification (LLOQ) were assigned a value of half the LLOQ, and measurements more than the upper limit of quantification (ULOQ) were assigned a value of twice the ULOQ for each marker.
Plasma sCD14 was measured according to the manufacturer’s protocol for commercial enzyme-linked immunosorbent assay (ELISA) ( Hycult Biotech, Wayne PA, USA).
Other tests
Human hepatitis B virus surface antigen (HBsAg) was tested by ELISA technique (Wan Tai Biomedical Co. Ltd, Beijing, China). CD4 cell counts were assessed by FACSCount (Becton, Dickinson and Co., San Jose, CA, USA).
Statistical analysis
Distributions of categorical variables were compared using chi-square test or Fisher’s exact test, and continuous varabiles were compared using t-test or Mann-Whitney U test. Log10-transformation was conducted for plasma cytokines levels for further statistical and regression analyses. A multiple logistic regression analysis with adjustment for age (as a contiuous variable), sex (0-male, 1-female), ethnicity (0-Han, 1-Dai, 2-Jingpo,3-Others, entered as dummy variables), current HIV RNA (0-Undetectable, 1-Detectable), current HCV RNA (0-Undetectable, 1-Detectable), current CD4 cell counts ((0-<200, 1-200-349, 2-≥350, entered as dummy variables), years on cART (as a contiuous variable) and ART regimen type ((0-NVP, 1-TDF, 2-Others, entered as dummy variables) was used to explore the association of liver fibrosis with the plasma level of each of the twenty-seven cytokines and sCD14. The odds ratio (OR) and 95% confidence interval (95%CI) represent the risk of higher FIB-4 with per one log-unit change in plasma cytokine concentration. Spearman correlations were computed to explore correlations between the plasma inflammatory biomarkers. Statistical significance was defined as P <0.05 and Bonferroni P <0.002 (0.05/27≈0.002) for multiple comparison adjustment. All statistical analyses were performed using R software (version 3.3.2).