Circulating macrophage inflammatory protein-1 β /IL-12p40 ratio predicts sofosbuvir-based treatment outcome in HCV- genotype 4 patients

2.1 Ethical committee approval

This retrospective study involving human participants was approved by the Medical Ethical Committee of the National Research Centre (Approval NO: 20131) in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

2.2 Patients’ sera

This study followed STARD 2015 reporting guidelines. Sera from 50 healthy individuals without any history of liver diseases and 450 chronic HCV patients were retrospectively acquired from the Medical Center of Excellence-National Research Centre, Egypt that was entitled for screening/treatment of HCV-patients during the national campaign led by the Egyptian Ministry of Health and Population in the period of 2018–2019. The diagnosis of chronic HCV was based on physical examination, biochemical testing of liver function and the presence of HCV-antibodies in sera using recombinant HCV antigen based test (Axium, USA). Positive patients were confirmed for their infection by RT-PCR and only patients who had the replicative viral genome in their sera were included in the study. Inclusion criteria included positive HCV antibodies/RNA and only sera with viral load ⩾ 127,137 IU/mL were included. Exclusion criteria included non-viral hepatitis; cirrhosis/carcinoma; co-infection with HBV/HIV, uncontrolled diabetes, current/previous treatment with immune-suppressants, and previous treatment with interferon- γ /ribavirin combination. The treatment protocol included a daily dose of 400 mg Sofosbuvir (SOF) and 60 mg Daclatasvir (DCV) for uninterrupted twelve weeks. The end point was sustained virologic response (SVR) defined as the absence of serum HCV-RNA three months minimum of treatment cessation [14].

2.3 Biochemical and hematological assessments

Serum levels of liver transaminases, albumin and total bilirubin were estimated Bio-diagnostic kits (Biogamma, Stanbio, Germany) according to user manual. Blood count of red and white blood cells, platelets and hemoglobin were estimated by complete blood count (CBC) analysis (Eurocount Plus, Medsource Ozone Biomedicals, India).

2.4 Quantification of viral RNA in patients’ sera

Viral RNA was extracted using QIAamp Viral RNA Kit (QIAGEN GmbH, Hilden, Germany) according to user manual. Purified RNA from each sample was used as template for PCR amplification using Verso SYBR Green One-Step qRT-PCR Kit Plus ROX (ThermoFisher Scientific, USA). Reaction mixture included 5 μ l RNA template; 70 nM final concentration of 5’-UTR genotype 4 specific primers F: 5’-TTCACGCAGAAAGCGTCT-3’ and R: 3’-GGTGCACGGTCTACGAG-5’; 1.25 μ l of RT enhancer; 12.5 μ l of 2X One-Step qPCR SYBR Mix; 0.25 μ l of Verso enzyme mix, and the volume was completed to 25 μ l with nuclease-free water. For amplification, the cycling conditions included a single round cDNA synthesis at 55 ∘ C for 15 min followed by polymerase activation cycle at 95 ∘ C for 15 min and 40 cycles of denaturation at 95 ∘ C for 15 sec; annealing at 55 ∘ C for 30 sec and extension at 72 ∘ C for 30 sec. After PCR amplification, a melting curve analysis was performed by one cycle denaturation at 95 ∘ C for 3 s; one cycle starting temp at 60 ∘ C for 30 sec and finally melting for 80 cycles at 60 ∘ C for 10 sec with 0.5 ∘ C increment/cycle. Fluorescence data were continuously collected during this heating to monitor the dissociation of the strands, and the derivative melting curves were obtained with the Rotor-Gene Q Series software 2.0.3 (QIAGEN GmbH, Hilden, Germany) [16].

2.5 Genotyping

HCV-infected sera were subjected to genotyping using Versant HCV Genotype Assay (Lipa; Bayer, Germany). Briefly, HCV RNA was extracted as aforementioned, followed by cDNA synthesis using biotinylated random primers specific for the 5’UTR of G4-HCV [17]. The generated biotinylated amplicons were hybridized to immobilized oligonucleotides probes specific for the 5’ UTR of different HCV genotypes that are bound to nitrocellulose strips by a poly (T) tail. After hybridization, unhybridized DNA was washed out from the strips that were then treated with alkaline phosphatase labeled streptavidin (conjugate) which bounds then to the biotinylated hybrid. The chromogenic substrate (BCIP/NBT) that allows the formation of a purple/brown precipitate upon degradation by alkaline phosphatase of the conjugate was used for visualization of the banding pattern on the strip.

2.6 Estimation of circulating chemokines

Serum levels of MIP-1 β and IL-12p40 were estimated using MIP-1 β (CCL4) Human Instant ELISA TM Kit (Invetrogen – ThermoFisher Scientific; USA, Cat # BMS2030INST) and IL-12 p40 Human ELISA Kit (Invetrogen – ThermoFisher Scientific; USA, Cat # KAC1561); respectively according to the manufacturer instructions.

2.7 Statistical analysis

Data were statistically analyzed using GraphPad InStat 3.0 and GraphPad Prism version 8.0.2 (GraphPad, San Diego, CA).

3.1 Baseline demographic and clinical characteristics of the study cohort

The baseline characteristics of the study cohort are summarized in Table 1. Sera of total 500 individuals categorized as 450 chronic HCV-patients (217 male and 233 female) and 50 healthy (25 males/females) were included in this study. The mean age of healthy and infected individuals was 40.13 ± 7.11and 44.02 ± 10.28 years; respectively. Due to the wide age range, the study cohort was stratified into two subgroups as below and above 45 years. Based on quantitative PCR results, viral loads were also stratified into low ( 800,000 IU/mL) that was detected in 61 patients, and high load (800,000 to $>$ 10 7 IU/mL) detected in 389 patients. Genotyping results showed the prevalence of subtype 4a as the most common among patients (88%), with lower frequencies of G4m (6.9%) and G4o (5.11%). Serum levels of ALT and AST were significantly higher in infected patients comparing to healthy individuals ( 0.0001 and 0.002; respectively), while no significant difference in total bilirubin and albumin levels was observed. Hematological parameters in infected patients including hemoglobin, RBCs, WBCs and platelet count were statistically insignificant comparing to healthy group. Circulating levels of MIP-1 β and IL-12p40 were significantly different in patients comparing to healthy control, where patients had higher levels of MIP-1 β ( 0.0001) and lower levels of IL-12p40 ( P = 0.0245).

OPEN IN VIEWER

OPEN IN VIEWER

Out of 450 chronic HCV-patients, 421 (93.56%) have achieved sustained virologic response (SVR) after the completion of 12-weeks treatment protocol, while 29 patients (6.44%) were relapsers as confirmed by detection of viral RNA three months after the treatment cessation. Comparing to baseline levels, patients who achieved SVR had significantly lower ( 0.0001) serum ALT and AST, and higher levels of serum albumin ( 0.0001), with no significant difference observed in total bilirubin levels. Blood count of RBCs, WBCs and platelets was significantly higher ( 0.0001) after treatment success, but no difference was detected in hemoglobin levels (Fig. 1A). ALT, total bilirubin as well as viral RNA loads were elevated in relapsers despite the absence of statistical difference, while levels of AST and serum albumin were significantly higher ( P = 0.008 and 0.0001; respectively). No significant changes were observed in hemoglobin levels or RBCs count in relapser patients, but a significant reduction in WBCs and platelets’ count ( P = 0.0025 and 0.0055; respectively) was observed after treatment failure (Fig. 1B). Levels of circulating MIP-1 β and IL-12p40 were significantly decreased in relapser patients ( P = 0.016, 0.0003; respectively), while a slight elevation of their mean levels was observed in SVR group without a significant difference, either when compared with pre-treatment levels or with relapsers. However, both chemokines were found to be strongly correlated in both groups, whether pre- or post-treatment (Table 2), which encouraged us to explore the possibility of calculated MIP-1 β /IL-12p40 ratios to clinically discriminate treated groups. Accordingly, this ratio was calculated and analyzed for variation between SVR and relapser individuals, and a significant difference of the baseline ratio has been observed, with relapser patients having a significantly higher mean ratio comparing to SVR group ( P = 0.0002), but no significant difference was found after treatment (Fig. 1C).

To assess the hepatic and immunological restorative effect of the treatment protocol, the clinical features of SVR group was compared to healthy control. Results showed that SVR patients had a significantly higher ( 0.0001) mean levels of ALT, AST, MIP-1 β and lower levels of albumin ( 0.0001), IL-12p40 ( P = 0.0008) and RBCs ( P = 0.016) and platelets count ( 0.0001) comparing to healthy individuals (Fig. 2) despite the viral clearance, but no difference in total bilirubin, hemoglobin or WBCs count was detected.

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

In order to determine the baseline features that might predict the treatment outcome, a multivariate logistic regression analysis of baseline demographic and clinical features was performed. The analysis results revealed that gender, age, viral load, albumin levels and MIP-1 β /IL-12p40 ratio can significantly predict treatment outcome ( P = 0.0114, 0.008, 0.04, 0.0004 and 0.0001; respectively), where male patients; age more than 45 years; high baseline viral load and chemokine ratios, and low levels of serum albumin are more likely associated with treatment failure. Meanwhile, viral subtype, baseline levels of ALT, AST, total bilirubin, WBCs, RBCs, hemoglobin and individual chemokines had no significant effect on the treatment outcome in all patients (Table 3).

3.4 Determination of the ideal threshold for treatment outcome prediction

A receiver operating characteristic (ROC) curve was calculated to define the ideal cut-off value of chemokine ratio that might predict the final treatment outcome. AUC for SVR prediction was 0.6917 ( P = 0.0108) with 95% CI: 0.5566 to 0.8268. The calculated threshold for predictive baseline MIP-1 β /IL-12p40 ratio was 8.245 with 93.33% sensitivity (95% CI: 78.68% to 98.82%) and 43.33% specificity (95% CI: 27.38% to 60.8%), where patients with ratios 8.245 were more likely to achieve SVR than those having ratios ⩾ 8.245 (Fig. 3A). Chi-square test results showed high predictive capacity of the calculated threshold ( 0.0001), with positive and negative predictive values of 93.1% (95% CI: 0.9026 to 0.9515) and 93.33% (95% CI: 0.7868 to 0.9882); respectively, with 99.49% (95% CI: 0.9816 to 0.9991) and 49.12% (95% CI: 0.3662 to 0.6174) sensitivity and specificity; respectively. The calculated ratio also showed a significant linear trend with strong association with treatment outcome; whether SVR or failure (Chi-square, df = 12.83, 1; P = 0.0003) (Fig. 3B).

3.5 Correlation of circulating chemokine ratios with pre/post-treatment clinical features

In patients who achieved SVR, the baseline chemokine ratio was found to be positively correlated with both viral load ( r = 0.8128, 0.0001) (Fig 4A) and WBCs count ( r = 0.4233, 0.0001) (Fig. 4B), while inversely correlated with ALT and AST ( r = - 0.9779, 0.0001 and r = - 0.9691, 0.0001; respectively) (Fig. 4C). A week but significant inverse correlation was also observed between the chemokine ratio and both albumin ( r = - 0.2245, 0.0001) (Fig. 4D) and hemoglobin levels ( r = - 0.1378, P = 0.0046) (Fig. 4E). In relapser patients, pre-treatment chemokine ratio was found to have positive correlations with albumin ( r = 0.6182, P = 0.0002) (Fig. 4F), total bilirubin ( r = 0.3292, P = 0.0406) (Fig. 4G), and WBC’s count ( r = 0.3498, P = 0.0314) (Fig. 4H), but no correlation was found with viral load or liver enzymes.

Post-treatment, the chemokine ratios were found to positively correlate with liver transaminases ( r = 0.35, 0.0001 for ALT and r = 0.2031, 0.0001 for AST) (Fig. 5A); albumin levels ( r = 0.1255, P = 0.005) (Fig. 5B) and WBCs count ( r = 0.16, P = 0.0005) (Fig. 5C) while inversely correlated with total bilirubin ( r = - 0.1215, P = 0.0063) (Fig. 5D) and RBCs count ( r = - 0.2405, 0.0001) in SVR patients (Fig. 5E). In relapsers, the ratio inversely correlated with viral load ( r = - 0.8238, 0.0001) (Fig. 5F) but positively correlated with serum albumin ( r = 0.3737, P = 0.0229) (Fig. 5G).