HCV genotype-specific correlation with serum markers: higher predictability for genotype 4a

Abstract

Background: Several factors have been proposed to assess the clinical outcome of HCV infection. The correlation of HCV genotypes to possible serum markers in clinical prediction is still controversial. The main objective of this study was to determine the existence of any correlation between HCV genotypes to viral load and different clinical serum markers.

Methods: We performed a prospective cross-sectional and observational study. About 3160 serum HCV RNA positive patients were chosen from 4020 randomly selected anti-HCV positive patients. Statistical analysis was performed using the SPSS 16 software package. ROC (receiver operating characteristics) curves were used to compare diagnostic values of serum markers to predict genotypes.

Results: The most prevalent genotype was 3a (73.9%) followed by 1a (10.7%), 4a (6.4%) and 3b (6.1%) in Pakistani population. No correlation was found between viral load and serum markers for genotype 3a in a large no. of sample (n = 2336). While significant correlation was observed between viral load and AST in genotype 3b, ALP with viral load and ALT for genotype 1a. Patients with genotype 4a showed a significant inverse correlation with viral load and Hb level and AST with ALP. For genotype 4a, AUC (area under the curve) of ALT, ALP, AST, bilirubin, Hb level and viral load was 0.790, 0.763, 0.454, 0.664, 0.458 and 0.872 respectively.

Conclusions: In conclusion, there was a significant variable response of HCV genotypes with serum markers. Severity of disease is independent of serum marker level in genotype 3a, while the liver damage in genotype 4a may associate with viral cytopathic effect as well as the immune-mediated process. An index using six serum markers may correctly predict genotype 4a in patients with ≥ 75% accuracy.

Figure 1

Variation of significant serum markers among genotypes. Box plots of four significant serum markers i.e. viral load, bilirubin level, Serum ALP and ALT in relation with eight different HCV genotypes. The line through the middle of the box is the median while the top and bottom of the box are 25^th and 75^th percentiles.

Figure 2

Correlation of serum markers with each other in HCV genotypes. Significant correlation between different serum markers in genotypes 1a, 3a, 3b and 4a was found. This can lead to different possible mechanisms of liver injury in different genotypes.

Figure 3

Association of ALP and ALT in HCV genotype 1a. A negative significant correlation was observed between ALP and ALT in patients with genotype 1a.

Figure 4

Receiver operator characteristic (ROC) curves of six serum markers. ROC curves were drawn to evaluate best cutoff points for predicting genotypes. ROC curves of serum markers in patients showed that viral load, ALP, ALT and bilirubin level can better predict genotype 4a at given cutoff values.