Comparison and application of a novel genotyping method, semiautomated primer-specific and mispair extension analysis, and four other genotyping assays for detection of hepatitis C virus mixed-genotype infections.

To date the true prevalence of hepatitis C virus (HCV) mixed-genotype infections has not been established mainly because currently available methods are not suitable for the detection of mixed genotypes in a viral population. A novel semiautomated genotyping method, primer-specific and mispair extension analysis (S-PSMEA), which is more reliable than other genotyping assays was developed for detection of HCV mixed-genotype infections. A genotype present at levels as low as 0.8% in a defined mix of HCV genotypes was detected, showing a 20-fold increase in sensitivity over that of direct DNA sequencing. A total of 434 HCV isolates were genotyped and analyzed for a comparative study of the accuracy between S-PSMEA and four current genotyping methods. The results showed that viruses in approximately 40% of the samples from this group determined to be infected with mixed genotypes by S-PSMEA were undetected by direct DNA sequencing due to its low sensitivity. Type-specific PCR, line probe assay, and restriction fragment length polymorphism analysis performed poorly, being able to identify only 38.5, 16.1, and 15.4% of mixed-genotype infections, respectively, that were detected by direct DNA sequencing. The prevalence of mixed-genotype infections detected by S-PSMEA was 7.9% (12 of 152 donors) among HCV-infected blood donors, 14.3% (15 of 105) among patients with chronic hepatitis C, and 17.1% (6 of 36) among thalassemia patients who had received multiple transfusions. The data lead us to conclude that HCV mixed-genotype infections are more common than previously estimated and that S-PSMEA may be the method of choice when detection of genotypes present at low levels in mixed-genotype infections is required due to its higher level of sensitivity.