1MRC University of Glasgow Centre for Virus Research, 8 Church Street Glasgow, UK G11 5JR.
Background and rationale of the study: High rates of sexually-transmitted infection and reinfection with hepatitis C (HCV) have recently been reported in HIV-infected men who have sex with men and reinfection has also been described in monoinfected injecting drug users. The diagnosis of reinfection has traditionally been based on direct Sanger sequencing of samples pre and post-treatment, but not on more sensitive deep sequencing techniques. We studied viral quasispecies dynamics in patients who failed standard of care therapy in a high-risk HIV-infected cohort of patients with early HCV infection to determine whether treatment failure was associated with reinfection or recrudescence of pre-existing infection. Paired sequences (pre- and post- treatment) were analysed. The HCV E2 hypervariable region-1 was amplified using nested RT-PCR with indexed genotype-specific primers and the same products were sequenced using both Sanger and 454 pyrosequencing approaches.
Results: Of 99 HIV-infected patients with acute HCV treated with 24-48 weeks of pegylated interferon alpha and ribavirin, 15 failed to achieve a sustained virological response (6 relapsed, 6 had a null response and 3 had a partial response). Using direct sequencing, 10/15 patients (66%) had evidence of a previously undetected strain post-treatment; in many studies, this is interpreted as reinfection. However, pyrosequencing revealed that 15/15 (100%) of patients had evidence of persisting infection. 6/15 (40%) patients had evidence of a previously undetected variant present in the post-treatment sample in addition to a variant that was detected at baseline. This could represent superinfection or a limitation of the sensitivity of pyrosequencing.
Conclusion: In this high-risk group, the emergence of new viral strains following treatment failure is most commonly associated with emerging dominance of pre-existing minority variants rather than re-infection. Superinfection may occur in this cohort but reinfection is over-estimated by Sanger sequencing. (Hepatology 2014;).