Sofosbuvir/Velpatasvir Failure in Hepatitis C Subtype 6f: A Paired Virological Analysis Before and After Therapy

INTRODUCTION: Pangenotypic direct-acting antivirals are presumed globally effective, but subtypes endemic to low- and middle-income countries are underrepresented in the trials and replicon systems that define resistance. In a Thai elimination program, sofosbuvir/velpatasvir (SOF/VEL) failures were concentrated in HCV subtype 6f. We investigated whether this reflects acquired resistance. METHODS: Within a province-wide SOF/VEL test-to-treat program, 124 treatment failures were referred for genotyping with the genotype unknown at referral. NS5A and NS5B were deep sequenced in paired pre-treatment and failure specimens from 29 patients infected with subtype 6f, giving codon-level amino acid composition at both time points. Subtype-specific failure burden was compared with the province's viremic subtype prevalence. RESULTS: Subtype 6f comprised 81.4% of genotyped failures but 34.4% of provincial viremic infections (odds ratio 8.4; 95% CI 5.0-14.1). Across the 29 patients, six of the seven NS5A drug-target positions were immobile, and six carried the reference residue; 27/29 patients carried NS5A-58 proline. Movement was confined almost entirely to NS5A-93, was bidirectional, and left the summed per-patient resistance burden unchanged (p = 0.47). Acquired NS5B S282T was excluded at a median detection limit of 0.06%, low enough to detect a 1% subpopulation in every specimen. The cohort differed from the reference genome at 12 of 1042 codons, and nucleotide diversity contracted under purifying selection without convergent substitutions. CONCLUSIONS: Subtype 6f fails SOF/VEL against a drug-target landscape of ordinary lineage sequence, present before treatment and unchanged after it. Sofosbuvir-based retreatment therefore remains appropriate, and a post-failure resistance report cannot distinguish virological from non-virological failure.