Therapeutic preparedness: DAA-resistant HCV variants in vitro and in vivo

Chronic HCV infection is one of the major global causes of chronic liver disease, cirrhosis, and HCC. With the approval of direct-acting antivirals (DAAs), sustained virologic response (SVR) rates of 90%–95% have been achieved across all HCV genotypes (GT). There are several reasons for virological treatment failure, such as drug-drug interactions or the presence of cirrhosis/HCC. Moreover, treatment failure is associated with the selection of resistance-associated substitutions (RASs) that confer resistance to DAAs.1 The WHO has set the target to eliminate HCV by 2030. Although the global HCV prevalence has decreased by 7 million since 2015, only a few countries are currently on track to reach the elimination targets. Global challenges include the availability of next-generation DAAs in resource-limited settings and the accessibility of vulnerable groups where reinfection rates are high [eg, people who inject drugs (PWIDs) or prisoners]. Moreover, a vaccine against HCV is not yet available. Other challenges exist in difficult-to-treat patients with multiple DAA pretreatments, cirrhosis, and those infected with GT3 or unusual subtypes. The importance of RASs depends primarily on the DAA regimen and the HCV GT. Due to the high antiviral activity and resistance barrier, general resistance testing is not necessary for first-line therapy with the pangenotypic regimens glecaprevir/pibrentasvir (G/P) or velpatasvir/sofosbuvir (VEL/SOF), but may be helpful for treatment optimization in GT3 and other unusual subtypes. For second-line treatment of patients with previous DAA treatment failure, voxilaprevir/velpatasvir/ sofosbuvir (VOX/VEL/SOF) is recommended with SVR rates above 90%, regardless of the presence of RASs. However, VOX/VEL/SOF was slightly less effective in patients with GT3 and cirrhosis in clinical and real-world studies and is not available in resource-limited settings.1 The current issue of Hepatology includes 2 studies that address unanswered questions in clinical translational HCV research and the role of RASs in treatment response. Both studies should increase our awareness of the high genetic variability of HCV, which might result in future limitations of the currently available DAA therapies. The first study conducted by Fernandez-Antunez et al provides important functional data regarding the selection of HCV resistance after multiple DAA treatments in cell culture. The authors establish a novel, highly efficient GT3a infectious cell culture system based on the chimpanzee infectious clone S52,2 requiring 31 mutations to reach high replication levels in Huh7.5 cells. They use this model to study how NS3, NS5A, and NS5B RASs influence the activity of second-generation DAA regimens such as VEL/SOF, G/P, or VOX/VEL/SOF and conducted DAA escape experiments and full-length next-generation-sequencing of the HCV ORF (open reading frame). They described the selection of resistant HCV variants and showed that NS5A RASs could compromise the efficacy of double-DAA pangenotypic regimens in GT3a.3 Successive treatment with the double-DAA regimens G/P and VEL/SOF resulted in the selection of , for example, L31F+Y93H in NS5A, which was associated with cross-resistance to other NS5A inhibitors. Consequently, G/P did not clear viral replication from VEL/SOF-treated isolates and vice versa. Interestingly, these observations are consistent with clinical practice, which showed that repetitions of NS5Ai-based treatments using double-DAA regimens are ineffective in the presence of NS5A RASs and that a switch of a DAA class not present in previous therapies should be considered if VOX/VEL/SOF is not available.1,4 Also, the high efficiency of VOX/VEL/SOF in cell culture in eradicating HCV variants, even in the presence of RASs, was in agreement with clinical observations.1,5 In clinical practice, G/P+SOF is another alternative to VOX/VEL/SOF for the retreatment of DAA failures that was not tested in the current study, but so far, it is not clear whether G/P+SOF may be superior to VOX/VEL/SOF.5 In contrast to clinical data,6 the current study surprisingly could not detect A30K in G/P treated cell culture isolates. The discrepancy could be due to the cell culture conditions or the clonal HCV isolate used. This is also a limitation of this study, as this experimental approach does not allow the investigation of patient isolates, and the highly heterogeneous HCV quasispecies cannot be represented using this approach. However, this study formally demonstrates that highly resistant, stable variants can be selected for each currently available second-generation DAA in clinical use, compensating fitness costs by, in part, complex patterns of secondary mutations, including SOF. By successive treatment, thereby, any dual drug-based regimen can be overcome. In addition, all selected variants provided cross-resistance to alternative drugs against the same target. The second study, conducted by Vo-Quang et al, makes a significant contribution to the virological characterization and provides insights into the treatment of patients infected with unusual HCV non-1a/1b genotype 1 subtypes. In a French cohort who failed to first-line DAA treatment, 47 patients (7%) were infected with unusual GT1 subtypes, mainly patients of African origin. In this subgroup, RASs were characterized by different sequencing approaches, and the outcome of retreatment was analyzed.7 Unusual GT1 and GT4 subtypes are rarely found in European or North American patients but are common in patients of African origin; unusual GT1 subtypes are particularly prevalent in sub-Saharan Africa. In addition, high numbers of baseline NS5A RASs as well as reduced SVR rates, were observed, especially when first-generation DAAs were used for treatment.8,9 The current study shows that the majority of patients with unusual GT harbor at least 2 NS5A RASs after DAA treatment failure, mostly at positions 31, 58, or 93. In patients with baseline samples available, unfortunately only 6 of 43 DAA treatments did not select additional RASs; therefore, the detected RASs can be considered naturally inherent in unusual subtypes. Interestingly, all patients achieved SVR after retreatment with optimal second-generation regimens, such as GLE/PIB or a triple DAA combination, such as VOX/VEL/SOF,7 with data available for 26 individuals. A limitation of this study is that follow-up samples after the end of DAA treatment were not available. It has been shown that the persistence of viral resistance, especially NS5A resistance, depends on HCV genotype and subtype,6 and it would have been interesting to know how RASs evolve over time in unusual GT1 suptypes. Moreover, the cohort was not very large, and most of the patients had failed to receive treatment with first-generation DAA regimens. Still, this study clearly demonstrates that unusual GT 1 isolates, which are frequently found in sub-Saharan Africa, indeed are more difficult to treat and that broad use of suboptimal treatment regimens could have the potential to raise multi-drug resistant variants, escaping therapies that are state-of-the-art in these countries. In summary, both studies make an important contribution to the investigation of the role of HCV RASs for the retreatment of multiple DAA failures as well as of patients infected with unusual subtypes. Although RASs analyses are usually not available in countries without access to second-generation DAA regimens, the results of the current studies can be used to derive general rules regarding the selection of RASs in GT3 and the prevalence of RASs in unusual subtypes that can then be applied by physicians in less developed countries. In the absence of second-generation DAA retreatment regimens in resource-limited settings, in addition to a DAA drug class switch, extending the duration of retreatment or administering ribavirin should also be considered to achieve the global HCV elimination targets. Importantly, the underestimation of the prevalence of unusual HCV subtypes, global mobility, the potential transmission of multi-drug resistant variants in vulnerable groups such as PWIDs, the lack of protective vaccines combined with the fact that only 3 targets for DAA therapy have been finally approved, and both studies in combination should raise our general awareness that the outstanding efficiency of current DAA regimens could be reduced by the enormous genetic heterogeneity and genetic flexibility of HCV.