THU-117-Evaluation of risk factors associated with failure to a first-line NS5A-containing regimen in HCV-infected patients naive to direct acting antivirals: Particular focus on natural resistance

Background and aims: This study aimed to evaluate the presence of natural resistance-associated- substitutions (RASs) and other pre-treatment risk-factors for failure in a large group of HCV-infected patients (pts) naive to direct-acting-antivirals (DAA) with an available outcome after their first-line NS5A inhibitor-containing regimen in Italy. Method: RASs in NS3/NS5A/NS5B (N = 1685/1497/1175) were analysed in 1947 DAA-naïve pts. Of them, 705 had an available outcome after a first-line NS5A-containing regimen recommended by the 2016/18 guidelines, with a baseline (BL) NS5A-test. HCV Sanger-sequencing was performed by home-made protocols. Potential differences between the sustained-virological-response (SVR) and virological-failure (VF) group were evaluated by Fisher’s exact test. A multivariable logistic-regression analysis was performed to define risk-factors associated to treatment-response. Results: Overall, 579/1947 (29.7%) pts showed at least one natural RASs, particularly NS5A-RAS was observed in 18.9% of pts. 705 pts (GT1a/b/g[200/214/1]-GT2a/c[84]-3a[141]-4a/d[65]) had an available outcome (656 with a SVR and 49 with a VF) after the following recommended NS5A- containing regimen: daclatasvir (DCV)/ledipasvir (LDV)/velpatasvir (VEL)+sofosbuvir (SOF)±ribavirin (RBV) (N = 125/130/161), 3D/2D (paritaprevir/ritonavir+ombitasvir ± dasabuvir)±RBV (N = 125/44), grazoprevir (GZR)+elbasvir (EBR)±RBV (N = 70), glecaprevir+pibrentasvir (G/P) (N = 50). By analysing retrospectively the BL samples, a higher prevalence of natural NS5A-RASs was observed before treatment in DAA-failures (18/49, 36.7%) vs SVR-pts (94/656, 14.3%; P < 0.001). Notably, ≥ 2 risk factors for failure were more frequently observed at BL among pts who experienced a VF to a DAA treatment (37/49, 75.5%) compared to those achieving SVR (295/656, 45.0%, P < 0.001). By multivariable logistic-regression high HCV-RNA, natural RAS, cirrhosis, previous IFN-failure were negatively associated with SVR (see figure). Interestingly, all 32 GT1-3 pts treated with G/P achieved SVR, with the exception of 1 GT3, who had a breakthrough and had at BL the NS5A RAS A30K and HCV-RNA > 800.000 IU/ml. All others were without (or only 1) risk-factor: notably none of them showed BL RASs regimen-related. Conclusion: The presence of specific pre-treatment risk-factor, such as RAS regimen-related, BL HCV-RNA > 800.000 IU/ml, cirrhosis and previous IFN-failure were associated with virological failure for some specific regimens and GTs.