Sofosbuvir/velpatasvir/voxilaprevir for patients with chronic hepatitis C virus infection previously treated with NS5A direct-acting antivirals: a real-world multicenter cohort in Taiwan

Real-world data are scarce about the effectiveness and safety of sofosbuvir/velpatasvir/voxilaprevir (SOF/VEL/VOX) for retreating East Asian patients with hepatitis C virus (HCV) infection who previously received NS5A direct-acting antivirals (DAAs). We conducted a multicenter study to assess the performance of SOF/VEL/VOX in patients who were not responsive to prior NS5A inhibitors in Taiwan. Between September 2021 and May 2022, 107 patients who failed NS5A inhibitor-containing DAAs with SOF/VEL/VOX salvage therapy for 12 weeks were included at 16 academic centers. The sustained virologic response at off-treatment week 12 (SVR12) was assessed in the evaluable (EP) and per-protocol (PP) populations. The safety profiles were also reported. All patients completed 12 weeks of treatment and achieved an end-of-treatment virologic response. The SVR12 rates were 97.2% (95% confidence interval (CI) 92.1–99.0%) and 100% (95% CI 96.4–100%) in EP and PP populations. Three (2.8%) patients were lost to off-treatment follow-up and did not meet SVR12 in the EP population. No baseline factors predicted SVR12. Two (1.9%) not-fatal serious adverse events (AE) occurred but were unrelated to SOF/VEL/VOX. Sixteen (15.0%) had grade 2 total bilirubin elevation, and three (2.8%) had grade 2 alanine transaminase (ALT) elevation. Thirteen (81.3%) of the 16 patients with grade 2 total bilirubin elevation had unconjugated hyperbilirubinemia. The estimated glomerular filtration rates (eGFR) were comparable between baseline and SVR12, regardless of baseline renal reserve. SOF/VEL/VOX is highly efficacious and well-tolerated for East Asian HCV patients previously treated with NS5A inhibitor-containing DAAs. The study was not a drug trial. There was no need for clinical trial registration.


Introduction
Although the global prevalence of chronic hepatitis C virus (HCV) infection has decreased from 0.9 to 0.7% between 2015 and 2020 following the widespread use of direct-acting antivirals (DAAs), it remains one of the major causes of cirrhosis, hepatocellular carcinoma (HCC), hepatic decompensation and liver transplantation [1][2][3]. Currently, the sustained virologic response (SVR) rates with DAAs for HCV are generally more than 95% in DAA-naïve patients, regardless of clinical trials or real-world studies [4][5][6][7][8][9][10][11][12][13][14][15]. While the treatment response for HCV in the era of interferon (IFN)-free DAAs is satisfactory, a small proportion of patients fail to respond to DAAs, resulting in the selection of resistance-associated substitutions (RASs) at the HCV non-structural (NS)3, NS5A, or NS5B regions that complicate the retreatment strategies. It is particularly relevant to patients exposed to NS5A inhibitors because (1) NS5A inhibitor is the backbone among most licensed DAA regimens; (2) the persistence of NS5A RASs is much longer than other classes of RASs following treatment failures [16]. Therefore, an effective means to eradicate HCV is needed in patients who are not responsive to previous NS5A inhibitors to meet the World Health Organization's (WHO) target for HCV elimination by 2030. Sofosbuvir/velpatasvir/voxilaprevir (SOF/VEL/VOX) is a fixed-dose combination (FDC) of HCV NS5B, NS5A, and NS3 inhibitors, which exhibits pangenotypic potency against HCV by once-daily, oral administration [17]. SOF/VEL/ VOX is indicated to retreat patients with compensated HCV disease previously treated with a DAA regimen containing an NS5A inhibitor (HCV genotype [GT] [1][2][3][4][5][6] or SOF without an NS5A inhibitor (HCV GT 1a or 3). To date, SOF/ VEL/VOX has been recommended by international guidelines as the first-line salvage therapy for DAA-experienced patients with HCV. [18][19][20] The phase 3 POLARIS-1 trial evaluated the performance of SOF/VEL/VOX for 12 weeks in patients with compensated HCV GT 1-6 infection previously treated with NS5A inhibitor-containing DAAs, demonstrating that the SVR 12 rates were 96% [21]. Furthermore, the SVR 12 rate of SOF/ VEL/VOX for 12 weeks was 97% in patients allocated to the deferred arm in the POLARIS-1 trial, which corroborated the excellent response in managing such patients [22]. Baseline HCV GT, cirrhosis, and RAS distribution did not affect the treatment responses [21,22]. The real-world studies from Western countries, including the U.S., Canada, Italy, Spain, and Australia, have reported the SVR 12 rates of SOF/ VEL/VOX for DAA-experienced patients with HCV ranged from 90 to 100%, and most patients tolerated treatment well [23][24][25][26][27][28][29]. In contrast, data regarding the performance of SOF/ VEL/VOX in the Asian population are limited. Only two small-scaled studies from Taiwan and Singapore recruited two and twenty-five NA5A inhibitor-experienced patients with HCV, in which the SVR 12 rates were 100% and 96%, respectively. [30,31] Furthermore, both studies recruited a significant portion of patients treated outside the label recommendation, including coadministration of ribavirin (RBV) and decompensated cirrhosis, making the clinical performance of SOF/VEL/VOX elusive in NS5A-inhibitor experienced Asian patients with HCV. Based on the abovementioned concerns, we conducted a multicenter prospective cohort study in Taiwan to confirm the effectiveness and tolerance of SOF/VEL/VOX for patients with HCV previously treated with NS5A inhibitor-containing DAAs.

Patients
Between September 2021 and April 2022, patients with chronic HCV infection and compensated liver disease aged ≥ 20 years who failed to respond to previous IFN-free DAAs containing an NS5A inhibitor were prospectively recruited at 16 academic centers in Taiwan. Chronic HCV infection was defined as the presence of HCV antibody (anti-HCV; Abbott HCV EIA 2.0, Abbott Laboratories, Abbott Park, Illinois, USA) and quantifiable serum HCV RNA level (Cobas TaqMan HCV Test v2.0, Roche Diagnostics GmbH, Mannheim, Germany, lower limit of quantification [LLOQ]: 15 IU/mL) for ≥ 6 months. Compensated liver disease was defined as patients without cirrhosis or with compensated cirrhosis (Child-Pugh A). Patients were excluded if they had active HCC or decompensated cirrhosis (Child-Pugh B or C).

Study design
We inquired about patient demographic data, including age, sex, previous anti-HCV treatment regimens, most recent virologic response to DAAs, history of HCC, diabetes mellitus (DM), and arterial hypertension (HTN). In addition, all participants underwent laboratory testing for hemogram, international normalized ratio (INR), serum albumin, total bilirubin (upper limit of normal [ULN]: 1.0 mg/dL), direct bilirubin, aspartate aminotransferase (AST), alanine aminotransferase (ALT) (ULN: 30 U/L for males and 19 U/L for females), estimated glomerular filtration rate (eGFR) as calculated using Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation, anti-HCV, hepatitis B surface antigen (HBsAg) (Abbott Architect HBsAg qualitative assay, Abbott Laboratories, Abbott Park, Illinois, USA) to confirm hepatitis B virus (HBV) coinfection, anti-HIV (Abbott Architect HIV Ag/Ab Combo, Abbott Laboratories, Abbott Park, Illinois, USA), to confirm HIV coinfection HCV RNA, and HCV genotype (Roche Cobas HCV GT, Roche Diagnostics GmbH, Mannheim, Germany, or Abbott RealTime HCV Genotype II, Abbott Laboratories, Abbott Park, Illinois, USA) [32][33][34]. The severity of hepatic fibrosis was graded by the fibrosis index based on four parameters (FIB-4) with cut-off values of < 1.45, 1.45-3.25, and > 3.25 [35]. Hepatic imaging studies, including ultrasonography, computed tomography or magnetic resonance imaging, were performed before treatment to detect active HCC or decompensated cirrhosis. Baseline HBV deoxyribonucleic acid (DNA) (Cobas AmpliPrep/Cobas Taqman HBV DNA test, v.2.0, LLOQ): 10 IU/mL) was checked in patients with HBV coinfection. Furthermore, HIV RNA (Cobas AmpliPrep TaqMan HIV-1 Test v2.0, Roche Diagnostics GmbH, Mannheim, Germany, LLOQ: 20 copies/mL) was checked in patients with HIV coinfection. Baseline RASs at the HCV NS3 and NS5A regions were analyzed by population sequencing with a cut-off detection level of 15% [36]. RAS analysis was performed at off-treatment week 12 if participants had virologic failures to SOF/VEL/VOX. Patients received SOF/VEL/VOX (Vosevi®, FDC 400/100/100 mg per tablet, Gilead Sciences, Carrigtohill, County Cork, Ireland) 1 tablet once daily with food for 12 weeks. Coadministration of RBV was permitted based on physicians' discretion. Before initiating SOF/VEL/VOX, the investigators performed drug-drug interaction (DDI) assessment for all comedications through the pre-defined study checklist [37]. If the comedications were contraindicated for use during SOF/VEL/VOX treatment, the investigators stopped the comedications or switched to other non-contraindicated agents. All patients underwent outpatient visits at on-treatment weeks 4 and 12 and off-treatment week 12. Serum HCV RNA levels were assessed at on-treatment week 12 and off-treatment week 12. Total bilirubin, direct bilirubin, AST, and ALT were assessed at on-treatment weeks 4 and 12. Hemogram, INR, albumin, total bilirubin, direct bilirubin, AST, ALT, eGFR, and hepatic imaging studies were assessed at off-treatment week 12. In patients with HBV coinfection, the serum HBV DNA levels were determined at on-treatment week 12 and off-treatment week 12. Additional HBV DNA testing was allowed during SOF/VEL/VOX treatment when indicated.

Effectiveness
We assessed the end-of-treatment virologic response and SVR at on-treatment week 12 and off-treatment week 12. Patients with serum HCV RNA levels > LLOQ at off-treatment week 12 (virologic failure) or did not have SVR 12 data (non-virologic failure) were considered failures to achieve SVR 12 . We defined two SVR 12 endpoints, which included the evaluable population (EP) in patients receiving at least one dose of SOF/VEL/VOX and the per-protocol population (PP) in patients undergoing virologic testing at off-treatment week 12 to confirm treatment responses.

Tolerance
At each outpatient visit, the investigators evaluated the constitutional adverse events (AEs), laboratory AEs, and serious AEs. Furthermore, they also assessed the causal relationship between SOF/VEL/VOX and various AEs, including early treatment discontinuation. We reported the proportion of patients with ≥ grade 2 total bilirubin or ALT elevation according to Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. The changes in eGFR from baseline to the off-treatment week 12 were also assessed. HBV reactivation was defined as an on-treatment ≥ 2 log 10 increase in HBV DNA level compared to the baseline level. HBV-associated hepatitis was defined as the presence of HBV reactivation in combination with ≥ grade 2 ALT elevation [38].

Statistical analysis
We used the Statistical Program for Social Sciences (SPSS Statistics Version 23.0, IBM Corp., Armonk, New York, USA) for all statistical analyses. Baseline characteristics were shown in median (interquartile range, IQR) and numbers (percentages) when appropriate. The virologic responses at on-treatment week 12 and off-treatment week 12 were shown in numbers (percentages) with a 95% confidence interval (CI). We demonstrated subgroup analyses of SVR 12 in number (percentage) with 95% CI according to the EP population with forest plots. The constitutional, laboratory, and serious AEs were shown in numbers (percentages). The eGFR at baseline and off-treatment week 12 was shown with box plots. The changes in eGFR between baseline and off-treatment week 12 in the overall population, patients with an eGFR ≥ 60 mL/min/1.73m 2 , and patients with an eGFR < 60 mL/min/1.73m 2 were compared with Wilcoxon signed-rank test. All statistical analyses were two-tailed, and a p-value of < 0.05 was considered statistically significant.

Discussion
Although the first-line DAA treatment is highly effective with an overall SVR 12 rate of > 95%, a minority of patients with HCV are not responsive to such treatment, mandating the need for salvage therapy to manage these patients in whom multiple RASs may emerge. Our results revealed that the SVR 12 rates were 97.2% and 100% in the EP and PP populations, comparable to the Western studies and metaanalyses reporting the SVR 12 rates of 90% to 100% [21][22][23][24][25][26][27][28][29]39]. Our results corroborated the SOF/VEL/VOX's excellent effectiveness in managing patients previously treated with NS5A inhibitors, regardless of ethnicity [30,31].
Our study revealed that no specific patient, viral, or treatment factors predicted treatment responses. Although 65% of our patients had baseline RASs at various NS3 and NS5A loci, none developed virologic failures after Fig. 3 Sustained virologic response (SVR) rates by evaluable population (EP) analysis according to viral and treatment factors. The position of the square indicates the sustained virologic response rate at off-treatment week 12 week (SVR 12 ) in each subgroup; the horizontal lines indicate 95% confidence intervals (Cis). The dotted vertical line represents the overall SVR 12 rate. Patients previously treated with SOF/ VEL plus RBV were categorized into the subgroup of SOF/ VEL. SOF, sofosbuvir; VEL, velpatasvir; VOX, voxilaprevir; DCV, daclatasvir; ASV, asunaprevir; PrOD, paritaprevir/ritonavir/ombitasvir plus dasabuvir; EBR, elbasvir; GZR, grazoprevir; LDV, ledipasvir; GLE, glecaprevir; PIB, pibrentasvir; RAS, resistance-associated substitution; NS, non-structural protein; HCV, hepatitis C virus; RNA, ribonucleic acid SOF/VEL/VOX, implying the combination of potent HCV NS3, NS5A, and NS5B agents can overcome complicated RAS profiles. However, HCV GT3 infection and previous null-response tended to have numerically lower response rates [36,[40][41][42]. Although some real-world studies indicated that patients with HCV GT3 infection, cirrhosis, previous SOF/VEL treatment, and multiple RASs might adversely affect the SVR 12 rate, the POLARIS-1 trial and the other real-world studies, including ours, did not show such association [21][22][23][24][25][26][27][28][29]42]. Onofrio et al. indicated that the SVR 12 rate of SOF/VEL/VOX might be lower if patients presented at least two unfavorable factors, such as HCV GT3 infection, cirrhosis, multiple RASs, previous SOF/ VEL treatment, and liver transplantation [26]. Because of unfavorable factors, the Italian and Canadian real-world studies reported that about 20% of patients received RBV in combination with SOF/VEL/VOX based on physicians' discretion. However, patients with combination therapy did not confer a better SVR 12 rate than those with SOF/VEL/ VOX alone [24,26]. While the EASL guidelines recommend SOF/VEL/VOX plus RBV for "difficult-to-cure" patients to intensify the retreatment responses, the AASLD guidelines do not endorse the universal combination except for cirrhotic patients with HCV GT3 infection [18,19]. Because the coadministration of RBV remains controversial for NS5A inhibitor-experienced patients, we herein propose to use SOF/VEL/VOX alone for 12 weeks to manage such patients based on the current label recommendation.
In line with the POLARIS-1 trial and real-world studies, we demonstrated that the tolerance of SOF/VEL/VOX was excellent, with the reported rates of AEs and serious AE of only 16.8% and 2.9%. There were no deaths or early SOF/VEL/VOX discontinuations due to AEs. Regarding  [15,45]. Furthermore, the APASL guidelines recommend the safe use of SOF-based DAAs without compromising efficacy in patients with renal impairment [46]. In patients with HBV coinfection, there was no HBV reactivation or HBV-related hepatitis during SOF/VEL/VOX. However, periodic surveillance of ALT and HBV DNA levels is needed to detect HBV reactivation/clinical hepatitis early and initiate NUC promptly [38]. Because more than half of our patients were aged more than 50 and may take comedications for comorbidities, careful DDI checks before SOF/VEL/VOX treatment are vital to secure on-treatment safety. While a large proportion of patients with chronic HCV infection reside in Asian-Pacific regions, our study provides encouraging results of SOF/VEL/VOX as an excellent rescue regimen if patients fail to respond to the first-line NS5Acontaining DAA regimens. By mass screening, efficient linkto-care, scale-up treatment uptake, and implementation of salvage treatment, most Asian-Pacific patients can speed up the cure to meet the WHO's HCV elimination goal by 2030.
The strengths of this study include the multicenter prospective cohort that recruited a sizable number of East Asian HCV patients previously treated with NS5A inhibitors and the clearly defined clinical as well as laboratory assessments to ensure outcome estimation. However, our study has several limitations. First, the number of patients with HCV GT3 infection, particularly those with cirrhosis, was relatively small. Therefore, we cannot assess the effects of these unfavorable factors on the effectiveness of SOF/VEL/ VOX. Second, although several studies reported that SOF/ VEL/VOX could be used in patients with Child-Pugh B or C cirrhosis, our study did not include patients with decompensated cirrhosis because SOF/VEL/VOX is not recommended for patients with decompensated cirrhosis [23,27,31,47].
In conclusion, our multicenter, prospective, real-world study demonstrates that SOF/VEL/VOX for 12 weeks is highly efficacious and well-tolerated for East Asian HCV patients with compensated liver diseases previously treated with NS5A inhibitor-containing DAAs.