Real-world safety and efficacy of paritaprevir/ritonavir/ombitasvir plus dasabuvir ± ribavirin in patients with hepatitis C virus genotype 1 and advanced hepatic fibrosis or compensated cirrhosis: a multicenter pooled analysis

Paritaprevir/ritonavir, ombitasvir, and dasabuvir (PrOD) with or without ribavirin shows favorable results in hepatitis C virus genotype 1 (HCV-1) patients in terms of safety and efficacy, but real-world data remain limited for those with advanced hepatic fibrosis (fibrosis 3, F3) or compensated cirrhosis (F4). A total of 941 patients treated in four hospitals (the Keelung, the Linkuo, the Chiayi and the Kaohsiung Chang Gung Memorial Hospital) through a nationwide government-funded program in Taiwan were enrolled. Patients with HCV and advanced hepatic fibrosis or compensated cirrhosis received 12 weeks of PrOD in HCV-1b and 12 or 24 weeks of PrOD plus ribavirin therapy in HCV-1a without or with cirrhosis. Advanced hepatic fibrosis or compensated cirrhosis was confirmed by either ultrasonography, fibrosis index based on 4 factors (FIB-4) test, or transient elastography/acoustic radiation force impulse (ARFI). The safety and efficacy (sustained virologic response 12 weeks off therapy, SVR12) were evaluated. An SVR12 was achieved in 887 of 898 (98.8%) patients based on the per-protocol analysis (subjects receiving ≥1 dose of any study medication and HCV RNA data available at post-treatment week 12). Child-Pugh A6 (odds ratio: 0.168; 95% confidence interval (CI): 0.043–0.659, p = 0.011) was the only significant factor of poor SVR12. Fifty-four (5.7%) patients were withdrawn early from the treatment because of hepatic decompensation (n = 18, 1.9%) and other adverse reactions. Multivariate analyses identified old age (odds ratio: 1.062; 95% CI: 1.008–1.119, p = 0.024) and Child-Pugh A6 (odds ratio: 4.957; 95% CI: 1.691–14.528, p = 0.004) were significantly associated with hepatic decompensation. In conclusion, this large real-world cohort proved PrOD with or without ribavirin to be highly effective in chronic hepatitis C patients with advanced hepatic fibrosis or compensated cirrhosis. However, Child-Pugh A6 should be an exclusion criterion for first-line treatment in these patients.

Hepatitis C virus (HCV) infection is a major cause of chronic liver disease, affecting approximately 150 million people worldwide 1 . Chronic infection with HCV leads to progressive hepatic fibrosis and cirrhosis in around 20% of patients, and 10-20% of cirrhotic patients will develop hepatocellular carcinoma (HCC) within 5 years [1][2][3] . This implies that HCV eradication is very important in preventing disease progression and associated morbidity and mortality. This is also essential in reducing the future health care burden in society.
Combination of interferon (IFN) and ribavirin (RBV) was the previous standard treatment for chronic hepatitis C. However, lower virologic response rate and lots of side effects with poor adherence limited the application of treatment, and resulted in low sustained virologic response (SVR) rate of 40-50% among patients with HCV genotype 1 (HCV-1) infection 4,5 . The recent advent of direct-acting antiviral agent (DAA) therapy has been widely acknowledged as a revolution in the field of HCV infection. In clinical trials, IFN-free regimens using second generation DAA combinations yield SVR rates above 90% in HCV-1 infected patients [6][7][8] . Due to high virologic response rates even in difficult-to-treat subgroups such as cirrhosis, non-responders to prior therapy, and transplant recipients, and less side effects with better tolerance, DAA has become the first-line therapy of HCV in the latest guidelines.
Paritaprevir/ritonavir, ombitasvir, and dasabuvir (PrOD)-based regimens for HCV-1 infection have been approved by the Food and Drug Administration (FDA) in the United States (US) since 2016. In one meta-analysis, the SVR rate in PrOD-based regimens with or without RBV can reach up to 94-100% in HCV-1a or HCV-1b patients with and without cirrhosis 7 . However, warning of severe liver injury, hepatic decompensation and even mortality during this treatment were reported and informed by US FDA in 2015 9 . As far, most of the published studies were performed in the US and in Europe; Asian data is still limited 10,11 . In Asia, HCV-1 is the most prevalent form, accounting for 60-70% of HCV infection in Taiwan 12 . Since January 2017, PrOD-based therapies have been reimbursed in Taiwan for HCV-1 patients with advanced hepatic fibrosis (fibrosis 3, F3) or compensated cirrhosis (F4) through a nationwide government-funded program. Thus, we conducted this study to evaluate the safety and efficacy of PrOD-based therapies in a large real-world cohort of patient with advanced hepatic fibrosis or compensated cirrhosis.

Methods
Patients and treatments. This was a retrospective cohort study of PrOD-based therapies in patients with HCV-1a or HCV-1b and advanced hepatic fibrosis or compensated cirrhosis from four hospitals in Taiwan (the Keelung, the Linkou, the Chiayi, and the Kaohsiung Chang Gung Memorial Hospital). In Taiwan, patients with HCV and advanced hepatic fibrosis or compensated cirrhosis were treated with DAA via a nationwide government-funded program since 2017. The patients should have positive HCV antibody or detectable HCV RNA in serum for more than 6 months before treatment. Patient who had any evidence of hepatic decompensation or previous exposure to DAA before PrOD-based therapies should be excluded. The provided regimens were 12 weeks of PrOD in HCV-1b with or without cirrhosis, 12 weeks of PrOD plus RBV therapy (PrOD + RBV/12w) in HCV-1a without cirrhosis, and 24 weeks of PrOD plus RBV therapy (PrOD + RBV/24w) in HCV-1a with cirrhosis, respectively. The severity of fibrosis was confirmed by either ultrasonography, fibrosis index based on 4 factors (FIB-4) test, FibroScan (Echosens, Paris, France), or acoustic radiation force impulse (ARFI) (Siemens AG, Erlangen, Germany). Advanced hepatic fibrosis or compensated cirrhosis (Metavir F3-F4) was defined as FIB-4 test ≧3.25, FibroScan ≧9.5 Kpa, or ARFI ≧1.81 m/s. Patients with BCLC advanced or terminal stage and/or limited life expectancy were excluded. Informed consent was obtained from all patients prior to registration into the program. Demographic data including patient characteristics, treatment information, laboratory studies, and adverse reactions were recorded. For patients with hepatitis B virus (HBV) coinfection, HBV reactivation was defined as either an increase in HBV DNA level of ≥1 log10 IU/mL in patients with baseline detectable HBV DNA level or the HBV DNA level became detectable in patients with undetectable baseline HBV DNA level. Clinically significant hepatitis was defined as an ALT level of ≥3 times upper limit of normal 13 . This study was approved by the Research Ethics Committee of Chang Gung Memorial Hospital and was conducted in accordance with the principles of Declaration of Helsinki and the International Conference on Harmonization for Good Clinical Practice.
Outcomes. Virologic response (VR) was defined as HCV RNA less than the lower limit of quantification (LLOQ) at week 2, week 4, week 8, and week 12. The primary outcome was SVR 12 rate, which was defined as the proportion of patients with HCV RNA < LLOQ at post-treatment week 12 in per-protocol population (subjects receiving ≥1 dose of any study medication and HCV RNA data available at post-treatment week 12). The secondary outcome was the early withdrawal rate, which was defined as the percentage of patients who failed to complete the course of PrOD-based therapies because of adverse events, comorbidity, or other reasons. Hepatic decompensation was defined as the presence of clinical events (variceal hemorrhage, and/or ascites, and/or hepatic encephalopathy) or biochemical evidence of worsening liver function (significantly increased total bilirubin >3 mg/dL and/or prolonged prothrombin time ≥3 seconds).
Serum HCV RNA levels were determined by COBAS TaqMan HCV Test (TaqMan HCV; Roche Molecular Systems Inc., Branchburg, N.J., lower limit of detection: 15 IU/ml), or Abbott RealTime HCV assay (ART; Abbott Molecular, Des Plaines, IL; lower limit of detection: 12 IU/ml). Genotyping of HCV was performed by reverse hybridization assay (Inno-LiPA TM HCV II; Innogenetics N.V., Gent, Belgium) using the HCV-Amplicor products, or RealTime Genotyping II RUO assay (Abbott Molecular, Des Plaines, IL). Serum HBV DNA levels were measured using the COBAS AmpliPrep-COBAS TaqMan HBV test (CAP-CTM; Roche Molecular Systems, Inc., Branchburg, NJ, USA), with a detection limit of 15 IU/ml. Statistical methods. We used statistical software (SPSS 15.0) for data analysis. Continuous data were expressed as mean ± standard deviation, and categorical data were expressed as number (percentage). In www.nature.com/scientificreports www.nature.com/scientificreports/ comparing different subgroups, chi-square test or Fisher exact test was used for categorical parameters, and Student's t-test or Mann-Whitney U test was used for continuous parameters where appropriate. Factors related to SVR 12 and hepatic decompensation were analyzed with univariate and stepwise multivariate logistic regression analyses, and the results were presented as odds ratios (OR) with 95% confidence intervals (CI). Paired t test was performed to compare the FIB-4 test between baseline and post-treatment week 12. All statistical tests were 2-tailed, and a p-value of less than 0.05 was considered statistically significant.

Results
Baseline characteristics. A total of 941 patients were enrolled in this study. The baseline characteristics of the study population are shown in Table 1. Eight hundred eighty-nine (94.5%) patients were infected with HCV-1b, whereas 52 (5.5%) were infected with HCV-1a. The mean age of HCV-1b patients was significantly older than that in HCV-1a patients without cirrhosis (p < 0.001) or with cirrhosis (p < 0.001). One hundred thirty-one (14%) patients had concomitant HCC, including 79 (8%) without viable tumors and 52 (6%) with viable tumors (active HCC) before PrOD-based therapies. Of the 79 patients without viable tumors, therapy for HCC included 35 local ablation, 31 resection, 7 multiple treatment, 4 transcather arterial chemoembolization and 2 liver transplantation.
Among 67 patients with HBV coinfection, 4 had long-term use of nucleos(t)ide analogue (NA) before PrOD-based therapies. The rest of the 63 patients included 30 (48%) with undetectable HBV DNA level and 33 (52%) patients with detectable HBV DNA level at baseline. Six patients received concomitant PrOD and NA treatment (3 with HBV DNA level of ≥2000 IU/ml and 3 with HBV DNA level <2000 IU/ml). After excluding these patients with prior or concomitant NA treatment and 1 patient with early withdrawal, 8 (14%) patients met the virologic criteria for HBV reactivation (Fig. 3). Of them, two patients (3.6%) with HBV reactivation and clinically significant hepatitis received NA immediately and did not develop hepatic decompensation.

Discussion
This is one of the largest real-world cohort studies enrolling HCV-1 patients with advanced hepatic fibrosis or compensated cirrhosis receiving PrOD-based therapies. Such studies are of great clinical importance, since safety and effectiveness are often lower than in clinical trials which usually include only highly selected patients 14 . In particular, safety concerns have been raised in cirrhotic patients treated with PrOD-based therapies. In our study, the overall SVR 12 rate was 98.8% in HCV-1b, 100% in HCV-1a without cirrhosis, and 96.4% in HCV-1a with cirrhosis. These results were comparable to those in previous clinical trials [15][16][17][18] .
Previous real-world studies have demonstrated that PrOD-based therapies achieved SVR 12 rate of 95% to 100% in HCV-1b with or without cirrhosis, 95% to 100% in HCV-1a without cirrhosis, and 92.6% to 93% in HCV-1a with cirrhosis [19][20][21][22][23][24] . However, African-American patients were reported to have a lower SVR 12 rate compared to Caucasian patients (89.8% vs. 92.8%; p = 0.003) from an US cohort 23 . In contrast, our data concurring with other studies conducted in East Asia 10,11 confirmed the effectiveness of PrOD without RBV in Asian patients with HCV-1b infection, and PrOD with RBV in HCV-1a infection.
Our study consisted of a large database of homogenous patients, thus we could identify the possible predictive factors of SVR 12 to PrOD-based therapies despite the high SVR 12 rate. In our study, multivariate analysis showed the presence of Child-Pugh A6 had a negative impact on SVR 12 independently. Although reasons for the lower SVR rate in Child-Pugh A6 patients need to be further clarified, several mechanisms may be proposed. Advanced cirrhotic changes and shunting could result in the inadequacy of drug delivery [25][26][27] , and uptake and metabolism might be affected by shunting and poor liver function 28,29 . In addition, viral clearance might be impaired as a result of immune defects caused by more advanced cirrhosis 30,31 . Furthermore, Child-Pugh A6 cirrhosis significantly worsens tolerability of treatment leading to high rates of treatment discontinuation.
Previous studies have reported that the rate of adverse reaction, the rate of severe adverse reaction and the early withdrawal rate during the PrOD-based therapy was 42% to 91%, 1.7% to 10.3% and 0% to 6.3%, respectively [19][20][21][22][23][24] . In our cohort, 54 patients (5.7%) prematurely discontinued the treatment due to serious adverse events or comorbidity of the underlying disease. This result was not inferior to those in previous studies even though all of our patients had advanced hepatic fibrosis or compensated cirrhosis. Importantly, hepatic decompensation is the most serious complication during the PrOD-based therapies. Overall, the incidence of hepatic decompensation in our cohort was 2% (n = 18), but there was no mortality associated with PrOD-based therapies. This was relatively different from one Romania cohort showing that the mortality rate had reached as high as 35% in the patients presenting with hepatic decompensation 24 . The causes of hepatic decompensation were related to the baseline characteristics of the patient, including advanced liver disease or with a history of hepatic decompensation, and idiosyncratic drug-related liver injury. In our study, multivariate analysis showed that old age and the  www.nature.com/scientificreports www.nature.com/scientificreports/ presence of Child-Pugh A6 predicted hepatic decompensation independently. When the patients developed liver decompensation during the PrOD-based therapies, although some patients could still achieve viral eradication despite the shorter treatment duration, the frequency and severity of adverse events would increase significantly,   www.nature.com/scientificreports www.nature.com/scientificreports/ and even life-threatening 32 . Thus, immediate discontinuation of PrOD is suggested when hepatic decompensation develops, and retreatment by another DAA is strongly recommended if needed.
The potential drug-drug interaction (DDI) is another disadvantage of PrOD, which inhibits multiple isoenzymes of the cytochrome P-450 family (CYP2C19 and CYP3A4) and may therefore affect the metabolism of many drugs 7 . These interactions may lead to unsafe levels of concomitant medications or loss of PrOD efficacy. Liu et al. recently reported that the prevalence of contraindications with PrOD was 13.3% in Taiwanese patients 33 . Importantly, elderly patients were more likely to have potential DDI due to increasing use of concomitant drugs 33 . It is highly recommended to increase awareness of potential DDI during PrOD-based therapies, especially in patients with impaired liver function and old age.
Of note, the US FDA has mandated the addition of a boxed warning to remind practitioners of the potential HBV reactivation and its related complications during IFN-free DAAs for HCV 34 . Our study showed that two patients (3.6%) HBV-coinfected patients with HBV reactivation presented clinically significant hepatitis but did not develop hepatic decompensation. This result was in accordance with a recent prospective study 35 , showing that most patients with evidence of HBV reactivation were asymptomatic and could be managed with watchful surveillance or oral antiviral agents for HBV.
Our study enrolled a large number of patients, giving us a chance to study the safety and efficiency of PrOD-based therapies in the elderly patients with advanced liver disease. However, there were still some limitations. First, due to its retrospective design, the mild to moderate adverse events might be underreported. Second, our study did not analyze the resistance-associated substitution (RAS) for all patients at the start of and when treatment failed. However, based on the evidence that HCV-1 patients treated with PrOD achieved similarly high SVR rates, regardless of the presence or absence of baseline RASs, current guidelines advised against routine check of baseline RASs in these patients 36 . Third, the number of patients in the HCV-1a subgroup was relatively low in this area. While our study appeared to be the largest cohort enrolling HCV-1a patients in Asia, since the prevalence of HCV-1a was low in this area.
In conclusion, our large real-world cohort suggests that PrOD with or without RBV is highly effective in Asian patients with advanced hepatic fibrosis or compensated cirrhosis, achieving an SVR 12 rate of 98.8%. Child-Pugh A6 not only correlated with poor SVR 12 rate but also predicted hepatic decompensation during PrOD-based therapies. Therefore, this should be an exclusion criterion for first-line treatment in these patients.