Grazoprevir-Elbasvir Combination Therapy for Treatment-Naive Cirrhotic and Noncirrhotic Patients With Chronic Hepatitis C Virus Genotype 1, 4, or 6 Infection: A Randomized Trial
Download the PDF here
Ann Intern Med. July 7 2015
Stefan Zeuzem, MD; Reem Ghalib, MD; K. Rajender Reddy, MD; Paul J. Pockros, MD; Ziv Ben Ari, MD; Yue Zhao, PhD; Deborah D. Brown, BS; Shuyan Wan, PhD; Mark J. DiNubile, MD; Bach-Yen Nguyen, MD; Michael N. Robertson, MD; Janice Wahl, MD; Eliav Barr, MD; and Joan R. Butterton, MD
Background: Novel interferon- and ribavirin-free regimens are needed to treat hepatitis C virus (HCV) infection.
Objective: To evaluate the safety and efficacy of grazoprevir (NS3/4A protease inhibitor) and elbasvir (NS5A inhibitor) in treatment-naive patients.
Design: Randomized, blinded, placebo-controlled trial. (ClinicalTrials.gov: NCT02105467)
Setting: 60 centers in the United States, Europe, Australia, Scandinavia, and Asia.
Patients: Cirrhotic and noncirrhotic treatment-naive adults with genotype 1, 4, or 6 infection.
Intervention: Oral, once-daily, fixed-dose grazoprevir 100 mg/elbasvir 50 mg for 12 weeks, stratified by fibrosis and genotype. Patients were randomly assigned 3:1 to immediate or deferred therapy.
Measurements: Proportion of patients in the immediate-treatment group achieving unquantifiable HCV RNA 12 weeks after treatment (SVR12); adverse events in both groups.
Results: Among 421 participants, 194 (46%) were women, 157 (37%) were nonwhite, 382 (91%) had genotype 1 infection, and 92 (22%) had cirrhosis. Of 316 patients receiving immediate treatment, 299 of 316 (95% [95% CI, 92% to 97%]) achieved SVR12, including 144 of 157 (92% [CI, 86% to 96%]) with genotype 1a, 129 of 131 (99% [CI, 95% to 100%]) with genotype 1b, 18 of 18 (100% [CI, 82% to 100%]) with genotype 4, 8 of 10 (80% [CI, 44% to 98%]) with genotype 6, 68 of 70 (97% [CI, 90% to 100%]) with cirrhosis, and 231 of 246 (94% [CI, 90% to 97%]) without cirrhosis. Virologic failure occurred in 13 patients (4%), including 1 case of breakthrough infection and 12 relapses, and was associated with baseline NS5A polymorphisms and emergent NS3 or NS5A variants or both. Serious adverse events occurred in 9 (2.8%) and 3 (2.9%) patients in the active and placebo groups, respectively (difference <0.05 percentage point [CI, -5.4 to 3.1 percentage points]); none were considered drug related. The most common adverse events in the active group were headache (17%), fatigue (16%), and nausea (9%).
Limitation: The study lacked an active-comparator control group and included relatively few genotype 4 and 6 infections.
Conclusion: Grazoprevir-elbasvir achieved high SVR12 rates in treatment-naive cirrhotic and noncirrhotic patients with genotype 1, 4, or 6 infection. This once-daily, all-oral, fixed-combination regimen represents a potent new therapeutic option for chronic HCV infection.
· Various oral interferon- and ribavirin-free regimens are becoming available to treat chronic hepatitis C virus (HCV) infection. A grazoprevir-elbasvir combination regimen has shown promise in phase 2 trials.
· This phase 3 trial found a once-daily grazoprevir-elbasvir regimen to be effective and well-tolerated in patients with HCV genotype 1, 4, or 6 infection. Outcomes were similar in patients with and without cirrhosis.
· The study did not include an active comparator, so how this regimen compares with others is unknown.
· Grazoprevir-elbasvir represents a new therapeutic option for chronic HCV infection.
Chronic hepatitis C virus (HCV) infection remains a growing cause of cirrhosis, hepatocellular carcinoma, hepatic decompensation, and liver transplantation (1). Effective therapy for HCV infection diminishes long-term liver-related complications and mortality (2). Convenient, oral, direct-acting antiviral regimens are being investigated for chronic HCV infection (3).
Grazoprevir is an NS3/4A protease inhibitor that has high potency in vitro against HCV genotype (GT) 1, GT2, GT4, GT5, and GT6 but is less active against GT3 (4). Grazoprevir retains substantial activity against resistance-associated variants (RAVs) commonly detected after failed therapy with first-generation protease inhibitors (4-5). Elbasvir is an NS5A inhibitor active against GT1, GT2a, GT3, GT4, GT5, and GT6, even in the presence of RAVs associated with failure of other NS5A inhibitors, such as daclatasvir and ledipasvir (6-7).
Grazoprevir-elbasvir has been evaluated in an extensive phase 2 clinical development program (5, 8-10). The C-WORTHY study indicated that grazoprevir-elbasvir with or without ribavirin for 12 weeks provided efficacious and well-tolerated therapy for monoinfected and HIV-co-infected patients, treatment-naive and treatment-experienced patients, and noncirrhotic and cirrhotic patients (9-10).
The objective of the phase 3 C-EDGE Treatment-Naive trial was to evaluate the efficacy and safety profile of a once-daily, fixed-dose, oral, 12-week regimen of grazoprevir-elbasvir without interferon or ribavirin in treatment-naive monoinfected patients with and without cirrhosis and with GT1, GT4, or GT6 infection.
The C-EDGE Treatment-Naive study was an international, randomized, blinded, placebo-controlled, parallel-group trial of a fixed-dose combination of grazoprevir 100 mg/elbasvir 50 mg for treatment-naive cirrhotic and noncirrhotic patients with chronic HCV GT1, GT4, or GT6 infections. A historical SVR12 rate was used as the comparator for efficacy. A deferred-treatment group was included as a concurrent placebo group to assess safety; after the follow-up period, placebo recipients received open-label grazoprevir-elbasvir so that all participants would receive therapy during the study.
Recruitment of Study Participants
Patients were recruited from general medical clinics at 60 trial centers: 4 in Australia, 4 in the Czech Republic, 5 in France, 5 in Germany, 5 in Israel, 3 in Puerto Rico, 3 in South Korea, 4 in Sweden, 3 in Taiwan, and 24 in the United States. Patients who fulfilled inclusion criteria were asked to participate in the trial. Selected clinical sites were experienced in the management and care of HCV-infected patients, with a history of successful study conduct and the capability for rapid enrollment. Sites were chosen to allow a wide geographic distribution and to ensure that requirements for minority representation, enrollment of patients with cirrhosis, and genotype distribution were met.
Adults (aged >18 years) with HCV RNA levels greater than 104 IU/mL were eligible. Hepatic fibrosis was staged by biopsy or noninvasive assessment (Appendix 1) (11). Exclusion criteria were decompensated liver disease, hepatocellular carcinoma, HIV or hepatitis B virus co-infection, uncontrolled diabetes mellitus (hemoglobin A1c level >10%), elevated prothrombin time unrelated to anticoagulation, creatinine clearance less than 50 mL/min, hemoglobin level less than 95 g/L, thrombocytopenia (platelet count <50 x 109 cells/L), aminotransferase levels more than 10 times the upper limit of normal, or hypoalbuminemia (albumin level <30 g/L). Enrollment was constrained to meet the following targets: 20% of the participants having cirrhosis and 15% having GT4 or GT6 infection. All participants provided written informed consent. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. Independent ethics committees reviewed and approved the protocol and applicable amendments for each institution.
Randomization and Masking
After stratification by presence or absence of cirrhosis and GT1, GT4, or GT6, patients were randomly assigned in a 3:1 ratio to receive immediate or deferred therapy with grazoprevir-elbasvir through a central interactive voice-response system according to a computer-generated random allocation schedule. Patients took 1 fixed-dose combination tablet of grazoprevir-elbasvir (immediate-treatment group) or matching placebo (deferred-treatment group) once daily at approximately the same time, without regard to food, for 12 weeks. Patients, clinical site, and sponsor personnel were blinded to treatment assignment (except for a separate unblinded medical team that monitored virologic failures and serious adverse events). Four weeks after completion of therapy, treatment allocation was unblinded, and patients in the deferred-treatment group then received open-label grazoprevir-elbasvir for 12 weeks. All patients were to be followed for 24 weeks after cessation of active study therapy (Figure 1).
This report describes the efficacy among patients enrolled in the immediate-treatment group through 12 weeks after treatment and the safety findings among patients enrolled in both groups through 14 days after the end of therapy in the initial treatment period. Efficacy and safety results for both groups through follow-up week 24 are still being collected and will be presented in a future report. The primary efficacy outcome variable was the proportion of patients in the immediate-treatment group achieving unquantifiable HCV RNA 12 weeks after the end of study treatment (SVR12). Virologic failures encompassed breakthrough viremia (confirmed HCV RNA level at or above the lower limit of quantification [LLOQ] during treatment after previously being below the LLOQ) and relapse (confirmed HCV RNA level at or above the LLOQ subsequent to cessation of study therapy after becoming undetectable at the end of treatment).
Viral and Resistance Assays
Plasma HCV RNA levels were measured by the COBAS AmpliPrep/COBAS TaqMan HCV test, version 2.0 (Roche Molecular Diagnostics, Branchburg, NJ), with an LLOQ of 15 IU/mL. Specimens for viral load measurements were collected at screening; baseline; treatment weeks 4, 8, and 12; and follow-up weeks 4, 12, and 24.
Circulating viral quasi-species at baseline or at the time of virologic failure underwent population sequencing with a detection limit for variants of approximately 25% prevalence (12). The complete NS3 and NS5A genes were amplified from samples with RNA levels of 1000 IU/mL or greater by using reverse transcription polymerase chain reaction (5, 12-13). Resultant amino acid sequences were compared with wild-type GT1a (H77; accession number NC004102), GT1b (Con1; AJ238799), GT4a (ED43; GU814265), or GT6a (EUHK2; Y12083) reference sequences.
To assess the effect of baseline NS3 variants, specific amino acid loci prone to selection by early-generation NS3/4A protease inhibitors (positions 36, 54, 55, 56, 80, 107, 122, 132, 155, 156, 158, 168, 170, and 175) were studied in replicon cell lines encoding mutations in a GT1a backbone (5, 14). These substitutions were categorized according to whether they conferred a greater than 5-fold reduced susceptibility to grazoprevir. Likewise, to assess the effect of baseline NS5A variants, amino acid loci selected by NS5A inhibitors (positions 28, 30, 31, 58, and 93) were categorized according to whether they conferred a greater than 5-fold reduced susceptibility to elbasvir in the replicon assay.
The C-EDGE Treatment-Naive study was designed to randomly assign approximately 400 patients, with 300 patients in the immediate-treatment group and 100 patients in the deferred-treatment group (which served as the placebo control group for the first 12 weeks). After a 4-week follow-up period, placebo recipients were unblinded at study week 16 and received open-label grazoprevir-elbasvir. The primary efficacy hypothesis exclusively applied to patients in the immediate-treatment group. Assuming a response rate of 85% or greater, the study had more than 99% power to demonstrate an SVR12 rate superior to the reference rate of 73% at an overall 1-sided α value of 0.025. The historical reference rate of 73% was derived from phase 3 trials of simeprevir/peginterferon + ribavirin in treatment-naive monoinfected patients, after adjustment for the expected proportion of cirrhotic patients and the anticipated improved tolerability with an interferon-free regimen (Appendix 1) (15-16).
The primary efficacy and safety analyses were performed on the full data set, which included all patients receiving at least 1 dose of the study treatment. The primary efficacy end point was prespecified as the proportion of patients with an HCV RNA level below the LLOQ 12 weeks after the end of treatment (SVR12) (17). Missing outcome data were imputed as failures unless the values immediately before and after the missing result were both successes, in which case the absent value was imputed as a success. The 95% CIs were computed by the conservative Clopper-Pearson method, which provides an exact CI based on the binomial distribution (18). A P value was calculated to support the primary hypothesis based on a 1-sided exact test. Because only a single primary efficacy hypothesis was stipulated by protocol, no multiplicity adjustments were needed. Secondary efficacy analyses were intended as supportive, hypothesis-generating estimations.
Serious adverse events occurring at any time during the study, and other adverse events up until 14 days after cessation of treatment, were to be reported. Investigators assessed the relationship of each adverse event to study therapy; adverse events were counted as drug related if judged as at least possibly related to the study drug. Frequencies of adverse events with corresponding 95% CIs were calculated for the grazoprevir-elbasvir and placebo groups during the treatment period through the first 14 days of posttherapy follow-up (see Appendix 1 for power calculations for differences in adverse-event frequencies between groups).
Role of the Funding Source
Merck is developing grazoprevir-elbasvir for treatment of HCV infection. The company contributed to the trial design, study execution and management, data collection, statistical analyses, and drafting of this report. The sponsor reviewed a penultimate version of the paper. All coauthors had access to the study data, approved the final report, and assume full responsibility for the veracity of the data and analyses. The lead/corresponding author had full access to all data and had final responsibility for the decision to submit the manuscript for publication.
Patient Accounting and Baseline Characteristics
Study enrollment began on 25 June 2014, and data collection for the primary efficacy end point concluded 4 March 2015. The study involved 60 centers in 9 countries on 4 continents. A total of 421 (90%) of 469 screened patients was enrolled and randomly assigned; 316 and 105 patients received grazoprevir-elbasvir or placebo, respectively, during the initial treatment period (Figure 2). Overall, the median age was 54 years (range, 20 to 78 years); 46% of patients were women, 37% were nonwhite, and 91% had GT1 infections (50% of which were GT1a). Most patients enrolled from Australia, Sweden, and the United States had GT1a infections, whereas most patients from the Czech Republic and Taiwan and all patients from Israel and the Republic of Korea had GT1b infections (Appendix Table 1). Most of the GT4-infected patients were enrolled from France and the United States, whereas most of the GT6-infected patients were enrolled from Taiwan and the United States. Baseline HCV RNA levels were greater than 800 000 IU/mL in 68% of patients. Of the 92 (22%) patients with cirrhosis, the diagnosis was biopsy proven in 26 (28%). The median platelet count and albumin level among cirrhotic patients were 124 x 109 cells/L (range, 49 to 298) and 41 g/L (range, 30 to 52), respectively (Table 1).
Viremia was rapidly suppressed with therapy (Appendix Table 2). Of the 316 patients in the immediate-treatment group, 299 (95% [CI, 92% to 97%]) achieved SVR12. SVR12 rates were 92% (144 of 157 [CI, 86% to 96%]) in patients with GT1a infection, 99% (129 of 131 [CI, 95% to 100%]) in those with GT1b, 100% (18 of 18 [CI, 82% to 100%]) in those with GT4, and 80% (8 of 10 [CI, 44% to 98%]) in those with GT6 (Figure 3). SVR12 was achieved in 97% (68 of 70 [CI, 90% to 100%]) of cirrhotic patients and 94% (231 of 246 [CI, 90% to 97%]) of noncirrhotic patients. Subgroup analyses did not identify meaningful effects of age, sex, race, ethnicity, or IL28B genotype on treatment outcome. SVR12 was achieved in 100% of patients with baseline HCV RNA levels of 800 000 IU/mL or less, compared with 92% of patients with baseline HCV RNA levels greater than 800 000 IU/mL, without overlapping 95% CIs (Figure 4 and Appendix Figure).
Virologic failure occurred in 13 patients (4%), including 1 breakthrough (GT1a infection at treatment week 8) and 12 relapses (9 patients with GT1a, 1 with GT1b, and 2 with GT6 infection) (Appendix Table 3). The 1 breakthrough patient appeared to be fully adherent with study medication dosing. Four additional patients (1.3%) did not achieve SVR12 (3 with GT1a and 1 with GT1b infection) because of 2 non-drug-related deaths, 1 drug-related adverse event, and 1 dropout. After exclusion of these nonvirologic failures, SVR12 was achieved in 94% of patients with GT1a and 99% of those with GT1b infection.
NS3 RAVs were detected at baseline in 86 of 151 (57%) and 25 of 129 (19%) patients with GT1a and GT1b infection, respectively. SVR12 was achieved in 83 of 86 (97%) and 58 of 65 (89%) patients with GT1a infection with or without these RAVs, respectively. The corresponding SVR12 rates among patients with GT1b infection were 24 of 25 (96%) and 104 of 104 (100%) (Table 2). Three patients (all with GT1b) had NS3 RAVs with greater than 5-fold shifts to grazoprevir, and all achieved SVR12 (Appendix Table 4).
NS5A RAVs were identified at baseline in 19 of 154 (12%) GT1a-infected patients. SVR12 was achieved in 11 of 19 (58%) of these patients compared with 133 of 135 (99%) patients without baseline NS5A RAVs. SVR12 in patients with baseline GT1a RAVs with a 5-fold or smaller shift to elbasvir was 90% (9 of 10) versus 22% (2 of 9) in patients with baseline GT1a RAVs with a greater than 5-fold shift (Appendix Table 5). An association of GT1a RAVs with virologic failure was observed only in patients with baseline viral loads greater than 800 000 IU/mL (Appendix Table 6). In 18 of 130 (14%) patients with GT1b infection and baseline NS5A RAVs, SVR12 rates were 94% (17 of 18) versus 100% (112 of 112) in those without baseline NS5A RAVs. Among the 6% (16 of 280) of GT1-infected patients with both NS3 and NS5A RAVs at baseline, SVR12 was achieved in 12 of 16 (75%) (Appendix Table 7).
At the time of virologic failure, RAVs were detected in most patients with GT1a infection (NS3: 6 of 10; NS5A: 10 of 10; both: 6 of 10), most commonly Q80K and D168A in NS3 and M28V/A/G, Q30H/L/R, L31M, and Y93H in NS5A (Appendix Table 3). The 1 patient with GT1b who had virologic failure exhibited T54S and V170I substitutions in NS3 and L31F and Y93H substitutions in NS5A at the time of failure.
In evaluable GT4-infected patients, NS3 RAVs, NS5A RAVs, and both NS3 and NS5A RAVs were present at baseline in 7 of 18 (39%), 9 of 18 (50%), and 2 of 18 (11%) patients, respectively (Appendix Table 7). All achieved SVR12. In evaluable GT6-infected patients, baseline NS3 RAVs and NS5A RAVs were present in 9 of 9 (100%) and 3 of 9 (33%) patients, respectively. SVR12 was achieved in 78% (7 of 9) of these patients, all of whom had baseline NS3 RAVs. At the time of failure, both GT6-infected patients with virologic failure had NS3 RAVs, and 1 also had an NS5A RAV (Appendix Tables 3, 4, 5, and 7).
Grazoprevir-elbasvir was generally well-tolerated (Table 3). The safety profile was similar in the active and placebo treatment groups (Appendix Table 8) and in cirrhotic and noncirrhotic patients (Appendix Table 9). Drug-related adverse events occurred in 114 (36.1%) and 41 (39.0%) patients in the active and placebo groups, respectively (difference, -2.9 percentage points [CI, -13.7 to 7.5 percentage points]). Serious adverse events during treatment and the first 14 follow-up days were reported in 9 (2.8%) and 3 (2.9%) patients in the active and placebo groups, respectively (difference <0.05 percentage points [CI, -5.4 to 3.1 percentage points]); none was considered drug related (Table 3). Two patients in the immediate-treatment group died: One GT1a-infected noncirrhotic patient died of a strangulated hiatal hernia between treatment weeks 2 and 4, and 1 GT1a-infected cirrhotic patient was found dead after treatment week 12 and before follow-up week 4 (death presumed to be due to an arrhythmia from autopsy-documented coronary disease). Neither death was considered drug related. The most common adverse events in the active-treatment group were headache (17%), fatigue (15%), and nausea (9%), with similar frequency in both groups (Table 3 and Appendix Table 10).
Treatment was discontinued because of adverse events in 3 (0.9%) grazoprevir-elbasvir recipients (2 patients with elevated aminotransferase levels and 1 patient with palpitations and anxiety on treatment day 4) and 1 (0.9%) placebo recipient (rash on treatment day 2). One cirrhotic and 3 noncirrhotic grazoprevir-elbasvir recipients (1.3%) developed late elevations of aminotransferase level more than 5 times the upper limit of normal, without an associated increase in bilirubin (Table 3). Two of these four patients discontinued treatment because of these elevations at treatment week 8 (cirrhotic patient) and week 10 (noncirrhotic patient), as stipulated by protocol. In both patients, aminotransferase elevations resolved rapidly after cessation of study therapy and SVR12 was achieved (Appendix 1).
The C-EDGE Treatment-Naive study demonstrated the efficacy of a 12-week regimen of grazoprevir-elbasvir, with an overall virologic failure rate of 4%. This phase 3 study supports the earlier phase 2 findings with grazoprevir-elbasvir (9-10). Although the current study had no active-control group, a historical comparator SVR rate of 73% was derived from previous studies involving simeprevir/peginterferon + ribavirin. Notably, we found similar efficacy in cirrhotic and noncirrhotic patients. Subgroup analyses showed no meaningful effects of age, sex, race, ethnicity, or IL28B genotype on treatment outcome but did suggest an effect of higher baseline viral load (>800 000 IU/mL), which appeared to vary by the presence of baseline NS5A RAVs.
The efficacy in GT1-infected patients was similar to the findings of the phase 2 C-WORTHY study, in which 12 weeks of grazoprevir-elbasvir without ribavirin resulted in SVR12 rates of 97% and 98% in cirrhotic and noncirrhotic patients, respectively (all of whom were treatment naive and GT1 infected) (9-10). Efficacy was also similar to that seen with 12 weeks of the fixed-dose combination of sofosbuvir-ledipasvir, which resulted in SVR12 rates of 96% to 99% in noncirrhotic and 94% in cirrhotic treatment-naive, GT1-infected patients (19-20).
Grazoprevir-elbasvir was highly efficacious in the small number of GT4-infected patients in this study (SVR12, 100% [18 of 18]). This high response rate is consistent with SVR12 rates seen in treatment-naive patients with GT4 infections from the combined phase 2/3 grazoprevir-elbasvir database (54 of 56 [96%]) (Rockstroh JK, Nelson M, Katlama C, Lalezari J, Mallolas J, Bloch M, et al. Efficacy and safety of grazoprevir and elbasvir in patients with hepatitis C virus and HIV co-infection [C-EDGE CO-INFECTION]: a phase 3 trial) (21) and is similar to SVR12 rates with paritaprevir-ritonavir-ombitasvir with or without ribavirin (91% to 100%) (22), sofosbuvir-ledipasvir for 12 weeks in co-infected patients (100% [8 of 8]) (23), and sofosbuvir-daclatasvir for 12 weeks (100% [3 of 3]) (24).
Few GT6-infected patients have been included in clinical trials to date. Sofosbuvir/peginterferon + ribavirin for 12 weeks resulted in an SVR12 rate of 100% (6 of 6) in GT6-infected patients (25), and sofosbuvir-ledipasvir for 12 weeks was associated with an SVR12 rate of 96% (24 of 25; 1 patient with relapse discontinued therapy at week 8) (26). In the C-EDGE Treatment-Naive study, the SVR12 rate was 80% (8 of 10), similar to the overall SVR12 of 80% (12 of 15) in treatment-naive patients from the combined phase 2/3 grazoprevir-elbasvir program (Rockstroh JK, Nelson M, Katlama C, Lalezari J, Mallolas J, Bloch M, et al. Efficacy and safety of grazoprevir and elbasvir in patients with hepatitis C virus and HIV co-infection [C-EDGE CO-INFECTION]: a phase 3 trial ). Conclusions regarding the efficacy of grazoprevir-elbasvir for GT6 infections are constrained by the small sample size. On the basis of available data, grazoprevir-elbasvir appears to possess clinically meaningful activity against GT6, although SVR12 rates have been consistently lower than those seen with GT1 and GT4.
As with other interferon- and ribavirin-free regimens, virologic failure in the C-EDGE Treatment-Naive study was predominantly due to relapse (12 of 13 virologic failures), with only 1 patient developing breakthrough viremia while receiving therapy. Among patients with GT1 and GT6 infections, failure was associated with baseline GT1a NS5A RAVs and emergence of NS3 or NS5A RAVs. Although the overall prevalence of baseline NS3-RAVs was approximately 40%, no association between baseline NS3 RAVs and virologic failure was evident. SVR12 was not affected by mutations with more than 5-fold shifts to grazoprevir in vitro, consistent with the results from C-WORTHY, in which baseline mutations to grazoprevir did not affect efficacy (9-10). In contrast, baseline NS3 RAVs may affect outcome for other protease inhibitor-based regimens, as exemplified by the negative effect of the common baseline Q80K polymorphism on the efficacy of simeprevir/peginterferon + ribavirin (SVR12, 84% without Q80K and 58% with Q80K) (27). In the phase 3 trials of ombitasvir-paritaprevir-dasabuvir with or without ribavirin, baseline samples from 47 patients with GT1a in whom treatment failed were compared with those from 94 demographically matched patients. At baseline, Q80K was detected in approximately 38% of patients and was noted to be about 2-fold more common in patients with virologic failures than responders (28). Specific virologic failure rates for patients with or without baseline Q80K were not reported, but the overall phase 3 SVR12 rates for treatment-naive noncirrhotic patients after 12 weeks of treatment with ombitasvir-paritaprevir-dasabuvir +ribavirin were 98.0% to 99.5% in GT1b-infected patients and 95.3% to 97.0% in GT1a-infected patients; without ribavirin, the corresponding SVR12 rates were 99.0% for GT1b-infected and 90.2% for GT1a-infected patients (29-30).
Baseline NS5A RAVs were detected in 13% of patients with GT1 infection. The data suggest an association between virologic failure and baseline NS5A RAVs, which was most apparent in the few GT1a-infected patients with baseline RAVs demonstrating a greater than 5-fold shift to elbasvir. Although the numbers of patients with baseline NS5A RAVs were small, the relationship between baseline GT1a RAVs and virologic failure was most pronounced in patients with high baseline viral load. A similar effect of baseline NS5A RAVs on efficacy has also been noted with other NS5A inhibitor regimens. In treatment-naive patients receiving sofosbuvir-ledipasvir, the response rate with and without NS5A RAVs varied by study. In ION-3, in which treatment-naive patients received sofosbuvir/ledipasvir with or without ribavirin for 8 weeks or sofosbuvir/ledipasvir for 12 weeks, NS5A RAVs were detected at baseline by deep sequencing in 18% of patients, of whom 90% achieved SVR12 (20). Specific virologic failure rates in patients with and without baseline RAVs by genotype were not reported, but 47% (9 of 19) of GT1a-infected patients and 25% (1 of 4) of GT1b-infected patients with virologic failure had baseline NS5A RAVs (20). In treatment-naive patients in ION-1 and ION-3 combined, 94% of patients with baseline NS5A RAVs versus 95% without RAVs achieved SVR12 after 8 weeks of treatment, whereas 99% with and without RAVs achieved SVR12 after 12 weeks of treatment (31). In treatment-experienced patients receiving 12 weeks of sofosbuvir-ledipasvir, 78% with NS5A RAVs versus 98% without RAVs achieved SVR12 (31). In pivotal trials of ombitasvir-paritaprevir-dasabuvir with or without ribavirin, baseline NS5A RAVs were approximately twice as common in patients with virologic failure as in responders (28). Separate virologic failure rates in patients with or without baseline NS5A RAVs against ombitasvir were not explicitly reported.
Coincident with virologic failure, GT1 NS3 or NS5A RAVs were detected by population sequencing in most GT1a-infected patients, most commonly NS3_Q80K and D168A and NS5A_M28V/A/G, Q30H/L/R, L31M, and Y93H. These findings may have implications for salvage therapy in the small percentage of patients who experience virologic failure while receiving grazoprevir-elbasvir.
The prognostic utility of pretreatment resistance testing requires continued scrutiny as the use of different classes of directly acting antiviral agents for chronic HCV infection becomes increasingly widespread (3-5, 7, 32). In this study, only 5.7% of patients with genotype 1a had baseline RAVs with greater than 5-fold decreased susceptibility to elbasvir, making up 2.8% of the total population. The cost-effectiveness of screening all patients for the small number who harbor highly resistant variants at baseline should be considered when practice guidelines are formulated.
The deferred-treatment group in this study allowed a concurrent placebo control group to evaluate safety. Grazoprevir-elbasvir was generally well-tolerated, with a similar safety profile in cirrhotic and noncirrhotic patients and in the active and placebo treatment groups. Serious adverse events were infrequent, with a similar incidence in the active and placebo groups. No serious adverse event was judged to be drug related by site investigators. Discontinuations for adverse events were likewise uncommon. The most common adverse events were headache, fatigue, and nausea, occurring with similar frequencies in noncirrhotic and cirrhotic patients and in the active and placebo groups.
Late elevations in aminotransferase levels greater than 5 times the upper limit of normal occurred in 4 patients (1.3%) in the immediate-treatment group; none was associated with hyperbilirubinemia. These abnormalities were reversible and had no major clinical consequences. The hepatic safety profile of grazoprevir-elbasvir in this study was similar to the findings of the phase 2 C-WORTHY study, in which fewer than 1% of patients experienced a late elevation in alanine aminotransferase level greater than 5 times the upper limit of normal (9-10).
The C-EDGE Treatment-Naive study confirmed that an oral fixed-dose combination of an NS3/4A protease inhibitor (grazoprevir) and an NS5A inhibitor (elbasvir) provides a highly efficacious and well-tolerated regimen for chronic HCV-GT1, GT4, or GT6 infection in treatment-naive, compensated cirrhotic and noncirrhotic patients. This once-daily, single-tablet, 2-drug combination constitutes a potent new therapeutic option for chronic HCV infection.
Appendix 1: Supplementary Information
Background and Summary
Chronic infection with hepatitis C virus (HCV) presents an expanding challenge worldwide. Without therapy, almost half of persistently infected patients will progress to cirrhosis, placing them at risk for hepatocellular carcinoma and hepatic decompensation requiring liver transplantation. As a consequence of contamination of blood supplies in the 1960s through 1992 and the aging of the HCV-infected population, the number of patients with HCV-related complications continues to increase dramatically. Effective HCV therapy with virologic cure can reduce long-term HCV complications and mortality.
Interferon- and ribavirin-free regimens can achieve sustained virologic response (SVR) rates >90% in treatment-naive populations in randomized controlled clinical trials. However, efficacy across HCV genotypes (GT), in certain historically difficult-to-treat patient subgroups, and in the presence of NS3 and/or NS5A variants differs by regimen.
C-EDGE Treatment-Naive demonstrated that an all-oral, once-daily, fixed-dose combination of a protease inhibitor (grazoprevir) plus an NS5A inhibitor (elbasvir) can be a highly efficacious and well-tolerated regimen for chronic HCV GT1, -4, or -6 infection in treatment-naive patients both with and without cirrhosis. Virologic failure was uncommon and was predominantly due to virologic relapse following completion of therapy. The presence of NS3 resistance-associated variants (RAVs) at baseline did not affect outcome, although NS5A RAVs at baseline were associated with lower rates of SVR at 12 weeks after the end of treatment in GT1a infections. Grazoprevir-elbasvir was associated with few serious adverse events and rare discontinuations, including in patients with compensated cirrhosis. Overall, grazoprevir-elbasvir provides a novel therapeutic option for treatment-naive HCV-infected patients with GT1a, -1b, -4, or -6.
Derivation of Historical SVR12 Reference Rate
The historical reference rate of 73% was derived from the phase 3 trials of simeprevir/PR in treatment-naive monoinfected patients after adjustment for the expected proportion of cirrhotic patients and the anticipated improved tolerability with an interferon-free regimen (15-16). In QUEST 1 and 2, SVR12 was achieved by 60% of patients with compensated cirrhosis and 82% of noncirrhotic patients. Assuming that 20% of C-EDGE Treatment-Naive patients had cirrhosis, a SVR of 78% was computed. A 5% decrease was then applied because of the expected improved tolerability.
Power Assumptions and Calculations for Safety Analyses
For power assumptions and calculations for safety analyses, see Appendix Tables 11 and 12.
Staging of Liver Injury
Hepatic fibrosis was staged by biopsy or noninvasive assessment. Cirrhosis was defined as a liver biopsy showing METAVIR stage F4 at any time prior to entry; transient elastography (Fibroscan) performed within 12 months of entry yielding a result >12.5 kPa; or biochemical markers of liver fibrosis (FibroTest or FibroSure) yielding a score of >0.75 along with an aspartate aminotransferase-platelet ratio index (APRI) >2. Patients were considered noncirrhotic if a liver biopsy performed within the previous 24 months did not reveal cirrhosis; a Fibroscan performed during the previous 12 months had a result of ≤12.5 kPa; or a FibroSure or FibroTest score was ≤0.48 with an APRI of ≤1 in the preceding 12 months.
Narratives of Patients With Late Aminotransferase Elevations 5 Times the Upper Limit of Normal (ULN)
In an earlier study of grazoprevir given with pegylated interferon plus ribavirin (8), some patients receiving doses of grazoprevir higher than the 100-mg dose used in the current study experienced late elevations in aminotransferases; therefore, liver safety signals were carefully monitored in this study. Four patients (1.3%) in the immediate-treatment group had late aminotransferase elevations > 5 times ULN. None of these cases were associated with elevated serum bilirubin. All cases were manageable and reversible and had no major clinical consequences.
Late ALT or AST elevations are defined as elevations in ALT or AST occurring after treatment week 4 among patients who had ALT or AST within normal limits between week 2 and week 4. The 4 cases with late elevations of aminotransferases observed in the current study are summarized below.
Elevation of ALT at Treatment Week 6
This 62-year-old noncirrhotic white woman with HCV GT1b was randomly assigned to the immediate-treatment group. While on study medication, her aminotransferases normalized from a baseline ALT/AST of 46/47 U/L to a nadir of 18/22 U/L at treatment week 2. At treatment week 6 she was noted to have an ALT increase to 376 U/L (11.4 times ULN) and an AST increase of 198 U/L (5.5 times ULN) associated with an eosinophil differential of 5.3% (ULN of 7.0%). She was asymptomatic, without nausea, vomiting, or upper right quadrant pain. She had not taken any new medications, consumed alcohol, or taken any over-the-counter or herbal remedies. The patient's ALT/AST and eosinophils decreased to within normal limits on continued therapy and she completed the full course of study drug and achieved SVR12.
Elevation of ALT at Treatment Week 8
This 52-year-old cirrhotic Asian woman with HCV GT1b infection was randomly assigned to the immediate-treatment group. While on study medication, her aminotransferases normalized from a baseline ALT/AST of 80/70 U/L to a nadir of 22/26 U/L at treatment week 2. At treatment week 8 she was noted to have an ALT of 199 U/L (6.0 times ULN) and an AST of 167 U/L (4.6 times ULN), with an increase in eosinophils to 8.8% (ULN 7.0). The patient had been seen in a local clinic and diagnosed with an upper respiratory infection. New concomitant medications included a theophylline derivative, loxoprofen (a nonsteroidal anti-inflammatory drug), an antacid, a mucolytic, and cough syrup. She had not consumed any alcohol and did not take any herbal remedies or other new prescription or over-the-counter medications except for those detailed above. She did not have any new or worsening nausea, vomiting, or upper right quadrant pain/tenderness. At treatment week 10 her ALT was 668 U/L, AST was 459 U/L, and eosinophils were 4.9%. The aminotransferases elevations met the protocol-specified criterion for study drug discontinuation. Her ALT/AST and eosinophils decreased to within normal limits after the discontinuation of study drug. The patient had completed 70 days of study drug therapy and subsequently achieved SVR12.
Elevation of ALT at Treatment Week 10
This 67-year-old noncirrhotic African American woman with HCV GT1b infection was randomly assigned to the immediate-treatment group. While on study medication, her aminotransferases normalized from a baseline ALT/AST of 94/99 U/L to a nadir of 18/26 U/L at treatment week 8. At treatment week 10 she was noted to have an ALT of 399 U/L (12 times ULN) and an AST of 338 U/L (9.4 times ULN) associated with an eosinophil differential of 5.1% (ULN 7.0%). She was asymptomatic and her physical examination was normal. She had not taken any new medications or taken any over-the-counter or herbal remedies; she had one glass of wine 4 days previously. Repeat lab test 5 days later demonstrated an ALT/AST of 474/352, which decreased the following day despite continued treatment to 439/310, respectively. Study drug was discontinued per the protocol stopping rules. Her ALT/AST resolved to within normal limits off treatment. The patient completed 78 days of therapy and subsequently achieved SVR12.
Elevation of ALT at Treatment Week 12
This 61-year-old noncirrhotic woman of Asian descent with HCV GT6 infection was randomly assigned to the immediate-treatment group. While on study medication, her aminotransferases normalized from a baseline ALT of 59 U/L and AST of 67 U/L to within normal limits. At treatment week 12 her ALT/AST increased to 170/133 U/L; eosinophils, prothrombin international normalized ratio, alkaline phosphatase and total bilirubin values remained within normal limits during treatment. The patient did not report new or worsening nausea, vomiting, jaundice, arthralgia, or upper right quadrant pain/tenderness. She denied taking any new medications, including herbal supplements and over the counter medications, and did not drink alcohol. An abdominal ultrasound showed no liver abnormalities. At follow-up week 4, laboratory testing showed the patient's ALT/AST levels had decreased to 63/65 U/L. However, HCV RNA testing showed virologic relapse with a viral RNA of 73,104 IU/mL.