Hepatitis C virus (HCV) genotype-1 is the most common

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2013;11:81 87 LIVER, PANCREAS, AND BILIARY TRACT Boceprevir With Peg Alfa-2a Ribavirin Is Effective for Previously Treated Chronic Hepatitis C Genotype 1 Infection STEVEN L. FLAMM,* ERIC LAWITZ, IRA JACOBSON, MARC BOURLIÈRE, CHRISTOPHE HEZODE, JOHN M. VIERLING, # BRUCE R. BACON,** CLAUS NIEDERAU, MORRIS SHERMAN, VENKATA GOTETI, HEATHER L. SINGS, RICHARD O. BARNARD, JOHN A. HOWE, LISA D. PEDICONE, MARGARET H. BURROUGHS, CLIFFORD A. BRASS, JANICE K. ALBRECHT, and FRED POORDAD *Division of Gastroenterology and Hepatology, Northwestern Feinberg School of Medicine, Chicago, Illinois; Department of Hepatology, Alamo Medical Research, San Antonio, Texas; Division of Gastroenterology and Hepatology, Center for the Study of Hepatitis C, Weill Cornell Medical College, New York, New York; Department of Hepato-Gastroenterology, Fondation Hôpital Saint Joseph, Marseille, France; Department of Hepatology, AP-HP, Hopital Henri Mondor, INSERM, U955, Université Paris-Est, Créteil, France; # Departments of Medicine and Surgery, Baylor College of Medicine, Houston, Texas; **Department of Internal Medicine, Division of Gastroenterology and Hepatology, Saint Louis University School of Medicine, St Louis, Missouri; Department of Medicine, Katholische Kliniken, Oberhausen, Germany; Department of Medicine, Toronto General Hospital, Toronto, Canada; Merck, Sharp & Dohme Corporation, Whitehouse Station, New Jersey; and Department of Hepatology and Liver Transplantation, Cedars-Sinai Medical Center, Los Angeles, California This article has an accompanying continuing medical education activity on page e5. Learning Objectives At the end of this activity, the successful learner will be able to apply data to treat previously treated patients with HCV genotype 1 with peg alfa-2a ribavirin and boceprevir. BACKGROUND & AIMS: The addition of boceprevir to therapy with peg alfa-2b and ribavirin results in significantly higher rates of sustained virologic response (SVR) in previously treated patients with chronic hepatitis C virus (HCV) genotype-1 infection, compared with peg alfa-2b and ribavirin alone. We assessed SVR with boceprevir plus peg alfa-2a ribavirin (PEG2a/R) in patients with identical study entry criteria. METHODS: In a double-blind, placebocontrolled trial, 201 patients with HCV genotype-1 who had relapsed or not responded to previous therapy were assigned to groups (1:2) and given a 4-week lead-in phase of PEG2a/R, followed by placebo plus PEG2a/R for 44 weeks (PEG2a/R) or boceprevir plus PEG2a/R for 44 weeks (BOC/PEG2a/R). The primary end point was SVR 24 weeks after therapy ended. RESULTS: The addition of boceprevir after 4 weeks of lead-in therapy with PEG2a/R significantly increased the rate of SVR from 21% in the PEG2a/R group to 64% in the BOC/PEG2a/R group (P.0001). Among patients with poor response to therapy ( 1-log 10 decline in HCV RNA at week 4), 39% in the BOC/PEG2a/R group had SVRs, compared with none of the patients in the PEG2a/R group. Among patients with good response to ( 1-log 10 decline), 71% in the BOC/PEG2a/R group had SVRs, compared with 25% in the PEG2a/R group. A 1-log 10 decline in HCV RNA at treatment week 4 was the strongest independent predictor of SVR, exceeding that of IL-28B genotype. Among 8 patients who began the study with HCV amino acid variants associated with boceprevir resistance, 3 (38%) achieved SVRs. Fifty percent of patients in the BOC/PEG2a/R group developed anemia (hemoglobin 10.0 g/dl), compared with 27% in the PEG2a/R group; 43% vs 21%, respectively, developed neutropenia (neutrophil count 750/mm 3 ). CONCLUSIONS: The addition of boceprevir after 4 weeks of lead-in therapy with PEG2a/R caused significantly higher rates of SVR in previously treated patients with chronic HCV genotype-1 infection, compared with patients given only PEG2a/R. ClinicalTrials.gov Identifier: NCT Keywords: Protease Inhibitor; Antiviral Therapy; Clinical Trial; Liver Disease. Hepatitis C virus (HCV) genotype-1 is the most common genotype in the United States, Europe, and many other countries and is the most difficult to eradicate with peg/ribavirin therapy. With peg alfa-2a (PEG2a) or alfa-2b and ribavirin (R) treatment, less than 50% of treatmentnaive genotype-1 patients achieve sustained virologic response (SVR). 1,2 Compared with treatment-naive patients, response rates are even lower for genotype-1 nonresponders to previous -ribavirin therapy, who are retreated with peg-ribavirin. 3 6 Boceprevir (BOC) is a peptidomimetic ketoamide protease inhibitor that binds reversibly to the HCV-NS3 active site. In phase 2 and 3 studies, addition of boceprevir to peg alfa-2b ribavirin in previously untreated (SPRINT-1 and 2) 7,8 and previous treatment-failure (RESPOND-2) 9 patients with HCV genotype-1 resulted in significantly higher rates of SVR as compared with peg alfa-2b ribavirin alone. In the RESPOND-2 trial, the rate of SVR was significantly higher in the 48-week boceprevir arm (66%) than in the control arm (21%, P.0001). 9 Because of the widespread use of both peg alfa-2b and alfa-2a in the treatment of HCV, this phase 3 study was designed to investigate the efficacy and safety of boceprevir when added to PEG2a/R backbone therapy in a patient population who met identical entry criteria as the RESPOND-2 Abbreviations used in this paper: AE, adverse event; BOC, boceprevir; BOC/PEG2a/R, boceprevir plus peg alfa-2a ribavirin; CI, confidence interval; HCV, hepatitis C virus; OR, odds ratio; PEG2a, peg alfa-2a; R, ribavirin; RAV, resistant associated variants; SVR, sustained virologic response by the AGA Institute /$

2 82 FLAMM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 1 Figure 1. Study design. trial that used peg alfa-2b, that is, in adults with chronic HCV genotype-1 with previously demonstrated responsiveness to peg and ribavirin but with failure to achieve SVR. Methods Patients From February 23, 2009, to June 22, 2009, 201 patients with HCV genotype-1 were screened from 53 sites in North America and Europe. Eligible patients met identical entry criteria as the RESPOND-2 trial. 9 Eligibility criteria included demonstrated responsiveness to (minimum duration of therapy of 12 weeks). We defined patients as having either nonresponse (ie, a decrease in the HCV RNA level of at least 2 log 10 IU/mL by week 12 but with a detectable HCV RNA level during the therapy period) or relapse (ie, an undetectable HCV RNA level at the end of treatment, without subsequent attainment of SVR). Eligibility criteria included an absolute neutrophil count 1500/mm 3 ( 1200/mm 3 for black patients), a platelet count 100,000/mm 3, and hemoglobin levels 12 g/dl for female patients and 13 g/dl for male patients. Exclusion criteria included hepatitis B or human immunodeficiency virus infection, any other cause of significant liver disease, decompensated liver disease, uncontrolled diabetes mellitus, severe psychiatric disorders, and/or active substance abuse. Pretreatment liver biopsy specimens were assessed for Metavir fibrosis scores and steatosis scores by a single pathologist who was unaware of the assignment of boceprevir or placebo. 9 Study Design The primary objective was to compare the rates of SVR achieved by using boceprevir in combination with open-label PEG2a (Pegasys; Hoffman-La Roche Ltd, Basle, Switzerland) and ribavirin with PEG2a/R plus placebo in adults with chronic HCV genotype-1 with previously demonstrated responsiveness to peg and ribavirin but with failure to achieve SVR. The study was conducted in accordance with principles of good clinical practice and was approved by the appropriate institutional review boards and regulatory agencies. Informed consent was obtained from each patient. All authors had access to the study data and have reviewed and approved the final manuscript. Patients were randomized in a 1:2 ratio by using an interactive voice response system with stratification by previous response to therapy (nonresponder or relapser) and HCV genotype (1a or 1b) as determined by Trugene HCV 5 NC (Bayer Healthcare, Tarrytown, NY) sequencing. Patients with HCV genotype-1 infection whose HCV subtype could not be classified were randomly assigned to one of the treatment groups. The HCV genotype-1 subtype was subsequently determined by means of sequencing of the nonstructural 5B (NS5B) region (Virco, Mechelen, Belgium). PEG2a was administered subcutaneously at 180 g once weekly. Ribavirin was administered by using weight-based dosing of mg/d (divided daily dose). Boceprevir was administered orally at a dose of 800 mg 3 times daily (to be taken with food and with an interval of 7 9 hours between doses) in 4 capsules of 200 mg each. Placebo was matched to boceprevir. The study was double-blinded regarding the administration of boceprevir. All patients received a 4-week lead-in with PEG2a/R (Figure 1). Subsequently, the control group received peg alfa- 2a ribavirin plus placebo for 44 weeks (PEG2a/R group), and the second group received boceprevir plus PEG2a/R for 44 weeks (BOC/PEG2a/R group). HCV RNA levels were measured with the use of the TaqMan 2.0 assay (Roche Diagnostics, Indianapolis, IN), which had lower limits of quantification and detection of 25 and 9.3 IU/mL, respectively; the lower limit of detection was used for decision-making at various points throughout the study. Measurements were obtained at the screening visits and at baseline; every 2 weeks through week 12 and then weeks 16, 20, 24, 30, 36, 42, and 48; and weeks 4, 12, and 24. The stopping rule applied to both groups was failure to achieve an undetectable HCV RNA level at week 12, which resulted in discontinuation of all treatment and advancement to. The study prospectively consented patients for pharmacogenomic testing and collected samples for biomarker identification. IL-28B genetic classification of patients was evaluated as a predictor of SVR by frequency tables and logistic regression models. IL-28B genotyping for markers rs , rs , and rs was performed by Gentris Clinical Genetics, Inc (Morrisville, NC) by using DNA Sanger sequencing technology. Several single nucleotide polymorphisms in linkage disequilibrium are in this region, and the principal one evaluated was the rs locus, as in other studies. 10,11 IL-28B was analyzed retrospectively. Safety Adverse events (AEs) were graded by investigators according to a modified World Health Organization grading

3 January 2013 BOCEPREVIR WITH PEGINTERFERON ALFA-2A RIBAVIRIN 83 system. Non life-threatening AEs were managed by dose reduction. The guidelines for a reduction of PEG2a/R were determined on the basis of the prescribing information from the Food and Drug Administration. Viral breakthrough and incomplete virologic response were defined as previously described. 12 If a patient had virologic breakthrough or an incomplete virologic response and rebound while receiving therapy, the protocol required discontinuation of boceprevir treatment, but PEG2a/R could be continued for up to 48 weeks with appropriate clinical. Plasma samples for the assessment of amino acid variants potentially associated with reduced susceptibility to boceprevir were collected. Postbaseline assessments for resistant associated variants (RAVs) were obtained for patients who did not achieve SVR in the BOC/PEG2a/R group. Statistics Analyses for the primary objective included data from all patients who received at least 1 dose of any study medication. The key secondary objective was to compare boceprevir in combination with PEG2a/R with PEG2a/R plus placebo in patients who completed the lead-in phase and received at least 1 dose of placebo or boceprevir. The primary efficacy end point was SVR, defined as undetectable plasma HCV RNA at week 24. The primary efficacy end point was summarized by using descriptive statistics for both treatment groups. SVR rates were based on a last observation carried forward approach, in which the week 12 HCV RNA result was carried forward for patients with a missing HCV RNA value at week 24. The primary statistical comparison was carried out by using a two-sided Cochran Mantel Haenszel 2 test, controlling for the baseline stratification factors. With 132 patients in the experimental group and 66 patients in the control group, the study would have 90% power to detect an improvement of 20% in the SVR rate between the 2 treatment groups, assuming a response rate of 10% in the control group and using a two-sided 2 test at alpha.05. To account for multiplicity between the primary and key secondary analyses, the key secondary analysis was conducted only if the significance of the primary comparison was established. Multivariate logistic regression analyses to evaluate baseline factors, including IL-28B rs genotype, as predictors of SVR and response were conducted. A separate model included early on-treatment response ( 1-log 10 decline vs 1-log 10 decline at week 4). Odds ratio (OR), 95% confidence interval (CI) of the ORs, and 2 P values from the logistic models are presented. Results Characteristics of the Study Patients A total of 292 patients were screened, and 201 were randomized and treated (Supplementary Figure 1). There were no significant differences in the baseline demographic characteristics between the treatment groups (Table 1). Seventy percent of patients (140/201) were male, and 90% (181/201) were nonblack. The mean age was 52.5 years (range, years), and the mean body mass index was 28 kg/m 2. Approximately half of patients (56%, 113/201) had HCV subtype 1a as ascertained by sequencing of the nonstructural 5B (NS5B) region. A total of 23% of patients had advanced fibrosis (Metavir fibrosis Table 1. Baseline Characteristics Characteristic, n (%) PEG2a/R (N 67) BOC/PEG2a/R (N 134) Mean age, y Male sex 43 (64) 97 (72) Race White 58 (87) 119 (89) Black 8 (12) 12 (9) Other 1 (1) 3 (2) Region North American 48 (72) 92 (69) European Union 19 (28) 42 (31) Mean body mass index, kg/m (4.6) 28.2 (4.5) (standard deviation) HCV subtype a 1a 38 (57) 75 (56) 1b 27 (40) 55 (41) 1 (missing or not determined) 2 (3) 4 (3) Alanine aminotransferase 47 (70) 97 (72) upper limit of normal range Platelet count 100,000 to 150,000/ L 10 (15) 30 (22) 150,000/ L 57 (85) 104 (78) Viral load 800,000 IU/mL 54 (81) 101 (75) Metavir fibrosis score b F0, F1, or F2 47 (70) 93 (69) F3 6 (9) 8 (6) F4 9 (13) 24 (18) Steatosis c 0 17 (25) 44 (33) 1 45 (67) 80 (60) Previous therapy PEG2a 40 (60) 76 (57) Peg alfa-2b 25 (37) 51 (38) Peg, type unknown 1 (1) 5 (4) Nonpegylated s 1 (1) 2 (1) Response to prior therapy Nonresponse 20 (30) 36 (27) Relapse 47 (70) 98 (73) IL-28B genotype N 51 N 95 CC 10 (20) 19 (20) CT 34 (67) 59 (62) TT 7 (14) 17 (18) a HCV subtype was ascertained by means of sequencing of nonstructural 5B (NS5B) region. b Five patients in PEG2a/R group and 9 patients in BOC/PEG2a/R group had missing data. c Five patients in PEG2a/R group and 10 patients in BOC/PEG2a/R group had missing data. score of 3/4), and 77% had a baseline viral load 800,000 IU/mL. The majority of patients (72%) had relapsed after previous HCV therapy. Among patients who consented to pharmocogenomic testing, the majority had the CT genotype (64%, 93/146), followed by CC (20%, 29/146) and TT (16%, 24/146). Efficacy SVR rates were significantly higher in patients receiving boceprevir compared with those treated with PEG2a/R alone (Figure 2), with overall SVR rates of 21% in both the primary and secondary end point analyses in the PEG2a/R arm (14/67) and 64% (86/134) to 66% (86/130) in the BOC/PEG2a/R arm,

4 84 FLAMM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 1 Figure 2. Patients with SVR, according to treatment group and analysis. respectively (P.0001). In the primary analysis population, end-of-treatment response rates were 42% (28/67) and 74% (99/134) and relapse rates were 33% (7/21) and 12% (11/95) in the PEG2a/R and BOC/PEG2a/R groups, respectively. Five patients (3 PEG2a/R, 2 BOC/PEG2a/R) had missing data at week 24; therefore, the week 12 value was carried forward. When those with missing data at week 24 were considered as treatment failures, SVR rates were 16% (11/67) in the PEG2a/R group and 63% (84/134) in the BOC/PEG2a/R group, respectively. The BOC/PEG2a/R group achieved a higher rate of SVR than the control group regardless of region. In North America and Europe, SVR rates were 59% (54/92) and 76% (32/42) in the BOC/PEG2a/R group and 17% (8/48) and 32% (6/19) in the PEG2a/R group. SVR rates for patients with a Metavir score of F0/1/2 were 21% (10/47) and 71% (66/93) and for patients with a Metavir score of F3/4 were 13% (2/15) and 50% (16/32) in the PEG2a/R and BOC/PEG2a/R groups, respectively. Rates of SVR were assessed by historical response to prior therapy (prior relapse and prior nonresponse) and by the log 10 decline in HCV RNA after the 4-week lead-in (ie, on-treatment response). The SVR rate for prior relapsers was 28% in the PEG2a/R group vs 70% in the BOC/PEG2a/R group. For prior nonresponders, the SVR rate was 5% for the PEG2a/R group vs 47% for the BOC/PEG2a/R group (Figure 2). Despite the criterion for study entry of having a documented prior response, 13% of historical relapsers (19/145) and 14% of historical nonresponders (8/56) had poor on-treatment response to, which was defined as a decrease in the HCV RNA level of l-log 10 IU/mL after the 4-week lead-in period. In this subgroup, SVR was achieved in none of the patients in the PEG2a/R group and in 39% of the patients in the BOC/ PEG2a/R group (Figure 2). Among patients with good ontreatment response to ( 1log 10 IU/mL decrease), SVR rates were 25% and 71% in PEG2a/R and BOC/PEG2a/R groups, respectively. Early response (undetectable HCV RNA at week 8, corresponding to 4 weeks of boceprevir) was associated with a high rate of SVR (89% [BOC/PEG2a/R] vs 44% [PEG2a/ R]). The proportion of patients achieving undetectable HCV RNA at week 8 was approximately 4-fold higher in patients receiving boceprevir (BOC/PEG2a/R, 54%, 72/134) compared with those who received PEG2a/R control (13%, 9/67). Among patients who consented to genomic testing, boceprevir-treated patients had higher SVR rates over control for all IL-28B genotypes (Supplementary Table 1), although the small number of patients with IL-28B data precludes the ability to draw robust conclusions. The highest SVR rate was observed for the less favorable TT genotype (82%, 14/17). SVR rates were highest for prior relapsers in each of the treatment groups for each of the IL-28B genotypes. In the BOC/PEG2a/R group, SVR rates for patients with good response to were 68% 86%, and for those who become undetectable by week 8, SVR rates were 85% 90%, regardless of IL-28B genotype. Population sequencing was used to detect RAVs previously identified to confer reduced susceptibility to boceprevir in HCV-infected patients. 13 Of the 134 patients randomized to the BOC/PEG2a/R group, 130 received at least 1 dose of boceprevir, and 33 of 44 patients who did not achieve SVR had postbaseline samples sequenced for RAVs. Among the non-svr patients whose samples were sequenced, 8 of 33 (24%) had RAVs detected on-treatment (Supplementary Table 2). RAVs were detected at baseline (ie, before therapy) in 8 patients from the 125 of 130 patients with DNA sequence available at baseline and who subsequently received boceprevir (Supplementary Table 2). Three of these 8 patients with baseline RAVs detected (38%) achieved SVR. The 5 patients with baseline RAVs detected who did not achieve SVR were all nonresponders in this study and discontinued by choice (1 patient), AEs (2 patients), and as a result of futility (2 patients). None of the 8 patients with baseline RAVs detected had additional RAVs detected while on boceprevir treatment. In general, patients with poor response to had a higher rate of RAVs. For those patients with sequencing data, 50% of those with poor response to (5/10) and 9% of those with good response to (2/22) had variants detected (1 patient had missing week 4 data). Multivariate logistic regression analyses were performed to identify predictors of SVR (Supplementary Table 3). Two factors were significantly associated with achievement of SVR: assignment to boceprevir treatment (OR, 12.0; P.001) and historical classification as a relapser (OR, 4.4; P.003). IL28B genotype was not predictive of SVR. As previously reported in SPRINT-2 and RESPOND-2, a 1-log 10 IU/mL decline at treatment week 4 was the strongest predictor of SVR. In the present study, when week 4 response was added to the model ( 1-log 10

5 January 2013 BOCEPREVIR WITH PEGINTERFERON ALFA-2A RIBAVIRIN 85 Table 2. AEs According to Treatment Group Table 2. Continued Event, n (%) PEG2a/R (N 67) BOC/PEG2a/R (N 134) Event, n (%) PEG2a/R (N 67) BOC/PEG2a/R (N 134) Median duration of study-drug exposure, d Death a 0 2 (1) Any AE 67 (100) 134 (100) Discontinuation owing to AE 3 (4) 23 (17) (any study drug) b Dose modification owing to AE (any 15 (22) 58 (43) study drug) bc Any serious AE 7 (10) 18 (13) Hematologic event Reduced neutrophil count b 12 (18) 42 (31) Grade 0: 1500 per mm 3 12 (18) 13 (10) Grade 1: per mm 3 24 (36) 32 (24) Grade 2: 750 to 1000 per 17 (25) 32 (24) mm 3 Grade 3: 500 to 750 per 12 (18) 38 (28) mm 3 Grade 4: 500 per mm 3 2 (3) 19 (14) Dose modification for neutropenia 9 (13) 23 (17) (any drug) Granulocyte-colony stimulating 8 (12) 19 (14) factor use (overall) Hemoglobin level b Grade 0: 11.0 g/dl 33 (49) 37 (28) Grade 1: 9.5 to 11.0 g/dl 23 (34) 55 (41) Grade 2: 8.0 to 9.5 g/dl 9 (13) 32 (24) Grade 3: 6.5 to 8.0 g/dl 2 (3) 9 (7) Grade 4: 6.5 g/dl 0 1 (1) Any erythropoietin use bd 20 (30) 63 (47) Management for hemogloblin 10g/dL, n/m (%) e Ribavirin dose reduction only 0/18 (0) 5/65 (8) Erythropoietin use only 5/18 (28) 19/65 (29) Both ribavirin dose reduction 10/18 (56) 37/65 (57) and erythropoietin use Neither 3/18 (17) 4/65 (6) Transfusion 0 5 (1) HCV RNA decline vs 1-log 10 HCV RNA decline at week 4), it was a stronger predictor of SVR than historical response (OR, 7.9; P.004 vs OR, 5.4; P.003) (Supplementary Table 3). Multivariate analysis to assess which baseline factors predict response showed that none of the baseline factors examined were significant predictors of response (Supplementary Table 3). Safety The study included a stopping rule whereby patients failing to achieve undetectable HCV RNA at treatment week 12 discontinued all therapy. Because of the lower response rate observed in the control group, the proportion of patients remaining on treatment at week 48 was substantially lower in the PEG2a/R group, compared with those in the BOC/PEG2a/R group (30% [20/67] vs 60% [80/134], respectively). Thus, the median duration of treatment was 3.2-fold longer in the boceprevir group, compared with control (Table 2). There was no significant difference in the proportion of patients reporting serious AEs between the treatment groups. A significantly higher proportion of patients in the BOC/ Common AE f Anemia 22 (33) 67 (50) Neutropenia 12 (18) 42 (31) Dysgeusia 10 (15) 52 (39) Diarrhea 5 (7) 30 (22) Rash 5 (7) 30 (22) Myalgia 5 (7) 25 (19) Leukopenia 2 (3) 20 (15) Vomiting 0 13 (10) a One death occurred 2 days after treatment (heart failure) and the other 15 days after treatment (Staphylococcus aureus bronchopneumonia). b 2 test with P.05 for comparison of BOC/PEG2a/R vs PEG2a/R. c Excludes patients who discontinued because of AEs. d Includes erythropoietin use in patients with hemoglobin level 10 g/dl. Five patients in PEG2a/R group and 7 patients in BOC/ PEG2a/R group who received erythropoietin had hemoglobin level 10 g/dl. e Includes all patients with hemoglobin level 10 g/dl up to 30 days after treatment end date. n, number of patients with indicated anemia management strategy; m, number of patients with hemoglobin level 10 g/dl up to 7 days after treatment end date. f Common AEs were those classified as being related to a study treatment and occurring with incidence of 10% or more in any group. Only common AEs for which 2 test with P.05 for comparison of BOC/PEG2a/R vs PEG2a/R are shown. All other common AEs (for which the treatment-related incidence was 10% or more in any group) can be found in Supplementary Table 4. PEG2a/R group discontinued treatment or had a dose modification owing to an AE ( 2 test with P.05 for a comparison of BOC/PEG2a/R vs PEG2a/R). There were 2 deaths among patients in the BOC/PEG2a/R group. One death occurred 2 days after treatment (heart failure) and the other 15 days after treatment (Staphylococcus aureus bronchopneumonia). Neutropenia was reported more frequently in patients receiving boceprevir than in patients in the PEG2a/R group (31% vs 18%). Neutropenia in both arms was effectively managed by using guidelines for dose modification (17% vs 13%) or granulocyte colony-stimulating factor treatment (14% vs 12%). The percentage of patients who met recommended criteria for dose reduction (neutrophil count 750 per mm 3 ) or dose interruption/discontinuation (neutrophil count 500 per mm 3 ) was significantly higher in the BOC/PEG2a/R group than in the PEG2a/R group (Table 2). Two patients in the BOC/PEG2a/R group had grade 4 neutropenia and discontinued treatment (nadirs of 0.0 and 100 per mm 3 ). In both cases, the neutropenia resolved. Serious AEs caused by infection were reported in 7 patients (5%) in the BOC/PEG2a/R group (0% in PEG2aR group), with risk factors of cirrhosis (3 patients) and diabetes (2 patients). The types of infection included influenza A, urosepsis, Staphylococcus aureus (3 patients), Chlamydia pneumoniae and Mycoplasma pneumoniae, and Haemophilus infection. None of the patients had had grade 3/4 neutropenia before the event. The frequencies of anemia (hemoglobin 10 g/dl) in the PEG2a/R and BOC/PEG2a/R groups were 27% (18/67) vs 50% (67/134). Anemia was effectively managed with ribavirin dose reduction alone (0% vs 8%), erythropoietin use alone (28% vs

6 86 FLAMM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 1 29%), or both (56% vs 57%). There were no serious AEs caused by anemia reported, and only 1 patient (BOC/PEG2a/R group) discontinued treatment because of anemia. In the BOC/ PEG2a/R group, patients with anemia had numerically higher SVR rates than those without anemia (53/65 [82%] vs 33/68 [49%]). In the PEG2a/R group, patients with and without anemia had SVR rates of 17% (3/18) and 22% (11/49), respectively. The most common AEs observed in both treatment groups were flu-like symptoms that are typically reported with peg-ribavirin therapy (Supplementary Table 4). A significantly higher proportion of patients in the BOC/PEG2a/R group reported dysgeusia, diarrhea, rash, myalgia, leukopenia, and vomiting than the PEG2a/R group (Table 2). There were no serious AEs or discontinuations caused by rash/skin eruptions. Discussion The present study demonstrates that adding boceprevir to PEG2a/R for retreatment of patients who have previously failed therapy results in a similar statistically significant and clinically meaningful increase in efficacy (64% vs 21%, P.0001) compared with retreatment with PEG2a/R alone. This benefit was demonstrated irrespective of historical response to prior therapy (prior relapse or prior nonresponse) and on-treatment responsiveness (good response or poor response). Patients with an undetectable HCV RNA at treatment week 8 were shown to have SVR rates of up to 89% when boceprevir was added to PEG2a/R vs 44% for PEG2a/R alone. The safety profile of both PEG2a/R and peg alfa-2b ribavirin is well defined. 12 Because of the widespread use of both pegs in the treatment of HCV, the present study provides important safety information regarding the addition of boceprevir to PEG2a/R. The most common AEs were those previously reported in the RESPOND-2 trial with boceprevir plus peg alfa-2b ribavirin, and no new safety findings were observed when PEG2a was used as the backbone of therapy. Of note, the rates of grade 3 or 4 neutropenia and rash were higher in the present study with PEG2a/R than in the RESPOND-2 study with peg alfa-2b ribavirin (grades 3/4 neutropenia, 43% vs 27%, respectively; rash, 22% vs 14%, respectively). Other AEs such as chills, pyrexia, arthralgia, and headache were reported less frequently in the current study. Of note, the rates of anemia, diarrhea, nausea, vomiting, and myalgia were similar between the boceprevir-containing groups of the 2 studies (difference 3%). In a previous phase 3 study of approximately 3000 treatment-naive genotype 1 infected patients treated with standard dose PEG2a/R or peg alfa-2b ribavirin, there were also differences noted for rates of certain AEs such as neutropenia and rash, suggesting that some of the differences in AE profiles in our study and the RESPOND-2 study could be attributed to the different s. 2 However, the present study was not designed to compare the safety of boceprevir when added to PEG2a/R vs peg alfa-2b ribavirin, so these data should be interpreted with caution. RAVs were detected by using population sequencing in 8 of 33 patients who did not achieve SVR. In this study, there was a lower percentage of non-svr patients with detectable RAVs compared with boceprevir studies in combination with PEG2a/R. However, the significance of this finding is confounded by the small numbers of non-svr patients in this study. The overall pattern of HCV genotype 1a and 1b RAVs detected was similar to previous boceprevir clinical studies. 7 9 Virologic failure with detectable RAVs in non-svr patients appeared to be inversely related to responsiveness and was also similar to other studies with boceprevir in combination with peg alfa-2b ribavirin. RAVs were detected in the quasispecies at baseline in 6% of patients, yet 38% (3 of 8) achieved SVR. Determination of the impact of preexisting RAVs on SVR rates for boceprevir plus PEG2a/R therapy would require a much larger sample size. In the present study, 13% of patients had poor response to. None of the poor responders in the control group achieved SVR, whereas 39% of those in the boceprevir group achieved SVR. This is similar to the RESPOND-2 trial in which SVR rates were 0% and 34% for patients who received control vs those who received boceprevir. 9 Consistent with the RESPOND-2 trial, patients with poor response to also had a higher likelihood of developing amino acid variants associated with reduced susceptibility to boceprevir. In addition, good response to was the strongest predictor of SVR. In contrast to the SPRINT-2 and RESPOND-2 studies, IL28B was not predictive of SVR. 14 Although the use of IL-28B testing is now common in clinical trials, its role in clinical practice is still evolving. In summary, the results of this phase 3 trial show that boceprevir is safe and effective when combined with PEG2a/R in difficult-to-treat patients. This study and the RESPOND-2 study are the first large phase 3 trials to demonstrate a direct acting antiviral agent may be combined with either of the commercially available pegs to significantly increase SVR in patients who failed prior therapy. Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at and at doi.org/ /j.cgh References 1. Zeuzem S. Heterogeneous virologic response rates to based therapy in patients with chronic hepatitis C: who responds less well? Ann Intern Med 2004;140: McHutchison JG, Lawitz EJ, Shiffman ML, et al. Peg alfa-2b or alfa-2a with ribavirin for treatment of hepatitis C infection. N Engl J Med 2009;361: Parise E, Cheinquer H, Crespo D, et al. Peg alfa-2a (40KD) (PEGASYS) plus ribavirin (COPEGUS) in retreatment of chronic hepatitis C patients, nonresponders and relapsers to previous conventional plus ribavirin therapy. Braz J Infect Dis 2006;10: Goncales FL Jr, Vigani A, Gonçales N, et al. Weight-based combination therapy with peg alpha-2b and ribavirin for naive, relapser and non-responder patients with chronic hepatitis C. Braz J Infect Dis 2006;10: Poynard T, Colombo M, Bruix J, et al. Peg alfa-2b and ribavirin: effective in patients with hepatitis C who failed alfa/ribavirin therapy. Gastroenterology 2009;136: Sherman M, Yoshida EM, Deschenes M, et al. Peg alfa-2a (40KD) plus ribavirin in chronic hepatitis C patients who failed previous therapy. Gut 2006;55: Kwo PY, Lawitz EJ, McCone J, et al. Efficacy of boceprevir, an NS3 protease inhibitor, in combination with peg alfa-2b and ribavirin in treatment-naive patients with genotype 1 hepatitis C infection (SPRINT-1): an open-label, randomised, multicentre phase 2 trial. Lancet 2010;376:

7 January 2013 BOCEPREVIR WITH PEGINTERFERON ALFA-2A RIBAVIRIN Poordad F, Bacon B, Bruno S, et al. Boceprevir for untreated chronic HCV genotype 1 infection. N Engl J Med 2011;364: Bacon B, Gordon S, Lawitz E, et al. Boceprevir for previously treated chronic HCV genotype 1 infection. N Engl J Med 2011; 364: Ge D, Fellay J, Thompson AJ, et al. Genetic variation in IL28B predicts hepatitis C treatment-induced viral clearance. Nature 2009;461: Thompson AJ, Muir AJ, Sulkowski MS, et al. Interleukin-28B polymorphism improves viral kinetics and is the strongest pretreatment predictor of sustained virologic response in hepatitis C virus-1 patients. Gastroenterology 2010;139: Susser S, Welsch C, Wang Y, et al. Characterization of resistance to the protease inhibitor boceprevir in hepatitis C virus-infected patients. Hepatology 2009;50: Ghany MG, Strader DB, Thomas DL, et al. Diagnosis, management, and treatment of hepatitis C: an update. Hepatology 2009; 49: Poordad F, Bronowicki JP, Gordon SC, et al. Factors that predict response of patients with hepatitis C virus infection to boceprevir. Gastroenterology 2012;143: Reprint requests Address requests for reprints to: Steven L. Flamm, MD, Arkes Building 19041, 676 North Saint Clair, Chicago, Illinois s-flamm@northwestern.edu; fax: (312) Acknowledgments The authors thank all the patients, health care providers, and investigators involved with the study. They also thank Ms Ruiyun Jiang (Merck) for statistical support and Ms Karyn Davis (Merck) for technical support in the preparation of this manuscript. Investigators, alphabetical by country: Belgium: J. Delwaide, H. Van Vlierberghe Canada: P. Marotta, M. Sherman France: L. Alric, S. Ben Ali, M. Bigard, M. Bourliere, N. Boyer-Darrigrand, J. Bronowicki, C. Hezode, P. Marcellin, M. Maynard-Muet, C. Trepo Germany: A. Lohse, M. Manns, C. Niederau, H. Wedemeyer, S. Zeuzem Italy: A. Alberti, G. Carosi, M. Colombo, M. Pirisi, M. Zuin United States: A. Al-Osaimi, B. Bacon, L. Balart, M. Bennett, M. Davis, S. Flamm, B. Freilich, J. Galati, G. Galler, A. Gibas, E. Godofsky, S. Gordon, J. Herrera, I. Jacobson, P. Kwo, E. Lawitz, W. Lee, J. Levin, V. Luketic, P. Martin, J. McCone, J. McHutchison, T. Morgan, A. Muir, M. Pauly, F. Poordad, N. Ravendhran, A. Reuben, R. Rubin, E. Schiff, W. Schmidt, M. Shiffman, M. Sulkowski, J. Vierling. The current affiliation for Dr Goteti and Dr Brass is Novartis, East Hanover, New Jersey. Dr Pedicone s current affiliation is Focus Medical Communications, Parsippany, New Jersey. Dr Poordad s current affiliation is Texas Liver Institute/Alamo Medical Research, University of Texas, San Antonio, Texas. Conflicts of interest The authors disclose the following: Dr Flamm has received grants from Merck, Vertex, Gilead, Novartis, Anadys, Achillion, Tibotec, and Bristol-Myers Squibb; consultancy fees from Vertex, Gilead, Boehringer Ingelheim, and Novartis; and payment for lectures including service on speakers bureaus from Merck and Vertex. Dr Lawitz has received clinical research grants from the following companies: Abbott Laboratories, Achillion Pharmaceuticals, Anadys Pharmaceuticals, Biolex Therapeutics, Boehringer Ingelheim, Bristol-Myers Squibb, Gilead Sciences, GlaxoSmithKline, GlobeImmune, Idenix Pharmaceuticals, Idera Pharmaceuticals, Inhibitex Pharmaceuticals, Medarex, Medtronic, Merck, Novartis, Pharmasset, Roche, Sanofi-Aventis, Schering-Plough, Santaris Pharmaceuticals, Scynexis Pharmaceuticals, Tibotec, Vertex Pharmaceuticals, ViroChem Pharma, and ZymoGenetics; and payment for lectures including service on speakers bureaus for Merck. Dr Jacobson has received consultancy fees from Bristol-Myers Squibb, Novartis, Gilead, Glaxo, Inhibitex, Merck, Pfizer, Vertex, GlobeImmune, Boehringer-Ingelheim, Pharmasset, Tibotec, Abbott, Roche/Genentech, Achillion, and GlaxoSmithKline; has grants/grants pending from Merck, Tibotec, Roche/Genentech, Pharmasset, Anadys, Boehringer Ingelheim, Novartis, Gilead, Vertex, GlobeImmune, Human Genome Sciences, and Bristol-Myers Squibb; payment for lectures including service on speakers bureaus from Merck, Vertex, Gilead, Bristol-Myers Squibb, and Roche/Genentech; and payment for development of educational presentations from Bristol-Myers Squibb, Gilead, and Vertex. Dr Bourlière has received consulting fees or honorarium and support to travel for meetings for the study or other purposes from Merck; reports board membership of Roche, Merck, Janssen, Vertex, Boehringer Ingelheim, Glaxo, Novartis, and Gilead; consultancy for Roche, Merck, Janssen, and Vertex; payment for lectures including service on speakers bureaus for Merck, Janssen, and Roche; and payment for development of educational presentations from Janssen, Roche, and Merck. Dr Hezode reports membership of the French National Board and payment for lectures including service on speakers bureaus for French National Meetings. Dr Vierling has received research grants, consulting fees or honorarium, and support for travel to meetings for the study or other purposes from Merck; consultancy fees from Bristol- Myers Squibb, Gilead, Vertex, and Roche; expert testimony for the Food and Drug Administration and Centers for Disease Control; has grants/grants pending from Exhalent, Hyperion, Ocera, Ikaria, Intercept, Mochida, Sundise, Abbott, Conatus, Gilead, GlobeImmune, Hyperion, Indenix-Novartis, Novartis, Pharmasset, Pfitzer, Roche, Vertex, ZymoGenetics, Bristol-Myers Squibb, and Johnson and Johnson; payment for continuing medical education lectures including service on speakers bureaus from Chronic Liver Diseases Foundation; patents (planned, pending, or issued) for hepatitis C virus riboprobes and S.C.I.D mouse model; royalties from Liver Immunology, 2nd edition, Gershwin, Vierling, and Manns, editors; travel/accommodations/meeting expenses from National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; is the Digestive Diseases Week Secretary-Treasurer, and is a member of Clinical Research Centers of America. Dr Bacon has received consultancy fees from Gilead, Three Rivers Pharmaceuticals, Valeant, Vertex, and Human Genome Sciences; has grants/grants pending from Roche, Gilead, Bristol-Myers Squibb, Three Rivers Pharmaceuticals, Valeant, Vertex, Human Genome Sciences, Wyeth, and Romark Laboratories; payment for lectures including service on speakers bureaus for Three Rivers Pharmaceuticals, Gilead, and Merck; and served on Data and Safety Monitoring Boards for Novartis, ISIS, Vertex and Gilead. Dr Niederau reports having received study grants, board membership fees, consultancy fees, grants pending, payment for lectures including service on speakers bureaus and travel/accommodations/meeting expenses from Merck. Dr Sherman has received grants from Merck, Roche, and Vertex; consulting fees or honorarium from Merck, Roche, and Vertex; and payment for lectures including service on speakers bureaus from Merck and Roche. Drs Brass, Pedicone, Albrecht, and Goteti are former employees of Merck Sharp & Dohme Corporation, a subsidiary of Merck & Co, Inc, and hold stock/stock options. Drs Sings, Barnard, Howe, and Burroughs are employees of Merck Sharp & Dohme Corporation, a subsidiary of Merck & Co, Inc, and hold stock and/or stock options. Dr Poordad has received consultancy fees from Merck, Vertex, Abbott, Gilead, Achillion, Genentech, and Tibotec; has grants and grants pending from Merck; and has received payment for the development of educational presentations and speaker fees from Merck, Genentech, Salix, and Gilead. Funding This study was sponsored by Schering-Plough (now part of Merck Sharp & Dohme Corporation, a subsidiary of Merck & Co, Inc).

8 January 2013 BOCEPREVIR WITH PEGINTERFERON ALFA-2A RIBAVIRIN 87.e1 1 Randomized but not treated 292 patients were screened 202 underwent randomization 90 Were excluded 78 Did not meet inclusion/exclusion criteria 7 Withdrew consent 2 Administrative 2 Did not wish to continue for reasons unrelated to assigned study treatment 1 Non-compliance with protocol 67 Were assigned to Group 1 (PEG2a/R) 134 Were assigned to Group 2 (BOC/PEG2a/R 20 Completed treatment 47 Discontinued treatment 3 Had adverse event 43 Had treatment failure 1 Withdrew consent 79 Completed treatment 55 Discontinued treatment 23 Had adverse event 21 Had treatment failure 2 Lost to 7 Did not wish to continue for reasons unrelated to assigned study treatment 2 Withdrew consent 3 Never entered 38 Discontinued HCV RNA negative at end of treatment = 42% (28/67) 64 Entered 26 Completed 24 wk 9 Never entered 11 Discontinued HCV RNA negative at end of treatment = 74% (99/134) 125 Entered 114 Completed 24 wk 14/67 (21%) Had SVR 86/134 (64%) Had SVR Supplementary Figure 1. Randomization, end-of-treatment response, and SVR among the study population.

9 87.e2 FLAMM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 1 Supplementary Table 1. SVR Rates by IL-28B Genotype and Selected Baseline Characteristics Treatment group CC CT TT SVR by IL-28B genotype, n/m (%) PEG2a/R 5/10 (50) 5/34 (15) 1/7 (14) BOC/PEG2a/R 12/19 (63) 38/59 (64) 14/17 (82) SVR by IL-28B rs genotype and prior response to therapy, n/m (%) Prior nonresponse Prior relapse Prior nonresponse Prior relapse Prior nonresponse Prior relapse PEG2a/R 0/1 (0) 5/9 (56) 1/13 (8) 4/21 (19) 0/2 (0) 1/5 (20) BOC/PEG2a/R 1/2 (50) 11/16 (69) 6/16 (38) 32/40 (80) 5/7 (71) 9/10 (90) SVR by IL-28B rs genotype and response at week 4, n/m (%) Poor response to Good response to Poor response to Good response to Poor response to Good response to PEG2a/R 0/0 (NA) 5/10 (50) 0/5 (0) 5/28 (18) 0/2 (0) 1/5 (20) BOC/PEG2a/R 0/3 (0) 12/15 (80) 4/7 (57) 34/50 (68) 2/3 (67) 12/14 (86) SVR by IL-28B rs genotype and HCV RNA detectability at week 8, n/m (%) Detectable HCV RNA at week 8 Undetectable HCV RNA at week 8 Detectable HCV RNA at week 8 Undetectable HCV RNA at week 8 Detectable HCV RNA at week 8 Undetectable HCV RNA at week 8 PEG2a/R 2/5 (40) 3/5 (60) 4/30 (13) 1/4 (25) 1/7 (14) 0/0 (NA) BOC/PEG2a/R 1/4 (25) 11/13 (85) 12/25 (48) 26/29 (90) 7/9 (78) 7/8 (88) NOTE. Prior relapse was defined as undetectable HCV RNA level at end of prior therapy without subsequent attainment of SVR. Prior nonresponse was defined as decrease in HCV RNA level of at least 2 log 10 IU/mL by week 12 of prior therapy but a detectable HCV RNA level throughout the course of prior therapy, without subsequent attainment of SVR. Poor response to was defined as decrease in HCV RNA of less than 1 log 10 IU/mL after 4-week lead-in period (treatment week 4). Good response to was defined as decrease in HCV RNA of 1 log 10 IU/mL or more after lead-in period. n, number of patients who achieved SVR; m, number of patients with indicated IL-28B genotype in the respective subgroup.

10 January 2013 BOCEPREVIR WITH PEGINTERFERON ALFA-2A RIBAVIRIN 87.e3 Supplementary Table 2. RAV Profile for Patients Who Did Not Achieve SVR (BOC/PEG2a/R Group) Response category n/m (%) Incomplete virologic response (IVR) 3/4 (75) Viral breakthrough (BT) 1/1 (100) Relapse (RL) 2/9 (22) Nonresponse (NR) 2/19 (11) Total 8/33 (24) By response to Poor response 5/10 (50) Good response 2/22 (9) RAV profile by HCV genotype 1 RAVs detected by patient (n 8) Response category RAVs detected Ontreatment Follow-up HCV subtype 1a HCV subtype 1b V36M/R155K IVR V36M/T54S/ IVR R155T/ R155K V36M RL V36M RL T54A IVR T54A NR V55A NR V170A BT NOTE. Incomplete virologic response is defined as increase of 1 log 10 IU/mL in HCV RNA from nadir with HCV RNA 1000 IU/mL; viral breakthrough is defined as undetectable HCV RNA on-treatment and subsequently had HCV RNA 1000 IU/mL; relapse is defined as undetectable HCV RNA at end of treatment but detectable HCV RNA at some point during the period; nonresponse is defined as detectable HCV RNA at end of treatment who were not otherwise categorized as breakthrough or incomplete virologic response (includes patients who discontinued because of futility, AEs, or patient choice). n, patients with RAVs; m, patients with resistance data available.

11 87.e4 FLAMM ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 11, No. 1 Supplementary Table 3. Logistic Regression for Predictors of Response Effect OR (95% CI) 2 P value Baseline predictors of SVR BOC/PEG2a/R vs PEG2a/R 12.0 ( ).0001 Baseline HCV RNA: 400,000 vs 400, ( ).06 Previous response: relapser vs nonresponder 4.4 ( ).003 Cirrhosis, no vs yes 2.3 ( ).2 BMI 30 vs 30 kg/m ( ).15 Genotype: 1b vs 1a/other 1.0 ( ).96 IL-28B rs genotype: CC vs TT 0.8 ( ).94 IL-28B rs genotype: CC vs CT 0.6 ( ).38 Race: nonblack vs black 0.63 ( ).54 Predictors of SVR including treatment week 4 response BOC/PEG2a/R vs PEG2a/R 19.0 ( ).001 Log decline in HCV RNA at week 4 ( 1 vs 1 log 10 decline at week 4) 7.9 ( ).004 Previous response: relapser vs nonresponder 5.4 ( ).003 Baseline HCV RNA: 400,000 vs 400, ( ).10 Cirrhosis, no vs yes 3.2 ( ).08 BMI 30 vs 30 kg/m ( ).10 Genotype: 1b vs 1a/other 1.1 ( ).87 Race: nonblack vs black 1.0 ( ) 1.0 IL-28B rs genotype: CC vs TT 0.7 ( ).94 IL-28B rs genotype: CC vs CT 0.6 ( ).47 Baseline predictors of response ( 1 log decline at week 4) Baseline HCV RNA: 400,000 vs 400, ( ).37 IL-28B rs genotype: CC vs TT 2.1 ( ).42 IL-28B rs genotype: CT vs TT 1.3 ( ).85 Previous response: relapser vs nonresponder 1.2 ( ).72 BMI 30 vs 30 kg/m ( ).74 Cirrhosis, no vs yes 0.6 ( ).50 Race: nonblack vs black 0.4 ( ).44 BMI, body mass index. Supplementary Table 4. Other Common AEs ( 10% Incidence in Either Group) Event PEG2a/R (N 67) BOC/PEG2a/R (N 134) Fatigue 36 (54) 66 (49) Headache 21 (31) 35 (26) Insomnia 20 (30) 29 (22) Nausea 18 (27) 51 (38) Influenza-like illness 18 (27) 34 (25) Irritability 16 (24) 29 (22) Dyspnea 16 (24) 26 (19) Cough 13 (19) 22 (16) Asthenia 12 (18) 28 (21) Decreased appetite 12 (18) 25 (19) Dry skin 11 (16) 20 (15) Arthralgia 11 (16) 14 (10) Dizziness 9 (13) 17 (13) Anxiety 9 (13) 14 (10) Pyrexia 7 (10) 17 (13) Pruritus 8 (12) 18 (13) Chills 8 (12) 14 (10) Depression 6 (9) 19 (14) Alopecia 5 (7) 22 (16) Muscle spasms 4 (6) 13 (10) Dyspnea exertional 4 (6) 13 (10) Sleep disorder 3 (4) 15 (11) NOTE. For these AEs, no statistically significant differences were noted for BOC/PEG2a/R vs PEG2a/R.