Guidance for Industry Chronic Hepatitis C Virus Infection: Developing Direct- Acting Antiviral Agents for Treatment DRAFT GUIDANCE

Guidance for Industry
Chronic Hepatitis C Virus
Infection: Developing Direct-
Acting Antiviral Agents for
Treatment
DRAFT GUIDANCE
This guidance document is being distributed for comment purposes only.
Comments and suggestions regarding this draft document should be submitted within 60 days of publication in the Federal Register of the notice announcing the availability of the draft guidance. Submit comments to the Division of Dockets Management (HFA-305), Food and Drug Administration, 5630 Fishers Lane, rm. 1061, Rockville, MD 20852. All comments should be identified with the docket number listed in the notice of availability that publishes in the Federal Register.
For questions regarding this draft document contact Jeffrey Murray at (301) 796-1500.
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
September 2010
Clinical Antimicrobial
I:\9216dft.doc
08/31/10
Guidance for Industry
Chronic Hepatitis C Virus
Infection: Developing Direct-
Acting Antiviral Agents for
Treatment
Additional copies are available from:
Office of Communications, Division of Drug Information
Center for Drug Evaluation and Research
Food and Drug Administration
10903 New Hampshire Ave., Bldg. 51, rm. 2201
Silver Spring, MD 20993-0002
Tel: 301-796-3400; Fax: 301-847-8714; E-mail: druginfo@fda.hhs.gov
http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
September 2010
Clinical Antimicrobial
TABLE OF CONTENTS
I.
INTRODUCTION
.............................................................................................................
1
II.
BACKGROUND
...............................................................................................................
2
III.
DEVELOPMENT PROGRAM
.......................................................................................
4
A.
General Considerations
.................................................................................................................
4
1.
Pharmacology/Toxicology Development Considerations
................................................................
4
2.
Nonclinical Virology Development Considerations
.........................................................................
5
a.
Mechanism of action
.................................................................................................................
5
b. Antiviral activity in cell culture
................................................................................................
5
c.
Resistance and cross-resistance
.................................................................................................
6
d. Combination antiviral activity
..................................................................................................
6
e.
Activity in animal models
.........................................................................................................
6
3.
Drug Development Population
.........................................................................................................
7
4.
Early Phase Clinical Development Considerations
.........................................................................
7
a.
First-in-human trials
..................................................................................................................
7
b. Phase 1b (proof-of-concept) trials
.............................................................................................
8
c.
Phase 2 trials and dose-finding
..................................................................................................
8
d. Combination therapy with multiple DAAs
.............................................................................
10
e.
Other phase 2 trial design considerations
................................................................................
11
5.
Efficacy Considerations
.................................................................................................................
11
6.
Safety Considerations
.....................................................................................................................
12
B.
Specific Efficacy Trial Design Considerations
...........................................................................
13
1.
Trial Design
...................................................................................................................................
13
2.
Trial Population
.............................................................................................................................
14
a.
Patient enrollment definition
...................................................................................................
14
b. Patient enrollment biopsy considerations
................................................................................
15
3.
Randomization, Stratification, and Blinding
..................................................................................
15
4.
Efficacy Endpoints
..........................................................................................................................
16
5.
Trial Procedures and Timing of Assessments
................................................................................
16
6.
Statistical Considerations
...............................................................................................................
16
a.
Analysis populations
...............................................................................................................
16
b. Efficacy analyses
.....................................................................................................................
16
c.
Handling of missing data
.........................................................................................................
17
d. Interim analyses and data monitoring committees
..................................................................
18
e.
Statistical analysis plan
...........................................................................................................
19
C.
Other Considerations
...................................................................................................................
19
1.
Clinical Virology Considerations
...................................................................................................
19
2.
PK/PD Considerations
...................................................................................................................
20
3.
Special Populations
........................................................................................................................
21
a.
Hepatic impairment
.................................................................................................................
21
b. HIV/HCV co-infected patients
................................................................................................
21
c.
Patients with decompensated cirrhosis
....................................................................................
22
d. Pediatric populations
...............................................................................................................
23
4.
Early Access/Treatment INDs
........................................................................................................
23
GLOSSARY OF ACRONYMS
.................................................................................................
25
REFERENCES
............................................................................................................................
26
Contains Nonbinding Recommendations
Draft — Not for Implementation
1 Guidance for Industry1
2 Chronic Hepatitis C Virus Infection: Developing Direct3 Acting Antiviral Agents for Treatment
4
5
6
7
8
This draft guidance, when finalized, will represent the Food and Drug Administration’s (FDA’s) 9
current thinking on this topic. It does not create or confer any rights for or on any person and 10
does not operate to bind FDA or the public. You can use an alternative approach if the approach 11
satisfies the requirements of the applicable statutes and regulations. If you want to discuss an 12
alternative approach, contact the FDA staff responsible for implementing this guidance. If you 13
cannot identify the appropriate FDA staff, call the appropriate number listed on the title page of 14
this guidance. 15
16 17 18 19 I. INTRODUCTION 20 21 This guidance provides recommendations for the development of direct-acting antiviral 22 agents (DAAs) regulated within the Center for Drug Evaluation and Research at the Food 23 and Drug Administration (FDA) for the treatment of chronic hepatitis C (CHC) infection. 24 For the purpose of this guidance, we define direct-acting hepatitis C virus (HCV) 25 antivirals as agents that interfere with specific steps in the HCV replication cycle through 26 a direct interaction with the HCV polyprotein and its cleavage products. This guidance is 27 intended to serve as a focus for continued discussions among the review divisions, 28 pharmaceutical sponsors, the academic community, and the public.2 The organization of 29 the guidance parallels the development plan for a particular drug or biologic.3 30 31 This guidance does not address the development of immune-based agents for the 32 treatment of HCV infection such as new interferon products. Therapeutics without 33 antiviral mechanisms intended to mitigate or reverse clinical or pathophysiological 34 outcomes of CHC, such as prevention of hepatocellular carcinoma (HCC), reversal of 35 fibrosis, or treatment of acute hepatitis C, are not addressed in this guidance. 36
1 This guidance has been prepared by the Division of Antiviral Products in the Center for Drug Evaluation and Research (CDER) at the Food and Drug Administration.
2 In addition to consulting guidance documents, sponsors are encouraged to contact the division to discuss specific issues that arise during the development of DAAs.
3 For the purposes of this guidance, all references to drugs include both human drugs and therapeutic biological products unless otherwise specified.
1
Contains Nonbinding Recommendations
Draft — Not for Implementation
37 Additionally, general issues of clinical trial design or statistical analyses for HCV trials 38 are not addressed in this guidance. Those topics are addressed in the ICH guidances for 39 industry E9 Statistical Principles for Clinical Trials and E10 Choice of Control Group 40 and Related Issues in Clinical Trials.4 This guidance also does not contain details 41 regarding nonclinical safety and toxicology studies. Such studies for direct-acting HCV 42 antivirals generally should be conducted in standard animal models as described in the 43 guidance for industry Nonclinical Safety Evaluation of Drug or Biologic Combinations. 44 45 FDA’s guidance documents, including this guidance, do not establish legally enforceable 46 responsibilities. Instead, guidances describe the Agency’s current thinking on a topic and 47 should be viewed only as recommendations, unless specific regulatory or statutory 48 requirements are cited. The use of the word should in Agency guidances means that 49 something is suggested or recommended, but not required. 50 51 52 II. BACKGROUND 53 54 HCV is a small positive-strand RNA virus in the Flaviviridae family. Six viral 55 genotypes, numbered 1 to 6, have been identified; some have been divided into multiple 56 subtypes (e.g., genotype 1 subtypes 1a and 1b). In the United States, genotype 1 is the 57 most common (70 to 90 percent), followed by genotypes 2 and 3. Other genotypes occur 58 uncommonly in the United States, but may predominate in other parts of the world. 59 60 In the United States, HCV infection causes 20 percent of all cases of acute viral hepatitis 61 and 70 to 90 percent of all cases of HCC. Estimates show nearly 3.2 million Americans 62 are chronically infected with HCV. CHC is currently the leading indication in the United 63 States for liver transplantation, and predictive modeling suggests that without effective 64 treatment interventions significant increases in CHC-associated liver morbidity, 65 mortality, and health care costs are likely (Kim 2002). 66 67 Current treatment of CHC typically is a pegylated interferon administered in combination 68 with ribavirin (Peg-Interferon/RBV), often referred to in hepatitis C clinical trials as 69 standard of care (SOC). The goal of treatment is sustained virologic response (SVR), 70 defined as undetectable plasma HCV RNA at week 24 following treatment cessation 71 (SVR24). Total duration of current treatment depends on genotype, with longer 72 treatment durations needed to achieve SVR for genotypes 1 and 4 and shorter treatment 73 durations needed for genotypes 2 and 3. SVR rates in treatment-naive patients receiving 74 Peg-Interferon/RBV typically are in the range of 40 percent to 45 percent for viral 75 genotype 1 and are 70 percent to 80 percent for genotypes 2 and 3 (Ghany, Stradler, et al.
4 We update guidances periodically. To make sure you have the most recent version of a guidance, check the FDA Drugs guidance Web page at http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm.
2
Contains Nonbinding Recommendations
Draft — Not for Implementation
76 2009).5 SVR rates for blacks and HIV co-infected patients with genotype 1 HCV are in 77 the range of 20 percent to 30 percent (in some studies less than 20 percent), which is 78 substantially lower than rates for whites and patients who are not co-infected. 79 80 On-treatment virologic measurements at early time points can predict the likelihood of 81 SVR and are often used to guide treatment duration. When treating with interferon-based 82 regimens, health care providers generally stop treatment if patients do not have at least a 83 2 log10 drop from baseline in HCV RNA at week 12 or do not have an undetectable HCV 84 RNA by week 24, because not meeting these interim virologic response criteria results in 85 a low likelihood of SVR. Three terms relating to on-treatment responses used in clinical 86 trials include: (1) rapid virologic response (RVR), meaning an undetectable HCV RNA 87 at 4 weeks of treatment; (2) complete early virologic response, meaning an undetectable 88 HCV RNA at week 12 of treatment; and (3) extended RVR, meaning an undetectable 89 HCV RNA at week 4 that persists through week 12. These measurements are sometimes 90 used to guide treatment duration strategies in clinical trials. 91 92 Even among patients who achieve SVR, liver injury may persist and hepatic 93 complications may occur; although the likelihood of hepatic complications appears to be 94 substantially reduced compared to patients who do not achieve SVR. Multiple 95 observational cohorts show correlations between SVR and improvements in clinical 96 outcomes of interest, such as development of HCC, hepatic events, fibrosis, and all-cause 97 mortality (Yoshida, Shiratori, et al. 1999; Yoshida, Arakawa, et al. 2002; Shiratori, Ito, et 98 al. 2005; Okanoue, Itoh, et al. 1999; Imai, Kawata, et al. 1998; Arase, Ikeda, et al. 2007; 99 Veldt, Heathcote, et al. 2007, Braks, Ganne-Carrie, et al. 2007; Bruno, Stroffolini, et al. 100 2007; Manos, Zhao, et al. 2009). Evaluating clinical outcomes from prospective, 101 randomized controlled clinical trials is challenging because of the difficulty of 102 maintaining patients on a randomized arm without intervening therapy for a sufficient 103 duration (many years) to identify late-occurring clinical events such as HCC. 104 105 Pegylated interferons and RBV are difficult to tolerate and have significant adverse event 106 profiles that limit treatment in many patients or result in substantial morbidity. 107 Therefore, new drugs are needed that increase SVR rates when added to current therapy, 108 that shorten the duration of interferon-based regimens, or that replace components of 109 current therapy in patients who cannot tolerate interferon or RBV. New drugs are also 110 needed to construct regimens in patients with decompensated cirrhosis and in patients 111 undergoing liver transplant. 112 113 Host factors, such as genetic polymorphisms, metabolic parameters, and viral factors 114 (i.e., genomic mutations), are being investigated for their roles in predicting response to 115 treatments for CHC. Recently, a genetic polymorphism near the IL-28B gene, encoding 116 interferon-l-3 (IFN-l-3), has been shown in several studies to predict an approximately 117 two-fold change in response to interferon-based treatment regimens in patients of
5 See also labeling information for PegIntron and Pegasys at http://www.accessdata.fda.gov/scripts/cder/drugsatfda/index.cfm?CFID=42688251&CFTOKEN=cea143f9 dc49c115-37E6D01E-0AF3-6971-CCAA04EECE6DE6A7.
3
Contains Nonbinding Recommendations
Draft — Not for Implementation
118 African-American and European ancestries (Ge, Fellay, et al. 2009). At least one test for 119 the IL-28B polymorphism is now available to physicians and for use in clinical protocols. 120 121 122 III. DEVELOPMENT PROGRAM 123 124 A. General Considerations 125 126 1. Pharmacology/Toxicology Development Considerations 127 128 Pharmacology/toxicology development for single direct-acting HCV antivirals should 129 follow existing guidances for drug development.6 130 131 Guidance suggests conducting nonclinical combination studies to support clinical trials 132 for combination drugs.7 However, similar to the approach used for HIV and oncology 133 drugs, we do not recommend that these nonclinical studies be conducted routinely for the 134 following reasons: 135 136 • In clinical practice, DAAs are likely to be used with multiple hepatitis C drugs, 137 including interferon and RBV and other DAAs, in multiple different 138 combinations; it would not be feasible to conduct animal studies for all potentially 139 relevant combinations 140 141 • Given the difficulty of conducting combination toxicologic studies that may 142 require multiple different drugs and multiple dose combinations, we believe that 143 nonclinical studies would be more interpretable and may offer more useful data 144 by looking at individual agents at multiple and higher doses 145 146 • Single- and multiple-dose drug-interaction trials in humans and in vitro metabolic 147 studies can screen for potential pharmacokinetic (PK) drug interactions that may 148 lead to safety issues 149 150 To support clinical trials evaluating 2 or more investigational DAAs for up to 90 days, we 151 recommend a minimum of 3 months repeat-dose nonclinical toxicity studies in a rodent 152 and nonrodent species for each individual agent. Longer term data on individual agents 153 (6-month rodent, 9-month nonrodent) can support longer duration combination clinical 154 trials, depending on the toxicity profile (see ICH M3(R2)). 155 156 Nonclinical combination studies of an investigational antiviral plus approved SOC (e.g., 157 Peg-Interferon/RBV) may not be needed unless data from nonclinical studies of an
6 See the ICH guidances for industry M3(R2) Nonclinical Safety Studies for the Conduct of Human Clinical Trials and Marketing Authorization for Pharmaceuticals and S6 Preclinical Safety Evaluation of Biotechnology-Derived Pharmaceuticals.
7 See the guidance for industry Nonclinical Safety Evaluation of Drug or Biologic Combinations.
4
Contains Nonbinding Recommendations
Draft — Not for Implementation
158 investigational antiviral drug suggest a potential for increased or synergistic toxicity with 159 the approved therapeutic agents. 160 161 2. Nonclinical Virology Development Considerations 162 163 DAAs for the treatment of CHC should be tested in cell culture for antiviral activity 164 before submission of an initial investigational new drug application (IND). Information 165 about pre-investigational new drug testing and information regarding appropriate 166 nonclinical assays is available from the FDA.8 Additional recommendations for general 167 antiviral drug development can be found in the guidance for industry Antiviral Product 168 Development — Conducting and Submitting Virology Studies to the Agency. 169 170 a. Mechanism of action 171 172 The mechanism by which a DAA exhibits anti-HCV activity should be investigated in 173 studies that include evaluation of the effect of the agent on relevant stages of the virus life 174 cycle. Mechanism of action investigations should include appropriate controls for 175 assessing the specificity of anti-HCV activity, which may include assessments of activity 176 against HCV proteins that are not targeted by the candidate agent, relevant host proteins, 177 or other viruses. 178 179 b. Antiviral activity in cell culture 180 181 The antiviral activity of a new agent should be characterized in cell culture to identify a 182 target plasma concentration for evaluation in HCV-infected patients. Antiviral activity of 183 candidate agents targeting nonstructural components should be assessed using HCV 184 genotype 1a- and 1b-derived replicon systems, and a 50 percent effective concentration 185 (EC50) determined. Nonclinical studies should include assessments of antiviral activity 186 against the major HCV genotypes and subtypes. Assessments of antiviral activity against 187 replication models using HCV components derived from multiple clinical isolates are 188 also recommended, because antiviral activity can vary for strains within each subtype. If 189 differences in susceptibility are observed for different clinical isolates within the same 190 viral genotype or subtype, additional genotypic and phenotypic characterizations should 191 be conducted to identify genetic polymorphisms that may affect HCV susceptibility to the 192 new agent. 193 194 The antiviral activity of agents that target HCV entry functions can be evaluated using 195 HCV pseudoparticle systems. Assessments of antiviral activity against HCV grown in 196 cell culture are recommended for any anti-HCV agent when appropriate. The cytotoxic 197 effects of the agent should be quantified directly in the cells used for assessing anti-HCV 198 activity, and a 50 percent cytotoxic concentration (CC50) and a therapeutic index should 199 be calculated. Cytotoxicity should also be assessed using various cell lines and primary 200 cells cultured under proliferating and nonproliferating conditions. Sequestration of the
8 See the FDA Web site http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandApproved/Approval Applications/InvestigationalNewDrugINDApplication/Overview/ucm077546.htm.
5
Contains Nonbinding Recommendations
Draft — Not for Implementation
201 agent by serum proteins should also be assessed and a serum-adjusted EC50 value 202 determined. We recommend evaluation of the agent’s antiviral activity at different 203 concentrations of human serum and extrapolation to a 100 percent human serum EC50 204 value. 205 206 c. Resistance and cross-resistance 207 208 The ability of HCV to develop resistance to a DAA when subjected to drug pressure 209 should be examined in appropriate cell culture models. Amino acid or nucleotide 210 substitutions associated with the development of resistance to the candidate agent should 211 be determined and validated by introducing the changes into the HCV genome and 212 determining the conferred fold-shift in susceptibility using appropriate cell culture and/or 213 biochemical assays. Results from these studies should be used to: (1) determine whether 214 the genetic barrier for resistance development is high or low; (2) predict whether the 215 genetic barrier for resistance may vary as a function of concentration of the new agent; 216 (3) reveal potential resistance pathways and the potential for cross-resistance with other 217 anti-HCV agents; and (4) support the agent’s hypothesized mechanism of action. 218 219 Resistance studies should include evaluation of the potential for cross-resistance, both to 220 approved agents and to agents in development, particularly focusing on those in the same 221 drug class. Although the mechanism of action for RBV remains unclear, RBV should be 222 included in assessments of cross-resistance for inhibitors that target the NS5B RNA223 dependent RNA polymerase. 224 225 d. Combination antiviral activity 226 227 Most, if not all, DAAs for HCV will be used to treat CHC in combination with other 228 approved drugs. Early in development, cell culture combination antiviral activity 229 relationships of the new agent and pegylated interferons and the new agent and RBV 230 should be characterized to determine whether the combination antiviral activity is 231 additive, synergistic, or antagonistic. Additional combination antiviral activity studies 232 with other candidate anti-HCV agents should be conducted if future combination therapy 233 with other agents is anticipated. For all combination antiviral activity assessments, 234 sponsors should provide combination index values when the two agents are combined at 235 or near their individual EC50 values, and studies should include controls for cytotoxicity. 236 Combination antiviral activity relationships for HIV and HCV agents with similar 237 mechanisms of action (e.g., nucleoside analogue polymerase/reverse-transcriptase 238 inhibitors, protease inhibitors) should also be assessed before testing combinations of the 239 agents in HIV/HCV co-infected patients. 240 241 e. Activity in animal models 242 243 Demonstration of anti-HCV activity in an animal model is not needed. However, if such 244 studies are conducted and provided in support of an anti-HCV therapy program, reported 245 data should include the HCV genotype/subtype used, time course plots of viral load data
6
Contains Nonbinding Recommendations
Draft — Not for Implementation
246 for each animal, and an assessment of resistance development that includes monitoring 247 the persistence of resistant virus in the absence of anti-HCV treatment. 248 249 3. Drug Development Population 250 251 Overall drug development programs should include a broad population as appropriate for 252 the characteristics of the antiviral agent. However, a DAA may have differential activity 253 against different HCV genotypes or subtypes; therefore, development can be targeted to a 254 specific genotype (e.g., genotype 1 versus genotype 2 or 3). We recommend including 255 patients diagnosed with compensated cirrhosis in phase 2 and phase 3 trials. Also, we 256 encourage the study of combinations of direct-acting HCV antivirals in patients with the 257 greatest need for new agents, such as patients who cannot tolerate interferon, patients for 258 whom interferon is contraindicated, transplant patients, and patients with decompensated 259 cirrhosis. DAAs can be studied in combination with other DAAs and with or without 260 SOC in HIV co-infected patients as soon as appropriate based on the availability of data. 261 Trials in the above-mentioned subgroups may need to be supported by preliminary data 262 from trials to define safety and pharmacokinetics, such as hepatic impairment trials and 263 drug-drug interaction trials (e.g., antiretrovirals for HIV, immunosuppresants for 264 transplant). 265 266 CHC is a disease that is present worldwide and clinical trials typically are conducted 267 internationally. However, trials should include adequate U.S. patient representation to 268 ensure applicability of trial results to the U.S. population. An adequate representation of 269 males and females, races, ages, and weights is recommended during drug development, 270 especially in phase 3 trials. Because race (e.g., black, Asian) and ethnicity (e.g., Latino) 271 affect response rates to interferon-based regimens, it is important to ensure that there is 272 sufficient diversity in clinical trial demographics to conduct meaningful analyses of such 273 groups. Furthermore, in addition to viral genotypes, host genotypes are emerging as 274 correlates of clinical response to antivirals and may partially explain differences in 275 response rates by race; therefore, collection of patient DNA is an important 276 consideration.9 277 278 4. Early Phase Clinical Development Considerations 279 280 The early clinical evaluation of new DAAs should follow a rational plan to provide 281 sufficient data to establish preliminary safety and activity to support phase 3 trials. 282 283 a. First-in-human trials 284 285 In general, we recommend single- and/or multiple-ascending-dose trials in healthy adult 286 subjects to assess safety and pharmacokinetics for the first-in-human trials. However, 287 single-dose and short multiple-dose PK trials (see below) can also be conducted in HCV288 infected patients, particularly if nonclinical data suggest a drug may be genotoxic or 289 otherwise unacceptable for studies in healthy volunteers.
9 See the guidance for industry Pharmacogenomic Data Submissions.
7
Contains Nonbinding Recommendations
Draft — Not for Implementation
290 291 b. Phase 1b (proof-of-concept) trials 292 293 The first proof-of-concept trial (meaning a trial in HCV-infected patients that 294 demonstrates initial activity as measured by reductions in HCV RNA from baseline 295 levels) should be a repeat-dose, randomized, dose-ranging, monotherapy trial, of 296 approximately 3 days duration, with collection of intensive PK, safety, and HCV RNA 297 decay data. Doses selected for phase 1b should be predicted to provide plasma drug 298 exposures expected to exceed, by several-fold, the protein binding-adjusted, cell culture 299 EC50 value of the agent for the relevant HCV genotype/subtype. Choice of doses should 300 also take into account safety margins identified in animal toxicology studies and in any 301 trials conducted in healthy volunteers. 302 303 Monotherapy exceeding 3 days is not recommended because data indicate resistant virus 304 is rapidly selected during monotherapy dosing with some DAA drug classes. 305 Furthermore, 3 days of monotherapy with a directly targeting anti-HCV agent is usually 306 sufficient for establishing proof of concept and for initial dose exploration. Selection of 307 resistance may limit the future utility of the new agent as well as other agents with similar 308 resistance pathways. In most cases, longer durations of monotherapy with directly 309 targeting anti-HCV agents are not appropriate because of resistance concerns, but can be 310 considered on a case-by-case basis depending on the characteristics of the individual 311 agent. In addition to limiting the duration of monotherapy, we recommend that phase 1 312 trials of initial activity be conducted in patients with CHC who are naïve to previous anti313 CHC therapy (including the agent under investigation), and who have minimal fibrosis 314 and no significant co-morbidities. Following demonstration of safety and antiviral 315 activity in treatment-naïve patients, sponsors can plan additional trials in treatment316 experienced patients. 317 318 Results from proof-of-concept trials can be used to guide dose selection for subsequent 319 phase 2 trials in which DAAs are studied for longer durations as part of a combination 320 regimen. We recommend that sponsors conduct mechanistic modeling of the 321 concentration-viral kinetics and the concentration-safety profile from phase 1 trials to 322 predict the most active and tolerable doses for study in phase 2. The mechanistic viral 323 kinetic model should describe time-dependent changes in HCV infection and the effect of 324 drug concentrations (Neumann, Lam, et al. 1998). The model should also include 325 components to describe virologic breakthrough, relapse, and long-term viral response 326 (e.g., SVR) to inform dose selection and treatment duration. In general, the model should 327 be used to inform dose selection and to reduce the risk of selecting for resistant virus 328 because of subtherapeutic exposure. 329 330 c. Phase 2 trials and dose-finding 331 332 A goal of early phase 2 trials is to begin to characterize the optimal dose and duration of 333 the DAA as part of combination regimens with regard to both activity and safety. 334
8
Contains Nonbinding Recommendations
Draft — Not for Implementation
335 The most straight-forward design for early phase 2 is a randomized controlled trial of 336 several doses of a DAA added to Peg-Interferon/RBV compared to a standard-of-care 337 regimen (consisting of Peg-Interferon/RBV). In general, trial patients should receive a 338 full course of treatment with the Peg-Interferon/RBV component (24 to 48 weeks 339 depending on early treatment responses); however, the DAA component can be 340 administered for shorter durations (e.g., 12 weeks, depending on results from phase 1b). 341 The dosing duration of the investigational agent in phase 2 trials should be based on 342 scientific and clinical rationale and not limited in duration only because long-term animal 343 toxicology studies have not been completed. 344 345 The U.S.-approved Peg-Interferon/RBV labels for treatment of HCV genotype 1 HCV 346 recommend 48 weeks of therapy; although in practice many clinicians shorten the course 347 for patients who have HCV RNA levels below the limit of detection at week 4 of 348 treatment. At present, the optimal duration of dosing a third drug in combination with 349 Peg-Interferon/RBV is not known and is likely to vary depending on characteristics of the 350 investigational agent and treatment population. Thus, various durations of treatment can 351 be evaluated in clinical trials. However, we generally recommend that phase 2 trials 352 include at least one treatment arm that evaluates 48 weeks of treatment with all 353 components of a regimen unless antiviral activity or safety data support a rationale for 354 shorter durations of the DAA component of the regimen. Evaluating shorter durations of 355 a regimen or a component of the regimen can also be accomplished by incorporating a 356 second randomization to assess treatment duration in those patients who have 357 demonstrated early virologic suppression. For example, one treatment strategy can allow 358 patients who reach undetectable HCV RNA by week 4 (RVR) and maintain undetectable 359 HCV RNA level at week 12 (extended RVR) to be re-randomized to receive a regimen of 360 24 versus 48 weeks in duration. Patients who do not attain extended RVR would receive 361 48 weeks of therapy in this example. 362 363 We recommend that sponsors conduct their first phase 2 combination trials with Peg364 Interferon/RBV in treatment-naïve patients as opposed to starting dose-finding in 365 treatment-experienced patients. Giving suboptimal doses to treatment-experienced 366 patients can further increase emergence of resistance and incomplete virologic response 367 to a DAA in combination with Peg-Interferon/RBV and this could jeopardize future 368 treatment regimens for those individuals. 369 370 Sustained virologic response should be the primary endpoint of the phase 2 trials; 371 however, analyses of 12 weeks of safety and antiviral activity data from the first 372 combination trial with Peg-Interferon/RBV in treatment-naïve patients can be used to 373 design larger phase 2b dose comparison trials to further characterize optimal dosing in 374 broader populations, including both treatment-naïve and treatment-experienced patients. 375 376 To provide the most meaningful comparisons for further development of a DAA, we 377 recommend phase 2 trial designs allow for direct comparisons between treatment arms 378 with respect to dose, strategy, and duration. For example, if two doses are evaluated, 379 both treatment doses should be evaluated for the same duration of therapy. 380
9
Contains Nonbinding Recommendations
Draft — Not for Implementation
381 Because the presence of an IL-28B genetic polymorphism has been shown to predict 382 substantial treatment response differences among patients receiving Peg-Interferon/RBV, 383 an effort should be made to collect samples for IL-28B testing at baseline to reduce the 384 potential for confounding in trial analyses. For example, in smaller dose-finding trials, 385 treatment arm imbalances in patients with the IL-28B polymorphism can confound 386 interpretation of trial results if sponsors do not consider the potential effect of this 387 predictive marker on treatment outcome. Sponsors should consider stratifying based on 388 IL-28B when DAAs are combined with Peg-Interferon/RBV in phase 2 and phase 3 389 trials. 390 391 d. Combination therapy with multiple DAAs 392 393 We encourage trials of DAAs with and without Peg-Interferon/RBV, depending on the 394 patient population. Trials of combinations of DAAs in patients who cannot tolerate 395 interferon or for whom interferon is contraindicated may address an unmet medical need. 396 Based on HCV replication dynamics in infected patients (Perelson 2009), the error-prone 397 nature of HCV genome replication, and the fact that the activity of a DAA is often 398 reduced by a single amino acid substitution in the drug target, multiple DAAs are needed 399 to suppress all pre-existing and emerging drug resistant variants to achieve SVR. At 400 present it is not known whether regimens that do not include interferon can produce SVR. 401 402 Ideally, agents with different mechanisms of action should be considered for combination 403 use. The information recommended to support combination trials using DAAs without 404 interferon and RBV includes: 405 406 • Combination antiviral activity data from cell culture 407 408 • Resistance and cross-resistance patterns for each agent in the combination 409 410 • Anti-HCV activity data from clinical trials (from short-term monotherapy trials or 411 from dose-finding in combination with Peg-Interferon/RBV) 412 413 • Some human safety data on each agent 414 415 • Justification for proposed doses based on clinical trials or other sources to indicate 416 doses chosen are likely to provide reasonable anti-HCV activity 417 418 • Drug-drug interaction data if the metabolism profiles suggest an interaction 419 potential between agents in the combination regimen 420 421 Some examples of potential designs for initial trials of combinations of DAAs include but 422 are not limited to the following: 423 424 • Randomized, controlled trials that compare short durations (less than 2 weeks) of 425 multiple DAAs in treatment-naïve patients followed by a full course of Peg426 Interferon/RBV either with or without one or more of the DAAs evaluated in the
10
Contains Nonbinding Recommendations
Draft — Not for Implementation
427 first 2 weeks. An approved Peg-Interferon/RBV regimen can be used as a control 428 arm. 429 430 • Randomized, controlled trials that compare several different dosing combinations 431 of multiple DAAs given for longer durations in treatment-naïve or -experienced 432 patients. This type of design includes frequent HCV RNA monitoring and 433 stopping rules for loss or lack of antiviral response. When enrolling treatment434 naïve patients or treatment-experienced patients who can tolerate interferon and 435 RBV, protocols can specify adding interferon and RBV to the DAA regimen after 436 a specified time point (e.g., 6 weeks) or at any other time if virologic rebound or 437 lack of complete virologic response is determined. 438 439 • A single-arm trial evaluating multiple doses of combination therapy before liver 440 transplant to study the overall antiviral effect before liver transplant and 441 potentially the effect on preventing infection of the transplanted liver. Response 442 rates can be compared to historical controls because transmission of HCV to a 443 transplanted liver in this setting is universal (Gane 2008), such that demonstrating 444 lack of infection in a substantial proportion of allograft recipients is meaningful. 445 446 Sponsors are encouraged to discuss with the FDA proposed development plans for 447 combination therapy of two or more DAAs. 448 449 e. Other phase 2 trial design considerations 450 451 Phase 2 trials can also be used to explore alternative dosing strategies of a DAA in 452 combination with other agents before confirmation of alternative dosing strategies in 453 larger phase 3 trials. Detailed rationale for an alternative dosing strategy should be 454 included with a phase 2 protocol submission. One example of an alternative dosing 455 strategy is a lead-in period with Peg-Interferon/RBV (before initiation of the new agent 456 as part of a three-drug therapy). One arm containing a lead-in period with Peg457 Interferon/RBV can be compared to another arm in which all drugs in the regimen were 458 started simultaneously. In theory, a lead-in strategy may be beneficial before starting a 459 DAA with a low genetic barrier to resistance because Peg-Interferon/RBV may reach a 460 steady-state by the time the new agent is added, reducing the possibility of combining the 461 agent in the setting of subtherapeutic Peg-Interferon/RBV exposures. The effects of 462 variations in dosing of a combination regimen, such as lead-in periods, can be explored in 463 phase 2 and confirmed in phase 3. 464 465 5. Efficacy Considerations 466 467 We recommend that sponsors analyze and provide summaries of SVR outcome data 468 (SVR12 and SVR24) from phase 2 to demonstrate that treatment responses are durable 469 and to allow for sample size calculations for phase 3 trials. 470 471 Sponsors can submit an NDA to gain approval of a drug in a single population, either 472 treatment-naïve or treatment-experienced patients. Such an application should include at
11
Contains Nonbinding Recommendations
Draft — Not for Implementation
473 least two adequate and well-controlled trials conducted in the proposed population 474 intended for labeling. Alternatively, sponsors can choose to pursue an indication for both 475 treatment-naïve and -experienced patients. In this circumstance, the NDA should contain 476 at least one adequate and well-controlled phase 3 trial in each patient population, with 477 adequate supporting data from phase 2 trials. 478 479 Trial designs for combinations of investigational DAAs without interferon and RBV 480 should include provisions for demonstrating that each component of the combination 481 therapy contributes to the desired effect. Establishing the contribution of each 482 component can be accomplished using factorial designs or modified factorial designs; 483 however, we acknowledge that factorial designs in which patients are randomized to only 484 one new DAA may not be appropriate because of emergence of resistance. As an 485 alternative to factorial designs, sponsors can show a DAA’s contribution toward efficacy 486 of a multiple DAA combination regimen using other types of data. Examples of data 487 supporting contribution of efficacy include but are not limited to the following: 488 489 • Cell culture data showing that DAA combinations slow or prevent the emergence 490 of resistance compared to single agents. 491 492 • Clinical trial data showing the efficacy of each new DAA in combination with 493 interferon and RBV. 494 495 • Comparisons of viral load reductions of short-term monotherapy trials (e.g., 3-day 496 trials) with viral load reductions of combination therapy in the same trial or across 497 other short-term trials. In this example, short-term viral load reductions in 498 patients given combination therapy with two DAAs should be substantially 499 greater than that observed in patients given the single agents. 500 501 • Early phase 2 clinical trial data showing that DAA combinations prevent or 502 reduce emergence of resistance. 503 504 Sponsors should consult 21 CFR 300.50 regarding combining drug products in a single 505 dosage form. 506 507 HCV treatment development plans may be eligible for consideration under 21 CFR part 508 312, subpart E, Drugs Intended to Treat Life-Threatening and Severely-Debilitating 509 Illnesses, fast track,10 or priority review if the specifics of the development plan justify 510 such an approach. 511 512 6. Safety Considerations 513 514 In general, we recommend that initial marketing applications for drugs intended to treat 515 CHC in patients without decompensated cirrhosis contain a safety database of
10 See the guidance for industry Fast Track Drug Development Programs — Designation, Development, and Application Review.
12
Contains Nonbinding Recommendations
Draft — Not for Implementation
516 approximately 1,000 to 1,500 patients exposed to the proposed dose and duration of 517 treatment. However, if significant safety signals emerge during drug development, the 518 safety database may need to be increased or specific safety studies may need to be 519 conducted. 520 521 For an indication in patients with decompensated cirrhosis or in patients who generally 522 have a high risk of morbidity and few if any treatment options, a safety database of 523 approximately 500 patients administered the DAA for the proposed dose and duration 524 may be sufficient for filing an NDA. We encourage sponsors to discuss their proposed 525 safety database before submitting an NDA. On occasion, specific findings in nonclinical 526 or clinical development may indicate the need for a database that is larger or longer in 527 duration to adequately evaluate potential drug toxicity. 528 529 We recommend that sponsors provide controlled and comparative safety data. Safety 530 data from uncontrolled protocols or treatment IND protocols may be useful, but often 531 lack the degree of detailed reporting obtained in controlled clinical trials. Moreover, the 532 assessment of causal relationships between a drug and an adverse event is more difficult 533 when relying on uncontrolled safety data and spontaneously occurring events or events 534 related to concurrent treatment or underlying illness may be attributed to the new drug. 535 536 B. Specific Efficacy Trial Design Considerations 537 538 1. Trial Design 539 540 Until the first DAA is approved, the recommended, and most straight-forward, design for 541 initial registration of a DAA is demonstration of superiority as an add-on to SOC, Peg542 Interferon/RBV, in a blinded comparison to placebo plus SOC. In the future, a 543 superiority design also can include a new drug as part of a four-agent regimen compared 544 to a three-agent regimen. Alternatively, an active-controlled noninferiority trial design 545 could be appropriate, comparing a new DAA plus Peg-Interferon/RBV to another 546 approved DAA (control) plus Peg-Interferon/RBV. The latter design is dependent on the 547 ability to define the contribution of the new active control to the Peg-Interferon/RBV 548 treatment so that a stringent noninferiority margin can be calculated. Sponsors 549 considering a noninferiority trial design should discuss in advance with the FDA 550 justification of the noninferiority margin, trial design, and the data analysis plan.11 551 552 Patients who achieve SVR should be followed for at least 3 years in larger phase 2 or 553 phase 3 trials to: (1) ensure durability of response; (2) determine whether subsequent 554 detection of HCV RNA represents outgrowth of pre-existing virus versus re-infection; 555 and (3) evaluate development of progressive liver disease and/or HCC. Long-term 556 follow-up can be provided as part of a postmarketing commitment following the initial 557 application.
11 For more information, see the draft guidance for industry Non-Inferiority Clinical Trials. When final, this guidance will represent the FDA’s current thinking on this topic. For the most recent version of a guidance, check the FDA Drugs guidance Web page at http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm.
13
Contains Nonbinding Recommendations
Draft — Not for Implementation
558 559 2. Trial Population 560 561 a. Patient enrollment definition 562 563 To be enrolled in a trial, patients should have CHC as documented by being tested: 564 565 • Positive for anti-HCV antibody, HCV RNA, or an HCV genotype at least 6 566 months before screening, and positive for HCV RNA and anti-HCV antibody at 567 the time of screening; or 568 569 • Positive for anti-HCV antibody and HCV RNA at the time of screening with a 570 liver biopsy consistent with chronic HCV infection (or a liver biopsy performed 571 before enrollment with evidence of CHC disease, such as the presence of fibrosis) 572 573 In addition to documentation of CHC, treatment-experienced patients should have 574 complete documentation of prior treatment history (including but not limited to 575 compliance with previous therapy and reasons for discontinuation), because these factors 576 may affect their response to retreatment. For the purpose of trial enrollment, the 577 following definitions are used to define the treatment experience of CHC patients, which 578 are based on previous responses to Peg-Interferon/RBV.12 579 580 • Naïve: received no prior therapy for HCV (including interferon or pegylated 581 interferon monotherapy) 582 583 • Null Responder13: less than 2 log10 reduction in HCV RNA at week 12 of a Peg584 Interferon/RBV 585 586 • Partial Responder: greater than or equal to 2 log10 reduction in HCV RNA at 587 week 12, but not achieving HCV RNA undetectable at end of treatment with a 588 Peg-Interferon/RBV 589 590 • Responder Relapser: HCV RNA undetectable at end of treatment with a 591 pegylated interferon-based regimen, but HCV RNA detectable within 24 weeks of 592 treatment follow-up 593 594 Note that HCV RNA undetectable for previous treatment response should have 595 been based on an assay that was considered sensitive at the time of treatment.
12 Patients who previously received interferon monotherapy or nonpegylated interferons plus RBV will be a diminishing proportion presenting for future trials. These patients can be categorized separately.
13 Other definitions for null response have been proposed, such as less than 1 log10 IU/mL decline in HCV RNA at week 4 of treatment. However, failure to achieve a greater than 2 log10 IU/mL HCV RNA decline at week 12 has typically been used as a treatment futility criterion and use of a null response definition of viral reduction less than 1 log10 IU/mL at week 4 causes a gap in classification for individuals with a viral load reduction greater than 1 log10 at week 4 but less than 2 log10 reduction at week 12.
14
Contains Nonbinding Recommendations
Draft — Not for Implementation
596 597 b. Patient enrollment biopsy considerations 598 599 Baseline biopsies can help to establish CHC diagnosis and can be useful for making 600 correlations between the amount of baseline fibrosis and subsequent treatment outcomes 601 such as SVR and occurrence of treatment-related adverse events. Correlations between 602 baseline fibrosis and efficacy or safety outcomes can provide useful information in 603 labeling. Sponsors should have a sufficient number of baseline biopsies throughout drug 604 development to explore correlations between fibrosis and outcomes. We recommend the 605 following regarding enrollment biopsies throughout drug development: 606 607 • For phase 1 trials in CHC patients and early phase 2 trials intended to evaluate 608 pharmacokinetics/pharmacodynamics (PK/PD) or initial efficacy and safety, a 609 liver biopsy may not be needed as long as patients fulfill the criteria for CHC 610 infection as described in the section above. 611 612 • For later phase 2 trials and phase 3 treatment-naïve trials, we recommend biopsies 613 within 2 to 3 years before enrollment. If cirrhosis has been previously 614 demonstrated on a biopsy, then biopsies obtained more than 3 years before 615 enrollment need not be repeated. 616 617 • For later phase 2 and 3 trials in treatment-experienced patients, a biopsy within 2 618 to 3 years may not be needed for trial enrollment; however, documentation of a 619 prior biopsy showing histological evidence of CHC should be available for 620 review. 621 622 • Biopsies can be waived for patients who would be placed at risk from the 623 procedure, such as patients with bleeding disorders. Inability to do a liver biopsy 624 should not exclude patients from a trial. 625 626 Noninvasive measures of hepatic fibrosis and disease activity assessments using 627 biochemical or scanning measurements are not considered validated and should not be a 628 substitute for the histological information yielded by liver biopsy. 629 630 3. Randomization, Stratification, and Blinding 631 632 We encourage sponsors to conduct double-blind trials whenever feasible. For add-on 633 superiority trials of a new DAA plus SOC compared to SOC alone, patients randomized 634 to SOC should receive a matching DAA placebo. It is appreciated that endpoints in these 635 trials are objective, but other aspects of the trial can be influenced by knowledge of 636 treatment assignment. In open-label protocols, patients may be more likely to drop out of 637 the trial if they know they are not receiving the new treatment or investigators could 638 provide different levels of encouragement to continue. 639 640 Sponsors should consider stratification of patients by important baseline factors such as 641 IL-28B polymorphisms, viral load (high or low), HCV genotype/subtype, and cirrhosis,
15
Contains Nonbinding Recommendations
Draft — Not for Implementation
642 because these baseline factors are predictive of SVR depending on the regimen and 643 population studied. In international trials, patients should be stratified by geographic 644 area. 645 646 4. Efficacy Endpoints 647 648 The primary endpoint for phase 3 studies should be SVR at 24 weeks after completion of 649 a scheduled course of therapy (SVR24). Viral RNA clearance should be measured using 650 a sensitive and specific quantitative assay. Before initiation of clinical trials, sponsors 651 should provide in their development plans the name and performance data for the assay 652 proposed for measuring HCV RNA viral load. 653 654 5. Trial Procedures and Timing of Assessments 655 656 Recommended key time points for measuring viral RNA are at weeks 4, 12, 24, and 48 or 657 at the end of therapy (which may occur at 24 or 48 weeks). Viral measurements at week 658 12 and 24 have been critical for deciding whether a full course of interferon/RBV is 659 justified. Week 4 and 12 measurements can be used in protocol decision making for 660 determining duration of a DAA or a regimen. 661 662 6. Statistical Considerations 663 664 a. Analysis populations 665 666 All patients who are randomized and receive at least one dose of assigned therapy during 667 the trial should be included in the primary efficacy analysis. If a substantial proportion of 668 patients exit the trial after randomization but before receiving treatment or if there is an 669 imbalance between treatment arms in the number of such patients, then sensitivity 670 analyses can be conducted imputing all or a proportion of those who exited as treatment 671 failures. 672 673 b. Efficacy analyses 674 675 The primary analysis endpoint should be SVR24, which measures the presence or 676 absence of viral RNA 24 weeks after completing a protocol-defined treatment course, and 677 this analysis determines whether effectiveness has been demonstrated.14 The primary 678 analysis should be adjusted for at least one or two of the most important covariates (e.g., 679 baseline HCV genotype, screening HCV RNA or IL-28B polymorphism). The covariates 680 that will be included in the primary analysis should be prespecified in the protocol. 681 682 For subgroup analyses, the analysis of SVR24 should be performed within important 683 demographic and baseline characteristics (e.g., geographic region (U.S., non-U.S.), sex, 684 race, age group, HCV genotype, screening serum HCV RNA, IL-28B status, baseline
14 Patients who discontinue therapy, for whatever reason, before the protocol-defined treatment duration can still be considered a responder if they have confirmed absence of HCV RNA 24 weeks after the originally planned treatment duration.
16
Contains Nonbinding Recommendations
Draft — Not for Implementation
685 weight, baseline body mass index, baseline alanine aminotransferase (ALT), baseline 686 liver histology, baseline fibrosis, and prior response to interferon/RBV-based regimens). 687 The purpose of these subgroup analyses basically is to evaluate the consistency of the 688 SVR24 endpoint result across these subgroups. It is important to recognize, however, 689 that simply by chance a hypothetical drug that has a homogeneous overall effect in a trial 690 population will almost invariably show statistically significant effects in some subgroups 691 and not in others in any given trial. Therefore, such subgroup results should be 692 interpreted with caution. 693 694 For meaningful subgroup analyses in treatment-experienced trials there should be 695 adequate representation from null responders, partial responders, and responder relapsers, 696 as appropriate for each drug based on activity observed in phase 2 data (phase 2 data may 697 suggest that it is futile to study certain categories of nonresponders in phase 3). 698 699 Secondary endpoints can include: 700 701 • Normalization of ALT levels 702 703 • The proportion of patients with RVR (undetectable HCV RNA after 4 weeks of 704 treatment) 705 706 • The proportion of patients with complete early virologic response (undetectable 707 HCV RNA after 12 weeks of treatment) 708 709 • The proportion of patients with undetectable levels of HCV RNA at the end of 710 treatment and 12 weeks after the end of treatment 711 712 • Relapse rates at 12 and 24 weeks after the end of treatment 713 714 However, secondary endpoints are not sufficient to support efficacy in the absence of an 715 effect on the primary endpoint. The protocol should propose a multiple testing strategy 716 for secondary endpoints that adjust for multiplicity to be applied after the result for the 717 primary endpoint is significant. 718 719 Patients who stop treatment because they did not completely suppress HCV RNA or had 720 rebound of HCV RNA after complete suppression should be regarded as failures in all 721 analyses. For patients who discontinue treatment early, sponsors should collect 722 information to determine if these patients switched treatments or added additional 723 therapy. This information can be used to understand reasons for discontinuation and how 724 patients will be included in the analysis. 725 726 c. Handling of missing data 727 728 For the primary analysis, sponsors should consider patients not to have achieved an SVR 729 if the patients discontinue from a trial before the end of the scheduled 24 week follow-up
17
Contains Nonbinding Recommendations
Draft — Not for Implementation
730 period and if the patients have missing HCV RNA values at the end of the scheduled 24 731 week follow-up period. 732 733 Sponsors should make every attempt to limit loss of patients from the trial, and when the 734 loss is unavoidable, to collect information that can help explain the cause of the loss and 735 the final status of the patient. Analyses excluding patients with missing data or other 736 post-treatment outcomes can be biased because patients who do not complete the trial 737 may differ substantially in both measured and unmeasured ways from patients who 738 remain in the trial. 739 740 A range of sensitivity analyses should be performed to demonstrate that the primary 741 analysis is robust to discontinuation and noncompliance. Sensitivity analyses can be 742 performed using various methods for imputing missing post-treatment virologic results at 743 24 weeks of follow-up. Examples include but are not limited to using results from any 744 available last post-treatment week in place of the 24-week follow-up visit or treating a 745 percentage of missing data as successes or failures based on the overall results in which 746 post-treatment data are available. 747 748 We recommend that sponsors collect detailed data on drug-adherence and confirmation 749 of reasons for discontinuation (e.g., opportunity to enter another trial offering a promising 750 new treatment, death or events leading to death, disease progression, adverse events, loss 751 to follow-up, withdrawal of consent, noncompliance, pregnancy, protocol violations, not 752 discontinued or not known to be discontinued but data were missing at the final visit). 753 The underlying reasons for discontinuation should be interpreted. For example, the 754 statistical analysis should include the number of patients who withdrew consent or were 755 lost to follow-up, or who had adverse events (e.g., nausea and diarrhea) that could have 756 been related to the treatment they were taking. 757 758 d. Interim analyses and data monitoring committees 759 760 If interim (or futility) analyses are performed, these analyses should be specified in the 761 statistical analysis plan (SAP). The purpose of the interim analysis should be stated in 762 the SAP. 763 764 The SAP should include provisions that ensure the interim analysis does not compromise 765 trial integrity. Sponsors should refer to ICH E9 when considering the use of interim 766 analyses in clinical trials. 767 768 Sponsors should consider using a data monitoring committee for phase 3 trials evaluating 769 treatments for CHC, particularly if there are potential safety issues with one or more 770 treatment arms. A detailed charter with the composition of the committee members and 771 the operational details should be provided for review.15 772
15 See the guidance for clinical trial sponsors Establishment and Operation of Clinical Trial Data Monitoring Committees.
18
Contains Nonbinding Recommendations
Draft — Not for Implementation
773 e. Statistical analysis plan 774 775 Before initiation of any phase 2b trial (larger phase 2 trial intended to be supportive of 776 efficacy for registration) or phase 3 trial, we recommend sponsors provide a detailed 777 SAP. The SAP can be either a separate document or be within the protocol. The SAP 778 should be considered as part of the protocol and ideally should be finalized together with 779 the protocol before patient enrollment. The SAP should have details on endpoint 780 ordering, the analysis population, the structure of statistical hypotheses to be tested, 781 methods and statistical models of analyses including the mathematical formulations, level 782 of significance or alpha-level, alpha adjustments for multiple comparisons and interim 783 analyses, and any planned covariates for the analyses. It is possible to modify an SAP as 784 long as the trial remains blinded, but sponsors should recognize that a detailed discussion 785 may be needed concerning data access and appropriate firewalls for maintaining the 786 integrity of the blind. 787 788 It is important that the SAP prospectively identify the covariates to be used in the 789 analysis. It is also important that the number of covariates be kept to a minimum and 790 limited to those that are expected to strongly influence outcome. 791 792 Center-by-treatment interaction should be investigated and reported to assess consistency 793 of the efficacy results. 794 795 C. Other Considerations 796 797 1. Clinical Virology Considerations 798 799 Proof-of-concept and efficacy trials should assess the development of HCV genotypic 800 resistance to the investigational agent. Resistance testing should be performed for 801 patients who demonstrate virologic breakthrough (defined as a greater than or equal to 1 802 log10 increase in HCV RNA above nadir, or detectable HCV RNA, while on treatment, 803 after an initial drop to below detection), an incomplete antiviral response (e.g., detectable 804 HCV RNA at end of treatment), a slow or plateau viral load decay phase, or virologic 805 relapse after treatment cessation. Any changes, including mixtures, in the amino acid 806 coding sequence of the targeted genome region present in on-treatment or follow-up 807 samples, but not in the baseline sample, should be reported as having developed during 808 therapy. In addition, baseline samples should be analyzed to identify HCV genetic 809 polymorphisms that are associated with differential antiviral activity with the new agent. 810 811 Viral resistance-associated polymorphisms or substitutions observed in clinical trials but 812 not identified and characterized in nonclinical virology experiments should be evaluated 813 phenotypically by introducing the changes into the HCV genome, and determining the 814 conferred fold-shift in susceptibility to the agent using appropriate cell culture and/or 815 biochemical assays. In addition, phenotypic analyses should be performed using baseline 816 and on-treatment clinical isolates from a subset of trial patients representative of the HCV 817 genetic diversity and virologic responses observed in clinical trials. 818
19
Contains Nonbinding Recommendations
Draft — Not for Implementation
819 Emerging data with new DAAs suggest resistance-associated substitutions may persist 820 for long periods of time in the absence of drug selection. Because DAAs within the same 821 drug class typically have overlapping resistance profiles, the persistence of resistance822 associated substitutions can significantly limit a patient’s future treatment options. 823 Therefore, patients who have detectable resistance-associated substitutions at treatment 824 cessation or follow-up should be followed for an extended period, preferably at least 1 825 year after treatment cessation, to assess the persistence of resistance-associated 826 substitutions. The potential persistence of resistance-associated substitutions should be 827 characterized for patients enrolled in phase 1 and phase 2 clinical trials so that 828 preliminary long-term follow-up data are obtained by the time of completion of phase 3 829 trials. Genotyping methodology should be capable of assessing the quantity of resistant 830 viruses during the outgrowth of wild-type virus. 831 832 Sponsors should consider genotyping regions outside the direct HCV genome target 833 depending on the characteristics of the antiviral agent and interactions of the target with 834 other viral proteins. In cases when resistance is suspected based on viral RNA kinetics, 835 but genotypic evidence of resistance is not detected, sponsors should also consider 836 performing additional genotypic analyses using a method sufficiently sensitive to detect 837 minority variants.16 838 839 2. PK/PD Considerations 840 841 Trials conducted in HCV-infected patients should include assessment of 842 pharmacokinetics and the relationship between exposure and virologic success and 843 toxicity in all patients. 844 845 Sponsors can use a combination of dense and sparse sampling throughout development to 846 characterize the pharmacokinetics of the investigational agent. For example, a dense 847 sampling schedule should be implemented in monotherapy trials. In longer term trials, 848 however, a dense sampling schedule might not be feasible. Alternatively, sparse 849 sampling from these trials can be combined with dense PK data from earlier trials for 850 analysis. Sparse PK samples should be obtained at the time of key virologic assessments, 851 such as weeks 4, 12, 24, and 48. These data can then be subjected to appropriate 852 population PK analysis.17 PK samples for evaluation of Peg-Interferon/RBV or any other 853 agent in the regimen should also be collected in trials of combination therapy to assist in 854 exposure-response analyses. 855 856 Sponsors can use the following two broad approaches to characterize the relationship 857 between exposure and viral kinetics or virologic success of the investigational agent, 858 depending on the stage of development and purpose of the analysis. Both approaches
16 Additional guidance for reporting HCV drug resistance can be found in the guidance for industry
Antiviral Product Development — Conducting and Submitting Virology Studies to the Agency: Guidance for Submitting HCV Resistance Data.
17 See the guidance for industry Population Pharmacokinetics.
20
Contains Nonbinding Recommendations
Draft — Not for Implementation
859 allow for exploration of relevant covariates. These analyses should also account for the 860 development of resistance to the investigational agent. 861 862 (1) To aid the design of phase 2b or phase 3 trials, with respect to dose and regimen 863 choice, a mechanistic approach relating drug concentrations and viral kinetics is 864 most appropriate. Specifically, sponsors should develop a viral kinetic model that 865 describes time-dependent changes in HCV infection during treatment, includes a 866 mechanistically appropriate targeted drug effect, and, includes components to 867 describe virologic breakthrough, relapse, and long-term viral response. 868 869 (2) When sufficient SVR12 or SVR24 data are available, a simplified analysis 870 relating proportion of patients with virologic success and appropriate exposure 871 variable (e.g., Cmin or area under curve) can be used to support evidence of 872 effectiveness and justify dose selection.18 873 874 3. Special Populations 875 876 Treatments for patients with hepatic impairment or pre- or post-transplant patients, 877 patients co-infected with HIV and HCV, and patients with decompensated cirrhosis are 878 unmet medical needs. We strongly encourage sponsors to discuss early in development 879 the process to determine appropriate timing for initiating trials in these populations. 880 881 a. Hepatic impairment 882 883 A hepatic impairment trial to inform the need for dose modifications should be conducted 884 early in development so that patients with hepatic impairment can be included in phase 2 885 and 3 trials, as appropriate. These data also can support use in pre- or post-transplant 886 patients. 887 888 b. HIV/HCV co-infected patients 889 890 It is estimated that nearly 30 percent of patients with HIV are co-infected with HCV 891 (Sulkowski 2008). Patients with HIV/HCV co-infection are at higher risk of more rapid 892 progression of liver disease than patients with HCV infection alone. In addition, 893 treatment responses (SVR24) with SOC in co-infection are generally less than responses 894 (SVR24) with HCV infection alone. 895 896 As needed, and based on a particular investigational drug’s metabolic profile, drug-drug 897 interaction trials should be conducted before trials in co-infected patients to support 898 concomitant dosing of a new HCV drug and antiretroviral drugs. 899 900 We strongly suggest that an initial NDA for the treatment of HCV contain some clinical 901 data on the HIV/HCV co-infected population at time of filing, including:
18 See the guidance for industry Exposure-Response Relationships — Study Design, Data Analysis, and Regulatory Applications.
21
Contains Nonbinding Recommendations
Draft — Not for Implementation
902 903 • Drug-drug interaction data with the most commonly used HIV drugs 904 905 • Safety data on a cohort of co-infected patients receiving the drug for the 906 recommended treatment duration 907 908 • Preliminary efficacy data characterizing, at minimum, on-treatment responses 909 910 With the above-mentioned preliminary data, labeling describing drug interactions and 911 preliminary safety data may be appropriate. For more extensive labeling that expands the 912 indication to the HIV co-infected population or includes a description of efficacy in the 913 co-infected population, a clinical trial demonstrating efficacy and safety in at least 300 914 co-infected patients may be appropriate. In some cases, single-arm prospective trials 915 (with historical controls) may be appropriate for the co-infected population if trials in the 916 HCV mono-infected population showed robust and substantial efficacy of the new DAA 917 added to SOC. Trials in co-infected patients should evaluate SVR at 24 weeks after end 918 of therapy as the primary efficacy endpoint. As part of the safety evaluation, loss of HIV 919 efficacy (rebounds in HIV viral RNA) should be assessed. 920 921 c. Patients with decompensated cirrhosis 922 923 SOC, interferon-based regimens are not considered appropriate for patients with 924 decompensated cirrhosis or for most patients pre- or post-liver transplant; therefore, 925 treatment with multiple investigational DAAs is likely to be needed to achieve viral 926 suppression. Because there are currently no HCV treatments in patients with 927 decompensated cirrhosis and because spontaneous resolution of HCV infection in this 928 population is consistently negligible, dose-response trials or historically controlled 929 efficacy and safety trials showing clinically significant SVRs may be appropriate to 930 expand the labeling for this population. However, as more drugs become available for 931 study in combination regimens, we will encourage comparative trials. SVR24 should be 932 the primary efficacy endpoint, but other important endpoints include progression of liver 933 disease, transplantation, and mortality. SVR24 is an important endpoint notwithstanding 934 disease progression requiring transplantation, because SVR24 will likely translate into 935 prevention of infection of a newly transplanted liver. 936 937 The contribution of each agent toward overall efficacy of a regimen should be 938 demonstrated, but can be based on data such as that discussed in section III.B.6, 939 Statistical Considerations. For example, trials showing the efficacy of one new DAA 940 added to Peg-Interferon/RBV in patients with compensated cirrhosis can serve as 941 supportive data for demonstrating contribution toward efficacy in other populations that 942 are more difficult to study. 943 944 Plans for expanded access trials or safety trials should also be considered for this 945 population early in development. 946
22
Contains Nonbinding Recommendations
Draft — Not for Implementation
947 d. Pediatric populations 948 949 Early trials of DAAs should enroll adult patients only, reserving pediatric exposure until 950 the pharmacokinetics, pharmacodynamics, and safety of the agent are reasonably well951 defined. Sponsors are encouraged to begin discussions of their pediatric formulation and 952 clinical development plan early in development, but pediatric clinical trials should be 953 initiated once phase 2 adult data characterizing the safety profile and initial antiviral 954 efficacy are available. If clinical trials in adults have demonstrated no safety concern 955 specific to a histologic stage, liver biopsies are not recommended for routine entry criteria 956 into pediatric trials. If biopsies are done because they are clinically indicated, biopsy data 957 should be provided. 958 959 4. Early Access/Treatment INDs 960 961 Some hepatitis C-infected patients who have not responded to approved treatments and/or 962 who are at substantial risk of liver disease progression may benefit from access to new 963 therapeutic options before their approval. Treatment INDs or other access protocols for 964 DAAs may be appropriate when sufficient clinical trial data have been generated to 965 characterize a reasonably safe and active dose of an investigational agent. Ideally, the 966 timing of a treatment IND would occur after phase 3 trials were fully enrolled or well 967 underway so as not to interfere with phase 3 drug development. Treatment INDs can 968 provide early access while phase 3 trials are being completed, analyzed, submitted, and 969 reviewed by the FDA. Alternatively, individual patient INDs and treatment access 970 protocols for intermediate size populations may be possible. In contrast to treatment 971 INDs for larger populations during or after phase 3 trials, access for intermediate size 972 populations (approximately 100 patients or fewer), can occur earlier in drug 973 development. 974 975 Historically, early access programs with HIV allowed many people to gain access to life976 saving drugs. However, for some individuals, early access to a drug resulted in what 977 amounted to sequential monotherapy and the emergence of multidrug resistance. 978 Because treatment of CHC requires multiple agents to achieve SVR and to reduce the 979 emergence of drug resistance to single agents or drug classes, treatment INDs that include 980 two or more investigational agents or that allow co-enrollment in several treatment IND 981 programs simultaneously are desirable, particularly for previous null responders or for 982 patients who cannot take interferon-based regimens. However, treatment use of multiple 983 investigational agents should be supported by: 984 985 • Data and rationale that characterize the potential for PK drug interactions and 986 potential for overlapping toxicity. Data to support dose modifications if drug 987 interactions are present. 988 989 • Information suggesting the potential for additive or synergistic activity and no or 990 minimal overlapping resistance profiles. 991
23
Contains Nonbinding Recommendations
Draft — Not for Implementation
992 Refer to section III.A.4.d., Combination therapy with multiple DAAs, for the data needed 993 to support treatment use of multiple investigational agents. 994
24
Contains Nonbinding Recommendations
Draft — Not for Implementation
995 GLOSSARY OF ACRONYMS 996 997 CHC chronic hepatitis C 998 DAA direct-acting antiviral agent 999 HCC hepatocellular carcinoma 1000 HCV hepatitis C virus 1001 HCV RNA hepatitis C virus ribonucleic acid 1002 HIV human immunodeficiency virus 1003 IFN interferon 1004 IL interleukin 1005 Peg pegylated 1006 RBV ribavirin 1007 RVR rapid virologic response 1008 SOC standard of care 1009 SVR sustained virologic response 1010 SVR24 sustained virologic response 24 weeks after stopping treatment 1011
25
Contains Nonbinding Recommendations
Draft — Not for Implementation
1012 REFERENCES 1013 1014 Arase, Y, K Ikeda, F Suzuki, Y Suzuki, S Saitoh, M Kobayashi, N Akuta, T Someya, R 1015 Koyama, T Hosaka, H Sezaki, M Kobayashi, H Kumada, 2007, Long-Term Outcome 1016 After Interferon Therapy in Elderly Patients With Chronic Hepatitis C, Intervirology, 1017 50(1):16-23. 1018 1019 Braks, RE, N Ganne-Carrie, H Fontaine, J Paries, V Grando-Lemaire, M Beaugrand, S 1020 Pol, JC Trinchet, 2007, Effect of Sustained Virological Response on Long-Term 1021 Clinical Outcome in 113 Patients With Compensated Hepatitis C-Related Cirrhosis 1022 Treated By Interferon Alpha and Ribavirin, World J Gastroenterol, Nov 14, 1023 13(42):5648-53. 1024 1025 Bruno, S, T Stroffolini, M Colombo, S Bollani, L Benvegnù, G Mazzella, A Ascione, T 1026 Santantonio, F Piccinino, P Andreone, A Mangia, GB Gaeta, M Persico, S Fagiuoli, 1027 PL Almasio; Italian Association of the Study of the Liver Disease (AISF), 2007, 1028 Sustained Virological Response to Interferon-Alpha Is Associated With Improved 1029 Outcome in HCV-Related Cirrhosis: A Retrospective Study, Hepatology, Mar, 1030 45(3):579-87. 1031 1032 Gane, EJ, 2008, The Natural History of Recurrent Hepatitis C and What Influences This, 1033 Liver Transpl., Suppl2: S36-44. 1034 1035 Ge, D, J Fellay, AJ Thompson, JS Simon, KV Shianna, TJ Urban, EL Heinzen, P Qiu, 1036 AH Bertelsen, AJ Muir, M Sulkowski, JG McHutchison, DB Goldstein, 2009, Genetic 1037 Variation in IL-28B Predicts Hepatitis C Treatment-Induced Viral Clearance, Nature, 1038 461:399-401. 1039 1040 Ghany, MG, DB Strader, DL Thomas, LB Seeff, 2009, Diagnosis, Management, and 1041 Treatment of Hepatitis C: An Update, Hepatology, 49:1335-1374. 1042 1043 Imai, Y, S Kawata, S Tamura, I Yabuuchi, S Noda, M Inada, Y Maeda, Y Shirai, T 1044 Fukuzaki, I Kaji, H Ishikawa, Y Matsuda, M Nishikawa, K Seki, Y Matsuzawa, 1998, 1045 Relation of Interferon Therapy and Hepatocellular Carcinoma in Patients With 1046 Chronic Hepatitis C, Osaka Hepatocellular Carcinoma Prevention Study Group, Ann 1047 Intern Med, Jul 15, 129(2):94-9, PMID: 9669992. 1048 1049 Kim, WR, 2002, The Burden of Hepatitis C in the United States, Hepatology, 1050 36(Suppl):S30-S34. 1051 1052 Manos, MM, WK Zhao, VA Shvachko, N Volkova, CP Quesenberry, Viral Hepatitis 1053 Registry, Kaiser Permanente Division of Research, Oakland, CA, 2009, Long Term 1054 Outcomes in Patients Treated With Peg-Interferon/Ribavirin Therapy for Hepatitis C: 1055 The Substantial Effect of Sustained Viral Response (SVR) on Liver Disease, Mortality 1056 and Diabetes, 13th International Symposium on Viral Hepatitis and Liver Disease, 1057 Abstract PL-3.
26
Contains Nonbinding Recommendations
Draft — Not for Implementation
1058 Neumann, AU, NP Lam, H Dahari, DR Gretch, TE Wiley, TJ Layden, AS Perelson, 1059 1998, Hepatitis C Viral Dynamics In Vivo and the Antiviral Efficacy of Interferon1060 alpha Therapy, Science, Oct 2, 282(5386):103-7. 1061 1062 Okanoue, T, Y Itoh, M Minami, S Sakamoto, K Yasui, M Sakamoto, K Nishioji, Y 1063 Murakami, K Kashima, 1999, Interferon Therapy Lowers the Rate of Progression to 1064 Hepatocellular Carcinoma in Chronic Hepatitis C but not Significantly in an Advanced 1065 Stage: A Retrospective Study in 1148 Patients, Viral Hepatitis Therapy Study Group, 1066 J Hepatol, Apr, 30(4):653-9, PMID: 10207807. 1067 1068 Perelson, AS, 2009, HCV Kinetics, 13th International Symposium on Viral Hepatitis and 1069 Liver Disease, Washington, DC. 1070 1071 Shiratori, Y, Y Ito, O Yokosuka, F Imazeki, R Nakata, N Tanaka, Y Arakawa, E 1072 Hashimoto, K Hirota, H Yoshida, Y Ohashi, M Omata; Tokyo-Chiba Hepatitis 1073 Research Group, 2005, Antiviral Therapy for Cirrhotic Hepatitis C: Association With 1074 Reduced Hepatocellular Carcinoma Development and Improved Survival, Ann Intern 1075 Med, Jan 18, 142(2):105-14, PMID: 15657158. 1076 1077 Sulkowski, MS, 2008, Viral Hepatitis and HIV Coinfection, J Hepatol, Feb, 48(2):3531078 67. 1079 1080 Veldt, BJ, EJ Heathcote, H Wedemeyer, J Reichen, WP Hofmann, S Zeuzem, MP 1081 Manns, BE Hansen, SW Schalm, HL Janssen, 2007, Sustained Virologic Response 1082 and Clinical Outcomes in Patients With Chronic Hepatitis C and Advanced Fibrosis, 1083 Ann Intern Med, 147:677-684. 1084 1085 Yoshida, H, Y Arakawa, M Sata, S Nishiguchi, M Yano, S Fujiyama, G Yamada, O 1086 Yokosuka, Y Shiratori, M Omata, 2002, Interferon Therapy Prolonged Life 1087 Expectancy Among Chronic Hepatitis C Patients, Gastroenterology, Aug, 123(2):4831088 91. 1089 1090 Yoshida, H, Y Shiratori, M Moriyama, et al., 1999, Interferon Therapy Reduces the Risk 1091 for Hepatocellular Carcinoma: National Surveillance Program of Cirrhotic and 1092 Noncirrhotic Patients With Chronic Hepatitis C in Japan, IHIT Study Group, 1093 Inhibition of Hepatocarcinogenesis by Interferon Therapy, Ann Intern Med, Aug 3, 1094 131(3):174-81. 1095
27