Obeticholic Acid for the Treatment of Nonalcoholic Steatohepatitis: Comparative Clinical Effectiveness and Value Evidence Report July 26, 2016 Institute for Clinical and Economic Review ©Institute for Clinical and Economic Review, 2016 AUTHORS Massachusetts General Hospital, Institute for ICER Staff Technology Assessment Modeling Group Steven D. Pearson, MD, MSc Jagpreet Chhatwal, PhD President, Institute for Clinical and Economic Review Assistant Professor, Harvard Medical School Senior Scientist, MGH Institute for Technology Reiner Banken, PhD, MSc Assessment Senior Fellow, Institute for Clinical and Economic Review Chin Hur, MD, MPH Associate Professor, Harvard Medical School Rick Chapman, PhD, MS Director, GI Health Outcomes Research Director of Health Economics, Institute for Clinical and Massachusetts General Hospital Economic Review Associate Director, MGH Institute for Technology Assessment Elizabeth T. Russo, MD Research Scientist, Institute for Clinical and Economic Matthew Klebanoff, BA Review Research Analyst, MGH Institute for Technology Assessment Anne M. Loos, MA Research Associate, Institute for Clinical and Economic Sumeyye K. Samur, PhD Review Post-doctoral Fellow, Harvard Medical School Research Fellow, MGH Institute for Technology Sarah K. Emond, MPP Assessment Chief Operating Officer, Institute for Clinical and Economic Review The role of the Massachusetts General Hospital, Institute for Technology Assessment is limited to the development of the cost- effectiveness model, and the resulting ICER reports do not Celia S. Segel, MPP necessarily represent the views of the MGH. Program Manager, New England CEPAC, Institite for Clinical and Economic Review Daniel A. Ollendorf, PhD Chief Scientific Officer, Institute for Clinical and Economic Review DATE OF PUBLICATION: July 26, 2016 We would also like to thank Erin Lawler, MA and Shanshan Liu, MS, MPH for their contributions to this report. ©Institute for Clinical and Economic Review, 2016 Page i Draft Evidence Report - OCA for the treatment of NASH About ICER The Institute for Clinical and Economic Review (ICER) is an independent non-profit research organization that evaluates medical evidence and convenes public deliberative bodies to help stakeholders interpret and apply evidence to improve patient outcomes and control costs. ICER receives funding from government grants, non-profit foundations, health plans, provider groups, and health industry manufacturers. For a complete list of funders, visit http://www.icer- review.org/about/support/. Through all its work, ICER seeks to help create a future in which collaborative efforts to move evidence into action provide the foundation for a more effective, efficient, and just health care system. More information about ICER is available at http://www.icer- review.org About New England CEPAC The New England Comparative Effectiveness Public Advisory Council (New England CEPAC) – a core program of ICER – provides a public venue in which the evidence on the effectiveness and value of health care services can be discussed with the input of all stakeholders. New England CEPAC seeks to help patients, clinicians, insurers, and policymakers interpret and use evidence to improve the quality and value of health care. The New England CEPAC is an independent committee of medical evidence experts from across New England, with a mix of practicing clinicians, methodologists, and leaders in patient engagement and advocacy. All Council members meet strict conflict of interest guidelines and are convened to discuss the evidence summarized in ICER reports and vote on the comparative clinical effectiveness and value of medical interventions. More information about New England CEPAC is available at http://icer-review.org/programs/new-england-cepac/. ©Institute for Clinical and Economic Review, 2016 Page ii Draft Evidence Report - OCA for the treatment of NASH External Input The following individuals and organizations provided input and feedback that helped guide the ICER team as we shaped our scope and report. None of these individuals or organizations is responsible for the final contents of this report or should be assumed to support any part of this report, which is solely the work of the ICER team and its affiliated researchers. External Input Received from: Clinical Experts: Naga Chalasani, MBBS, Indiana University School of Medicine Kathleen Corey, MD, MPH, Massachusetts General Hospital* Keith Lindor, MD, Mayo Clinic Daniel Pratt, MD, Massachusetts General Hospital* Harvard Pilgrim HealthCare Intercept Pharmaceuticals * Also provided input into the development of the economic model ©Institute for Clinical and Economic Review, 2016 Page iii Draft Evidence Report - OCA for the treatment of NASH Table of Contents Executive Summary ............................................................................................................................ ES1 1. Background ........................................................................................................................................ 1 1.1 Introduction ................................................................................................................................. 1 2. The Topic in Context .......................................................................................................................... 4 3. Summary of Coverage Policies ........................................................................................................... 8 4. Comparative Clinical Effectiveness .................................................................................................... 9 4.1 Overview ...................................................................................................................................... 9 4.2 Methods ....................................................................................................................................... 9 4.3 Results ........................................................................................................................................ 12 5. Other Benefits or Disadvantages ..................................................................................................... 21 6. Comparative Value ........................................................................................................................... 22 6.1 Overview .................................................................................................................................... 22 6.2 Prior Published Evidence on Costs and Cost-Effectiveness of Obeticholic Acid ........................ 22 6.3 Incremental Costs per Outcome Achieved ................................................................................ 22 6.4 Potential Budget Impact ............................................................................................................ 34 6.5 Value-based Benchmark Prices .................................................................................................. 36 6.6 Summary and Comment ............................................................................................................ 36 7. Voting Results .................................................................................................................................. 39 7.1 About the New England CEPAC Process .................................................................................... 39 7.2 Clinical Effectiveness Voting Results .......................................................................................... 41 7.3 Care Value Voting Results .......................................................................................................... 41 8. Roundtable Discussions and Key Policy Implications....................................................................... 42 References ........................................................................................................................................... 46 Appendix A. Evidence Review Methods and PRISMA .......................................................................... 52 Appendix B. Clinical Guidelines ............................................................................................................ 57 Appendix C. Previous Systematic Reviews and Technology Assessments .......................................... 58 Appendix D. Ongoing Studies............................................................................................................... 59 Appendix E. Summary Evidence Tables ............................................................................................... 60 ©Institute for Clinical and Economic Review, 2016 Page iv Draft Evidence Report - OCA for the treatment of NASH Appendix F. Policy Roundtable Conflict of Interest Disclosures .......................................................... 65 Appendix G. Public Comments............................................................................................................. 67 ©Institute for Clinical and Economic Review, 2016 Page v Draft Evidence Report - OCA for the treatment of NASH List of Acronyms Used in this Report ALP Alkaline phosphatase ALT Alanine aminotransferase AST Aspartate aminotransferase CEA Cost-effectiveness analysis CC Compensated cirrhosis CT Computed tomography CRN Clinical Research Network FDA US Food and Drug Administration FLINT Farsenoid X Receptor Ligand Obeticholic Acid in Nonalcoholic Steatohepatitis Treatment GGT Gamma-glutamyl transferase HCC Hepatocellular carcinoma HDL High-density lipoprotein HOMA-IR Homeostasis model assessment of insulin resistance ICER Increased cost effectiveness ratio ITT Intention-to-treat LDL Low-density lipoprotein NAFLD Nonalcoholic fatty liver disease NAS NAFLD activity score NASH Nonalcoholic steatohepatitis OCA Obeticholic acid PBC Primary biliary cholangitis PBO Placebo PDUFA Prescription Drug User Fee Act PICOTS Population, Intervention, Comparator, Outcome, Timing, Setting PRISMA Preferred Reporting Items for Systematic Reviews and Meta-Analyses QALY Quality-adjusted life year QoL Quality of life RCT Randomized controlled trial SF-36 36-Item Short Form Health Survey US United States UPSTF United States Preventive Services Task Force VA Department of Veterans Affairs VLDL Very low-density lipoprotein ©Institute for Clinical and Economic Review, 2016 Page vi Draft Evidence Report - OCA for the treatment of NASH Executive Summary Background Nonalcoholic steatohepatitis (NASH) is a form of nonalcoholic fatty liver disease (NAFLD) that can progress to cirrhosis, liver failure, and cancer. It is defined by an accumulation of triglycerides in the cells of the liver with inflammation and ballooning of the liver cells with or without fibrosis. The development of NAFLD and NASH is closely linked to obesity, which has reached epidemic proportions in the US.1 NAFLD is estimated to be present in up to 30% of the population (or 80 million adults) and NASH in around 5% (or 15 million adults) in the US alone.2,3 Among the 25 million Americans with diabetes, around 18 million are thought to have NAFLD, and 63-87% of patients with both diabetes and NAFLD may have NASH.4,5 Because it can be challenging to distinguish NASH from NAFLD without a liver biopsy, some of the results we examine from the evidence base include NAFLD-specific rather than NASH-specific outcomes. Current treatment of NASH is limited to lifestyle interventions (e.g., diet, exercise, and/or behavioral change), control of diabetes and/or other metabolic disturbances, and liver-directed pharmacotherapy.6 Significant weight loss has been shown to reduce the symptoms of and/or resolve NASH, but is difficult to maintain for many individuals.3 Vitamin E is considered the liver- specific first-line treatment of NASH, but it does not improve fibrosis and it may increase the risk of prostate cancer.3,7 There is therefore clinical interest in other treatment options that can address the symptoms and progression of NASH. Topic in Context The natural history of NASH is highly variable between individuals. In a longitudinal study of 103 patients with sequential liver biopsies in the absence of effective treatment, fibrosis stage progressed in 37%, remained stable in 34%, and regressed in 29%, with a mean interval of around three years between biopsies.8 Approximately 11% of NASH patients progress to cirrhosis over a 15- year period.3 As NASH is largely asymptomatic,3 cirrhosis can develop without any prior diagnosis. About 7% of patients with NASH cirrhosis will develop hepatocellular carcinoma (HCC) over 6.5 years of follow-up.2 Overall, NASH patients have a doubling of cardiovascular risk and a more than tenfold increased risk of liver-related death.6 Between 2004 and 2013, NASH has become the second leading etiology of liver disease among adults awaiting liver transplantation in the United States (Figure 2) and is expected to become the most common indication for liver transplantation in the United States between 2020 and 2025.9 Resource utilization for HCC is also largely driven by NASH, with NAFLD/NASH being the most ©Institute for Clinical and Economic Review, 2016 Page ES1 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents common underlying etiologic risk factor (59%) for HCC in the United States between 2002 and 2008, followed by diabetes (36%) and hepatitis C virus infection (22%).10 Experts estimate that a very large percentage of the 15 million adults in the United States who are estimated to be afflicted with NASH are unaware of their condition.2,3 A study in a primary care setting indicated that only 14% of probable NAFLD patients received some form of treatment in this primary care setting and that only 3% of high risk patients are seen by specialists.11 A cost- effectiveness analysis for screening for NASH in the high risk population of patients with type 2 diabetes concluded that “screening for NASH may improve liver related outcomes, but is not cost- effective at present, due to side effects of therapy. As better tolerated treatments for NASH become available, even with modest efficacy, screening for NASH will become cost-effective.”5 Obeticholic Acid Obeticholic acid (OCA; Ocaliva™, Intercept Pharmaceuticals, Inc.) is a selective agonist of the farnesoid X receptor (FXR), a bile acid nuclear receptor. Its activity on lipid and glucose metabolism and hepatic inflammation makes it an interesting candidate as a pharmacologic agent for treating NASH. 12 A US-based Phase II trial of NASH treatment with obeticholic acid showed an improvement in liver histology including fibrosis over a period of 72 weeks. In January 2015, OCA received a US Food and Drug Administration (FDA) breakthrough designation for treatment of NASH with concomitant liver fibrosis 13, and a 5-year Phase III trial was started in September 2015. Interim findings from this Phase III trial are expected to be available around March 2017. OCA received FDA approval for the treatment of primary biliary cholangitis (PBC) on May 27, 2016. Clinical interest in its potential off-label use for NASH is likely to be great given the unmet medical need and the lack of other approved treatments. Comparative Clinical Effectiveness To inform our analysis of the comparative clinical effectiveness of OCA as an off-label treatment for NASH, we abstracted evidence from available clinical studies, whether in published, unpublished, or abstract form, concerning key clinical benefits and surrogate outcomes of clinical benefit as well as potential harms and drug-related adverse events. The timeframe for our search spanned the period from January 1996 to June 20, 2016. Our literature search identified 105 potentially relevant references, of which two publications and four abstracts met our inclusion criteria; these citations related to two individual studies. These two industry-sponsored Phase II studies were double-blind, placebo-controlled, multicenter RCTs that examined OCA use among adults with NAFLD (Table ES1). As mentioned previously, the Phase III trial of OCA in NASH (REGENERATE) is ongoing and interim results are not yet available. ©Institute for Clinical and Economic Review, 2016 Page ES2 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents We rated one RCT publication, the FLINT trial, to be of good quality. 14 However, interpretation of the trial of non-cirrhotic NASH patients is limited by its having been stopped early when interim analysis suggested a benefit with OCA. This prevented 64 (23%) of the patients from receiving a post-treatment biopsy to assess fibrosis. We rated the other publication of the NCT00501592 trial by Mudaliar et al. to be of fair quality because the study arms in this investigation of OCA use among diabetic patients with NAFLD (n=64) were not randomized evenly and follow-up was limited to six weeks.15 Four abstracts16-19 provided supplemental results to the FLINT trial, and these are described separately in keeping with the ICER grey literature policy (http://icer- review.org/methodology/icers-methods/icer-value-assessment-framework/grey-literature-policy/). ©Institute for Clinical and Economic Review, 2016 Page ES3 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Table ES1. Key Trials Patient Key Trials Treatment Comparator Harms Characteristics Mean age: 52 OCA 25 mg daily Placebo Pruritus: 23% FLINT14 Percent male: 34% (n=141; ITT* n=110) (n=142; ITT* n=109) vs. 6% Mean weight: Administered for 72 wks w/ 24 wks follow-up (p<0.0001) Phase II ~98kg Primary outcome: ≥2 point decrease in centrally scored NAS w/o worsening Double-blind RCT Hyperlipidemic: fibrosis 72 wks (OCA 45% vs. PBO 21%) Multicenter 62% RR 1.9 (95% CI 1.3-2.8); p=0.0002 ITT analysis Diabetic: 53% Secondary outcomes: Vitamin E last 6 -Mean change in NAS (-1.7 vs. -0.7) mos: 22% RR -0.9 (95% CI -1.3 to -0.5); p<0.0001 Antilipidemic last 6 -Patients w/ improved fibrosis (35% vs. 19%) mos: 48% RR 1.8 (95% CI 1.1-2.7); p=0.004 Definite -Resolution of NASH (22% vs. 13%) steatohepatitis: RR 1.5 (95% CI 0.9-2.6); p=0.08 80% -Mean change (baseline to 72 wks): -Mean change (baseline to 72 wks): Mean NAFLD ALT -38 ALT -18 (p<0.001) score: 5.2 AST -27 AST -10 (p=0.0001) ALP +12 ALP -6 (p<0.0001) GGT -37 GGT -6 (p<0.0001) Total cholesterol 0.16 Total cholesterol -0.19 (p=0.0009) HDL -0.02 HDL 0.03 (p=0.01) HOMA-IR 15 HOMA-IR 4 (p=0.01) Weight (kg) -2.3 Weight (kg) 0.0 (p=0.008) NCT00501592 by Mean age: 52 OCA 25 mg daily (n=20) Placebo (n=23) Any AEs (OCA Mudaliar et al.15 Percent male: 53% or 25 mg Mean weight: OCA 50 mg daily (n=21) vs. 50 mg vs. Phase II ~106kg PBO): Double-blind RCT Diabetic: 100% Administered for 6 wks 45% vs. 76% Multicenter Primary outcomes: vs. 61% -Percent change in low-dose glucose infusion rate Treatment- (OCA 24.5 vs. PBO -5.5); p=0.011 related AEs: -Percent change in high-dose glucose infusion rate 5% vs. 38% (OCA 15.0 vs. PBO -5.4); p=0.025 vs. 26% Secondary outcomes: change in mean Secondary outcomes: change in mean Pruritus: values 25 mg/50 mg values (p-value 25 mg/p-value 50 mg) 0 vs. 5% vs. AST 5 (0.12/0.73) 9% AST -2/5 ALT 11 (0.003/0.84) ALT -10/10 ALP 0 (0.003/<0.001) ALP 14/27 GGT 5 (<0.001/<0.001) GGT -37/-22 Total cholesterol 8 (0.08/0.15) Total cholesterol 18/13 HDL 0 (0.42/0.01) HDL -2/-6 Weight (0.096/0.008) Weight 1/1.9 ITT = intent-to-treat; DB = double-blind; LTSE = long-term safety extension; ULN = upper limit of normal; UDCA = ursodeoxycholic acid; OCA = obeticholic acid; ALP = alkaline phosphatase; ALT = alanine aminotransferase; AST = aspartate aminotransferase; GGT = gamma-glutamyl transferase; HDL = high-density lipoprotein; AEs = adverse events; HOMA-IR = Homeostasis model assessment of insulin resistance *ITT population was defined in FLINT trial as those 219 patients who received both baseline and 72-week follow-up biopsies ©Institute for Clinical and Economic Review, 2016 Page ES4 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Results Clinical Effectiveness The NAFLD activity score, or NAS, measures the presence and the degree of NASH and is based on histologic assessment of liver biopsies: it is the unweighted sum of scores given for steatosis (0-3), lobular inflammation (0-3), and hepatocellular ballooning (0-2). Fibrosis is measured separately, as it is the result of inflammation and damage to hepatocytes. 20 In the FLINT trial, patients in the OCA arm had greater mean change in NAS (-1.7 vs. -0.7 for placebo; p<0.0001). These statistically- significant improvements in NAS are currently of unknown clinical significance, however. Some patients in both treatment groups experienced histologic resolution of NASH by 72 weeks of therapy, but these changes did not differ statistically between groups.14 Among the patients in the FLINT trial with most severe NASH at baseline (stage 2-3 fibrosis or stage 1 with diabetes, obesity, or ALT ≥ 60 U/L), a poster presentation reported that significantly more patients treated with OCA experienced two or more points of improvement in NAS compared with the placebo group (50% vs. 31%; p=0.001). A greater proportion of patients in this subpopulation also experienced resolution of NASH if they were treated with OCA (18% vs. 6.5%; p=0.03).18 The ultimate goal of NASH therapy is to prevent cirrhosis from developing, which also requires preventing fibrosis. In the FLINT trial, improvement in fibrosis was observed in 35% of patients treated with OCA versus 19% for placebo (rate ratio [RR] 1.8; 95% CI 1.1-2.7; p=0.004).14 Again in patients with the most severe NASH at baseline, a poster presentation reported more patients in the OCA group experiencing regression of fibrosis by at least one stage (39% vs. 22% for placebo; p=0.012), and fibrosis progressed in fewer patients treated with OCA than placebo (16% vs. 29%; p=0.047).18 Liver enzymes were lowered in patients on OCA compared to the placebo group in both trials. In the FLINT trial the changes in alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma-glutamyl transferase (GGT) were statistically significantly in favor of OCA, 14 whereas in the other trial, ALT and GGT were statistically significantly lower among patients receiving 25 mg OCA. 15 However, the clinical significance of these changes is uncertain beyond their use in tracking disease progression. Health-related quality of life measures were assessed in the FLINT trial using Short Form [SF]-36 questionnaires for physical and mental well-being. 14 Neither the treatment nor placebo group showed a change from baseline over the course of the 72 weeks of treatment for either component of the SF-36. As NASH is relatively asymptomatic in its earlier stages, these findings are not surprising. ©Institute for Clinical and Economic Review, 2016 Page ES5 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Harms As NASH patients already have a doubling of cardiovascular risk, 6 the effects of OCA on dyslipidemia are of particular interest. Patients in the FLINT trial experienced a small but statistically significant increase in total cholesterol and LDL, as well as a decrease in HDL, while taking OCA.14 Among diabetic patients with NAFLD, Mudaliar et al. found that patients treated with OCA (25 mg or 50 mg) for six weeks did not have statistically significantly elevated total cholesterol levels, but HDL levels were significantly lower.15 It is not clear whether this is because patients with diabetes are less susceptible to the lipid effects of OCA, if more of these patients were taking statins at baseline, or whether the drug has an impact on lipids only after a longer course of therapy. OCA treatment is associated with increased pruritus among those in the treatment arms (FLINT: 23% vs. 6%; p<0.0001), but this reportedly led to little treatment discontinuation. 14 In the smaller Mudaliar trial lasting only six weeks, the incidence of pruritus was actually higher in the placebo group compared to the group of patients receiving 25 mg of OCA. 15 Controversies and Uncertainties NASH is currently an off-label indication for OCA, which makes its use in this clinical setting more susceptible to hypothetical benefit and anecdotal supporting evidence. The published evidence base for using OCA in NASH is currently very slim. The current studies do not directly describe the impact of OCA on cirrhosis and the outcome measures used have not yet been validated as surrogate outcomes. The clinical significance of several of the secondary outcomes is uncertain: what is the clinical impact of lowering liver aminotransferase or of lowering HDL levels? Are there undesirable and unintended consequences of initiating patients on statin therapy to manage dyslipidemia associated with OCA therapy, or are these offset by long-term gains in preventing end- stage liver disease? These are questions that remain unanswered given the limited evidence base. Another consideration is that the side effect of pruritus is noxious, whereas the symptoms of NASH are quiescent for many years. This raises the question of long-term adherence to oral therapy taken daily to suppress a chronic condition with few symptoms until late stages of the disease. Comparative Clinical Effectiveness: Summary and Comment Given the limited evidence base and uncertainty regarding the long-term clinical effects of changes in surrogate endpoints and conflicting physiological outcomes while taking the drug (e.g., insulin resistance in the Mudaliar vs. FLINT trials), we assign an ICER evidence rating of “Insufficient,” or “I” for the use of obeticholic acid as an off-label treatment for adults with nonalcoholic steatohepatitis with fibrosis. Additional trials are underway (REGENERATE and CONTROL) and should be examined carefully to further characterize the effectiveness of OCA activity on NASH. ©Institute for Clinical and Economic Review, 2016 Page ES6 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Other Benefits or Disadvantages OCA offers the potential of oral therapy to slow or suspend the progression of NASH. Currently, best practices include lifestyle modification (e.g., weight loss) and managing cardiovascular risk. Other oral agents (e.g., vitamin E, pioglitazone) have unclear effectiveness given some safety concerns about their widespread use (i.e., increase in all-cause mortality and weight gain as a side effect). Approval of OCA gives practitioners an additional oral agent to deploy in preventing end- stage liver disease, but the data currently available do not demonstrate impact on long-term outcomes. Comparative Value A cost-effectiveness analysis was conducted by developing a microsimulation model that simulated the long-term outcomes of patients receiving OCA as observed in the Phase II FLINT study; as a comparator, we also simulated the placebo arm of the trial. Model parameters were estimated from published studies and calibrated when assumptions were required. The model compares two different strategies for treating NASH in hypothetical patients: OCA, and standard care, which could include treatment with Vitamin E in both the OCA and standard care strategies. The outcomes of the model included total costs, quality-adjusted life years (QALYs), incremental cost-effectiveness ratios, transplant-free survival, and cumulative incidence of advanced disease stages. Outputs from this model were also used to inform a population-based analysis of the one- and five- year potential budgetary impact of OCA at a national level. Potential budgetary impact included estimates of costs saved from averted liver-related events (e.g., transplant, HCC) and was calculated assuming an uptake pattern for OCA for off-label use. Details on methods and inputs for all analyses can be found in the full report and appendices. Incremental Costs per Outcomes Achieved: Results In comparison with placebo (i.e., usual care), treatment with OCA would decrease the 15-year cumulative incidence of decompensated cirrhosis from 10% to 8.8%, hepatocellular carcinoma from 4.7% to 4.2%, liver transplant from 0.9% to 0.8%, and liver-related deaths from 12.9% to 11.3%, respectively. In addition, treatment with OCA increased 15-year transplant-free survival from 68.6% to 69.9%. Compared with placebo, treating 10,000 patients using OCA could prevent 120 cases of decompensated cirrhosis, 50 cases of hepatocellular carcinoma, 10 liver transplants and 160 liver- related deaths. The average (undiscounted) life years per patient in placebo versus OCA were 16.45 and 17.36 (increment = 0.91 years), respectively. The corresponding discounted QALYs were 10.91 and 11.02 (increment = 0.11 years), respectively. The average lifetime discounted cost per patient treated ©Institute for Clinical and Economic Review, 2016 Page ES7 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents with placebo was $70,300. Using the wholesale acquisition cost of OCA of $69,350/year, average lifetime cost of patients in the OCA arm was $371,000 (increment of $300,700). The incremental cost-effectiveness ratio (ICER) of OCA was approximately $2.75 million per QALY gained (Table ES2). Table ES2. Cost-Effectiveness of OCA When the Annual Cost of OCA is $69,350 per Year Placebo OCA Undiscounted Life Years 16.45 17.36 Discounted QALYs 10.91 11.02 Discounted Total Cost $70,300 $371,000 ICER ($/QALY) 2,748,300 Next we conducted a price threshold analysis to determine the price of OCA that would meet commonly cited thresholds for cost-effectiveness (Table ES5). Using willingness-to-pay thresholds of $50,000, $100,000 and $150,000 per QALY gained, the maximum annual price of OCA was $2,654, $3,889, and $5,124, respectively (Table ES3). Table ES3. OCA Price Threshold Analysis for NASH Patients Willingness to Annual Price of OCA Pay ($/QALY) $50,000 $2,654 $100,000 $3,889 $150,000 $5,124 Sensitivity Analysis We conducted one-way sensitivity analysis to identify the most sensitive model parameters. We found that cost-effectiveness ratios were most sensitive to the cost of OCA and percentage of patients who had NASH resolution in the placebo and OCA arms. However, the ICERs remained above $500,000 / QALY across the plausible range of values. Potential Budget Impact: Results We used the cost-effectiveness model to estimate the potential total budgetary impact of OCA for NASH patients, based on assumed patterns of product uptake. We used an estimate of NASH prevalence in the US from a review by Yeh et al.,21 who found estimates in the literature of from 3.5% to 5%. For this analysis, we used the lower estimate of 3.5% of the US population having NASH. Applying this prevalence to the projected 2016 US population would imply approximately 11.3 million individuals with NASH. Because of the difficulty in definitively diagnosing NASH (requiring liver biopsy) and the current lack of effective medical treatments, we assumed that the vast majority of these patients would not be diagnosed at this time. We assumed that 5% of the population with NASH would have been diagnosed and therefore eligible for treatment. Applying ©Institute for Clinical and Economic Review, 2016 Page ES8 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents this percentage resulted in a candidate population size of approximately 567,000 individuals in the US. Using ICER’s methods for estimating budget impact, we assumed a low scenario of 10% uptake pattern for OCA in NASH patients after 5 years. We assumed that uptake would be low for this drug because, while there is a lack of effective therapeutic alternatives for NASH patients, the use of OCA for these patients would be off-label, at least at the beginning of this time frame. Over the entire five-year time horizon, we estimated that “unmanaged” uptake would lead to approximately 56,700 persons taking OCA. Across the full five-year time horizon, the weighted potential budgetary impact (i.e., adjusted for differing periods of drug utilization and associated cost-offsets) is approximately $95,400 per patient. Total potential budgetary impact over five years is approximately $5.4 billion, with an average budget impact per year of approximately $1.08 billion (Table ES4). This annualized potential budget impact is 120% of the budget impact threshold of $904 million for a new drug that should trigger policy action to manage affordability. Table ES4. Estimated Total Potential Budget Impact (BI) of OCA Analytic Horizon = 1 Year Analytic Horizon = 5 Years Eligible Number Annual BI per Total BI Number Weighted BI Average BI per Population Treated Patient* (millions) Treated per Patient* year (millions) OCA 567,000 11,340 $69,500 $788.6 56,700 $95,400 $1,082 *Weighted budget impact calculated by subtracting cost offsets from drug costs for one-year horizon. For five- year horizon, drug costs and cost offsets apportioned assuming 20% of patients in uptake target initiate therapy each year. Those initiating in year 1 receive full drug costs and cost offsets, those initiating in year 2 receive 80% of drug costs and cost offsets, etc. Value-based Benchmark Prices Value-based price benchmarks were not calculated for OCA in the treatment of NASH, given the preliminary nature of the currently available data and that the lack of a current FDA indication for NASH. Comparative Value: Summary and Comment We conducted a cost-effectiveness analysis by developing a microsimulation model that simulated the long-term outcomes of NASH patients receiving OCA compared to placebo. We estimated that, in comparison with placebo, treatment with OCA would marginally decrease the 15-year cumulative incidence of decompensated cirrhosis, hepatocellular carcinoma, liver transplant, and liver-related deaths. In addition, treatment with OCA slightly increased 15-year transplant-free survival from 68.6% to 69.9%. Using the wholesale acquisition cost of OCA of $69,350/year, the incremental cost- effectiveness of OCA was estimated to be approximately $2.75 million per QALY gained. The use of ©Institute for Clinical and Economic Review, 2016 Page ES9 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents OCA in NASH patients exceeds commonly used willingness-to-pay thresholds of $100,000 to $150,000/QALY gained. The results were most sensitive to the price of OCA, and would achieve a cost-effectiveness ratio of $150,000/QALY only when the annual price of OCA is below approximately $5,100 per year (more than a 90% discount from the list price). We also made the assumption that patients who experienced NASH “remission” in the model would not later relapse to progressive disease. In reality, it is likely that a subset of these patients would later relapse. Due to this assumption, our model may overstate the cost-effectiveness of OCA. Another limitation is that our model draws on hepatitis C data to inform cost and quality-of-life parameters. Given the lack of economic and quality of life data for NASH health states, our model assumed that NASH-associated cirrhosis and hepatitis-associated cirrhosis involve similar treatment costs and that NASH patients experience similar decrements in quality of life as in hepatitis C patients. However, sensitivity analysis on these parameters showed that model outcomes were robust to uncertainty surrounding these cost and quality-of-life inputs. We also used the cost-effectiveness model to estimate the potential total budgetary impact of OCA for NASH patients over five years. Assuming that “unmanaged” uptake would lead to 10% of eligible patients (or approximately 56,700 persons) taking OCA, total potential budgetary impact over five years is approximately $5.4 billion, with an average budget impact per year of approximately $1.08 billion. This annualized potential budget impact is 120% of the budget impact threshold of $904 million for a new drug, suggesting the possibility of a need for policy interventions even at a relatively low rate of uptake. We used data from a Phase II study that estimated the efficacy of OCA in NASH patients. An ongoing Phase III study will provide more robust data in the future, and the model should be updated when the results from this study become available. ©Institute for Clinical and Economic Review, 2016 Page ES10 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 1. Background 1.1 Introduction Background Nonalcoholic steatohepatitis (NASH) is a form of nonalcoholic fatty liver disease (NAFLD) that can progress to cirrhosis, liver failure, and cancer. It is defined by an accumulation of triglycerides in the cells of the liver with inflammation and ballooning of the liver cells with or without fibrosis. Both NAFLD and NASH are highly prevalent. NAFLD is estimated to be present in up to 30% of the population (or 80 million adults) and NASH in around 5% or 15 million adults in the US alone.2,3 Among the 25 million Americans with diabetes, around 18 million are thought to have NAFLD, and 63-87% of patients having both diabetes and NAFLD may have NASH.4,5 High fructose intake coupled with a sedentary lifestyle are associated with higher incidence rates, especially for NASH. NASH is closely linked to the metabolic syndrome, defined by the presence of three or more of the following: abdominal obesity, hypertriglyceridemia, low high-density lipoprotein (HDL) levels, hypertension, and an elevated fasting plasma glucose.3 The rise in obesity and diabetes is contributing to an increase in NASH incidence worldwide.1 Current treatment of NASH comprises lifestyle interventions (e.g., diet, exercise, and/or behavioral change), control of the metabolic syndrome, and liver-directed pharmacotherapy.6 Obeticholic acid is a selective agonist of the farnesoid X receptor (FXR), a bile acid nuclear receptor. Its activity on lipid and glucose metabolism and hepatic inflammation makes it interesting candidate as a pharmacologic agent for treating NASH.12 A US-based Phase II trial of treatment of NASH with obeticholic acid (FLINT trial, NCT01265498) has shown an improvement in liver histology including fibrosis over a period of 72 weeks. In January 2015, obeticholic acid (OCA) received a US Food and Drug Administration (FDA) breakthrough designation for treatment of NASH with concomitant liver fibrosis13 and a 5-year Phase III trial was started in September 2015 (REGENERATE trial, NCT02548351) Interim findings from this Phase III trial are expected to be available around March 2017. OCA for the treatment of primary biliary cholangitis (PBC) was given a priority review by FDA, which approved this indication under the Prescription Drug User Fee Act (PDUFA) on May 27, 2016. Since OCA receives market access for PBC, the clinical interest in its potential off-label use for NASH is likely to be great given the unmet medical need and the lack of other approved treatments. Scope of the Assessment The proposed scope for this assessments is described below using the PICOTS (Population, Intervention, Comparators, Outcomes, Timing, and Settings) framework. Evidence was culled from ©Institute for Clinical and Economic Review, 2016 Page 1 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Phase II or III randomized controlled trials and comparative cohort studies as well as high-quality systematic reviews and meta-analyses where available. We supplemented our review of published studies with data from conference proceedings, regulatory documents, information submitted by manufacturers, and other grey literature when the evidence met ICER standards (for more information, see http://icer-review.org/methodology/icers-methods/icer-value-assessment- framework/grey-literature-policy/). Analytic Framework The analytic framework for this assessment is depicted in Figure 1. Figure 1. Analytic Framework Adverse Events Off-label treatment with • Pruritus Obeticholic Acid • Dyslipidemia Intermediate Clinical and Patient Relevant Outcomes Outcomes Biopsy confirmed Impact on NASH • Cirrhosis NASH patients with Impact of fibrosis • Liver transplantation fibrosis • Mortality • CVD Mortality Populations The population of focus for the review included adults age ≥18 with biopsy confirmed NASH and fibrosis. Interventions The intervention of interest was treatment with obeticholic acid administered as oral tablets in doses of 10 or 25 mg once daily. Comparators The comparator was usual care, including lifestyle interventions and treatment with vitamin E. ©Institute for Clinical and Economic Review, 2016 Page 2 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Outcomes This review examined key clinical outcomes related to NASH and its treatment, including surrogate outcomes in available clinical trials. Outcomes of interest included: • Impact on NASH (improvement, resolution) • Measures of liver fibrosis • Cirrhosis • Liver transplantation • Survival • Health-related quality of life • Adverse events (e.g., pruritus, effects on blood lipids) Timing Evidence on intervention effectiveness and harms was derived from studies of any duration. Settings All relevant settings were considered, including inpatient, clinic, and office settings. ©Institute for Clinical and Economic Review, 2016 Page 3 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 2. The Topic in Context The natural history of NASH is highly variable between individuals. In a longitudinal study of 103 patients with sequential liver biopsies in the absence of effective treatment, fibrosis stage progressed in 37%, remained stable in 34%, and regressed in 29%, with a mean interval of around three years between biopsies.8 Approximately 11% of NASH patients progress to cirrhosis over a 15- year period.3 As NASH is largely asymptomatic,3 cirrhosis can develop without any prior diagnosis. About 7% of patients with NASH cirrhosis will develop hepatocellular carcinoma (HCC) over 6.5 years of follow-up.2 While the risk of developing HCC in cirrhotic NASH patients seems lower than in cirrhotic hepatitis C patients,22 HCC can also occur in a substantial proportion of NASH patients in the absence of cirrhosis,2 especially among diabetic patients.23 Overall, NASH patients have a doubling of cardiovascular risk and a more than tenfold increased risk of liver-related death.6 Current treatment of NASH comprises lifestyle interventions (e.g., diet, exercise, and/or behavioral change), control of the metabolic syndrome, and liver-directed pharmacotherapy.6 Weight loss does appear to be highly effective for treating NASH. In a prospective study of 293 patients with biopsy- proven NASH, NASH resolved in 58% of patients who lost more than 5% of body weight over a period of 52 weeks. In patients who lost more than 10% of their body weight, NASH resolved in 90% and fibrosis regressed in 45%.24 After bariatric surgery, steatosis, steatohepatitis, and fibrosis appear to improve or completely resolve.7 Pioglitazone, a drug used in the treatment of diabetes, has shown to be useful for treating NASH in non-diabetic patients, but the long term safety and efficacy of this approach has not been established.7 Vitamin E is considered the liver-specific first- line treatment of NASH but does not improve fibrosis and may increase the risk of prostate cancer.3,7 Data from the Centers for Disease Control and Prevention show that liver disease is the 12th leading cause of death in the United States.25 This liver-related mortality results from complications of chronic liver disease. Between 1988 and 2008, NAFLD increased from 46.8% of chronic liver disease cases to 75.1%.26 Between 2004 and 2013, NASH has become the second leading etiology of liver disease among adults awaiting liver transplantation in the United States (Figure 2) and is expected to become the most common indication for liver transplantation in the United States between 2020 and 2025.9 Resource utilization for HCC is also largely driven by NASH, with NAFLD/NASH being the most common underlying etiologic risk factor (59%) for HCC in the United States between 2002 and 2008, followed by diabetes (36%) and hepatitis C virus infection (22%).10 ©Institute for Clinical and Economic Review, 2016 Page 4 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Figure 2. Annual Trends in New Liver Transplant Waitlist Registrations in the US27 Considering the important disease burden of NASH, the evidence base for treatment is very poor compared to other chronic liver diseases.28,29 However in recent years, clinical trials for NASH have increased dramatically.30 The database ClinicalTrials.Gov contains currently 159 open studies for NASH including 16 Phase III trials.31 Intercept and Genfit seem to be the only pharmaceutical companies with Phase III NASH drug candidates – respectively, OCA and elafibranor – with the elafibranor results expected first according to some business analysts.32 NASH’s dynamic nature with spontaneous regression and slow asymptomatic evolution represents a great challenge for clinical trials.33 The FDA and the American Association for Study of Liver Diseases (AASLD) jointly sponsored a workshop in September 2013 to discuss specific challenges and opportunities to facilitate development of therapeutics for NASH.34 There are currently no validated surrogate endpoints that meet the evidentiary burden to qualify as a generally accepted endpoint for NASH trials.35 The accelerated approval pathway used by FDA is based on surrogate outcomes that are “reasonably likely, based on epidemiologic, therapeutic, pathophysiologic, or other evidence, to predict clinical benefit or on the basis of an effect on a clinical endpoint other than survival or irreversible morbidity.”36 While there is a consensus to use histology-based endpoints as reasonable endpoints in NASH trials, the most appropriate choice of these endpoints is subject to debate, with trials using variations of outcome on NAFLD activity score (NAS) – based on steatosis, ballooning of hepatocytes and lobular inflammation – and outcomes based on fibrosis.20 For example, the outcome of the Phase III trial of elafibranor uses a primary endpoint of NASH resolution without worsening of fibrosis (NCT02704403), while the co-primary endpoints of the Phase III trial for OCA are both a liver fibrosis ©Institute for Clinical and Economic Review, 2016 Page 5 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents improvement with no worsening of NASH and NASH resolution with no worsening of liver fibrosis (NCT02548351). With fibrosis stage being the strongest predictor for disease-specific mortality in NASH,37 the main author of the publication defining NAS in 200520 states 10 years later that “assessing fibrosis change as a primary outcome provides the clearest answer to the question of clinically relevant therapeutic response, although it may come at the cost of longer and/or larger trials.”38 NAS alone is not predictive of clinical outcomes, and therefore changes in NAS on therapy are probably not an adequate reasonable surrogate endpoint for drug approval.39 In the absence of disease-specific FDA approved therapeutics, the editors of the journal Hepatology consider the use of weight loss therapy in NASH to be “transformed from one of relatively ineffective lifestyle advice to providing evidence-based effective weight loss interventions, including dietician consultation, meal replacement, medications, and endoscopic intervention. This will change what is typically a relatively unsatisfying clinical interaction into a full plan for care with multiple visits, specific interventions, and the ability to monitor responses biochemically and via noninvasive assessment of fibrosis.”40 Dietary changes and lifestyle modifications are currently and will continue to be the first-line therapy for patients with NASH. Even with advances in lifestyle and weight loss interventions, those individuals with advanced liver disease or judged to be at high risk of progression to cirrhosis are in need of pharmacological therapy.41 Since OCA received market access for PBC on May 29, 2016, this medication is accessible for off- label use for NASH. While pharmaceutical companies are not allowed to promote their medications for off-label use, FDA does not limit or control how medications are prescribed by physicians once the medications are available on the market.42 A very large percentage of the 15 million adults in the United States alone who are estimated to be afflicted with NASH2,3 are not aware of their condition. Among 127 patients with hepatic steatosis found incidentally on abdominal computed tomography (CT), only 29 (22%) patients had their diagnosis entered into their medical record by the primary care provider, none had documentation of the NAFLD fibrosis score, and none were referred for specialist evaluation or for liver biopsy. Fourteen patients (11%) at high risk for advanced hepatic fibrosis were identified by calculating the NAFLD fibrosis score.43 Gastroenterologists and hepatologists also frequently diverge from published practice guidelines for the management of NASH. Although liver biopsy remains the gold standard to diagnose NASH, less than 25% of respondents routinely require it to make the diagnosis of NASH.44 In a primary care setting, the Michael E DeBakey VA Medical Center in Houston, Texas, 19,692 patients with elevated liver enzymes were identified from a total of 120,226 patients who consulted between 2004 and 2009. Of these, 450 were randomly selected for detailed chart review using the Computerized Patient Record System, and 251 patients were identified with probable NAFLD. For only 99 patients (39.4%), the medical record mentioned abnormal ALT, with 54 patients (21.5%) ©Institute for Clinical and Economic Review, 2016 Page 6 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents identified as potentially having NAFLD. Thirty-seven patients (14.7%) were counseled on diet and exercise, and 26 (10.4%) were referred to a specialist. Among those at a high risk of fibrosis (NAFLD fibrosis score >0.675), only 3% of patients were referred to specialists.11 This study indicates that only around 14% of probable NAFLD patients received some form of treatment in this primary care setting and that only 3% of high risk patients are seen by specialists. The 2012 practice guideline for NAFLD recommends that “screening for NAFLD in adults attending primary care clinics or high-risk groups attending diabetes or obesity clinics is not advised at this time due to uncertainties surrounding diagnostic tests and treatment options, along with lack of knowledge related to the long-term benefits and cost-effectiveness of screening. (Strength – 1, Evidence -B)”7 A cost-effectiveness analysis for screening for NASH in the high risk population of patients with type 2 diabetes concludes that “screening for NASH may improve liver related outcomes, but is not cost-effective at present, due to side effects of therapy. As better tolerated treatments for NASH become available, even with modest efficacy, screening for NASH will become cost-effective.”5 ©Institute for Clinical and Economic Review, 2016 Page 7 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 3. Summary of Coverage Policies Given the very recent FDA approval of obeticholic acid, no coverage policies were available at the time of this report. This section will be updated as coverage policies become available. ©Institute for Clinical and Economic Review, 2016 Page 8 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 4. Comparative Clinical Effectiveness 4.1 Overview To inform our analysis of the comparative clinical effectiveness of OCA as an off-label treatment of NASH, we abstracted evidence from available clinical studies, whether in published, unpublished, or abstract form. The intervention of interest was treatment with obeticholic acid administered as oral tablets in doses of 10 or 25 mg once daily. As described previously in the Topic in Context section, the comparators of interest included usual care, including lifestyle interventions and treatment with vitamin E. Our review focused on key clinical benefits and surrogate outcomes of clinical benefit as well as potential harms and drug- related adverse events:  Clinical Benefits o Fibrosis (as described by histologic assessment of biopsy specimens) o Cirrhosis o Liver transplantation o Survival o Impact on NASH (as described by NAFLD activity score and histologic assessment) o Other measures of liver function (AST, ALT, GGT, ALP, etc.) o Health-related quality of life (recorded with standardized and validated questionnaires administered at serial time points)  Harms o Dyslipidemia o Incidence of pruritus o Other possible treatment-related events Stratified results of these clinical benefits are provided within each reported outcome whenever possible, and other subgroup analyses (e.g., for diabetic patients) are presented separately. 4.2 Methods We included evidence from Phase II and III randomized controlled trials (RCTs) and supplemented our review of published studies with data from conference proceedings, regulatory documents, information submitted by manufacturers, and other grey literature that met ICER standards for review (for more information, see http://icer-review.org/methodology/icers-methods/icer-value- assessment-framework/grey-literature-policy/). ©Institute for Clinical and Economic Review, 2016 Page 9 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Data Sources and Searches Procedures for the systematic literature review assessing the evidence on OCA for NASH followed established best methods used in systematic review research.45 We conducted the review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.46 The PRISMA guidelines include a checklist of 27 items, further detail of which is available in Appendix Figure A1. The timeframe for our search spanned the period from January 1996 to June 20, 2016 and focused on MEDLINE, EMBASE, and Cochrane-indexed articles. We limited each search to studies of human subjects and excluded articles indexed as guidelines, letters, editorials, narrative reviews, case reports, or news items. To supplement the above searches and ensure optimal and complete literature retrieval, we performed a manual check of the references of recent relevant reviews and meta-analyses. Further details on the search algorithm, methods for study selection, data extraction, quality assessment, assessment for publication bias, and our approach to meta-analyses of the data are available in Appendix A. Assessment of Level of Certainty in Evidence We used the ICER Evidence Rating Matrix (see Figure 3) to evaluate the evidence for a variety of outcomes. The evidence rating reflects a joint judgment of two critical components: a) The magnitude of the difference between a therapeutic agent and its comparator in “net health benefit” – the balance between clinical benefits and risks and/or adverse effects AND b) The level of certainty in the best point estimate of net health benefit.47 ©Institute for Clinical and Economic Review, 2016 Page 10 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Figure 3. ICER Evidence Rating Matrix ©Institute for Clinical and Economic Review, 2016 Page 11 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 4.3 Results Study Selection Our literature search identified 105 potentially relevant references (see Appendix A, Figure A1), of which two publications and four abstracts met our inclusion criteria; these citations related to two individual studies. Primary reasons for study exclusion included animal studies and absence of information on the outcomes of interest. Details of the included studies are described in Appendix E and the two key trials are summarized below in Table 1. Key Studies We identified two studies of interest for this review. Summarized in Table 1, these Phase II studies were double-blind, placebo-controlled, multicenter RCTs that examined OCA use among adults with NAFLD. Farsenoid X Receptor Ligand Obeticholic Acid in NASH Treatment (FLINT) Trial14 The FLINT trial enrolled 283 patients with histologic evidence of non-cirrhotic, non-alcoholic steatohepatitis into a 72-week course of receiving either 25 mg daily OCA or placebo. Participants had baseline biopsies within 90 days of randomization into the study and again at the end of week 72, except for 64 patients who were excluded from the end-of-treatment biopsy when the data safety and monitoring board recommended not performing the biopsy after the superiority boundary was crossed (significantly different decrease in NAFLD activity score [NAS] – the primary outcome – between the OCA group and the placebo group: 43% vs. 21%; p=0.0024). All participants received recommendations on healthy lifestyle behaviors and appropriate management of hypertension, hyperlipidemia, and diabetes. Non-histologic outcomes were also assessed 24 weeks after the final dose was administered. Efficacy and Safety of the Farnesoid X Receptor Agonist Obeticholic Acid in Patients with Type 2 Diabetes and Fatty Liver Disease15 A trial conducted by Mudaliar and colleagues15 enrolled 64 patients with type 2 diabetes and NAFLD, which was defined by elevated aminotransferases, hepatomegaly detected with imaging, and/or histologic evidence from a biopsy done in the prior five years. Participants were randomly assigned to receive 25 mg OCA, 50 mg OCA, or placebo daily for six weeks. In order to determine glucose sensitivity, participants were admitted for a hyperinsulinemic-euglycemic clamp procedure before the first and after the last dose of the treatment. During the procedure, patients received first a low-dose infusion rate of insulin followed by a high-dose infusion rate of insulin while glucose measurements were taken every 5-10 minutes to maintain euglycemia. The resulting glucose infusion rate was the primary outcome for this trial. ©Institute for Clinical and Economic Review, 2016 Page 12 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Two key differences distinguish the populations in the FLINT and Mudaliar trials. In the former, all of the participants have NASH and 53% have diabetes, whereas in the Mudaliar trial, all of the participants have NAFLD and diabetes. Both trials examined changes in liver enzymes and cholesterol levels while on therapy as secondary outcomes; additional outcomes and study descriptors are provided in Appendix E. Quality of Individual Studies Using criteria from US Preventive Services Task Force (USPSTF), we rated one RCT publication, the FLINT trial, to be of good quality.14 We judged this publication investigating OCA use among adults with NASH (n=283) to be of good quality because it was double-blind with comparable patient characteristics in each study arm at baseline, and the authors used valid instruments to evaluate outcomes with no differential attrition observed. Interpretation of the trial is limited by its having been stopped early when interim analysis suggested a benefit with OCA. This prevented 64 (23%) of the patients from receiving a post-treatment biopsy to assess fibrosis. We rated the other publication of the NCT00501592 trial by Mudaliar et al. to be of fair quality because the study arms in this investigation of OCA use among diabetic patients with NAFLD (n=64) were not randomized evenly and follow-up was limited to six weeks.15 The male to female ratios, baseline glucose levels, and baseline concomitant medications in each of the treatment arms were dissimilar. In addition, the description of the analyses and results in the publication limited interpretation of the results (e.g., intention-to-treat analysis was not used in reporting primary outcome, and only 69% of the participants were retained for this outcome). Four abstracts16-19 provided supplemental results to the FLINT trial, and these are unrated in keeping with the ICER grey literature policy (http://icer-review.org/methodology/icers- methods/icer-value-assessment-framework/grey-literature-policy/). ©Institute for Clinical and Economic Review, 2016 Page 13 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Table 1. Key Trials Key Trials Patient Characteristics Treatment Comparator Harms Mean age: 52 OCA 25 mg daily Placebo Pruritus: 23% vs. 6% FLINT14 Percent male: 34% (n=141; ITT* n=110) (n=142; ITT* n=109) (p<0.0001) Mean weight: ~98kg Administered for 72 wks w/ 24 wks follow-up Phase II Hyperlipidemic: 62% Primary outcome: ≥2 point decrease in centrally scored NAS w/o Double-blind Diabetic: 53% worsening fibrosis 72 wks (OCA 45% vs. PBO 21%) RCT Vitamin E last 6 mos: 22% RR 1.9 (95% CI 1.3-2.8); p=0.0002 Multicenter Antilipidemic last 6 mos: Secondary outcomes: ITT analysis 48% -Mean change in NAS (-1.7 vs. -0.7) Definite steatohepatitis: RR -0.9 (95% CI -1.3 to -0.5); p<0.0001 80% -Patients w/ improved fibrosis (35% vs. 19%) Mean NAFLD score: 5.2 RR 1.8 (95% CI 1.1-2.7); p=0.004 -Resolution of NASH (22% vs. 13%) RR 1.5 (95% CI 0.9-2.6); p=0.08 -Mean change (baseline to 72 wks): -Mean change (baseline to 72 ALT -38 wks): AST -27 ALT -18 (p<0.001) ALP +12 AST -10 (p=0.0001) GGT -37 ALP -6 (p<0.0001) Total cholesterol 0.16 GGT -6 (p<0.0001) HDL -0.02 Total cholesterol -0.19 HOMA-IR 15 (p=0.0009) Weight (kg) -2.3 HDL 0.03 (p=0.01) HOMA-IR 4 (p=0.01) Weight (kg) 0.0 (p=0.008) NCT00501592 Mean age: 52 OCA 25 mg daily (n=20) Placebo (n=23) Any AEs (OCA 25 mg by Mudaliar et Percent male: 53% or vs. 50 mg vs. PBO): al.15 Mean weight: ~106kg OCA 50 mg daily (n=21) 45% vs. 76% vs. 61% Diabetic: 100% Treatment-related Phase II Administered for 6 wks AEs: Double-blind Primary outcomes: 5% vs. 38% vs. 26% RCT -Percent change in low-dose glucose infusion rate Pruritus: Multicenter (OCA 24.5 vs. PBO -5.5); p=0.011 0 vs. 5% vs. 9% -Percent change in high-dose glucose infusion rate (OCA 15.0 vs. PBO -5.4); p=0.025 Secondary outcomes: change in Secondary outcomes: change mean values 25 mg/50 mg in mean values (p-value 25 mg/p-value 50 mg) AST -2/5 AST 5 (0.12/0.73) ALT -10/10 ALT 11 (0.003/0.84) ALP 14/27 ALP 0 (0.003/<0.001) GGT -37/-22 GGT 5 (<0.001/<0.001) Total cholesterol 18/13 Total cholesterol 8 (0.08/0.15) HDL -2/-6 HDL 0 (0.42/0.01) Weight 1/1.9 Weight (0.096/0.008) ITT = intent-to-treat; DB = double-blind; LTSE = long-term safety extension; ULN = upper limit of normal; UDCA = ursodeoxycholic acid; OCA = obeticholic acid; ALP = alkaline phosphatase; ALT = alanine aminotransferase; AST = aspartate aminotransferase; GGT = gamma-glutamyl transferase; HDL = high-density lipoprotein; AEs = adverse events; HOMA-IR = Homeostasis model assessment of insulin resistance *ITT population was defined in FLINT trial as those 219 patients who received both baseline and 72-week follow-up biopsies ©Institute for Clinical and Economic Review, 2016 Page 14 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Clinical Benefits A detailed review of each clinical outcome of interest is presented in the sections that follow. The most important clinical outcome, resolution of fibrosis, was reported in the FLINT trial publication as a secondary outcome but not in the other trial publication or affiliated abstracts. Although neither study shared the same primary outcomes (FLINT: decrease in NAS; Mudaliar: change in glucose infusion rate), both reported secondary outcomes of mean changes in liver enzymes, cholesterol, and weight.14-18 Liver enzymes include aminotransferases, ALT and AST, and other enzymes, GGT and ALP, which are elevated when the liver is inflamed. They can indicate liver function and be used by clinicians to track disease progression. These studies reported no data on cirrhosis, liver transplantation, or survival. Resolution of Fibrosis The ultimate goal of NASH therapy is to prevent cirrhosis from developing, which also means preventing fibrosis. Patients with NASH are at higher risk of developing significant fibrosis than patients with simple steatosis.3 At the same time, some patients with NASH experience improvement or stabilization of fibrosis without intervention.48 Resolution or improvement in fibrosis was presented in the FLINT trial as a secondary outcome (improvement for OCA: 35% vs. 19% for placebo; rate ratio [RR] 1.8; 95% CI 1.1-2.7; p=0.004).14 Among the patients in the FLINT trial who had the most severe NASH at baseline (stage 2-3 fibrosis or stage 1 with diabetes, obesity, or ALT ≥ 60 U/L), an accompanying poster by Neuschwander-Tetri examined changes in fibrosis within this subpopulation (OCA: n=85; placebo: n=77).18 More patients in the OCA group experienced regression of fibrosis by at least one stage (39% vs. 22% for placebo; p=0.012). Similarly, fibrosis progressed for fewer patients treated with OCA than placebo (16% vs. 29%; p=0.047). This outcome was not reported in other abstracts or publications included in the evidence review. NAFLD Activity Score The NAFLD activity score, or NAS, is based on histologic assessment of liver biopsies: it is the unweighted sum of scores given for steatosis (0-3), lobular inflammation (0-3), and hepatocellular ballooning (0-2). Fibrosis is measured separately as it the result of inflammation and damage to hepatocytes.20 In the FLINT trial, these scores were given centrally by pathologists blinded to the treatment arm.14 NAS scores were a key component of the primary outcome: ≥2-point decrease without worsening fibrosis after 72 weeks of OCA versus placebo (45% vs. 21%; RR 1.9; 95% CI 1.3- 2.8; p=0.0002). When measured as a single outcome, patients in the OCA arm had greater mean change in NAS (-1.7 vs. -0.7 for placebo; p<0.0001). These statistically-significant improvements in ©Institute for Clinical and Economic Review, 2016 Page 15 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents NAS are of unknown clinical significance, however, since improvement in NAS does not always lead to a reduction in NASH diagnoses.14 Some patients in both treatment groups experienced histologic resolution of NASH by 72 weeks of therapy, but these changes were not statistically significant. Among the patients in the FLINT trial with most severe NASH at baseline, a Neuschwander-Tetri poster reported that significantly more patients treated with OCA experienced two or more points of improvement in NAS compared with the placebo group (50% vs. 31%; p=0.001).18 More of these patients in this subpopulation also experienced resolution of NASH if they were treated with OCA (18% vs. 6.5%; p=0.03). Liver Enzymes Serum aminotransferases (ALT and AST) and other liver enzymes (GGT and ALP) are elevated in liver inflammation, which is an earlier stage of the pathway to cirrhosis. However, elevations in these enzymes are not indicative of NASH by themselves, since they can be elevated in many other conditions and normal in patients with histologic evidence of NASH. While the FLINT patients were taking OCA, they had significantly lower levels of ALT, AST, and GGT than the patients receiving placebo.14 This suppression of enzyme levels became indistinguishable between the two arms at week 96 when the drug was no longer being administered (stopped at week 72). An inverse of this relationship was seen with ALP, which was significantly higher among the patients receiving OCA and then indistinguishable from baseline at week 96. Although a secondary outcome in the FLINT trial, the clinical significance of altering liver enzymes by 12 to 38 U/L is uncertain beyond its use in tracking disease progression. The trial by Mudaliar in which NAFLD patients with diabetes were treated with six weeks of OCA at 25 mg or 50 mg doses versus placebo also reported some small, but statistically-significant, decreases (-10 to -37 U/L) in ALT and GGT among patients receiving 25 mg OCA at day 43 compared to baseline measurements.15 The patients receiving 50 mg experienced a mean decrease only in GGT (-22 U/L; p<0.001 compared to placebo). There was no significant change in levels of AST regardless of treatment arm. As in the FLINT trial, ALP levels increased for both OCA treatment groups compared to patients in the placebo arm (25 mg: +14 U/L; p=0.003; 50 mg: +27 U/L; p<0.001). Weight Body weight decreased while patients in the FLINT trial took OCA (mean change of -2.3kg vs. no change in the placebo arm; p=0.008), but this gravitated back to baseline once treatment was stopped at week 72.14 Weight loss of this degree was associated with some benefit in lowering systolic blood pressure (mean change of -4mmHg vs. -1mmHg in the placebo arm; p=0.05). Similarly, diabetic patients with NAFLD in the Mudaliar trial lost more weight (-1.9% change in mean ©Institute for Clinical and Economic Review, 2016 Page 16 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents body weight; p=0.008) if they had been treated with 50 mg OCA for six weeks compared to the placebo group.15 This same relationship was not observed for the patients taking the 25 mg dose. A subgroup analysis of the FLINT trial examined several clinical outcomes by presence or absence of weight loss during the trial. These outcomes were reported in affiliated abstracts.16,17 The NAS was significantly lower for patients who experienced weight loss during the 72-week trial period, whether in the OCA or placebo treatment arm (mean change in NAS for OCA patients with weight loss -2.4 vs. without weight loss -1.2; p<0.001; placebo patients -1.4 vs. -0.4; p=0.006).16 LDL levels were higher among patients who received OCA and lost weight during the trial compared to patients who did not lose weight (23 mg/dL vs. 0; no significance testing). However, patients who lost weight while taking placebo had lower LDL levels than patients on placebo who did not lose weight (-17 mg/dL vs. -2 mg/dL; no significance testing). Although the changes in LDL were not tested statistically within treatment arm, they were tested across treatment arm within the subpopulation who lost weight and found to be significantly different (+23 mg/dL vs. -17 mg/dL for OCA and placebo, respectively; p<0.001).17 Health-Related Quality of Life Health-related quality of life measures were assessed in the FLINT trial using SF-36 questionnaires for physical and mental well-being.14 Neither treatment nor placebo group showed a change from baseline over the course of the 72 weeks of treatment for either component of the SF-36. Of note, NASH, as with NAFLD, is often asymptomatic and identified incidentally when testing is performed for an unrelated condition, so these findings are unsurprising.2 Insulin Sensitivity As an activator of the farnesoid X nuclear receptor, OCA can promote insulin sensitivity.49 Diabetes is also an independent risk factor for advancing fibrosis and disease progression in NAFLD and NASH.50 The FLINT trial included as a secondary outcome a fasting homeostasis model of insulin resistance (HOMA-IR).14 Somewhat surprisingly, patients treated with OCA had a greater increase in hepatic insulin resistance than those in the placebo arm (p=0.01). This is contrary to the findings demonstrated in the six-week long Mudaliar trial of OCA treatment of NAFLD patients with diabetes.15 The authors suggest that the difference may be accounted for by adaptive mechanisms in settings of longer-term farsenoid X receptor activation. The primary outcome was the percent change in low- and high-dose glucose infusion rates between the treatment arms. For both low- and high-dose glucose infusion rates, the percent increase was significantly larger for patients treated with 25 mg OCA compared to placebo (low-dose: 28.0 vs. -5.5; p=0.019; high-dose: 18.3 vs. -5.4; p=0.036) than it was for 50 mg OCA compared to placebo (low-dose: 20.1 vs -5.5; p=0.60; high- dose: 10.8 vs. -5.4; p=0.076). ©Institute for Clinical and Economic Review, 2016 Page 17 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Other Subgroup Analyses In addition to the subgroup analyses presented within the Clinical Benefit subsections above, we review here the impact of OCA on the subpopulation of NASH patients with diabetes, which describes 53% of the patients in the FLINT trial.14 As summarized, the results of the Mudaliar trial fit within this subpopulation, too, although histologic outcomes were not examined and the proportion of NASH patients was not reported.15 The supplemental post-hoc analyses accompanying the FLINT trial revealed that of the patients in the ITT sample (n=219), the patients without diabetes at baseline (n=103) were less likely than the patients with diabetes (37% vs. 53%) to experience histologic improvement while taking OCA; histological improvement was defined as a decrease of at least two points in the NAS without worsening fibrosis at week 72. 14 Diabetes also had an impact on measures of treatment effect. The odds ratio (OR) for histological improvement with OCA versus placebo in patients without diabetes (OR 2.0; 95% CI 0.8-4.7) was not statistically significant, whereas the OR for patients with diabetes was significant (OR 4.6; 95% CI 2.0-10.6). Furthermore, only patients with advanced beta cell loss (n=60) had significantly improved histology (OR 5.3; 95% CI 2.1-13.3), whereas the ORs for the subpopulations with early beta cell loss (n=23), intact beta cells (n=69), and those who were insulin sensitive (n=17) were smaller and not statistically significant. Another subgroup post-hoc analysis of the FLINT trial examined if the use of antidiabetic or antilipidemic medication used concomitantly to OCA influenced the effects of OCA on histological improvement in patients. Concomitant antidiabetic or antilipidemic medications had no significant influence on histological outcomes.19 Harms We describe the most commonly reported types of harms associated with OCA therapy: dyslipidemia, pruritus, and other treatment-related adverse events. Dyslipidemia When the lipophilic bile acid OCA binds to farsenoid X nuclear receptors, both hepatic gluconeogenesis and circulating triglycerides are inhibited. These helpful actions occur because the liver synthesizes fewer lipids and upregulates clearance of VLDL. Unfortunately, at the same time, activating the farsenoid X nuclear receptor also increases HDL clearance because it speeds up reverse cholesterol transport (through upregulation of hepatic scavenger receptors). As with the liver enzymes, patients in the FLINT trial experienced a small, but statistically significant, increase in total cholesterol and LDL and decrease in HDL while taking OCA.14 This effect disappeared once the drug was stopped, and clinical experts differ about the clinical impact of worsening lipid profiles that might be managed with statin therapy. ©Institute for Clinical and Economic Review, 2016 Page 18 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents For the subpopulation with most severe NASH at baseline in the FLINT trial, Neuschwander-Tetri reported in a poster that statin therapy initiated for those in the OCA treatment arm reduced LDL levels to those seen among the patients who were treated with statins at baseline.18 Among diabetic patients with NAFLD, Mudaliar et al. found that patients treated with OCA (25 mg or 50 mg) for six weeks did not have statistically significantly elevated total cholesterol levels.15 HDL levels, on the other hand, were significantly lower (-6 mg/dL; p=0.01) for patients treated with 50 mg OCA at day 43 than they were at baseline compared to patients in the placebo arm. Patients treated with 25 mg OCA had no change in HDL levels. It is not clear whether this is because patients with diabetes are less susceptible to the lipid effects of OCA, if more of these patients were taking statins at baseline, or whether the drug has an impact on lipids only after a longer course of therapy. Pruritus and Other Adverse Events OCA treatment is associated with increased pruritus among those in the treatment arms (FLINT: 23% vs. 6%; p<0.0001), but this reportedly led to little treatment discontinuation.14 The Mudaliar trial of diabetic patients with NAFLD reported greater incidence of patients with any kind of adverse events among the patients treated with 50 mg OCA (76%) compared with 25 mg (45%) and placebo (61%).15 Similar absolute differences were seen when treatment-related adverse events were reported, although incidence in the 25 mg OCA group was low: 50 mg OCA (38%) compared with 25 mg (5%) and placebo (26%). In either scenario, the lower dose of OCA was associated with fewer adverse events, including pruritus (25 mg: 0%, 50 mg: 5%, placebo 9%). Controversies and Uncertainties NASH is currently an off-label indication for OCA, which makes its use in this clinical setting more susceptible to hypothetical benefit and anecdotal supporting evidence. The published evidence base for using OCA in NASH is very slim and excludes the findings from a completed Phase II trial conducted among Japanese patients with NASH that failed to meet its primary endpoint (≥ 2-point improvement in NAS without worsening fibrosis). Data from that trial have not been published and are available only from press releases and online news aggregators.51,52 This placebo-controlled trial was similar to FLINT in that daily OCA was administered (doses were 10 mg, 20 mg, or 40 mg) for 72 weeks. Additional trials are underway (REGENERATE and CONTROL) and should be examined carefully to further characterize the effectiveness of OCA activity on NASH. The current studies do not directly describe the impact of OCA on cirrhosis. The clinical significance of several of the secondary outcomes is uncertain: what is the clinical impact of lowering liver aminotransferase or of raising HDL levels to the degree described in the sections above? Are there undesirable and unintended consequences of initiating patients on statin therapy to manage ©Institute for Clinical and Economic Review, 2016 Page 19 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents dyslipidemia associated with OCA therapy or are these offset by long-term gains in preventing end- stage liver disease? These are questions that remain unanswered given the limited evidence base. Another consideration is that the side effect of pruritus is noxious, whereas the symptoms of NASH are quiescent for many years. This raises the question of long-term adherence to oral therapy taken daily to suppress a chronic condition with few symptoms until late stages of the disease. Summary Given the limited evidence base and uncertainty regarding the long-term clinical effects of changes in surrogate endpoints and conflicting physiological outcomes while taking the drug (e.g., insulin resistance in the Mudaliar vs. FLINT trials), we assign an ICER evidence rating of “Insufficient,” or “I” for using obeticholic acid as an off-label treatment for adults with nonalcoholic steatohepatitis with fibrosis. ©Institute for Clinical and Economic Review, 2016 Page 20 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 5. Other Benefits or Disadvantages Our reviews seek to provide information on other benefits or disadvantages offered by the intervention to the individual patient, caregivers, the delivery system, other patients, or the public that would not have been considered as part of the evidence on comparative clinical effectiveness. Examples include but are not limited to: 1. Methods of administration that improve or diminish patient acceptability and adherence 2. A public health benefit, e.g., reducing new infections 3. Treatment outcomes that reduce disparities across various patient groups 4. More rapid return to work or other positive effects on productivity (if not considered a benefit as part of comparative clinical effectiveness) 5. New mechanisms of action for treatments of clinical conditions for which the response to currently available treatments vary significantly among patients for unknown reasons (substantial heterogeneity of treatment effect) OCA offers the potential of oral therapy to slow or suspend the progression of NASH. Currently, best practices include lifestyle modification (e.g., weight loss) and managing cardiovascular risk. Other oral agents (e.g., vitamin E, pioglitazone) have unclear effectiveness given some safety concerns about their widespread use (i.e., increase in all-cause mortality and weight gain as a side effect). Approval of OCA gives practitioners an additional oral agent to deploy in preventing end- stage liver disease, but the data currently available do not demonstrate impact on long-term outcomes. ©Institute for Clinical and Economic Review, 2016 Page 21 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 6. Comparative Value 6.1 Overview The primary aim of this analysis was to estimate the cost-effectiveness of obeticholic acid (OCA) treatment for patients with NASH. We conducted a cost-effectiveness analysis (CEA) by developing a microsimulation model that simulated the long-term outcomes of patients receiving OCA as observed in the Phase II FLINT study; as a comparator, we also simulated the placebo arm of the trial. Model parameters were estimated from published studies and calibrated where assumptions were required. The outcomes of the model included total costs, quality-adjusted life years (QALYs), incremental cost-effectiveness ratios, transplant-free survival, and cumulative incidence of advanced disease stages. 6.2 Prior Published Evidence on Costs and Cost-Effectiveness of Obeticholic Acid We did not identify any published articles or public presentations pertaining to the cost and/or cost- effectiveness of OCA for treatment of NASH patients. To the best of our knowledge, this report is the first analysis that estimates the cost-effectiveness and long-term impact of OCA for the treatment of patients with NASH. 6.3 Incremental Costs per Outcome Achieved Cost-Effectiveness Model: Methods Model Structure We developed an individual-level state-transition model (i.e., a microsimulation model) to assess two different strategies for treating NASH in hypothetical patients: OCA, and standard care, which could include treatment with Vitamin E in both the OCA and standard care strategies. We simulated a hypothetical cohort of patients with fibrosis stages F1–F3 to estimate the CEA of OCA in NASH patients. Health states in our model included: NASH fibrosis states prior to cirrhosis (F0–F3), compensated NASH cirrhosis, decompensated NASH cirrhosis, HCC and liver transplant. Patients with decompensated cirrhosis and HCC would be eligible for liver transplantation. Possible causes of death included liver-related mortality, cardiovascular mortality, and all-cause mortality. Liver-related mortality can occur because of advanced fibrosis (F3), compensated cirrhosis, decompensated cirrhosis, HCC or liver transplantation. All patients with NASH have a higher cardiovascular mortality compared to the general population, which was based on published ©Institute for Clinical and Economic Review, 2016 Page 22 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents studies53 and incorporated into the model. Background mortality risk was based on patients’ age and sex and estimated from US life tables.54 Transplant patients have higher risk of mortality for the first year than in subsequent years. Figure 4. Model Structure: Natural History of NASH Disease Abbreviations: F0–F4=fibrosis stages; CC=compensated cirrhosis; DC=decompensated cirrhosis; HCC= hepatocellular carcinoma For each treatment regimen, a hypothetical patient cohort began distributed across the three fibrosis states (F1–F3). Patients remained in their fibrosis state until they experienced: A) progression in disease toward cirrhosis, B) regression in disease, or C) death from all-cause mortality. Patients continued to receive OCA in the intervention group as long as they were in fibrosis stages F1, F2 or F3. Patients that progressed to compensated cirrhosis (CC, F4) stopped taking the treatment. We estimated overall average patient survival, total costs, and QALYs for each treatment strategy, as well as the incremental cost-effectiveness ratio (ICER) comparing the two strategies. Model cycle length was one year, except for the first cycle, which was assumed to be 72 weeks, to align with the duration of the FLINT study. To model the efficacy of OCA treatment, we used the primary endpoint used in the Phase III REGENERATE trial. Consequently, the primary outcome in our model was the percentage of patients who achieved NASH resolution without worsening of fibrosis. ©Institute for Clinical and Economic Review, 2016 Page 23 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Target Population The population modeled in the analysis will be patients 18 years or older with histologic evidence of NASH with fibrosis (F1, F2 or F3) diagnosed by a liver biopsy (Table 2). The baseline parameters in the model were based on data from the NASH Clinical Research Network (CRN) Study for patients who had definite steatohepatitis.55 As it is likely that payer coverage of OCA will be contingent on a prior attempt at lifestyle intervention, we assumed that patients in the model had made an attempt at lifestyle intervention but did not achieve meaningful success. Table 2. Model Cohort Characteristics Value Mean age 49 (18-75) NASH fibrosis stage distribution F1 39% F2 27% F3 34% Sex: Female / male 66% / 34% Source: NASH CRN Study55 Treatment Strategies The interventions of interest were OCA compared to usual supportive care. The intervention of interest was treatment with OCA administered as oral tablets in a dose of 25 mg once daily. The comparator was usual supportive care, including treatment with vitamin E. Key Model Choices and Assumptions  The intervention of interest was treatment with OCA administered as oral tablets in a dose of 25 mg once daily.  The natural history model of NASH progression to cirrhosis and liver-related death was constructed incorporating data from published sources.  The model structure was adapted from a previously published model and published data regarding the natural history of NASH.5  OCA efficacy was estimated based on analysis of one US-based Phase II trial (FLINT, NCT01265498), which showed an improvement in liver histology, including fibrosis, over a period of 72 weeks.  There was no data to inform the model beyond week 72; therefore, we made biologically and clinically plausible assumptions beyond week 72, which were additionally confirmed by clinical experts.  Key adverse events related to OCA are included in the model, including pruritus and dyslipidemia. ©Institute for Clinical and Economic Review, 2016 Page 24 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents  The model included grade 3/4 adverse events only, as less severe events are not expected to substantially impact patient health or costs. The model therefore included all grade 3/4 events that occur in at least 5% of patients.  Costs included those of current and subsequent treatment, management of adverse events, ongoing NASH-related care, and management of advanced disease outcomes such as hepatocellular carcinoma.  We utilized a health system perspective (i.e., focus on direct medical care costs only) and a lifetime horizon, modeling patients from treatment initiation until death.  Results were expressed primarily in terms of the incremental cost per quality-adjusted life year (QALY) gained relative to the standard treatment strategy.  A 3% annual discount rate for both costs and QALYs, with a half-cycle correction, were used in the model analyses. Clinical Inputs The primary outcome in our model was resolution of NASH without worsening of fibrosis. From the post-hoc analysis of the FLINT study, 19% of patients in the OCA arm achieved the primary outcome versus 8% in the placebo arm.2 Adverse Events The model includes pruritus and dyslipidemia as adverse events. In addition, we considered three levels of pruritus, defined as mild, moderate and severe. A decrement to quality of life was applied when a patient experienced pruritus. There was an additional cost for pruritus treatment, which included the cost of physician office visits and anti-pruritic drugs. Pruritus did not affect the transitions between health states. We also included dyslipidemia as an adverse event, which did not have impact on quality of life or transitions between health states. However, it did lead to additional costs for management of dyslipidemia, including a physician office visit and medication. Table 3. Adverse Event Inputs Adverse Event Patients Affected Reference OCA Pruritus – mild 6.3% FLINT Table 4 (OCA) Pruritus – moderate 14.8% FLINT Table 4 (OCA) Pruritus – severe 2.1% FLINT Table 4 (OCA) Dyslipidemia 51% FLINT Study Standard Care Pruritus – mild 4.2% FLINT Table 4 (Placebo) Pruritus – moderate 2.1% FLINT Table 4 (Placebo) Pruritus – severe 0% FLINT Table 4 (Placebo) Dyslipidemia 35% FLINT Study ©Institute for Clinical and Economic Review, 2016 Page 25 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Drug Utilization The model used a 25 mg OCA dose, as this was the dosage used in the FLINT trial. Costs Health state costs associated with advanced stages of the disease were based on reported costs for patients with hepatitis C virus infection.56 The cost of early stages of NASH was assumed to be similar to that of hepatitis C patients having mild, moderate and advanced fibrosis. Table 4 summarizes the costs associated with each health state. All costs were converted to a 2015 baseline using the medical care Consumer Price Index. For patients who have pruritus, we assumed there will be additional costs for two primary care visits; these costs are based on the fees associated with HCPCS code 99213 in the physician fee schedule.3 We also applied the cost of one year of hydroxyzine treatment, based on the Red Book value for a 25 mg dose three times per day for one year.57 Table 4. Cost Inputs Associated with Health States and Management of Adverse Events Parameter Values F0 $728 (±25%) F1 $728 (±25%) F2 $737 (±25%) F3 $1496 (±25%) Compensated cirrhosis $5,752 (±25%) Decompensated cirrhosis $40,141 (±25%) Hepatocellular carcinoma $88,383(±25%) Liver transplant-1st year $179,080 ($134,310-$739,100) Liver transplant-subsequent year $44,074 (±25%) Cost of OCA (wholesale acquisition cost) $69350 (±25%) Dyslipidemia treatment $191 (±25%) Dyslipidemia doctor’s office visit $52 (±25%) Cost of Pruritus (doctor’s office visit) $103 (±25%) Cost of Pruritus (ongoing hydroxyzine treatment) $712 (±25%) Utilities We assigned health-related quality-of-life (QOL) utilities to each patient in the model, with 0 denoting death and 1 denoting perfect health. Health state utilities from publicly available literature (Table 5) were used, with consistent values across treatments evaluated in the model. Because NASH-specific utilities by different stages of disease were not available, we used the utilities of health states for patients with hepatitis C. Specifically, we used health-state specific utility weights ©Institute for Clinical and Economic Review, 2016 Page 26 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents from a previously published study using the EuroQol-5D,58,59 and adjusted these weights to the US population norm (Table 6).60 We further applied a disutility for patients who experience pruritus; to determine the overall utility for a patient with pruritus, we took the product of the health state utility and the pruritus utility. Table 5. Utilities for Health States and Adverse Events Base Case Health State (Range) Health States58 F0 0.85 (0.84-0.99) F1 0.84 (0.84-0.99) F2 0.84 (0.84-0.99) F3 0.84(0.84-0.99) Compensated cirrhosis 0.90 (0.81-0.99) Decompensated cirrhosis 0.80 (0.57-0.99) Hepatocellular carcinoma 0.79 (0.54-0.99) Transplant-first year 0.84 (0.77-0.93) Transplant-subsequent year 0.93 (0.84-0.99) Adverse events (multiplicative factor)61 Pruritus – mild 0.93 (±25%) Pruritus – moderate 0.87 (±25%) Pruritus – severe 0.79 (±25%) Table 6. Health-Related Quality-of-Life Utilities of the United States Population Age Group Male Female 20–29 0.928 0.913 30–39 0.918 0.893 40–49 0.887 0.863 50–59 0.861 0.837 60–69 0.84 0.811 70–79 0.802 0.771 80–89 0.782 0.724 Source: Hanmer et al.60 Transition Probabilities We used a meta-analysis by Singh et al.62 to estimate progression and regression in NASH fibrosis stages. Because patients with NASH have higher cardiovascular risk compared to the general population, we calibrated background mortality such that the overall survival for the modeled NASH patients matched that of a cohort published by Adams et al. that followed patients for 16 years.63 ©Institute for Clinical and Economic Review, 2016 Page 27 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents To do so, we simulated 16-year survival in patients followed in Adams et al. with the following baseline characteristics: age 50 years; 58% female, NASH fibrosis of F0 36%, F1, 18%, F2, 15%, F3, 18%, F4, 13%. We found that the model-predicted overall survival closely matched survival reported by Adams et al. (Figure 5). The mortality associated with decompensated cirrhosis, hepatocellular carcinoma and liver transplantation were assumed to be similar to that in patients with hepatitis C who developed these advanced outcomes.64 Figure 5. Comparison of Overall Survival of NASH Patients Predicted by the Model and Adams et al. 100% 80% 60% Survival 40% Adams et al 20% NASH Model 0% 0 2 4 6 8 10 12 14 16 Year We estimated transition probabilities between health states from previously published studies. For progression and regression of fibrosis, we used a meta-analysis by Singh et al.62 which provided fibrosis progression/regression at the end of 7.1 years. We converted these probabilities to annual probabilities. Table 7 below provides the parameter values. For advanced stages of NASH, we extracted transition probabilities from published sources. ©Institute for Clinical and Economic Review, 2016 Page 28 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Table 7. Transition Probabilities in the NASH Model Annual transition probability Parameters (Range) F0 to F162 0.095 (±25%) F0 to F262 0.010 (±25%) F0 to F362 0.002 (±25%) F0 to compensated cirrhosis62 0.095 (±25%) F1 to F062 0.049 (±25%) F1 to F262 0.077 (±25%) F1 to F362 0.034 (±25%) F2 to F062 0.021 (±25%) F2 to F162 0.165 (±25%) F2 to F362 0.051 (±25%) F2 to compensated cirrhosis62 0.043 (±25%) F3 to F162 0.116 (±25%) F3 to F262 0.067 (±25%) F3 to compensated cirrhosis62 0.118 (±25%) Compensated cirrhosis to F362 0.059 (±25%) Compensated cirrhosis to decompensated cirrhosis22,65,66 0.058 (±25%) Compensated cirrhosis to HCC67 0.026 (±25%) Compensated cirrhosis to LRD22,28,66 0.021 (±25%) Decompensated cirrhosis to HCC67 0.026 (±25%) Decompensated cirrhosis to liver transplantation68,69 0.023 (±25%) Decompensated cirrhosis to liver-related death70 0.130 (±25%) HCC to liver transplantation69,71 0.040 (±25%) HCC to liver-related death72 0.427 (±25%) Liver transplant (first year for DC) to liver-related death73,74 0.094 (±25%) Liver transplant (first year for HCC) to liver-related death74 0.101 (±25%) Sensitivity Analyses We performed one-way sensitivity analyses to identify the key drivers of model outcome variability. Incremental Costs per Outcomes Achieved: Results Figures 6-10 show 15-year cumulative incidence of decompensated cirrhosis, hepatocellular carcinoma, liver transplants and liver-related deaths in the simulated cohort of patients treated with OCA versus placebo. In comparison with placebo, treatment with OCA would decrease the 15- year cumulative incidence of decompensated cirrhosis from 10% to 8.8%, hepatocellular carcinoma from 4.7% to 4.2%, liver transplant from 0.9% to 0.8%, and liver-related deaths from 12.9% to 11.3%, respectively. In addition, treatment with OCA increased 15-year transplant-free survival from 68.6% to 69.9% (Figure 10). Compared with placebo, treating 10,000 patients using OCA could ©Institute for Clinical and Economic Review, 2016 Page 29 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents prevent 120 cases of decompensated cirrhosis, 50 cases of hepatocellular carcinoma, 10 liver transplants and 160 liver-related deaths. Figure 6. Cumulative Incidence of Decompensated Cirrhosis in NASH Patients Treated with OCA versus Placebo Decompensated Cirrhosis 10.0% 10.0% 8.8% 8.0% Placebo Cumulative Incidences OCA 6.0% 4.0% 2.0% 0.0% 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Year Figure 7. Cumulative Incidence of Hepatocellular Carcinoma in NASH Patients Treated with OCA versus Placebo Hepatocellular Carcinoma 5.0% 4.7% 4.5% Placebo 4.2% 4.0% Cumulative Incidences 3.5% OCA 3.0% 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Year ©Institute for Clinical and Economic Review, 2016 Page 30 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Figure 8. Cumulative Incidence of Liver Transplants in NASH Patients Treated with OCA versus Placebo Liver Transplant 1.0% 0.9% 0.9% Placebo 0.8% 0.8% OCA Cumulative Incidences 0.7% 0.6% 0.5% 0.4% 0.3% 0.2% 0.1% 0.0% 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Year Figure 9. Cumulative Incidence of Liver-Related Deaths in NASH Patients Treated with OCA versus Placebo Liver Related Death 15.0% 12.9% 12.0% Placebo 11.3% Cumulative Incidences OCA 9.0% 6.0% 3.0% 0.0% 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Year ©Institute for Clinical and Economic Review, 2016 Page 31 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Figure 10. Transplant-Free Survival in NASH Patients Treated with OCA versus Placebo 100% 80% 69.9% Transplant Free Survival Placebo 60% OCA 68.6% 40% 20% 0% 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Year The average (undiscounted) life years per patient in placebo versus OCA were 16.45 and 17.36 (increment = 0.91 years), respectively. The corresponding discounted QALYs were 10.91 and 11.02 (increment = 0.11 years), respectively. The average lifetime discounted cost per patient treated with placebo was $70,300. Using the wholesale acquisition cost of OCA of $69,350/year, average lifetime cost of patients in the OCA arm was $371,000 (increment of $300,700). The incremental cost-effectiveness ratio (ICER) of OCA was $2.75 million per QALY (Table 8). Table 8. Cost-Effectiveness of OCA When the Annual Cost of OCA is $69,350 per Year Placebo OCA Undiscounted Life Years 16.45 17.36 Discounted QALYs 10.91 11.02 Discounted Total Cost $70,300 $371,000 ICER ($/QALY) 2,748,300 Next we conducted a price threshold analysis to determine the price of OCA that would meet commonly cited thresholds for cost-effectiveness. Using willingness-to-pay thresholds of $50,000, $100,000 and $150,000 per QALY gained, the maximum annual price of OCA was $2,654, $3,889, and $5,124, respectively (Table 9). ©Institute for Clinical and Economic Review, 2016 Page 32 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Table 9. OCA Price Threshold Analysis for NASH Patients Willingness to Annual Price of OCA Pay ($/QALY) $50,000 $2,654 $100,000 $3,889 $150,000 $5,124 Sensitivity Analysis We conducted one-way sensitivity analysis to identify the most sensitive model parameters. We plotted the tornado diagram showing 15-most sensitive parameters (Figure 11). We found that ICERs were most sensitive to the cost of OCA and percentage of patients who had NASH resolution in placebo and OCA arm. However, the ICERs remained above $500,000 / QALY. Figure 11. Tornado Diagram Showing 15 Most Sensitive Model Parameters Cost: OCA $15,000 $100,000 NASH Resolution for OCA (%) 0.237 NASH Resolution for Placebo (%) 0.06 0.10 QoL: CC 0.81 0.99 QoL: F1 0.99 0.84 HR: Cardiovascular Mortality 1.875 3.125 QoL: F0 0.99 0.84 QoL: DC 0.57 0.99 QoL: F2 0.84 0.99 Multiplicative Factor: Moderate Pruritus 0.92 0.80 Moderate Pruritus for OCA (%) 0.11 0.19 QoL: F3 0.99 0.84 QoL: HCC 0.54 0.99 Severe Pruritus for OCA(%) 0.016 0.027 Mild Pruritus for OCA (%) 0.048 0.080 500,000 1,500,000 2,500,000 3,500,000 4,500,000 Incremental Cost-Effectiveness Ratio ©Institute for Clinical and Economic Review, 2016 Page 33 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 6.4 Potential Budget Impact We also used the cost-effectiveness model to estimate the potential total budgetary impact of OCA for NASH patients, based on assumed patterns of product uptake. Potential Budget Impact Model: Methods We used results from the same model employed for the cost-effectiveness analyses to estimate total potential budget impact. Potential budget impact was defined as the total incremental cost of the OCA therapy for the treated population, calculated as incremental health care costs (including drug costs) minus any offsets in these costs from averted disease progression. All costs were undiscounted and estimated over one- and five-year time horizons. The five-year timeframe was of primary interest, given the potential for cost offsets to accrue over time. The potential budget impact analysis included the entire candidate population for treatment, which was considered to be adult NASH patients with fibrosis stages F1–F3. To estimate the size of the potential candidate population for OCA, we first applied the estimated prevalence of NASH in the United States. We used an estimate of NASH prevalence in the US from a review by Yeh et al.,21 who found estimates in the literature of from 3.5% to 5%. For this analysis, we used the lower estimate of 3.5% of the US population having NASH. Applying this prevalence to the projected 2016 US population would imply approximately 11.3 million individuals with NASH. Because of the difficulty in definitively diagnosing NASH (requiring liver biopsy) and the current lack of effective medical treatments, we assumed that the vast majority of these patients would not be diagnosed at this time. We assumed that 5% of the population with NASH would have been diagnosed and therefore eligible for treatment. Applying this percentage resulted in a candidate population size of approximately 567,000 individuals in the US. ICER’s methods for estimating potential budget impact and calculating value-based benchmark prices are described in detail elsewhere. Briefly, our calculations assumed that the utilization of new drugs occurs without any payer, provider group, or pharmacy benefit management controls in place, to provide an estimate of “unmanaged” drug uptake by five years after launch. In general, we examined six characteristics of the drug or device and the marketplace to estimate “unmanaged” uptake. These characteristics are listed below: • Magnitude of improvement in clinical safety and/or effectiveness • Patient-level burden of illness • Patient preference (ease of administration) • Proportion of eligible patients currently being treated • Primary care versus specialty clinician prescribing/use • Presence or emergence of competing treatments of equal or superior effectiveness ©Institute for Clinical and Economic Review, 2016 Page 34 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Based on our assessment of these criteria, we assigned a new drug or device to one of four categories of unmanaged drug uptake patterns: 1) very high (75% uptake by year 5); 2) high (50% uptake by year 5); 3) intermediate (25% uptake by year 5); and 4) low (10% uptake by year 5). In this analysis, we assumed a 10% uptake pattern for OCA in NASH patients. We assumed that uptake would be low for this drug because, while there is a lack of effective therapeutic alternatives for NASH patients, the use of OCA for these patients would be off label, at least at the beginning of this time frame. Using this approach to estimate potential budget impact, we then compared our estimates to a budget impact threshold that represents a potential trigger for policy mechanisms to improve affordability, such as changes to pricing, payment, or patient eligibility. As described in ICER’s methods presentation, this threshold is based on an underlying assumption that health care costs should not grow much faster than growth in the overall national economy. From this foundational assumption, our potential budget impact threshold is derived using an estimate of growth in US gross domestic product (GDP) +1%, the average number of new drug approvals by the FDA each year, and the contribution of spending on retail and facility-based drugs to total health care spending. Calculations are performed as shown in Table 10. For 2015-16, therefore, the five-year annualized potential budget impact threshold that should trigger policy action to manage affordability is calculated to total approximately $904 million per year for new drugs. Table 10. Calculation of Potential Budget Impact Threshold Item Parameter Estimate Source 1 Growth in US GDP, 2015-2016 (est.) +1% 3.75% World Bank, 2015 2 Total health care spending ($) $3.08 trillion CMS NHE, 2014 3 Contribution of drug spending to total health 13.3% CMS National Health care spending (%) Expenditures (NHE), Altarum Institute, 2014 4 Contribution of drug spending to total health $410 billion Calculation care spending ($) (Row 2 x Row 3) 5 Annual threshold for net health care cost $15.4 billion Calculation growth for ALL new drugs (Row 1 x Row 4) 6 Average annual number of new molecular 34 FDA, 2014 entity approvals, 2013-2014 7 Annual threshold for average cost growth per $452 million Calculation individual new molecular entity (Row 5 ÷ Row 6) 8 Annual threshold for estimated potential $904 million Calculation budget impact for each individual new molecular entity (doubling of Row 7) ©Institute for Clinical and Economic Review, 2016 Page 35 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Potential Budget Impact Model: Results Table 11 below presents the potential budget impact of one year and five years of OCA in the candidate population, assuming the uptake patterns previously described. Results are presented for both one-year and five-year time horizons. Results from the potential budget impact model showed that, with the uptake pattern assumptions mentioned above, an estimated 11,340 individuals would receive OCA in the first year. After one year of treatment, with net annual costs of approximately $69,500 per patient, one-year budget impact is estimated to be $788.6 million. Over the entire five-year time horizon, we estimated that “unmanaged” uptake would lead to approximately 56,700 persons taking OCA. Across the full five-year time horizon, the weighted potential budgetary impact (i.e., adjusted for differing periods of drug utilization and associated cost-offsets) is approximately $95,400 per patient. Total potential budgetary impact over five years is approximately $5.4 billion, with an average budget impact per year of approximately $1.08 billion. This annualized potential budget impact is 120% of the budget impact threshold of $904 million for a new drug. Table 11. Estimated Total Potential Budget Impact (BI) of OCA Analytic Horizon = 1 Year Analytic Horizon = 5 Years Eligible Number Annual BI per Total BI Number Weighted BI Average BI per Population Treated Patient* (millions) Treated per Patient* year (millions) OCA 567,000 11,340 $69,500 $788.6 56,700 $95,400 $1,082 *Weighted budget impact calculated by subtracting cost offsets from drug costs for one-year horizon. For five- year horizon, drug costs and cost offsets apportioned assuming 20% of patients in uptake target initiate therapy each year. Those initiating in year 1 receive full drug costs and cost offsets, those initiating in year 2 receive 80% of drug costs and cost offsets, etc. 6.5 Value-based Benchmark Prices Value-based price benchmarks were not calculated for OCA in the treatment of NASH, given the preliminary nature of the currently available data and that there is no current FDA indication for NASH. 6.6 Summary and Comment We conducted a cost-effectiveness analysis by developing a microsimulation model that simulated the long-term outcomes of NASH patients receiving OCA compared to placebo. We estimated that, in comparison with placebo, treatment with OCA would marginally decrease the 15-year cumulative incidence of decompensated cirrhosis from 10% to 8.8%, hepatocellular carcinoma from 4.7% to ©Institute for Clinical and Economic Review, 2016 Page 36 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 4.2%, liver transplant from 0.9% to 0.8%, and liver-related deaths from 12.9% to 11.3%, respectively. In addition, treatment with OCA increased 15-year transplant-free survival from 68.6% to 69.9%. Using the wholesale acquisition cost of OCA of $69,350/year, the incremental cost- effectiveness of OCA was estimated to be approximately $2.75 million per QALY. The use of OCA in NASH patients would not be considered cost-effective using the commonly used willingness-to-pay thresholds of $100,000 to $150,000/QALY gained. The results were most sensitive to the price of OCA, and would achieve a cost-effectiveness ratio of $150,000/QALY when the annual price of OCA is below approximately $5,100 per year. As with any model, ours has some limitations. First, the only comparator against OCA was placebo, but in reality there are a few treatment options available to patients with NASH. Current guidelines from the American Association for the Study of Liver Diseases (AASLD) recommend pioglitazone as a treatment for NASH,7 but it is poorly tolerated because of side effects including weight gain and exacerbation of congestive heart failure.75 Vitamin E is also recommended, but a meta-analysis found that it has no histologic benefits,76 and its use has been associated with increased all-cause mortality.77 In addition, lifestyle intervention has been shown to cause remission of NASH in a high proportion of patients; however, this analysis excludes this therapy because we assumed that OCA would be prescribed to individuals with inadequate results from lifestyle intervention.78 Most notably, prospective studies indicate that bariatric surgery may be a highly effective at inducing NASH remission;79,80 however, there are no randomized controlled trial data to confirm the effectiveness of surgery for treating NASH, and AASLD guidelines do not recommend bariatric surgery specifically for this purpose. If our analysis included any of these NASH treatments—even those with limited effectiveness—OCA would have become even less cost-effective. Second, we used data from a Phase II study that estimated the efficacy of OCA in NASH patients. An ongoing Phase III study will provide more robust data in the future, and our model will be updated when the results from this study become available. We also made the assumption that patients who experienced NASH “remission” in the model would not later relapse to progressive disease. In reality, it is likely that a subset of these patients would later relapse. Due to this assumption, our model may overstate the cost-effectiveness of OCA. Another limitation is that our model draws on hepatitis C data to inform cost and quality-of-life parameters. Given the lack of economic and quality of life data for NASH health states, our model assumed that NASH-associated cirrhosis and hepatitis-associated cirrhosis involve similar treatment costs and that NASH patients experience similar decrements in quality of life as in hepatitis C patients. Sensitivity analysis on these parameters showed that model outcomes were robust to uncertainty surrounding these cost and quality-of-life inputs. We also used the cost-effectiveness model to estimate the potential total budgetary impact of OCA for NASH patients over five years. Assuming that “unmanaged” uptake would lead to 10% of eligible ©Institute for Clinical and Economic Review, 2016 Page 37 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents patients (or approximately 56,700 persons) taking OCA, total potential budgetary impact over five years is approximately $5.4 billion, with an average budget impact per year of approximately $1.08 billion. This annualized potential budget impact is 120% of the budget impact threshold of $904 million for a new drug. **** This is the first ICER review of obeticholic acid for NASH. ©Institute for Clinical and Economic Review, 2016 Page 38 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 7. Voting Results 7.1 About the New England CEPAC Process During public meetings of the New England CEPAC, the Council deliberates and votes on key questions related to the systematic review of the clinical evidence, an economic analysis of the applications of the medical technologies or treatments under examination, and the supplementary information presented. Council members are selected for three year terms and are intentionally selected to represent a range of expertise and diverse perspectives. To maintain the objectivity of New England CEPAC and ground the conversation in the interpretation of the published evidence, members are not pre-selected based on the topic being addressed. Acknowledging that any judgment of evidence is strengthened by real-life clinical and patient perspectives, clinical representatives with expertise in the subject matter are recruited for each meeting topic and provide input to Council members before the meeting to help clarify their understanding of the interventions being analyzed in the evidence review. The same clinical experts serve as a resource to the Council during their deliberation, and they help form recommendations with the Council on ways the evidence can be applied to policy and practice. At each meeting, after the Council votes, a Policy Roundtable discussion is held with the Council, clinical experts, and representatives from relevant manufacturers, provider groups, payers, and patient groups. This is intended to bring stakeholders into the discussion on how best to apply the evidence to guide patient education, clinical practice, and coverage and public policies. Participants on Policy Roundtables are selected for their expertise on the specific meeting topic, are different for each meeting, and do not vote on any questions. At the July 15, 2016 meeting, the Council discussed issues regarding the application of the available evidence to help patients, providers, and payers address the important questions related to the treatment of primary biliary cholangitis and non-alcoholic steatohepatitis with obeticholic acid. Following an evidence presentation and public comments, the Council voted on key questions concerning the clinical effectiveness and value of obeticholic acid. These questions are developed by the ICER research team for each assessment, with input from the New England CEPAC Advisory Board to ensure that the questions are framed to address the issues that are most important in applying the evidence to support clinical practice and medical policy decisions. The voting results are presented in the section below, along with comments reflecting considerations mentioned by the Council members during the voting process. In its deliberations and voting related to value, the Council made use of a value assessment framework with four different components of care value, a concept which represents the long-term perspective, at the individual patient level, on patient benefits and the incremental costs to achieve ©Institute for Clinical and Economic Review, 2016 Page 39 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents those benefits. The four components of care value are comparative clinical effectiveness, incremental cost per outcomes achieved, additional benefits or disadvantages, and contextual considerations regarding the illness or therapy. Figure 12. Care Value Framework There are four elements to consider when deliberating on care value: 1. Comparative clinical effectiveness is a judgment of the overall difference in clinical outcomes between two interventions (or between an intervention and placebo), tempered by the level of certainty possible given the strengths and weaknesses of the body of evidence. New England CEPAC uses the ICER Evidence Rating Matrix as its conceptual framework for considering comparative clinical effectiveness. 2. Incremental cost per outcomes achieved is the average per-patient incremental cost of one intervention compared to another to achieve a desired “health gain,” such as an additional stroke prevented, case of cancer diagnosed, or gain of a year of life. Alternative interventions are compared in terms of cost per unit of effectiveness, and the resulting comparison is presented as a ratio: a “cost per outcome achieved.” Relative certainty in the cost and outcome estimates continues to be a consideration. As a measure of incremental costs per outcomes achieved, ICER follows common academic and World Health Organization (WHO) standards by using cost per quality-adjusted life years (QALYs) and adopting thresholds at $100,000 per QALY and $150,000 per QALY as guides to reasonable ratios of incremental costs per outcomes achieved. 3. Other benefits or disadvantages refers to any significant benefits or disadvantages offered by the intervention to the individual patient, caregivers, the delivery system, other patients, or the public that would not have been considered as part of the evidence on comparative clinical effectiveness. Examples of other benefits include mechanisms of treatment delivery that require many fewer visits to the clinician’s office, treatments that reduce disparities ©Institute for Clinical and Economic Review, 2016 Page 40 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents across various patient groups, and new potential mechanisms of action for treating clinical conditions that have demonstrated low rates of response to currently available therapies. Other disadvantages could include increased burden of treatment on patients or their caregivers. For each intervention evaluated, it will be open to discussion whether other benefits or disadvantages such as these are important enough to factor into the overall judgment of care value. There is no quantitative measure for other benefits or disadvantages. 4. Contextual considerations include ethical, legal, or other issues (but not cost) that influence the relative priority of illnesses and interventions. Examples of contextual considerations include whether there are currently any existing treatments for the condition, whether the condition severely affects quality of life or not, and whether the condition affects priority populations. There is no quantitative measure for the role of contextual considerations in an overall judgment of care value. 7.2 Clinical Effectiveness Voting Results For patients with nonalcoholic steatohepatitis (NASH) and fibrosis, is the evidence adequate to demonstrate a net health benefit with the addition of obeticholic acid to usual care (e.g., lifestyle interventions, treatment with vitamin E, etc.)? Yes: 0 votes No: 14 votes Comments: The New England CEPAC voted unanimously that the evidence was insufficient to demonstrate a net health benefit of OCA in comparison to usual care for NASH, given the preliminary nature of the clinical trial data for the condition and the ongoing status of a large Phase III trial with interim data not expected to be available until 2017. 7.3 Care Value Voting Results Given the available evidence for patients with NASH, what is the care value* of adding obeticholic acid to UDCA alone? NO VOTE TAKEN Comments: Because the panel voted unanimously that there was not sufficient evidence to ascertain clinical effectiveness in NASH, there was no vote on the care value of OCA. ©Institute for Clinical and Economic Review, 2016 Page 41 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 8. Roundtable Discussions and Key Policy Implications Following its deliberation on the evidence, the New England CEPAC engaged in a moderated discussion about the use of obeticholic acid for the treatment of NASH with a Policy Roundtable that included two clinical experts, one patient, a public payer representative, a private payer representative and a manufacturer representative. This discussion reflected multiple perspectives and opinions, and therefore, none of the recommendations below should be taken as a consensus view held by all participants. The names of the Policy Roundtable participants are shown below. All participants completed potential conflict-of-interest forms in advance of the meeting; their written statements can be found in Appendix F. Roundtable Participant Association Kathleen Corey, MD, MPH Director, MGH Fatty Liver Clinic, Massachusetts General Hospital Gastrointestinal Unit Judith Donovan Patient with PBC Barbara Henry, RPh Senior Clinical Pharmacy Coordinator, Harvard Pilgrim Health Care Juan Carlos Lopez-Talavera, MD, PhD Senior Vice President, Global Medical Affairs, Intercept Pharmaceuticals Daniel S. Pratt, MD Clinical Director, Liver Transplantation; Director, Autoimmune & Cholestatic Liver Center, Massachusetts General Hospital Tom Simpatico, MD Chief Medical Officer, Vermont Department of Health Access The roundtable discussion was facilitated by Dr. Steven Pearson, MD, MSc, President of ICER. The main themes and recommendations from the discussion are organized by audience and summarized below. ©Institute for Clinical and Economic Review, 2016 Page 42 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Patients Work with researchers and manufacturers to include patient-relevant outcomes in clinical studies, particularly measures of quality of life as well as measures and determinants of treatment adherence. Quality of life metrics are important to support decision-making by patients, clinicians, and insurers, but are usually not required for market authorization. Collaborative efforts between researchers, manufacturers, and patient groups are needed for producing a wide range of evidence. Considering the largely asymptomatic nature of NASH, adherence to treatment should be part of the broader range of evidence. Consider participating in clinical trials. The evidence base for treating NASH is relatively poor compared to other chronic liver diseases. NASH is a unique disease with a growing burden on individuals and health systems across the country. With increasing recognition of the importance of the disease, the number of clinical trials for NASH has increased dramatically in recent years. Yet the stigma associated with the disease may prevent many individuals from volunteering for clinical study. Patients can help speed the regulatory and payer consideration of novel medications by participating in these trials. Clinicians Refrain from administering OCA to patients with NASH outside of the clinical trial environment. All participants at the NE CEPAC meeting unanimously agreed that the current evidence is insufficient to demonstrate a net health benefit of OCA in comparison to usual care due to the preliminary nature of the trial data. The evidence is therefore not adequate to support off-label use, even for a select number of patients. Clinicians should instead encourage patients to participate in clinical research programs that can help establish a more solid evidence base for OCA. Improve communication, care infrastructure, and care coordination between primary and specialty care. With a growing patient population and new possible treatments for NASH, clinicians need to plan for managing the demand for services. In particular, the referral process from primary to specialty care should be clear, and the role of both primary care physicians and specialists in ongoing monitoring and management of comorbid conditions should be clarified. ©Institute for Clinical and Economic Review, 2016 Page 43 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Insurers In anticipation of the possibility of OCA gaining FDA approval for use in NASH, insurers should explore opportunities for indication-specific pricing as one mechanism for addressing the tension between initial pricing of OCA for an orphan condition and later use for a different indication among a large patient population. Due to its asymptomatic nature in earlier stages, a small percentage of the estimated 10 million NASH patients in the US are aware of their disease. Increasing awareness of the condition and efforts to encourage screening are likely to change this picture moving forward, so insurers will likely need to trigger policy actions to manage the affordability of OCA. Given that OCA’s price for PBC is set, insurers and manufacturers should explore possible steps to manage OCA’s affordability in the much larger NASH population. These could include agreement on a different price specific to the NASH indication or a weighted price for both indications. A recent ICER Policy Summit report provides a historical perspective on indication-specific pricing as well as guidance on potential policy solutions: http://icer-review.org/material/isp-white-paper/ Manufacturers Work with researchers to integrate non-invasive diagnostic tools alongside liver biopsy in clinical trials to advance knowledge on their potential use for patient selection and follow-up. Currently, the diagnosis of NASH can only be made with histological evidence from a liver biopsy. Biopsies are also necessary for assessing the progression of the disease and assessing the effectiveness of treatment. Considering the invasive nature of liver biopsy and the sheer number of NASH patients, routine use of serial biopsy will be challenging. Evidence on the accuracy and clinical utility of non-invasive tools for diagnosis, prognosis, and monitoring is clearly necessary. Integrating the use of non-invasive tests alongside liver biopsy in manufacturer-sponsored trials represents a pragmatic and cost-effective way to produce this evidence. Explore opportunities for indication-specific pricing as one mechanism for addressing the tension between initial pricing of OCA for an orphan condition and later use for a different indication among a large patient population. Anticipating the possibility of OCA gaining FDA approval for use in NASH, discussions on reimbursement will be enhanced if the manufacturer engages early on in discussions with insurers on innovative payment mechanisms such as indication-specific pricing. Such a collaborative ©Institute for Clinical and Economic Review, 2016 Page 44 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents partnership can assist in aligning the price with value to patients, which often improves both affordability and access to medication. Clinical Research Community Include formal and intensive weight loss programs as a formal comparator intervention in trials. When this is not possible, describe patient eligibility criteria to enable future indirect comparisons. NASH is substantially improved or completely resolved in a large percentage of patients who lose at least 7-10% of their weight. Formal and intensive weight loss programs are a first-line treatment for all patients with NASH. The incremental value of treatment with pharmaceuticals should be tested through formal comparison to weight-loss interventions alone in future trials. Even when this is not possible, clear descriptions of patient eligibility criteria and baseline characteristics in separate trials of medication and weight-loss interventions can enable indirect comparisons of their effects. Include outcomes of cardiovascular-related morbidity and mortality as well as overall mortality in trials for NASH. Patients with NASH are at increased risk of cardiovascular disease, and the clinical impact of OCA’s short-term effects on blood lipid levels is currently unknown. Clinicians need to understand the impact of OCA on cardiovascular-related morbidity well as all-cause mortality in order to have a complete picture of the long-term net health benefits of OCA in comparison to usual care. Develop evidence on non-invasive diagnostic tools to advance knowledge on their potential use for patient selection and follow-up. OCA is one of several compounds being tested for their efficacy and safety in NASH. Evidence on the accuracy and clinical utility of non-invasive tools for diagnosis, prognosis, risk stratification, and monitoring is urgently needed to support the use of these treatments. Integrating the use of non- invasive tests in future clinical study should aid in our understanding of their potential use as a complement to liver biopsy or possible replacement in some circumstances. ©Institute for Clinical and Economic Review, 2016 Page 45 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents References 1. Filozof C, Goldstein BJ, Williams RN, Sanyal A. Non-Alcoholic Steatohepatitis: Limited Available Treatment Options but Promising Drugs in Development and Recent Progress Towards a Regulatory Approval Pathway. Drugs. 2015;75(12):1373-1392. 2. Rinella ME. Nonalcoholic fatty liver disease: a systematic review. JAMA. 2015;313(22):2263- 2273. 3. Torres DM, Harrison SA. Nonalcoholic Fatty Liver Disease. In: Feldman M, Friedman LS, Brandt LJ, eds. Sleisenger and Fordtran's gastrointestinal and liver disease : pathophysiology/diagnosis/management. Tenth edition. ed. Philadelphia, PA: Saunders/Elsevier; 2016:1428-1441. 4. Bazick J, Donithan M, Neuschwander-Tetri BA, et al. Clinical Model for NASH and Advanced Fibrosis in Adult Patients With Diabetes and NAFLD: Guidelines for Referral in NAFLD. Diabetes Care. 2015;38(7):1347-1355. 5. Corey KE, Klebanoff MJ, Tramontano AC, Chung RT, Hur C. Screening for Nonalcoholic Steatohepatitis in Individuals with Type 2 Diabetes: A Cost-Effectiveness Analysis. Dig Dis Sci. 2016. 6. Hardy T, Anstee QM, Day CP. Nonalcoholic fatty liver disease: new treatments. Curr Opin Gastroenterol. 2015;31(3):175-183. 7. Chalasani N, Younossi Z, Lavine JE, et al. The diagnosis and management of non-alcoholic fatty liver disease: practice Guideline by the American Association for the Study of Liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association. Hepatology. 2012;55(6):2005-2023. 8. Adams LA, Sanderson S, Lindor KD, Angulo P. The histological course of nonalcoholic fatty liver disease: a longitudinal study of 103 patients with sequential liver biopsies. J Hepatol. 2005;42(1):132-138. 9. Charlton MR, Burns JM, Pedersen RA, Watt KD, Heimbach JK, Dierkhising RA. Frequency and outcomes of liver transplantation for nonalcoholic steatohepatitis in the United States. Gastroenterology. 2011;141(4):1249-1253. 10. Sanyal A, Poklepovic A, Moyneur E, Barghout V. Population-based risk factors and resource utilization for HCC: US perspective. Curr Med Res Opin. 2010;26(9):2183-2191. 11. Blais P, Husain N, Kramer JR, Kowalkowski M, El-Serag H, Kanwal F. Nonalcoholic fatty liver disease is underrecognized in the primary care setting. Am J Gastroenterol. 2015;110(1):10-14. 12. Gawrieh S, Chalasani N. Pharmacotherapy for Nonalcoholic Fatty Liver Disease. Semin Liver Dis. 2015;35(3):338-348. 13. Intercept. Intercept Receives Breakthrough Therapy Designation From FDA for Obeticholic Acid for Nonalcoholic Steatohepatitis (NASH) With Liver Fibrosis. 2015; http://ir.interceptpharma.com/releasedetail.cfm?ReleaseID=893699. Accessed 2016-05-07. 14. Neuschwander-Tetri BA, Loomba R, Sanyal AJ, et al. Farnesoid X nuclear receptor ligand obeticholic acid for non-cirrhotic, non-alcoholic steatohepatitis (FLINT): a multicentre, randomised, placebo-controlled trial. Lancet. 2015;385(9972):956-965. 15. Mudaliar S, Henry RR, Sanyal AJ, et al. Efficacy and safety of the farnesoid X receptor agonist obeticholic acid in patients with type 2 diabetes and nonalcoholic fatty liver disease. Gastroenterology. 2013;145(3):574-582.e571. ©Institute for Clinical and Economic Review, 2016 Page 46 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 16. Hameed B, Terrault N, Gill RM, et al. Clinical and metabolic effects associated with weight loss and obeticholic acid in nonalcoholic steatohepatitis (NASH). Hepatology. 2015;62(13). 17. Hameed B, Terrault N, Gill R, et al. Separate and combined effects of obeticholic acid and weight loss in nonalcoholic steatohepatitis (NASH). Journal of hepatology. 2015;62(22). 18. Neuschwander-Tetri B, Sanyal A, Loomba R, et al. Obeticholic acid for NASH: Benefits in a high risk subgroup and the effects of concomitant statin use. Journal of Hepatology. 2015;62:S272. 19. Kowdley KV, Abdelmalek MF, McCullough A, et al. Evaluation of effects of concomitant medications for NASH and associated comorbidities on histological improvements with obeticholic acid. Gastroenterology. 2016;150(4):S1144. 20. Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology. 2005;41(6):1313-1321. 21. Yeh M, Raju M, Younes M. Pathology of nonalcoholic steatohepatitis. http://emedicine.medscape.com/article/2038493-overview. Accessed May 25, 2016. 22. Hui JM, Kench JG, Chitturi S, et al. Long-term outcomes of cirrhosis in nonalcoholic steatohepatitis compared with hepatitis C. Hepatology. 2003;38(2):420-427. 23. Khan FZ, Perumpail RB, Wong RJ, Ahmed A. Advances in hepatocellular carcinoma: Nonalcoholic steatohepatitis-related hepatocellular carcinoma. World J Hepatol. 2015;7(18):2155-2161. 24. Vilar-Gomez E, Martinez-Perez Y, Calzadilla-Bertot L, et al. Weight Loss Through Lifestyle Modification Significantly Reduces Features of Nonalcoholic Steatohepatitis. Gastroenterology. 2015;149(2):367-378.e365; quiz e314-365. 25. Xu J, Murphy SL, Kochanek KD, Bastian BA. Deaths: Final Data for 2013. Natl Vital Stat Rep. 2016;64(2):1-119. 26. Younossi ZM, Stepanova M, Afendy M, et al. Changes in the prevalence of the most common causes of chronic liver diseases in the United States from 1988 to 2008. Clin Gastroenterol Hepatol. 2011;9(6):524-530 e521; quiz e560. 27. Wong RJ, Aguilar M, Cheung R, et al. Nonalcoholic steatohepatitis is the second leading etiology of liver disease among adults awaiting liver transplantation in the United States. Gastroenterology. 2015;148(3):547-555. 28. Mahady SE, George J. Management of nonalcoholic steatohepatitis: an evidence-based approach. Clin Liver Dis. 2012;16(3):631-645. 29. Corey KE, Rinella ME. Medical and Surgical Treatment Options for Nonalcoholic Steatohepatitis. Dig Dis Sci. 2016;61(5):1387-1397. 30. Rinella ME. PBC and NASH: Progress on the Horizon. ACG LGS Regional Post-Graduate Course. New Orleans, Louisiana. March 2016. 2016; http://s3.gi.org/meetings/lgs2016/16ACG_LGS_Regional_0010_final.pdf. Accessed 2016-05-09. 31. Gov CT. Open phase III studies in NASH. https://clinicaltrials.gov/ct2/results?term=Nonalcoholic+Steatohepatitis&recr=Open&phase=2. Accessed 2016-05-15. 32. BIOTECHS N. Phase 3 NASH Trials: Genfit's Drug Will Beat Intercept's OCA To Market. 2015; http://www.nashbiotechs.com/nash-biotech-analysis/point-of-view/phase-3-nash-trials- genfits.html. Accessed 2016-05-14. 33. Day CP, Anstee QM. Foreword. Nonalcoholic Fatty Liver Disease. Semin Liver Dis. 2015;35(3):203-206. 34. Kowdley KV, Wilson LA, Van Natta ML, Pai RK, Sanyal AJ. Efficacy and safety of vitamin E for nonalcoholic steatohepatitis: Combined analysis of three controlled trials. Journal of hepatology. 2015;62(22). ©Institute for Clinical and Economic Review, 2016 Page 47 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 35. Sanyal AJ, Miller V. Regulatory Science and Drug Approval for Alcoholic and Nonalcoholic Steatohepatitis. Gastroenterology. 2016. 36. Office UOotFRaGP. 21 CFR 601.41,. http://www.ecfr.gov/cgi-bin/text- idx?SID=5bd53ffe761893165cff6d45f277907c&mc=true&node=se21.7.601_141&rgn=div8. Accessed 2016-05-07. 37. Ekstedt M, Hagstrom H, Nasr P, et al. Fibrosis stage is the strongest predictor for disease-specific mortality in NAFLD after up to 33 years of follow-up. Hepatology. 2015;61(5):1547-1554. 38. Kleiner DE, Bedossa P. Liver Histology and Clinical Trials for Nonalcoholic Steatohepatitis- Perspectives from 2 Pathologists. Gastroenterology. 2015;149(6):1305-1308. 39. Ratziu V, Goodman Z, Sanyal A. Current efforts and trends in the treatment of NASH. J Hepatol. 2015;62(1 Suppl):S65-75. 40. Mehal WZ. We are not in kansas anymore: The unfamiliar landscape of NASH. Hepatology. 2016;63(3):682-684. 41. Ratziu V. Novel Pharmacotherapy Options for NASH. Dig Dis Sci. 2016;61(5):1398-1405. 42. Wittich CM, Burkle CM, Lanier WL. Ten common questions (and their answers) about off-label drug use. Mayo Clinic proceedings. 2012;87(10):982-990. 43. Wright AP, Desai AP, Bajpai S, King LY, Sahani DV, Corey KE. Gaps in recognition and evaluation of incidentally identified hepatic steatosis. Dig Dis Sci. 2015;60(2):333-338. 44. Rinella ME, Lominadze Z, Loomba R, et al. Practice patterns in NAFLD and NASH: real life differs from published guidelines. Therapeutic advances in gastroenterology. 2016;9(1):4-12. 45. Cook DJ, Mulrow CD, Haynes RB. Systematic reviews: synthesis of best evidence for clinical decisions. Annals of internal medicine. 1997;126(5):376-380. 46. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. International journal of surgery (London, England). 2010;8(5):336-341. 47. Ollendorf DA, Pearson SD. An integrated evidence rating to frame comparative effectiveness assessments for decision makers. Med Care. 2010;48(6 Suppl):S145-152. 48. Singh S, Khera R, Allen AM, Murad MH, Loomba R. Comparative effectiveness of pharmacological interventions for nonalcoholic steatohepatitis: A systematic review and network meta-analysis. Hepatology. 2015;62(5):1417-1432. 49. Trauner M, Halilbasic E, Kazemi-Shirazi L, et al. Therapeutic role of bile acids and nuclear receptor agonists in fibrosing cholangiopathies. Dig Dis. 2014;32(5):631-636. 50. Hossain N, Afendy A, Stepanova M, et al. Independent predictors of fibrosis in patients with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol. 2009;7(11):1224-1229, 1229.e1221- 1222. 51. Intercept Pharmaceuticals Announces Results of Phase 2 Trial of OCA in NASH Patients in Japan [press release]. October 28, 2015 2015. 52. EPGonline.org. INT 747 (obeticholic acid) fails to achieve goals in Japanese study for non- alcoholic steatohepatitis (NASH)- Intercept Pharma + Sumitomo - 30-Oct-2015. 2015; http://www.epgonline.org/news/int-747-obeticholic-acid-fails-to-achieve.cfm. 53. Misra VL, Khashab M, Chalasani N. Non-Alcoholic Fatty Liver Disease and Cardiovascular Risk. Current gastroenterology reports. 2009;11(1):50-55. 54. Arias E. United states life tables, 2011. National vital statistics reports: from the Centers for Disease Control and Prevention, Natio. 2015;64(11):1-63. 55. Brunt EM, Kleiner DE, Wilson LA, Belt P, Neuschwander-Tetri BA. Nonalcoholic fatty liver disease (NAFLD) activity score and the histopathologic diagnosis in NAFLD: distinct clinicopathologic meanings. Hepatology. 2011;53(3):810-820. ©Institute for Clinical and Economic Review, 2016 Page 48 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 56. Saab S, Gordon S, Park H, Sulkowski M, Ahmed A, Younossi Z. Cost‐effectiveness analysis of sofosbuvir plus peginterferon/ribavirin in the treatment of chronic hepatitis C virus genotype 1 infection. Alimentary pharmacology & therapeutics. 2014;40(6):657-675. 57. Truven Health Analytics. RedBook Online. 2016; http://www.redbook.com/redbook/online/. 58. Chong CAKY, Gulamhussein A, Heathcote EJ, et al. Health-state utilities and quality of life in hepatitis C patients. The American journal of gastroenterology. 2003;98(3):630-638. 59. Siebert U, Sroczynski G, Rossol S, et al. Cost effectiveness of peginterferon-2b plus ribavirin versus interferon -2b plus ribavirin for initial treatment of chronic hepatitis C. Gut. 2003;52(3):425. 60. Hanmer J, Lawrence WF, Anderson JP, Kaplan RM, Fryback DG. Report of nationally representative values for the noninstitutionalized US adult population for 7 health-related quality-of-life scores. Medical Decision Making. 2006;26(4):391-400. 61. Hawe E, McBride D, Balp MM, Tian H, Halliday A, Stull DE. EQ-5D Utilities in Chronic Spontaneous/Idiopathic Urticaria. PharmacoEconomics. 2016;34(5):521-527. 62. Singh S, Allen AM, Wang Z, Prokop LJ, Murad MH, Loomba R. Fibrosis progression in nonalcoholic fatty liver vs nonalcoholic steatohepatitis: a systematic review and meta-analysis of paired-biopsy studies. Clin Gastroenterol Hepatol. 2015;13(4):643-654.e641-649; quiz e639-640. 63. Adams LA, Lymp JF, St Sauver J, et al. The natural history of nonalcoholic fatty liver disease: a population-based cohort study. Gastroenterology. 2005;129(1):113-121. 64. White DL, Kanwal F, El-Serag HB. Association between nonalcoholic fatty liver disease and risk for hepatocellular cancer, based on systematic review. Clin Gastroenterol Hepatol. 2012;10(12):1342-1359.e1342. 65. Mahady SE, Wong G, Craig JC, George J. Pioglitazone and vitamin E for nonalcoholic steatohepatitis: a cost utility analysis. Hepatology. 2012;56(6):2172-2179. 66. Sanyal AJ, Banas C, Sargeant C, et al. Similarities and differences in outcomes of cirrhosis due to nonalcoholic steatohepatitis and hepatitis C. Hepatology. 2006;43(4):682-689. 67. Ascha MS, Hanouneh IA, Lopez R, Tamimi TA, Feldstein AF, Zein NN. The incidence and risk factors of hepatocellular carcinoma in patients with nonalcoholic steatohepatitis. Hepatology. 2010;51(6):1972-1978. 68. Davis KL, Mitra D, Medjedovic J, Beam C, Rustgi V. Direct economic burden of chronic hepatitis C virus in a United States managed care population. J Clin Gastroenterol. 2011;45(2):e17-24. 69. Lang K, Danchenko N, Gondek K, Shah S, Thompson D. The burden of illness associated with hepatocellular carcinoma in the United States. J Hepatol. 2009;50(1):89-99. 70. Ratziu V, Bonyhay L, Di Martino V, et al. Survival, liver failure, and hepatocellular carcinoma in obesity-related cryptogenic cirrhosis. Hepatology. 2002;35(6):1485-1493. 71. Saab S, Hunt DR, Stone MA, McClune A, Tong MJ. Timing of hepatitis C antiviral therapy in patients with advanced liver disease: a decision analysis model. Liver Transpl. 2010;16(6):748- 759. 72. Fattovich G, Pantalena M, Zagni I, et al. Effect of hepatitis B and C virus infections on the natural history of compensated cirrhosis: a cohort study of 297 patients. Am J Gastroenterol. 2002;97(11):2886-2895. 73. Eckman MH, Talal AH, Gordon SC, Schiff E, Sherman KE. Cost-effectiveness of screening for chronic hepatitis C infection in the United States. Clin Infect Dis. 2013;56(10):1382-1393. 74. U.S_Department. US Department of Health and Human Services. Unadjusted patient survival, deceased donor liver transplants survival at 3 months, 1 year, 5 years, and 10 years . http://www.srtr.org/annual_Reports/archives/2009/2009_Annual_Report/914a_rec- dgn_li.htm. Updated 2010. Accessed May 19, 2015. ©Institute for Clinical and Economic Review, 2016 Page 49 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 75. Dyson JK, Anstee QM, McPherson S. Republished: Non-alcoholic fatty liver disease: a practical approach to treatment. Postgraduate medical journal. 2015;91(1072):92-101. 76. Musso G, Gambino R, Cassader M, Pagano G. A meta-analysis of randomized trials for the treatment of nonalcoholic fatty liver disease. Hepatology. 2010;52(1):79-104. 77. Miller ER, 3rd, Pastor-Barriuso R, Dalal D, Riemersma RA, Appel LJ, Guallar E. Meta-analysis: high-dosage vitamin E supplementation may increase all-cause mortality. Annals of internal medicine. 2005;142(1):37-46. 78. Promrat K, Kleiner DE, Niemeier HM, et al. Randomized controlled trial testing the effects of weight loss on nonalcoholic steatohepatitis. Hepatology. 2010;51(1):121-129. 79. Lassailly G, Caiazzo R, Buob D, et al. Bariatric Surgery Reduces Features of Nonalcoholic Steatohepatitis in Morbidly Obese Patients. Gastroenterology. 2015;149(2):379-388; quiz e315- 376. 80. Mathurin P, Hollebecque A, Arnalsteen L, et al. Prospective study of the long-term effects of bariatric surgery on liver injury in patients without advanced disease. Gastroenterology. 2009;137(2):532-540. 81. U.S. Preventive Services Task Force. Procedure Manual. Agency for Healthcare Research and Quality;2008. ©Institute for Clinical and Economic Review, 2016 Page 50 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents APPENDICES ©Institute for Clinical and Economic Review, 2016 Page 51 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Appendix A. Evidence Review Methods and PRISMA Table A1. PRISMA 2009 Checklist # Checklist item TITLE Title 1 Identify the report as a systematic review, meta-analysis, or both. ABSTRACT Structured summary 2 Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number. INTRODUCTION Rationale 3 Describe the rationale for the review in the context of what is already known. Objectives 4 Provide an explicit statement of questions being addressed with reference to participants, interventions, comparisons, outcomes, and study design (PICOS). METHODS Protocol and registration 5 Indicate if a review protocol exists, if and where it can be accessed (e.g., Web address), and, if available, provide registration information including registration number. Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) and report characteristics (e.g., years considered, language, publication status) used as criteria for eligibility, giving rationale. Information sources 7 Describe all information sources (e.g., databases with dates of coverage, contact with study authors to identify additional studies) in the search and date last searched. Search 8 Present full electronic search strategy for at least one database, including any limits used, such that it could be repeated. Study selection 9 State the process for selecting studies (i.e., screening, eligibility, included in systematic review, and, if applicable, included in the meta-analysis). Data collection process 10 Describe method of data extraction from reports (e.g., piloted forms, independently, in duplicate) and any processes for obtaining and confirming data from investigators. Data items 11 List and define all variables for which data were sought (e.g., PICOS, funding sources) and any assumptions and simplifications made. ©Institute for Clinical and Economic Review, 2016 Page 52 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Risk of bias in individual 12 Describe methods used for assessing risk of bias of individual studies (including specification of whether this was done studies at the study or outcome level), and how this information is to be used in any data synthesis. Summary measures 13 State the principal summary measures (e.g., risk ratio, difference in means). Synthesis of results 14 Describe the methods of handling data and combining results of studies, if done, including measures of consistency (e.g., I2) for each meta-analysis. Risk of bias across studies 15 Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies). Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified. RESULTS Study selection 17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram. Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations. Risk of bias within studies 19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12). Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot. Synthesis of results 21 Present results of each meta-analysis done, including confidence intervals and measures of consistency. Risk of bias across studies 22 Present results of any assessment of risk of bias across studies (see Item 15). Additional analysis 23 Give results of additional analyses, if done (e.g., sensitivity or subgroup analyses, meta-regression [see Item 16]). DISCUSSION Summary of evidence 24 Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers). Limitations 25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias). Conclusions 26 Provide a general interpretation of the results in the context of other evidence, and implications for future research. FUNDING Funding 27 Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review. From: Moher D, Liberati A, Tetzlaff J, Altman DG. The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 6(6): e1000097. doi:10.1371/journal.pmed1000097 ©Institute for Clinical and Economic Review, 2016 Page 53 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Search Strategies Table A2: Search Strategy of Medline 1996 to Present with Daily Update, EBM Reviews - Cochrane Database of Systematic Reviews, EBM Reviews - Cochrane Central Register of Controlled Trials 1 exp Fatty Liver/ 15,607 2 ((fatty and (liver* or hepat*)) or steatohepat* or NAFL* or NASH*).mp. 37,963 3 1 or 2 38,175 4 (obeticholic acid or OCA or INT-747).mp. 665 5 3 and 4 32 Date of Search: April 12, 2016 Table A3: Search Strategy of Embase on April 12, 2016 #4 #3 AND [humans]/lim AND [english]/lim NOT [medline]/lim 73 #3 #1 AND #2 152 #2 'obeticholic acid' OR oca OR 'int 747' 2,774 #1 'fatty liver'/exp OR (fatty AND (liver* OR hepat*) OR steatohepat* OR nafl*or AND nash*) 49,819 Study Selection We performed screening at both the abstract and full-text level. Two investigators screened abstracts identified through electronic searches according to the inclusion and exclusion criteria described earlier. All exclusions were validated by a third reviewer. We did not exclude any study at abstract-level screening due to insufficient information. For example, an abstract that did not report an outcome of interest would be accepted for further review in full text. We retrieved the citations that were accepted during abstract-level screening for full text appraisal. Two investigators reviewed full papers and provided justification for exclusion of each excluded study; a third investigator resolved any discrepancies in selection as necessary. Data Extraction and Quality Assessment Summary tables of extracted data are available in Appendix E. We abstracted data from conference abstracts and posters affiliated with publications included in the evidence review. We used criteria published by the U.S. Preventive Services Task Force (USPSTF) to assess the quality of RCTs and comparative cohort studies, using the categories “good,” “fair,” or “poor.”81 Guidance for quality ratings using these criteria is presented below. Good: Meets all criteria: Comparable groups are assembled initially and maintained throughout the study; reliable and valid measurement instruments are used and applied equally to the groups; interventions are spelled out clearly; all important outcomes are considered; and appropriate attention is paid to confounders in analysis. In addition, intention to treat analysis is used for RCTs. ©Institute for Clinical and Economic Review, 2016 Page 54 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Fair: Studies were graded "fair" if any or all of the following problems occur, without the fatal flaws noted in the "poor" category below: Generally comparable groups are assembled initially but some question remains whether some (although not major) differences occurred with follow-up; measurement instruments are acceptable (although not the best) and generally applied equally; some but not all important outcomes are considered; and some but not all potential confounders are addressed. Intention to treat analysis is done for RCTs. Poor: Studies were graded "poor" if any of the following fatal flaws exists: Groups assembled initially are not close to being comparable or maintained throughout the study; unreliable or invalid measurement instruments are used or not applied equally among groups (including not masking outcome assessment); and key confounders are given little or no attention. For RCTs, intention to treat analysis is lacking. Assessment of Bias As part of our quality assessment, we evaluated the evidence base for the presence of potential publication bias. Given the emerging nature of the evidence base for newer treatments, we performed an assessment of publication bias using the clinicaltrials.gov database of trials. We scanned the site to identify studies completed more than two years ago that would have met our inclusion criteria and for which no findings have been published. Any such studies identified provided qualitative evidence for use in ascertaining whether there was a biased representation of study results in the published literature. Although this process did not culminate in the suggestion of a publication bias, we are aware of one completed study in Japan on use of OCA among NASH patients that has not been published. Although this study was not listed in ClinicalTrials.Gov, the European manufacturer disseminated press releases about the trial, which was sponsored by a Japanese pharmaceutical company. Data Synthesis and Statistical Analyses Given the small numbers of relevant studies for OCA in NASH, we judged there to be no role for formal meta-analysis to generate pooled estimates of treatment effect. Figure A1. PRISMA Flow Chart Showing Results of Literature Search for OCA in PBC and NASH ©Institute for Clinical and Economic Review, 2016 Page 55 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents 234 potentially relevant references screened 176 citations excluded 18 Population 16 Intervention/Comparator 77 Study Type 58 references for full 65 Duplicates text review 36 citations excluded (not a population of interest, unrelated to OCA, abstracts/posters with duplicated data, 22 TOTAL inappropriate study type) 3 RCTs (1 PBC, 2 NASH) 19 conference abstracts/posters (15 PBC, 4 NASH) ©Institute for Clinical and Economic Review, 2016 Page 56 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Appendix B. Clinical Guidelines The American Gastroenterological Association, the American Association for the Study of Liver Diseases, the American College of Gastroenterology (2012) http://www.gastrojournal.org/article/S0016-5085(12)00494-5/pdf For treatment of biopsy-confirmed NASH, AASLD guidelines recommend Vitamin E as a first-line pharmacotherapy for non-diabetic patients. Vitamin E should be administered daily at a dose of 800 IU/day. It is not recommended for use by patients with diabetes, those with NAFLD who have not had a liver biopsy to confirm NASH, or those with NASH cirrhosis. Pioglitazone, a thiazolidnedione, may be used in the treatment of NASH, though its long-term safety and efficacy has not been well studied in non-diabetic patients. UDCA is not recommended for treatment of NASH. Bariatric surgery is not contraindicated for otherwise obese patients with NASH or NAFLD without established cirrhosis, but should not be considered a specific treatment option for NASH. Routinely repeated liver biopsies for patients with NASH are not recommended. World Gastroenterology Organisation (2014) http://journals.lww.com/jcge/Fulltext/2014/07000/World_Gastroenterology_Organisation_Global. 4.aspx The World Gastroenterology Organisation provides recommendations for lifestyle and pharmacologic interventions for management of NASH. The WGO recommends proper control of diabetes, hyperlipidemia, and cardiovascular risks for patients with NASH, noting that use of atorvastatin and pravastatin have shown improvements for patients with NASH. Weight loss of 5%- 10% is recommended in addition to regular exercise. Vitamin E or pentoxifylline can be added, but are considered to be experimental. Bariatic surgery may be an option for patients with morbid obesity but should be considered early on, as surgery is often not an option for patients with cirrhosis. In cases of liver failure, liver transplantation is successful, but NASH may recur after transplant. ©Institute for Clinical and Economic Review, 2016 Page 57 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Appendix C. Previous Systematic Reviews and Technology Assessments Because the approval for obeticholic acid is pending at the time of this report, we were not able to identify any previous health technology assessments of OCA for NASH. However, Singh et al. published a systematic review summarized in the following paragraph. No other systematic reviews were identified through the literature search.  Sing S, Allen AM, Wang Z, et al. Fibrosis progression in nonalcoholic fatty liver vs. nonalcoholic steatohepatitis: A Systematic review and meta-analysis of paired biopsy results. Clinical Gastroenterology and Hepatology. 2015;13:643-654.48 Singh et al. conducted a systematic review of cohort studies and RCTs conducted among adult patients with histologic diagnosis of NAFLD with a repeat biopsy performed at least a year later. The primary outcome of intereste was estimating fibrosis progression rate (FPR) of patioents with NAFLD, nonalcoholic fatty liver (NAFL), and NASH with baseline 0 fibrosis. FPR was calculated by number of stages of change between biopsy time periods. From 1,994 unique studies, the authors ultimately included 11 observational studies (NAFLD n=411; NASH n=261). Although two RCTs were identified, these were excluded because FPR could not be obtained. Nearly half of the patients in the observational studies had diabetes. At baseline, the stages of fibrosis were distributed as follows: Stage 0 35.8%, Stage 1 32.5%, Stage 2 16.7%, Stage 3 9.3%, and Stage 4 5.7%. Upon follow- up examination 2145.5 person-years later, 33.6% had progressed by at least one stage of fibrosis since baseline, 43.1% remained stable, and 22.3% had improvement in fibrosis stage. Seven studies provided NASH-specific data for 116 patients: 34.5% had progressive fibrosis, 38.8% remained stable, and 26.7% had improvement in fibrosis. A meta-analysis of NASH patients with baseline F0 (n=21) revealed an annual FPR of 0.14 stages (95% CI 0.07-0.21), which calculates to ~7 years to progress to Stage 1. Among all patients, factors associated with progressive fibrosis included hypertension and a low AST:ALT ratio at the time of baseline biopsy. ©Institute for Clinical and Economic Review, 2016 Page 58 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Appendix D. Ongoing Studies Title/ Trial Sponsor Study Design Comparators Patient Population Primary Outcomes Estimated Completion Date REGENERATE Phase 3 OCA 10mg vs. Adults with NASH Histologic October 2021 (NCT02548351) Double-blinded OCA 25g vs. improvement Multicenter Placebo Liver-related Randomized Global RCT clinical outcomes Phase 3 Study to Evaluate the Impact on NASH with Fibrosis of Obeticholic Acid Treatment CONTROL Phase 2 OCA 5mg + Adults with NASH LDL August 2016 (NCT02633956) Double-blinded Atorvastatin vs. Multicenter OCA 10mg + Combination OCA and RCT Atorvastatin vs. Statins for Monitoring OCA 25mg + of Lipids Atorvastatin vs. Placebo Source: www.ClinicalTrials.gov (NOTE: studies listed on site include both clinical trials and observational studies) ©Institute for Clinical and Economic Review, 2016 Page 59 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Appendix E. Summary Evidence Tables Author, Pub. Intervention Inclusion and Year Study (n) Dosing Exclusion Patient Characteristics Outcomes Harms (Trial) Design Schedule Criteria Quality rating Publication in Phase II 1) OCA 25mg Adults ≥ 18 1) 2) Primary outcome: Pruritus, n (%): Lancet RCT po qd (n=141; years Mean age, yrs 52 51 ≥2-point decrease in centrally scored NAFLD 1) 33 (23%) Multicenter 110 in ITT Histological e/o Male, n (%) 43 53 Activity Score w/o worsening fibrosis by 2) 9 (6%) Neuschwander- Double- analysis) definite or (30) (37) treatment end (ITT results), n (%): p<0.0001 Tetri 2015 blinded borderline Mean weight, 100 96 1) 50 (45%) ITT analysis 2) Placebo NASH based on kg (SD) (23) (18) 2) 23 (21%) Nausea, (FLINT) same size liver bx ≤90d Hyperlipidemia, 87 86 RR 1.9 (95% CI 1.3-2.8); p=0.0002 vomiting, tablet daily before n (%) (62) (61) diarrhea, n: good (n=142; 109 in randomization Diabetes, n (%) 75 74 No change in various subgroup analyses. 1) 12 ITT analysis) NAFLD activity (53) (52) 2) 12 score ≥4, with Mean ALT, U/L 83 82 Secondary outcome: Administered ≥1 in each (SD) (49) (51) Mean change in NAFLD score (SD): for 72 weeks component of Mean ALP, U/L 82 81 1) -1.7 (1.8) with 24 weeks the score (SD) (29) (25) 2) -0.7 (1.8) post- Mean tot chol, 4.9 4.8 RR -0.9 (95% CI -1.3 to -0.5); p<0.0001 administration Excluded µmol/L (SD) (1.2) (1.2) follow-up cirrhosis, other Resolution of NASH, n (%): Mean HDL, 1.1 1.1 causes of liver 1) 22 (22%) µmol/L (SD) (0.3) (0.4) dz, substantial 2) 13 (13%) Vitamin E in 29 32 EtOH RR 1.5 (95% CI 0.9-2.6); p=0.08 last 6 months, (21) (23) consumption, n (%) and other Patients w/ improved fibrosis, n (%): Antilipidemic 72 64 confounding 1) 36 (35%) agent in last 6 (51) (45) conditions 2) 19 (19%) months, n (%) (listed in RR 1.8 (95% CI 1.1-2.7); p=0.004 Definite 114 111 protocol) steatohepatitis, (81) (79) Mean change in values from baseline to 72 n %) weeks (SD): Mean NAFLD 5.3 5.1 1) 2) p-value score (SD) (1.3) (1.3) ALT (U/L) -38 (47) -18 (44) <0.001 ©Institute for Clinical and Economic Review, 2016 Page 60 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Author, Pub. Intervention Inclusion and Year Study (n) Dosing Exclusion Patient Characteristics Outcomes Harms (Trial) Design Schedule Criteria Quality rating AST (U/L) -27 (37) -10 (31) 0.0001 ALP (U/L) 12 (26) -6 (20) <0.0001 GGT (U/L) -37 (70) -6 (48) <0.0001 Tot chol 0.16 -0.19 0.0009 (mmol/L) (1.07) (0.96) HDL -0.02 0.03 0.01 (mmol/L) (0.20) (0.19) HOMA-IR 15 (50) 4 (29) 0.01 Weight (kg) -2.3 0.0 0.008 (6.7) (6.1) SF-36 0 (7) -1 (7) 0.22 physical SF-36 0 (9) 1 (9) 0.65 mental Abstract in See FLINT See FLINT See FLINT See FLINT Weight loss, n (%): Hepatology 1) 43 (42) Secondary 2) 29 (30) Hameed, 2015 analysis of p=0.08 200 patients (FLINT) with baseline Subanalysis: and end of Change in NALFD Activity Score by weight loss treatment status liver biopsy 1) 2) W/ wt loss -2.4 -1.4 w/o wt loss -1.2 -0.4 p-value <0.001 0.006 Change in ALT by weight loss status 1) 2) W/ wt loss -42 -27 ©Institute for Clinical and Economic Review, 2016 Page 61 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Author, Pub. Intervention Inclusion and Year Study (n) Dosing Exclusion Patient Characteristics Outcomes Harms (Trial) Design Schedule Criteria Quality rating w/o wt loss -34 -11 p-value 0.15 0.01 Change in LDL by weight loss status 1) 2) p-value W/ wt loss 22 -17 <0.001 Poster in J of See FLINT See FLINT See FLINT See FLINT Change in LDL (mg/dL) by weight loss status Hepatology 1) 2) p-value Secondary W/ wt loss 23 -17 <0.001 Hameed, 2015 analysis of w/o wt loss 0 -2 NR 200 patients (FLINT) with baseline Change in HDL by weight loss status and end of 1) 2) p-value treatment W/ wt loss -0.6 3.6 NR liver biopsy Abstract in J of See FLINT See FLINT See FLINT See FLINT R= received any treatment; NR = not received Gastroenterology Medication Histological OCA Placebo Secondary improvement Kowdley, 2016 analysis to Antidiabetic Fibrosis R *37% 18% determine if NR 33% 22% (FLINT) the use of Primary R 54%ⱡ 19% endpoint NR 42% 26% other Insulin Fibrosis R 46% 15% concomitant NR 32% 20% medications Primary R 54%ⱡ 15% influenced the endpoint NR 47%* 25% effects of OCA Antilipidemic Fibrosis R 36%* 17% on histological NR 34% 23% improvement Primary R 50%ⱡ 17% in patients endpoint NR 47% 33% with NASH Statin Fibrosis R 40%* 17% NR 30% 22% Primary R 55%◊ 15% endpoint NR 41% 31% ©Institute for Clinical and Economic Review, 2016 Page 62 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Author, Pub. Intervention Inclusion and Year Study (n) Dosing Exclusion Patient Characteristics Outcomes Harms (Trial) Design Schedule Criteria Quality rating *p<0.05; ⱡp<0.01; ◊p<0.0001 Poster in J of See FLINT See FLINT See FLINT See FLINT Subanalysis: Hepatology ≥2 point improvement in NAFLD activity score: Subanalysis of 1) 50% Neuschwander- pts w/ more 2) 31% Tetri, 2015 severe NASH p=0.001 (stage 2-3 (FLINT) fibrosis or NASH resolution: stage 1 1) 18% fibrosis w/ 2) 6.5% DM, obesity, p=0.03 or ALT≥60 1) n=85 Fibrosis regression ≥1 stage: 2) n=77 1) 39% 2) 22% p=0.012 Fibrosis progression: 1) 16% 2) 29% p=0.047 Statin therapy initiated on OCA reduced LDL levels to levels seen among pts treated w/ statins at baseline. Publication in Phase II 1) 25mg OCA Patients w/ Male, n (%): Primary outcomes: Pt w/ any AEs, Gastroenterology RCT qd for 6 wks Type 2 DM and 1) 14 (70) Percent change in low-dose glucose infusion rate n (%): Multicenter (n=20) NAFLD 2) 9 (43) (SD): 1) 9 (45) Mudaliar, 2013 Double- 3)10 (43) 1) 28.0 (40.2); p=0.019 2) 16 (76) blinded 2) 50mg OCA Excluded high 2) 20.1 (32.6); p=0.60 3) 14 (61) (NCT00501592) qd for 6 wks AST, ALT, bili, Mean age, years: 1+2) 24.5 (36.6); p=0.011 (n=21) DM agents 1) 52.7 3) -5.5 (35.9) Pt w/ tx- fair (not other than 2) 50.5 related AEs, n randomized 3) Placebo qd metformin or 3) 53.1 Percent change in high-dose glucose infusion (%): evenly, poor for 6 wks sulfonylureas, rate (SD): 1) 1 (5) labeling, does (n=23) EtOH/substance Mean body weight, kg (SD): 1) 18.3 (36.3); p=0.036 2) 8 (38) ©Institute for Clinical and Economic Review, 2016 Page 63 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Author, Pub. Intervention Inclusion and Year Study (n) Dosing Exclusion Patient Characteristics Outcomes Harms (Trial) Design Schedule Criteria Quality rating not disclose ITT) abuse in prior 2 1) 108.6 (23.0) 2) 10.8 (21.8); p=0.076 3) 6 (26) yrs, heart/renal 2) 106.4 (25.1) 1+2) 15.0 (30.4); p=0.025 disease 3) 104.2 (25.6) 3) -5.4 (24.3) Pruritus, n (%): 1) 0 (0) Glucose, mg/dL Secondary outcomes: 2) 1 (5) 1) 149 Change in mean LFTs (U/L) and lipids (mg/dL): 3) 2 (9) 2) 132 1) 2) 3) 3) 159 AST -2 5 5 Infections, n (p=0.12) (p=0.73) (%): ALT -10 10 11 1) 5 (22) (p=0.003) (p=0.84) 2) 2 (10) ALP 14 27 0 3) 2 (10) (p=0.003) (p<0.001) GGT -37 -22 5 GI disorder, n (p<0.001) (p<0.001) (%): tchol 18 13 8 1) 1 (5) (p=0.08) (p=0.15) 2) 6 (29) HDL -2 -6 0 3) 4 (17) (p=0.42) (p=0.01) Change in FGF19 (ng/L): 1) 85; p=0.006 2) 176; p<0.0001 3) 7 Percent change in mean body weight: 1) 1; p=0.096 2) 1.9; p=0.008 3) ~0 Mean change in ELF score (SD): 1) -0.2 (0.4); p=0.004 2) 0.03 (0.8); p=0.21 3) 0.3 (0.5) ©Institute for Clinical and Economic Review, 2016 Page 64 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Appendix F. Policy Roundtable Conflict of Interest Disclosures Conflict of Interest Disclosure Language: Do you have a relationship with a health care company (life sciences manufacturer, health plan, health plan association, life sciences manufacturer association, etc.) that falls under any of the following categories? A relationship extends to immediate family members and/or any entity in which you have an interest. Please mark all that apply within the last 12 months. Multiple selections are permitted. Kathleen Corey, MD, MPH, MMSc √ Manufacturer support of research in the clinical Director, MGH Fatty Liver Clinic area of this meeting in which you are participating Massachusetts General Hospital Gastrointestinal Unit Please describe: I am a principle investigator for trials including those from Novartis, Intercept, Galectin, Conatus and Tobira. Judith Donovan I have no conflicts of interest as defined in the above Patient with PBC categories Barbara Henry, RPh √ Receipt or potential receipt of anything of monetary Lead Clinical Pharmacy Specialist value, including but not limited to, salary or other Harvard Pilgrim Health Care payments for services such as consulting fees or honoraria in excess of $5,000 Please describe: Salary from a Health Plan; Employee of Health Plan Juan Carlos Lopez-Talavera, MD, √ Receipt or potential receipt of anything of monetary PhD value, including but not limited to, salary or other Senior Vice President, Global payments for services such as consulting fees or Medical Affairs honoraria in excess of $5,000 Intercept Pharmaceuticals √ Equity interests such as individual stocks, stock options or other ownership interests in excess of $10,000. Ownership of stock in a mutual fund over which an individual has no trading control does not count toward this item. √ Manufacturer support of research in the clinical ©Institute for Clinical and Economic Review, 2016 Page 65 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents area of this meeting in which you are participating √ Any other relationship that could reasonably be considered a financial conflict of interest Please describe: I am actively employed by Intercept Pharmaceuticals. Daniel Pratt, MD I have no conflicts of interest as defined in the above Clinical Director, Liver categories Transplantation; Director of the Autoimmune Cholestatic Liver Center Massachusetts General Hospital Thomas Simpatico, MD √ Receipt or potential receipt of anything of monetary Chief Medical Officer value, including but not limited to, salary or other Vermont Department of Health payments for services such as consulting fees or Access honoraria in excess of $5,000 Please describe: Salary from a Public Health Plan; Employee of Public Health Plan ©Institute for Clinical and Economic Review, 2016 Page 66 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Appendix G. Public Comments This section includes summaries of the public comments prepared for the New England CEPAC Public Meeting on July 15, 2016 in Portland, ME. These summaries were prepared by those who delivered the public comments at the meeting. Conflict of interest disclosures are included at the bottom of each statement. (1) DONNA R CRYER, JD, PRESIDENT & CEO, GLOBAL LIVER INSTITUTE SUMMARY REMARKS FOR REPORT I am Donna Cryer, CEO of the Global Liver Institute, a patient who has lived with biliary disease and liver transplant for over 20 years. I serve on the ABIM GI Board, Executive Committee of the Liver Forum, WHO/PAHO Technical Expert Panel, and FDA patient representative, in fact serving on the OCA advisory committee. I am not hard to find. Value of ICER: Illegitimate You were not mandated by Congress, clinicians don’t look to you instead of their medical societies, patients haven’t delegated our voice or confidence to determine affordability. Due to tortured terminology, muddled methodology, and lack of transparency, even members of your advisory board have criticized you. Purpose of ICER: Payer Voice You’ve admitted that timing is to give payers leverage to negotiate with manufacturers. Be honest about your purpose/effect– justify access restrictions, be self-appointed “pricing police” and provide a chilling effect to innovation which we certainly cannot afford in liver disease. Stakeholder Engagement: Substantially None There are no patients on the voting panel. Scheduling the policy roundtable after the vote is also anti- informed engagement. Examples of patient engagement -- UNOS, FDA, PCORI, National Health Council and Faster Cures. PBC Comments OCA for PBC falls below your budget impact threshold. Question 1 invalid unless panel purports to have superior clinical expertise than the FDA. Question 2 usurps the personalized judgement and shared decisionmaking of physicians and patients. NASH Comments There are no FDA-approved therapies for NASH. Your questions 3 and 4 are premature and should be removed. Conflict of Interest Disclosure. My husband and I serve on several boards of nonprofit organizations that receive funding from healthcare companies as defined above. Neither of us receive any funds directly or indirectly for work related to PBC or NASH. The Global Liver Institute has received funding in the preceding 12 months from Abbvie, Amgen, Genentech, and Gilead. CryerHealth has received money from Amgen and Esperion. ©Institute for Clinical and Economic Review, 2016 Page 67 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents (2) TERRY WILCOX, EXECUTIVE DIRECTOR, PATIENTS RISING NEW ENGLAND CEPAC PUBLIC COMMENT SUMMARY Patients Rising believes value frameworks like the one ICER is trying to develop are by any objective measure meant to control access first. Not advance treatment first. We gathered in Portland, Maine at the New England CEPAC to discuss two liver dieases PBC and NASH. One disease is being treated effectively by current therapy for many patients. The other has no effective treatment for the most vulnerable cases. NASH is set to pass Hepatitis C as the leading cause of liver transplants by 2020. By 2025 more than 25 million Americans will be living with the disease, 10% of those being children; mostly Hispanic. Tina Dooley is a PBC patient from Houston. She meets the criteria for OCA as she is only a partial responder to the standard of care. Patients like Tina deserve fair access to the best of medical innovation that is right for them. If we as a society focus on the precision capabilities of medical innovation, figure out how to monitor the efficacy of these treatments in real world settings and ultimately create value structures that look at costs in totality across all sectors of healthcare – now that is a care model Patients Rising can stand with. We can do better than the value structure being used in a country with the poorest overall cancer survival rates in Western Europe. We can win the economic battle without sacrificing the access to new therapies patients in this country need. Conflict of Interest Disclosure. Patients Rising has received funding from Amgen, Celgene and PhRMA for specific projects: Voices of Value, Right Patient, Right Treatment, Right Now and Check Yourself to Protect Yourself screenings. We have no conflict of interest relationship with Intercept Pharmaceuticals. (3) JUAN CARLOS LOPEZ-TALAVERA, SVP, HEAD GLOBAL MEDICAL AFFAIRS ICER COMMENTS SUMMARY Intercept is dedicated to developing innovative treatments for progressive, non-viral liver diseases with high unmet need, and we are committed to pricing our medicines based on the value they provide to patients and the healthcare system. Since ICER announced its intended review of Ocaliva (obeticholic acid) in PBC and NASH, we have had a constructive dialogue with the organization regarding their proposed economic models, and provided our feedback on its draft report. PBC is a rare, autoimmune liver disease that progresses to severe life-threatening complications such as liver cirrhosis, liver failure, and ultimately death from liver failure or hepatocellular carcinoma in the absence of liver transplantation. ©Institute for Clinical and Economic Review, 2016 Page 68 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents Despite revisions in the final draft report, ICER’s PBC model still does not accurately reflect the risks associated with disease progression, costs of managing complications, or the impact of PBC on patient quality of life. Thus, ICER’s resulting analysis of Ocaliva significantly undervalues the benefit it provides to patients. While Intercept has completed Phase 3 development of Ocaliva in PBC, it is still being evaluated for NASH in the Phase 3 REGENERATE trial. It is inappropriate to conduct a value analysis of a drug candidate prior to having data from a Phase 3 program. We were pleased to see that ICER’s final draft report acknowledged that the introduction of a value-based price benchmark for OCA in NASH is premature at this time. Finally, while we appreciate ICER’s effort, the evaluation of Ocaliva was compromised by a lack of patient perspective, as reflected in the critical public comments from patient advocacy groups. Conflict of Interest Disclosure. I am actively employed by Intercept Pharmaceuticals. ©Institute for Clinical and Economic Review, 2016 Page 69 Draft Evidence Report - OCA for the treatment of NASH Return to Table of Contents