Cost-Effectiveness of Boceprevir or Telaprevir for Untreated Patients with Genotype 1 Chronic Hepatitis C

Transcription

1 Cost-Effectiveness of Boceprevir or Telaprevir for Untreated Patients with Genotype 1 Chronic Hepatitis C Calogero Camma, 1 Salvatore Petta, 1 Marco Enea, 2 Raffaele Bruno, 3 Fabrizio Bronte, 1 Vincenza Capursi, 2 Americo Cicchetti, 4 Giorgio L. Colombo, 5 Vito Di Marco, 1 Antonio Gasbarrini, 6 and Antonio Craxì, 1 on behalf of the WEF Study Group Randomized controlled trials (RCTs) show that triple therapy (TT) with peginterferon alpha, ribavirin, and boceprevir (BOC) or telaprevir (TVR) is more effective than peginterferon-ribavirin dual therapy (DT) in the treatment of previously untreated patients with genotype 1 (G 1 ) chronic hepatitis C (CHC). We assessed the cost-effectiveness of TT compared to DT in the treatment of untreated patients with G 1 CHC. We created a Markov Decision Model to evaluate, in untreated Caucasian patients age 50 years, weight 70 kg, with G 1 CHC and Metavir F2 liver fibrosis score, for a time horizon of 20 years, the cost-effectiveness of the following five competing strategies: 1) boceprevir response-guided therapy (BOC- RGT); 2) boceprevir IL28B genotype-guided strategy (BOC-IL28B); 3) boceprevir rapid virologic response (RVR)-guided strategy (BOC-RVR); 4) telaprevir response-guided therapy (TVR-RGT); 5) telaprevir IL28B genotype-guided strategy (TVR-IL28B). Outcomes included life-years gained (LYG), costs (in 2011 euros) and incremental cost-effectiveness ratio (ICER). In the base-case analysis BOC-RVR and TVR-IL28B strategies were the most effective and cost-effective of evaluated strategies. LYG was 4.04 with BOC-RVR and 4.42 with TVR-IL28B. ICER compared with DT was per LYG for BOC-RVR and per LYG for TVR-IL28B. The model was highly sensitive to IL28B CC genotype, likelihood of RVR and sustained virologic response, and BOC/TVR prices. Conclusion: In untreated G 1 CHC patients age 50 years, TT with first-generation protease inhibitors is costeffective compared with DT. Multiple strategies to reduce costs and improve effectiveness include RVR or genotype-guided treatment. (HEPATOLOGY 2012;00: ) The estimated global prevalence of hepatitis C virus (HCV) infection is 2.2%, corresponding to about 130 million HCV-positive persons worldwide, most of whom are chronically infected. 1 A recent revision 2 reported that the estimated prevalence of HCV infection in Europe ranges from 0.6% to 5.6%. This is of increasing interest because HCV is a leading cause of both cirrhosis and hepatocellular carcinoma (HCC) in Western countries. The prevalence of HCV-related cirrhosis and its complications will continue to increase through the next decade, and will mostly affect those above age Considering the burden of HCV-related cirrhosis and its complications, the achievement of a sustained virologic response (SVR) is a very important surrogate outcome in the management of chronic hepatitis C (CHC) patients. In fact, viral eradication prevents the development of cirrhosis 4 and its complications, such as esophageal varices 5 and HCC, 6 and leads to a decrease in liver-related death. 7 Abbreviations: BOC, boceprevir; CHC, chronic hepatitis C; DT, dual therapy; G 1, genotype 1; ICER, incremental cost effectiveness ratio; PI, protease inhibitors; PEG-IFN, pegylated interferon; RBV, ribavirin; TVR, telaprevir. From the 1 Sezione di Gastroenterologia, Di.Bi.M.I.S., University of Palermo, Italy; 2 Dipartimento di Scienze Statistiche e Matematiche S. Vianelli, University of Palermo, Palermo, Italy; 3 Division of Infectious and Tropical Diseases, Foundation IRCCS San Matteo Hospital, University of Pavia, Italy; 4 Universita Cattolica del Sacro Cuore, Facolta di Economia, Roma, Italy; 5 University of Pavia, School of Pharmacy, Italy; 6 Universita Cattolica del Sacro Cuore, Facolta di Medicina e Chirurgia, Gastroenterologia Roma, Italy. Received December 20, 2011; accepted March 12, This study was entirely funded by 3P Solution. The funding agency was not involved in the study design or its execution, data management or analysis, article preparation or review, or the decision to submit the article for publication. 1

2 2 CAMMa ET AL. HEPATOLOGY, Month 2012 Dual therapy (DT) with peginterferon alpha (PEG- IFN) and weight-based ribavirin (RBV) for 48 weeks in untreated patients with genotype 1 (G 1 ) CHC, the most prevalent HCV genotype in Western countries, has achieved an SVR rate of about 40%-50%. 8 However, recent data have shown that the presence of the favorable IL28B polymorphism and/or the achievement of rapid virologic response (RVR) identifies patients with higher likelihood of SVR (about 80%) after DT. 9,10 A better response with new agents could significantly reduce the disease burden in coming years. Two NS3-NS4 HCV protease inhibitors (PI), telaprevir (TVR) and boceprevir (BOC), have recently been developed and are ready for use in clinical practice. Randomized controlled trials (RCTs), namely SPRINT2 and ADVANCE, 11,12 showed that these drugs, in combination with DT, achieve SVR in about 65%-75% of untreated G 1 CHC patients. Although these results are very encouraging, the use of these new drugs in clinical practice needs to be carefully evaluated because of such factors as the tolerability profile, the issue of drug-drug interaction, the induction of viral mutations of uncertain significance, and high costs. Because of a lack of consensus regarding the use of this new class of drugs, 13,14 we sought to determine the cost-effectiveness of DT and of triple therapy (TT) with BOC or TVR in untreated patients with G 1 CHC. Materials and Methods Dual Therapy (DT) We defined DT as: 2) PEG-IFN alpha-2b at a dose of 1.5 lg/kg subcutaneously weekly in combination with weight-based oral RBV at a total dose of 600 to 1,400 mg per day; or 2) PEG-IFN alpha-2a at a dose of 180 lg per week in combination with RBV (1,000 mg per day for body weight <75 kg or 1,200 mg per day for body weight 75 kg). Patients were treated with DT according to available evidence 15,16 and recently published EASL guidelines 17 on HCV treatment. Data derived from indirect comparison meta-analysis of 28 published RCTs of PEG-IFN alpha (2a or 2b) and RBV in the treatment of untreated G 1 CHC patients showed no differences between the two PEG- IFNs used, with a mean rate of SVR of 45% (Supporting Fig. 1). Due to the similar effectiveness of the two PEG-IFNs, in this study we defined DT as treatment with any PEG-IFN alpha and weight-based RBV. Supporting Fig. 2A summarizes DT strategy. Effectiveness of Five HCV Protease Inhibitor Treatment Strategies We evaluated five different strategies for the HCV- PI-based antiviral treatment of untreated G 1 CHC patients. Supporting Fig. 2B-E summarizes these strategies. Boceprevir Response-Guided Therapy (BOC-RGT) (Supporting Fig. 2B). Data from the SPRINT-2 RCT 11 were utilized for this analysis. PEG-IFN alpha- 2b was dosed at 1.5 lg/kg subcutaneously weekly, in combination with weight-based oral RBV at a total dose of 600 to 1,400 mg per day. After 4 weeks of PEG-IFN and RBV (lead-in period), all patients received 24 weeks of BOC 800 mg administered orally three times a day, in combination with PEG-IFN and RBV (PR). For patients with extended RVR (ervr), defined as undetectable HCV RNA at treatment weeks 8 through 24 (44% of patients), all three drugs were stopped at week 28. For those without ervr, defined as detectable HCV RNA at week 4 but undetectable at week 24, TT was continued for another 8 weeks. These patients then received an additional 12 weeks of PR, based on analyses done by the U.S. Food and Drug Administration (FDA) 18 and by the European Medicines Agency. 19 This strategy differs from the SPRINT-2 RCT design. Overall, this strategy is defined as boceprevir response-guided therapy (BOC-RGT). In the SPRINT-2 RCT, all treatments were discontinued and considered futile if HCV RNA was detectable at week 24. An SVR rate of 67% associated with the BOC- RGT strategy was derived from the nonblack cohort of the SPRINT-2 RCT. 11 Boceprevir IL28B Genotype-Guided Strategy (BOC-IL28B) (Supporting Fig. 2C). G 1 CHC patients with the IL28B C/C genotype were treated with DT for 48 weeks. The SVR rate of 80% associated with DT treatment was derived from a study by Address reprint requests to: Prof. Calogero Camma, Sezione di Gastroenterologia, University of Palermo, Italy, Piazza delle Cliniche, 2, Palermo, Italy. calogero.camma@unipa.it; fax: þ Copyright VC 2012 by the American Association for the Study of Liver Diseases. View this article online at wileyonlinelibrary.com. DOI /hep Potential conflict of interest: Carlo Camma and Antonio Craxl have received consulting fees from Roche and Merck-Sharp-Dohme. Additional Supporting Information may be found in the online version of this article.

3 HEPATOLOGY, Vol. 000, No. 000, 2012 CAMMa ET AL. 3 Ge et al. 9 All patients with the C/T or T/T genotypes received BOC-RGT. The SVR rate of 62% associated with BOC-RGT strategy in this subgroup of patients was derived from analysis of BOC by the FDA. 18 The 20% of IL28B C/C genotype patients treated with DT and who failed to achieve SVR were retreated with 4 weeks of DT followed by 44 weeks of BOC in combination with PR, achieving an SVR rate of 66% according to RESPOND-2 RCT. 20 Boceprevir RVR-Guided Strategy (BOC-RVR) (Supporting Fig. 2D). All patients received PEG-IFN alpha-2b at the standard dose of 1.5 lg per kg of body weight per week in combination with weightbased RBV for 4 weeks (lead-in period). After the 4- week PR lead-in period, patients who achieved RVR were treated with DT. There was an SVR rate of 85% in patients who achieved RVR, according to available evidence on G 1 CHC patients achieving RVR and treated with DT. 15,16 Patients who did not achieve RVR received BOC- RGT. The SVR rate of 64.4% associated with BOC- RGT was derived from the subgroup of nonblack, no- RVR patients in the SPRINT-2 RCT. 11 The 15% of RVR patients treated with DT and who failed to achieve SVR were retreated with 4 weeks of DT followed by 44 weeks of BOC in combination with PR, achieving an SVR rate of 66% according to RESPOND-2 RCT. 20 Telaprevir Response-Guided Therapy (TVR-RGT) (Supporting Fig. 2E). Data from the ADVANCE RCT 12 were utilized for this analysis. The proposed dosing regimen for TVR was 750 mg given three times daily for 12 weeks (T12) in combination with standard doses of PEG-IFN-2a (180 lg/wk) and RBV (1000 mg per day for body weight <75 kg or 1,200 mg per day for body weight 75 kg) for 24 weeks (T12/PR24) or 48 (T12/PR48) weeks, depending on treatment response. Untreated patients who achieved an ervr, defined as undetectable HCV RNA at weeks 4 and 12, received the T12/PR24 regimen (58% of patients), whereas subjects who failed to achieve ervr received the T12/PR48 regimen (42% of patients). All treatments were discontinued for futility if the HCV RNA was greater than 1,000 IU per ml at week 12, or if HCV RNA decline was lower than 2 log10 at week 12. This strategy is defined as telaprevir response-guided therapy (TVR-RGT). The SVR rate of 74.5% associated with TVR-RGT was derived from the ADVANCE RCT. 12 Telaprevir IL28B Genotype-Guided Strategy (TVR-IL28B) (Supporting Fig. 2F). G 1 CHC patients with the IL28B C/C genotype were treated with DT for 48 weeks. The SVR rate of 80% associated with DT treatment was derived from a study by Ge et al. 9 All patients with the IL28B C/T or T/T genotypes received TVR-RGT. The SVR rate of 74.6% associated with TVR-RGT in this subgroup of patients was also derived from analyses of TVR RCTs done by the FDA. 21 The 20% of IL28B C/C genotype patients treated with DT and who failed to achieve SVR, were retreated with 12 weeks of TVR in combination with PR, followed by 36 weeks of DT, achieving an SVR rate of 66.2% according to REALIZE RCT. 22 Structure of the Model Our base-case was a hypothetical cohort of Caucasian male patients, 50 years old, with a weight of 70 kg, G 1 CHC, and F2 Metavir liver fibrosis score. We assumed that SVR eliminates the risk of developing progressive liver disease. 4-7 We used a semi-markov model to simulate the natural history of CHC in the nonresponder proportion of each treatment strategy over a 20-year time horizon (Table 1) (Supporting Material 1). Figure 1 shows the basic format of the model, in which the transition between the health states was analyzed in 1-year (duration of the treatment cycles) periods (Markovian cycles). Every patient who completed each cycle in any state other than death received 1 life year-gained (LYG). We used LYG as the main measure of effectiveness. Because of relevant differences in the quality of life estimates (utility) reported worldwide 31 in patients with CHC, we assessed quality-adjusted life year (QALY) as a secondary measure of effectiveness. Model creation and analyses were performed using R (R Foundation for Statistical Computing, Vienna, Austria) 32 and Microsoft Excel 2007 (Microsoft, Redmond, WA). Resource Consumption and Costs We conducted a cost analysis of the five treatment strategies, calculating all costs in 2011 euros. 33 Total cost per strategy was the unit cost multiplied by the quantity used. The mean annual cost per treated patient was calculated by taking into account only the direct medical costs, according to the analysis perspective. In particular, the drug costs (PEG-IFN, RBV, BOC, TVR) and the costs associated with disease progression (e.g., diagnostic exams, visits, hospitalization) were considered. The cost of the drugs (PEG-IFNa-2a, PEG-IFN a-2b, and RBV) was calculated based on the

4 4 CAMMa ET AL. HEPATOLOGY, Month 2012 Table 1. Base-Case Values and Ranges Used in Sensitivity Analyses Annual Probability Initial State State after Transition Base case Lower limit Upper limit Ref (a) Probability of Liver Disease Progression Chronic Hepatitis C! Compensated cirrhosis ! Liver-related death! Non liver-related death Compensated Cirrhosis! Decompensated cirrhosis ! Hepatocellular carcinoma ! Liver-related death ! Non liver-related death Decompensated Cirrhosis! Hepatocellular carcinoma ! Liver transplantation ! Liver-related death ! Nonliver-related death Hepatocellular Carcinoma! Liver transplantation ! Liver-related death for BCLC A ! Liver-related death for BCLC B ! Liver-related death for BCLC C ! Liver-related death for BCLC D ! Nonliver-related death Liver Transplantation! Liver-related death ! Nonliver-related death *Averaged over 3 years (b) Weekly Cost of Drugs (2011 euros) Peginterferon a-2a þ ribavirin peginterferon a-2a ribavirin Peginterferon a-2b þ ribavirin peginterferon a-2b ribavirin Peginterferon aþribavirinþtelaprevir 3, , , Estimated (34) peginterferon a-2a ribavirin telaprevir 3, , , Estimated (34) Peginterferon aþribavirinþboceprevir 1, , Estimated (34) peginterferon a-2b ribavirin boceprevir , Estimated (34) (c) Annual costs for disease progression (2011 euros) (c) Annual costs for disease progression (2011 euros) Chronic hepatitis C , 36 Compensated cirrhosis , 36 Decompensated cirrhosis 4, , , , 36 Hepatocellular carcinoma 6, , 36 Liver transplantation 90, , , , 36 Liver transplantation after first year 5, , , , 36 (d) Cost discounting rate, % (e) Base-case value of quality of life Long-term symptoms (Q values) Well 1 Chronic hepatitis C Compensated cirrhosis Decompensated cirrhosis Hepatocellular carcinoma Liver transplantation Liver transplantation after first year pharmacy purchase price. Although the cost of PEG- IFN alpha 2a and PEG-IFN alpha 2b is slightly different, in order to obtain only one cost for DT we considered the mean cost of the two PEG-IFNs. Pricing for both TVR and BOC has not yet been available in Europe, whereas it has been established in the U.S.

5 HEPATOLOGY, Vol. 000, No. 000, 2012 CAMMa ET AL. 5 Fig. 1. Schematic of the Markov model. Every year patients can move between heath states in the model according to a defined transition rate. Specifically, the wholesale acquisition cost (WAC) of 12 weeks of TVR in the U.S. is $49, (equal to 37, in 2012 euros at time of analysis, or 3, weekly). 34 Boceprevir WAC in the U.S. for 24 and 32 weeks of treatment is $26, and $35,200.00, respectively (equal to 20, and 26,755.84, respectively, in 2012 euros at time of analysis, or weekly). 34 Estimates of annual direct costs for each health state included the frequency and costs of inpatient and outpatient visits, diagnostic and laboratory testing, medications, and procedures. These costs were updated based on a previous study 35 in which the medical resource use associated with each disease state was estimated based on the DRG tariffs 36 and national ambulatory fees. The drug costs and costs associated with disease progression are reported in Table 1. Future costs and life-years were discounted at 3% per year. Cost-Effectiveness Analysis We calculated the incremental cost-effectiveness ratio (ICER) among the five treatment strategies in a short- and long-term scenario. In the short-term scenario, the ICERs among the different options were calculated using difference in drug costs in 2011 euros divided by the difference in effectiveness in SVR rate. In the long-term scenario, the ICERs among the different options were calculated using difference in the costs associated with disease progression in 2011 euros divided by the difference in LYG or QALY effectiveness. To evaluate the quality of life, a specific utility value, calculated by means of the Health Utility Index (HUI-Mark III), 37 was associated with each health state considered in the model (Table 1). We performed one-way, two-way, and probabilistic sensitivity analyses to explore the impact of variables on results (Supporting Material 1). Results Base-Case Analysis. The results of our base-case analysis in the short-term scenario are shown in Table 2a and Fig. 2A, which depict the drug costs versus the proportion of patients who achieve SVR among the competing strategies. The DT strategy with PEG-IFN alpha and RBV costs 12,673 per average patient treated and achieved SVR in 45.8% of the cohort. It was therefore the least expensive, but also the least effective, of the five competing strategies. The introduction of PI in all patients (BOC-RGT or TVR-RGT therapies) increased the drug cost significantly, with a significant increase in effectiveness. Specifically, compared with DT the BOC-RGT and the TVR-RGT strategies had an ICER of 85,650 and 118,000, respectively. The BOC-RGT strategy was more expensive and less effective than the BOC-RVR strategy, and was therefore dominated. Similarly, the BOC-IL28B strategy was more expensive than, and similarly effective as, the BOC-RVR strategy and was therefore

6 6 CAMMa ET AL. HEPATOLOGY, Month 2012 Table 2. Results of Cost-Effectiveness Analyses: (a) Short-Term Scenario: Sustained Virological Response (SVR); (b) Long- Term Scenario: Life Year Gained (LYG); (C) Long Term Scenario: Quality-Adjusted Life Year (QALY) (a) Short-Term Scenario Treatment Strategies Costs in 2011 Euros SVR (%) ICER/SVR Base-Case Analysis (2011 Euros) Dual therapy 12, Boceprevir response-guided therapy 30, ,650 Boceprevir IL28B genotype-guided strategy 28, ,500 Boceprevir RVR-guided strategy 27, ,960 Telaprevir response-guided therapy 46, ,000 Telaprevir IL28B genotype-guided strategy 37, ,600 (b) Long-Term Scenario Treatment Strategies Costs in 2011 Euros LYG ICER/LYG Base-Case Analysis (2011 Euros) Dual therapy 18, Boceprevir response-guided therapy 34, ,428 Boceprevir IL28B genotype-guided strategy 31, ,936 Boceprevir RVR-guided strategy 30, ,304 Telaprevir response-guided therapy 49, ,204 Telaprevir IL28B genotype-guided strategy 39, ,455 (c) Long term scenario Treatment Strategies Costs in 2011 Euros QALY ICER/QALY Base-Case Analysis (2011 Euros) Dual therapy 18, Boceprevir response-guided therapy 34, ,520 Boceprevir IL28B genotype-guided strategy 31, ,004 Boceprevir RVR guided-strategy 30, ,650 Telaprevir response-guided therapy 49, ,755 Telaprevir IL28B genotype-guided strategy 39, ,414 dominated. Specifically, compared with DT the BOC- RVR strategy had an ICER of 56,960. Similar results were obtained for TVR treatment. The TVR-RGT strategy was more expensive than, and less effective as, the TVR-IL28B and was therefore dominated. Specifically, compared with DT the TVR- IL28B strategy had an ICER of 74,600. In the long-term scenario, the total cost and both the LYG and the QALY of the different strategies were calculated (Table 2b,c). Figure 2B depicts total costs versus LYG among the competing strategies. Considering BOC, the BOC-RGT strategy was more expensive and less effective than the BOC-RVR strategy and was therefore dominated. When comparing BOC-IL28B to BOC-RVR strategy, under basecase conditions, the LYG was 4.03 and 4.04, respectively. Total costs were 31,469 for the BOC-IL28B strategy and 30,542 for BOC-RVR strategy. The ICERs compared with DT were 8,936 and 8,304 for the BOC-IL28B strategy and BOC-RVR strategy, respectively. Thus, at our base-case costs, the BOC- RVR strategy was more cost-effective than the BOC- IL28B strategy. Similar results were obtained considering QALY instead of LYG. Similarly, the TVR-RGT strategy was more expensive and less effective than the TVR-IL28B strategy and was therefore dominated. Specifically, compared with DT the TVR-IL28B strategy had an ICER of 11,455 for LYG and 6,414 for QALY. One-Way Sensitivity Analyses. One-way sensitivity analyses were done for the two dominant strategies: BOC-RVR and TVR-IL28B. Figure 3 summarizes the results of one-way sensitivity analyses, using tornado diagrams. The costs of HCV PI had a great effect on their cost-effectiveness. In fact, sensitivity analysis with a hypothesized variation of TVR/BOC costs of 630% showed that ICER for LYG ranges significantly from 5,519 to 12,824, and from 8,105 to 16,272 in the BOC-RVR (Fig. 3A) and TVR-IL28B (Fig. 3B) strategies, respectively. For the sensitivity analysis of disease costs, a variation of 20% was assessed and the model was insensitive. With an increase in the disease costs, the ICER for LYG marginally improves in both BOC-RVR (Fig. 3A) and TVR-IL28B (Fig. 3B) strategies. Similar variations were found for both strategies by applying a discount rate ranging from 0% to 5%.

7 HEPATOLOGY, Vol. 000, No. 000, 2012 CAMMa ET AL. 7 Fig. 2. (A) Drug costs (in 2011 euros) and effectiveness evaluated as sustained virologic response (SVR). (B) Disease costs (in 2011 euros) and effectiveness evaluated as life year gained (LYG). The cost-effectiveness of HCV PI was sensitive to change (625%) in transition probabilities from CHC to compensated cirrhosis, but insensitive to change (625%) in transition probabilities from compensated to decompensated cirrhosis (625%). With a faster progression of the disease, the ICER for LYG marginally improves in both the BOC-RVR (Fig. 3A) and TVR-IL28B (Fig. 3B) strategies. A decrease in the SVR rate from 74.6% to 60% in the TVR-IL28B strategy, and from 69.6% to 60% in the BOC-RVR strategy, a plausible expectation in real clinical practice, had a high effect on cost-effectiveness. The ICER for LYG ranged from 9,172 to 18,577 in the BOC-RVR strategy (Fig. 3A) and from 12,189 to 31,031 in the TVR-IL28B strategy (Fig. 3B). Considering the BOC-RVR strategy, the results were highly sensitive to an increase in the RVR rate of between 10% and 40%. In this analysis the ICER for LYG ranged from 6,407 to 13,112 (Fig. 3A). Finally, considering the TVR-IL28B strategy, with a variation of the prevalence of IL28B CC from 10% in black patients to 60% in Asiatic patients, ICER for LYG ranged from 6,630 to 18,699, making it a highly sensitive model (Fig. 3B). Two-Way Sensitivity Analyses. Two-way sensitivity analyses were done for the two dominant strategies, simultaneously considering alternative assumptions about liver disease progression, IL28B genotype, efficacy of antiviral treatment (SVR and RVR), and both drug and disease costs. The results were sensitive to changes in our basecase assumptions for both BOC-RVR (Fig. 4A) and TVR-IL28B (Fig. 4B) strategies. For example, in the BOC-RVR strategy, when we simultaneously considered alternative assumptions about SVR associated with the price of boceprevir, and SVR associated with RVR, even modest changes in our base-case assumption substantially increased the ICERs (Fig. 4A). Similarly, in the TVR-IL28B strategy, when we simultaneously considered alternative assumptions about SVR associated with the price of telaprevir, and SVR associated with IL28B genotype, even modest changes in our base-case assumption substantially increased the ICERs (Fig. 4B). Probabilistic Sensitivity Analysis. Varying all variables simultaneously in the Monte Carlo simulation, and using a willingness-to-pay threshold of 25,000 per LYG, TVR-IL28B and BOC-RVR strategies were cost-effective in 76.6% and 93.6% of the simulations, respectively (Fig. 5). In more than 10,000 interactions, the TVR-IL28B strategy resulted in a gain in LYG 6,274 times, whereas BOC-RVR strategy was preferred 3,726 times. The BOC-RVR strategy had a 34.8% chance of being both more effective and less expensive than the TVR-IL28B strategy. Discussion We found that first-generation HCV PI improved survival by about 4 years, and QALY by about 7 years, in untreated patients with G 1 CHC by increasing the SVR rate about 25%. This gain came at a relatively low cost, resulting in an ICER per QALY of less than 10,000, a cost lower than the generally accepted societal threshold for willingness to pay, 38 and indicating that the cost-effectiveness of HCV PI is favorable. The

8 8 CAMMa ET AL. HEPATOLOGY, Month 2012 Fig. 3. Tornado diagram for the one-way sensitivity analyses of variables that influenced the incremental cost-effectiveness ratio (ICER) per additional life year gained (LYG) of BOC RVR-guided strategy compared with DT (A) and of TVR IL28B genotype-guided strategy compared with DT (B). The horizontal axis represents the incremental cost effectiveness ratio for life year gained (LYG). The width of the bars illustrates the range of cost per additional LYG of BOC RVR-guided strategy (A) and TVR-L28B genotype-guided strategy (B) compared with DT. Upper and lower limits of values evaluated in sensitivity analysis are indicated next to the bars. The bars are ordered from greatest width at the top to least width at the bottom. The ICERs, over the full range of assumptions evaluated, remained < and < per LYG for BOC-RVR and TVR-IL28B strategies, respectively. robustness of these results was confirmed in the probabilistic sensitivity analyses. The cost-effectiveness of first-generation HCV PI was sensitive to BOC and TVR prices, likelihood of RVR and SVR rates, and IL28B CC prevalence. The 2011 update of the practice guidelines by the American Association of the Study of Liver Disease on Hepatitis C 13 does not recommend any selective allocation of patients to triple treatment with firstgeneration HCV PI. In contrast, Aronsohn and Jensen 14 recently proposed a needs-based allocation system in which priority would be given to the sickest patients first, i.e., patients with advanced fibrosis. Therefore, a debated question is if all untreated patients should be treated with TT, especially considering the increase in costs due to HCV PI, in an era in which resource scarcity will be a prominent issue. Our study provides further evidence that a selective allocation system, fulfilling the moral framework of distributive justice, 39 could be a solution for this allocation dilemma and increase cost-effectiveness. Therefore, the key issue is the identification of the best strategy for HCV PI use. We estimated, for the first time to our knowledge, the cost-effectiveness of adding first-generation HCV PI to DT, performing a consistent comparison of five different strategies. Our analyses suggest that BOC- RVR is dominant compared with both BOC-RGT and BOC-IL28B strategies. Similarly, the TVR-IL28B strategy is dominant compared with the TVR-RGT strategy. As a result, we suggest using BOC with RVRguided and TVR with genotype-guided strategies. It is worth underlining that compared with a universal treatment with HCV PI of all untreated G 1 CHC patients, using selective treatment strategies guided by RVR or IL28B genotype we are able to avoid exposure to HCV PI in 25%-33% of patients, respectively, reducing costs and risks, and improving benefits. Therefore, we recommend using PI-free strategies as first-line therapy in patients with IL28B CC genotype or in those who achieve RVR. Regarding cost saving, the cost to treat 1,000 untreated patients

9 HEPATOLOGY, Vol. 000, No. 000, 2012 CAMMa ET AL. 9 Fig. 4. Tornado diagram for the two-way sensitivity analyses of variables that influenced the incremental cost-effectiveness ratio (ICER) per additional life year gained (LYG) of BOC RVR-guided strategy compared with DT (A) and of TVR-IL28B genotype-guided strategy compared with DT (B). The horizontal axis represents the incremental cost effectiveness ratio for life year gained (LYG). The width of the bars illustrates the range of cost per additional LYG of BOC RVR-guided strategy (A) and TVR-L28B genotype-guided strategy (B) compared with DT. Upper and lower limits of values evaluated in sensitivity analysis are indicated next to the bars. The bars are ordered from greatest width at the top to least width at the bottom. The ICERs, over the full range of assumptions evaluated, remained < and < per LYG for BOC-RVR and TVR-IL28B strategies, respectively. with the TVR-RGT strategy would be 46.6 million versus 37.4 million for TVR-IL28B a 24% difference. Based on similar calculations, the BOC-RGT strategy cost is 30.8 million versus 27.6 million for BOC-RVR an 11% difference (Fig. 6). Although the proposed algorithms are useful tools for decision making, the treatment strategy should be carefully agreed upon with the patient, taking into account all the different factors that can interfere with treatment response. In particular, the choice of treatment should Fig. 5. Cost-effectiveness acceptability curve.

10 10 CAMMa ET AL. HEPATOLOGY, Month 2012 Fig. 6. Cost to treat 1,000 untreated patients with genotype 1 chronic hepatitis C with boceprevir or telaprevir as part of triple therapy. be targeted to obtain the best possible treatment in the individual patient without any economic analysis affecting the clinical value and ethical impact of this decision. Therefore, efforts aimed at identifying strong SVR predictors to optimize and personalize DT in untreated patients could be also applied in the HCV PI era to identify patients at higher likelihood of responding to DT, reserving TT only for patients with a low likelihood of response to DT. Thus, stratification of patients according to other variables interfering with SVR achievement (e.g., viral load, steatosis, fibrosis, insulinresistance, vitamin D) could further improve the costeffectiveness of our strategies. Another relevant issue arising from our analysis is that RVR and IL28B genotype-guided strategies are attractive not only in terms of cost-effectiveness but also in terms of safety, because TVR and BOC are associated with more adverse reactions, lower tolerability, and induction of viral mutations. Adverse reactions of TT, leading to treatment discontinuation in 15%- 20% of patients included in RCTs, increase management costs, and are expected to be more frequent in real clinical practice. This issue could also lead to a higher treatment discontinuation and inefficacy in patients who could respond well to DT, finally reducing the likelihood of SVR in the individual patient. TT may cause the selection of drug-resistant viral mutations, which at least hypothetically might compromise future treatment of TT nonresponders to next-generation HCV PI. With this in mind, sparing the use of first-generation HCV PI could further strengthen the usefulness of our proposed strategies. This study has several caveats. The first is that efficacy data are derived from registration trials of HCV PI. In fact, data from RCTs are not directly transferable to clinical practice due to the fact that trial patients are healthier, more adherent, and more intensively monitored. Second, we used the utilities considered acceptable for an Italian population. However, it is well known that utilities may vary widely across different patient subgroups and depend critically on quality-oflife assumptions. 31 Accordingly, we assessed LYG, but not QALY, as the primary measure of effectiveness. Third, other important limitations were the transition probabilities from CHC to cirrhosis, which were assumed to remain constant over time, and assumptions about changes in both RVR and SVR rates, in IL28B C/C prevalence, and in TVR and BOC costs. However, the results were robust under a broad probabilistic sensitivity analysis. Another methodological issue of the current model arises in the use of aggregate rather than individual patient data. Thus, our results reflect group averages rather than individual data. In particular, the lack of data on the SVR rate of patients treated with HCV PI according to IL28B genetic polymorphisms together with other predictors of response (age, body mass index, cirrhosis, and viral load) could also affect the accuracy of the results. More detailed treatment comparisons could be achieved with an analysis of individual patient data. Finally, the study s perspective was not societal. Therefore, our analysis was limited to direct medical costs: indirect costs such as lost productivity and caregiver salaries were not included. In conclusion, we found that TT, including first-generation HCV PI, is a highly cost-effective treatment in untreated G 1 CHC patients. Allocation systems based on RVR or genetic testing strategies are likely to be costeffective compared with universal treatment of all patients with HCV PI, also potentially reducing the risk of side effects and viral resistance associated with treatment. Acknowledgment: Author contributions: C. Camma, S. Petta, M. Enea, F. Bronte, V. Capursi, R. Bruno, A. Cicchetti, G.L. Colombo, V. Di Marco, A. Gasbarrini, and A. Craxì had full control of the study design, data analysis and interpretation, and preparation of article. All authors were involved in planning the analysis and drafting the article. The final draft article was approved by all the authors. References 1. Global burden of disease (GBD) for hepatitis C. Global Burden Of Hepatitis C Working Group. J Clin Pharmacol 2004;44: Hepatitis B and C in the EU neighbourhood: prevalence, burden of disease and screening policies. European Centre for Disease Prevention and Control, September 2010.

11 HEPATOLOGY, Vol. 000, No. 000, 2012 CAMMa ET AL Davis GL, Alter MJ, El-Serag H, Poynard T, Jennings LW. Aging of hepatitis C virus (HCV)-infected persons in the United States: a multiple cohort model of HCV prevalence and disease progression. Gastroenterology 2010;138: Camma C, Di Bona D, Schepis F, Heathcote EJ, Zeuzem S, Pockros PJ, et al. Effect of peginterferon alfa-2a on liver histology in chronic hepatitis C: a meta-analysis of individual patient data. HEPATOLOGY 2004;39: Bruno S, Crosignani A, Facciotto C, Rossi S, Roffi L, Redaelli A, et al. Sustained virologic response prevents the development of esophageal varices in compensated, Child-Pugh class A hepatitis C virus-induced cirrhosis. A 12-year prospective follow-up study. HEPATOLOGY 2010;51: Camma C, Giunta M, Andreone P, Craxì A. Interferon and prevention of hepatocellular carcinoma in viral cirrhosis: an evidence-based approach. J Hepatol 2001;34: Singal AG, Volk ML, Jensen D, Di Bisceglie AM, Schoenfeld PS. A sustained viral response is associated with reduced liver-related morbidity and mortality in patients with hepatitis C virus. Clin Gastroenterol Hepatol 2010;8: McHutchison JG, Lawitz EJ, Shiffman ML, Muir AJ, Galler GW, McCone J, et al; IDEAL Study Team. Peginterferon alfa-2b or alfa-2a with ribavirin for treatment of hepatitis C infection. N Engl J Med 2009;361: Ge D, Fellay J, Thompson AJ, Simon JS, Shianna KV, Urban TJ, et al. Genetic variation in IL28B predicts hepatitis C treatment-induced viral clearance. Nature ;461: Thompson AJ, Muir AJ, Sulkowski MS, Ge D, Fellay J, Shianna KV, et al. Interleukin-28B polymorphism improves viral kinetics and is the strongest pretreatment predictor of sustained virologic response in genotype 1 hepatitis C virus. Gastroenterology 2010;139: e Poordad F, McCone J Jr, Bacon BR, Bruno S, Manns MP, Sulkowski MS, et al; SPRINT-2 Investigators. Boceprevir for untreated chronic HCV genotype 1 infection. N Engl J Med 2011;364: Jacobson IM, McHutchison JG, Dusheiko G, Di Bisceglie AM, Reddy KR, Bzowej NH, et al; ADVANCE Study Team. Telaprevir for previously untreated chronic hepatitis C virus infection. N Engl J Med 2011;364: Ghany M, Nelson D, Strader D, Thomas D, Seeff LB. An update on treatment of genotype 1 chronic hepatitis c virus infection: 2011 practice guidelines by the American Association for the Study of Liver Diseases. HEPATOLOGY 2011;54: Aronsohn A, Jensen D. Distributive justice and the arrival of direct-acting antivirals: who should be first in line? HEPATOLOGY 2011;53: Moreno C, Deltenre P, Pawlotsky JM, Henrion J, Adler M, Mathurin P. Shortened treatment duration in treatment-naive genotype 1 HCV patients with rapid virological response: a meta-analysis. J Hepatol 2010;52: National Institute for Health and Clinical Excellence (NICE) technology appraisal guidance Peginterferon alfa and ribavirin treatment of chronic hepatitis C. Issue date: September European Association for the Study of the Liver. EASL Clinical Practice Guidelines: management of hepatitis C virus infection. J Hepatol 2011; 55: FDA Antiviral Drugs Advisory Committee Meeting. Boceprevir. Available at: MeetingMaterials/Drugs/AntiviralDrugsAdvisoryCommittee/UCM pdf. Accessed August 3, Victrelis [boceprevir]. European Medicines Agency. Available at: Accessed August 17, Bacon BR, Gordon SC, Lawitz E, Marcellin P, Vierling JM, Zeuzem S, et al; HCV RESPOND-2 Investigators. Boceprevir for previously treated chronic HCV genotype 1 infection. N Engl J Med 2011;364: FDA Antiviral Drugs Advisory Committee Meeting. Telaprevir. Available at: MeetingMaterials/Drugs/AntiviralDrugsAdvisoryCommittee/UCM pdf. 22. Zeuzem S, Andreone P, Pol S, Lawitz E, Diago M, Roberts S, et al; REALIZE Study Team. Telaprevir for retreatment of HCV infection. N Engl J Med 2011;364: Drummond AJ, Ho SY, Phillips MJ, Rambaut A. Relaxed phylogenetics and dating with confidence. PLoS Biol 2006;4:e Freeman AJ, Dore GJ, Law MG, Thorpe M, Von Overbeck J, Lloyd AR, et al. Estimating progression to cirrhosis in chronic hepatitis C virus infection. HEPATOLOGY 2001;34: Camma C, Di Marco V, Cabibbo G, Latteri F, Sandonato L, Parisi P, et al. Survival of patients with hepatocellular carcinoma in cirrhosis: a comparison of BCLC, CLIP and GRETCH staging systems. Aliment Pharmacol Ther 2008;28: Angelico M, Cillo U, Fagiuoli S, Gasbarrini A, Gavrila C, Marianelli T, et al. Liver Match Investigators. Liver Match, a prospective observational cohort study on liver transplantation in Italy: study design and current practice of donor-recipient matching. Dig Liver Dis 2011;43: Guido M, Fagiuoli S, Tessari G, Burra P, Leandro G, Boccagni P, et al. Histology predicts cirrhotic evolution of post transplant hepatitis C. Gut 2002;50: Benvegnù L, Noventa F, Bernardinello E, Pontisso P, Gatta A, Alberti A. Evidence for an association between the aetiology of cirrhosis and pattern of hepatocellular carcinoma development. Gut 2001;48: D Amico G, Garcia-Tsao G, Pagliaro L. Natural history and prognostic indicators of survival in cirrhosis: a systematic review of 118 studies. J Hepatol 2006;44: Siebert U, Sroczynski G, Rossol S, Wasem J, Ravens-Sieberer U, Kurth BM, et al; German Hepatitis C Model (GEHMO) Group; International Hepatitis Interventional Therapy (IHIT) Group. Cost effectiveness of peginterferon alpha-2b plus ribavirin versus interferon alpha-2b plus ribavirin for initial treatment of chronic hepatitis C. Gut 2003;52: Salomon JA, Weinstein MC, Hammitt JK, Goldie SJ. Cost-effectiveness of treatment for chronic hepatitis C infection in an evolving patient population. JAMA 2003;290: R Development Core Team (2011). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN , URL ISTAT, Last accessed, February Wolters Kluwer Health. Medi-Span Master Drug Data Base v2.5 (MDDB). Accessed September 20, Munari L, Picciotto A. Recombinant interferon alfa-2b therapy for chronic hepatitis C in Italy: an economic analysis. FORUM Trends Exp Clin Med 1996;6: Conferenza delle Regioni e Provincie Autonome. Tariffa Unica Concenzionale per le prestazioni di assistenza ospedaliera. Regole e tariffe valide per il Rome, 15 December Sullivan SD, Craxì A, Alberti A, Giuliani G, De Carli C, Wintfeld N, et al. Cost effectiveness of peginterferon alfa-2a plus ribavirin versus interferon alfa-2b plus ribavirin as initial therapy for treatment-naïve chronic hepatitis C. Pharmacoeconomics 2004;22: Jonsson B. Changing health environment: the challenge to demonstrate cost-effectiveness of new compounds. Pharmacoeconomics 2004; 22(Suppl 4): Callahan D. Health Technology Assessment Implementation: The Politics of Ethics. Med Decis Making 2012;32:E13-E19.