A pragmatic and cost-effective strategy of a combination therapy of interferon alpha-2b and ribavirin for the treatment of chronic hepatitis c

A pragmatic and cost-effective strategy of a combination therapy of interferon alpha-2b and ribavirin for the treatment Markus Sagmeistera,b, John B. Wongb, Beat Mullhauptc and Background Combination of interferon (IFN) alpha and combination therapy dominates all other strategies. In ribavirin is considered the standard treatment for patients genotype 1 infection, 48 weeks of combination therapy for with chronic hepatitis C. While combination therapy is week-24 responders only prolongs life expectancy at a more effective than IFN alone, the optimal management of favourable cost-effectiveness ratio (CE) of 7135 euros per combination treatment remains uncertain.
quality-adjusted life year (QALY). Taking response factorsother than genotype into account does not add to the Objective To assess a pragmatic and cost-effective effectiveness or cost effectiveness.
strategy for the therapy of treatment-naive patients withchronic hepatitis C.
Conclusion Treating non-cirrhotic patients with chronic hepatitis C according to genotype only is most cost Design Markov model on original data of two randomized effective independent of the number of other known response factors. Eur J Gastroenterol Hepatol 13:483±488& 2001 Lippincott Williams & Wilkins Methods A validated computer simulation model was applied to non-cirrhotic hepatitis C virus (HCV)-infected European Journal of Gastroenterology & Hepatology 2001, 13:483±488 patients. Patient characteristics and ef®cacy of treatment Keywords: cost-effectiveness, costs, hepatitis C, interferon, ribavirin were extracted from two randomized trials reporting on 1445 non-cirrhotic patients. Different strategies were Division of Gastroenterology and Hepatology, University Hospital, Zurich, Switzerland; bDivision of Clinical Decision Making, Informatics and Telemedicine, compared separately for genotype 1 and genotype non-1 New England Medical Center, Tufts University School of Medicine, Boston, USA; (mostly genotype 2/3) infections: (1) no treatment; (2) IFN and cDivision of Gastroenterology and Hepatology, Department of Medicine, for 48 weeks (if at 12 weeks HCV RNA undetectable); (3)IFN and ribavirin for 24 weeks; (4) IFN and ribavirin for Correspondence to M. Sagmeister MD, Division of Gastroenterology and Hepatology, University Hospital, RaÈmistrasse 100, 8091 Zurich, Switzerland 48 weeks; (5) IFN and ribavirin for 48 weeks (if at 24 weeks Tel: ‡41 255 11 11; e-mail: [email protected] HCV RNA undetectable). All strategies were tested for Supported, in part, by unrestricted grants from Essex Chemie AG Lucerne, different combinations of known response factors.
Switzerland, and Roche Pharma (Schweiz) AG, Reinach, Switzerland.
Received 1 August 2000 Revised 8 September 2000 Results In genotype non-1 infection, 24 weeks of tion of interferon alpha-2b with ribavirin (IFN-R; IFN People infected with hepatitis C virus (HCV) will 3 MIU sc TIW and ribavirin 1000/1200 mg orally once develop chronic hepatitis in 80% of cases Most studies report a risk of 10±20% of developing cirrhosis over a period of 20±30 years With established Although combination therapy is more costly than cirrhosis, the annual risk of developing hepatocellular therapy with IFN alone, Younossi et al. have carcinoma (HCC) rises considerably to 1±5% shown a favourable CE of IFN-R compared with IFN.
The most favourable strategy was to perform viral Until recently, 48 weeks of interferon (IFN; 3 million genotyping to determine the duration of IFN-R with units subcutaneously three times a week) was the only genotype 1 patients receiving 48 weeks of IFN-R and effective strategy for the treatment of chronic hepatitis all others receiving 24 weeks of therapy. Wong et al.
C, with sustained response rates (SR) reaching only also demonstrated a favourable CE of IFN-R.
15±20% Although IFN is costly, the cost-effective- ness ratio (CE) for IFN falls within the limit of other Poynard et al. con®rmed IFN-R as the ®rst-line well accepted therapies Recent randomized treatment of chronic hepatitis C, and introduced ®ve trials have shown a higher effectiveness of a combina- favourable independent response factors that are asso- 0954±691X & 2001 Lippincott Williams & Wilkins 484 European Journal of Gastroenterology & Hepatology 2001, Vol 13 No 5 ciated with a sustained response: genotype 2 or 3, ribavirin 1000/1200 mg orally once daily for baseline viral load less than 3.5 million copies/ml, no or portal ®brosis only, female gender, and age less than 3. Combination therapy II (Comb II): after 24 weeks of 40 years. Those patients who are HCV-negative at combination therapy, viral response is determined week 24 and have less than four favourable response and only responders continue treatment for an factors should be treated for an additional 24 weeks.
additional 24 weeks (IFN 3 MIU sc TIW and Poynard et al. argued that an approach that considered ribavirin 1000/1200 mg orally once daily for 48 only virological characteristics (as suggested by Younos- weeks, if at 24 weeks HCV RNA [PCR] undetect- si et al. might be an oversimpli®cation and may lead to wrong treatment decisions in certain patient 4. Combination therapy III (Comb III): patients are treated for 48 weeks (IFN 3 MIU sc TIW and ribavirin 1000/1200 mg orally once daily for The aim of our study was to assess and optimize the CE of different schemes of combination therapy with 5. No treatment: natural history without antiviral treat- IFN-R. This evaluation, re¯ecting actual treatment patterns and costs, was aimed at ®nding a pragmatic, cost-effective strategy for treatment of patients with The patient characteristics and the effectiveness of treatment strategies were extracted from a data set of 1445 patients with mild or moderate hepatitis from two Using standard decision analysis software (Decision randomized trials comparing IFN with IFN-R Maker 7.0 we applied an already established and The mean patient age was 42.2 years (G1, 43.2 years; G validated computer simulation model The mod- non-1, 41.3 years); 34.9% were female; 66.3% had el is based on a Markov simulation of the natural moderate hepatitis; 64.6% of the patients were infected history of hepatitis C. Cohorts of patients move through with G1 and 32.0% were infected with genotype G2 or prede®ned health states over time until all patients have entered the dead state. Patients may progress over time by given transition probabilities to more advanced We de®ned an SR as viral negativity (HCV RNA disease states or death. Morbidity and mortality from negative by PCR, i.e. < 100 copies/ml) 24 weeks after hepatitis C, and mortality from other causes as occur in treatment. For our calculations, the SRs shown in the general Swiss population, were considered. The original model was extended to include therapeutic options for HCC (orthotopic liver transplantation and partial hepatectomy), re¯ecting current treatment mod- The study included total direct costs for the year 1998 from a societal perspective, but excluded indirect costs, such as lost productivity. To calculate the present value Because viral genotype is the most important indepen- of future costs, we used a 3% annual discount rate. The dent response factor and is usually determined cost evaluation followed actual treatment patterns for before any antiviral treatment we examined Switzerland. The costs of complications from HCV (the ®ve different strategies in two patient groups: those annual frequency and type of medical care for each with genotype 1 (G1) and those with genotype non-1 disease state) were estimated by a panel of Swiss (G non-1, mostly genotype 2/3). We considered only physicians (two gastroenterologists and three hepatolo- those patients with mild to moderate hepatitis (includ- gists). Costs for GP surgery visits and laboratory tests ing bridging ®brosis) but not cirrhosis, because the were based on reimbursement tariffs and the costs optimal antiviral treatment of cirrhotic patients is still for drugs were based on current retail prices.
1. IFN monotherapy (IFN): re¯ecting current recom- Probability of sustained viral response (%)Ã mendations the viral response is determined after 12 weeks of IFN treatment. Only responders continue antiviral treatment for an additional weeks, if at 12 weeks HCV RNA [PCR] undetect- ÃData extracted from 1445 patients with mild or moderate hepatitis C treated in 2. Combination therapy I (Comb I): patients are IFN, interferon; tion therapy I; Comb II, combination therapy II; treated for 24 weeks (IFN 3 MIU sc TIW and Combination therapy for chronic hepatitis C Sagmeister et al. 485 Costs of monitoring and antiviral treatment discounted with a rate of 3%. Strategies that are The costs of treatment re¯ected current recommenda- dominated (lower costs and higher effectiveness versus tions for Switzerland and included a liver biopsy, other strategies) were excluded from further evalua- HCV RNA testing, genotyping and GP surgery visits tions. Similarly, strategies were excluded by extended dominance (a mix of other strategies has lower costs The average length of stay per complication was based A panel of Swiss physicians (two gastroenterologists on the database of the Swiss Hospital Association and three hepatologists) estimated the quality of life of (VESKA) and on separate evaluations based on a the health states using a time trade-off technique consecutive series of liver transplantations (1995±1997) Using these estimates, we assigned a utility weight (0 at the University Hospital, Berne (E.L. Renner, unpub- death, 1 perfect health) to each health state, and used lished data). Information on individual cost elements was obtained mainly from VESKA and separate calcula- tions of two large hospitals. The cost of a liver trans- plantation was based on an analysis of the Swiss We tested the sensitivity of our results to the in¯uence hospital institute (SKI) We updated these data to of Poynard's favourable response factors varied re¯ect the current shorter average length of stay discount rates, progression rate of cirrhosis and costs and the increased costs of hospital care and data are comparable to those reported in the USA All costs are calculated in euros (mean 1999 exchange rate euro/USD: 1.067).
G1 patients had a lower likelihood of sustained re- The marginal CE is expressed as the additional costs sponse than G non-1 patients, resulting in a lower life per additional quality-adjusted life year (QALY) gained, expectancy. The lower SR of G1 patients and longer IFN, interferon; IFN-R, interferon alpha-2b with ribavirin.
Hepatocellular carcinoma (no OP), following years Hepatocellular carcinoma (PH), following years Hepatocellular carcinoma (OLT), following years OP, operation; PH, partial hepatectomy; OLT, orthotopic liver transplantation.
486 European Journal of Gastroenterology & Hepatology 2001, Vol 13 No 5 Quality-of-life adjustments (utility weights) G1 virus, Comb II is the most favourable strategy: it should prolong life and be cost effective.
Comb I dominates all other strategies by reducing costs and extending quality-adjusted life expectancy. Again, compared with 24 weeks of combination treatment, the quality-of-life decrements of treating all patients for 48 weeks (Comb III) exceeded the additional long-term quality-of-life bene®ts. Compared with a strategy of no antiviral therapy (natural history) or treating only 24- week responders for 48 weeks (Comb II), Comb I prolonged life and reduced costs. Thus for G non-1 OLT, orthotopic liver transplantation; PH, partial hepatectomy; IFN, interferon; patients, Comb I is the optimal strategy, lengthening IFN-R, interferon alpha-2b with ribavirin.
duration of antiviral treatment resulted in higher life- To assess the sensitivity of our results to Poynard's time costs compared with G non-1 patients approach of favourable response factors we tested each strategy (separately for G1 and G non-1 patients) for the effect of the number of favourable response factors present (c.f. Regardless of the number The IFN strategy showed a favourable marginal CE of favourable response factors (except genotype) pre- (discounted) of 4886 euros (additional costs per QALY sent, our results remained stable, with Comb II being gained). Nevertheless, compared with IFN, Comb II the most cost-effective strategy for patients with G1 should prolong life expectancy by 1.8 years on average and Comb I being most cost effective for G non-1 at a still favourable CE ratio (discounted) of 7135 euros patients. Therefore, consideration of known response (additional costs per QALY gained). Treating all G1 factors, other than genotype, does not increase effec- patients for 48 weeks (Comb III) rather than stopping tiveness (QALY) or CE of the treatment strategies treatment in 24-week non-responders (Comb II), how- ever, led to a loss of quality-adjusted life expectancy, as the quality-of-life decrements associated with treatment We varied the discount rate from 0% to 5%, reduced of all patients for 48 weeks outweighed the long-term the histological progression rates to cirrhosis by 33%, quality-of-life bene®ts obtained in the additional re- and changed long-term disease costs by Æ 25%. In all sponders Thus, for patients infected with cases, our strategies remained below 24 960 euros per ÃStrategies are in order of the rising lifetime costs.
QALY, quality-adjusted life year; CE, marginal cost-effectiveness ratio; IFN, interferon; Comb I, combination therapy I; DOM, dominated by other strategies; Comb II, combination therapy II; Comb III, combination therapy III.
Combination therapy for chronic hepatitis C Sagmeister et al. 487 Sensitivity analysis of favourable response factors most cost-effective strategy. G1 patients should be (a) Genotype 1 with one or two favourable response factors treated for 48 weeks only if a virological response has occurred after 24 weeks of treatment. For G non-1 patients, IFN-R for 24 weeks dominated all other strategies, including no antiviral therapy.
Our results con®rm and extend the study by Younossi et al. We showed that for G1 patients the CE is improved substantially by continuing combination ther- (b) Genotype 1 with three or four favourable response factors apy for a further 24 weeks in those patients who are virological responders at week 24. Moreover, taking response factors other than genotype into account, as suggested by Poynard et al. did not improve the CE Thus, deciding on the length of combi- nation therapy in non-cirrhotic patients based solely on the presence of genotypes non-1 and 1 and, in the case of the latter, stopping treatment if HCV RNA in serum is still detectable after 24 weeks of treatment, seems (c) Genotype non-1 with two or three favourable response factors simple, effective and cost effective.
Effectiveness of therapy was expressed as QALY gained. When quality adjustments were made, quality- of-life decrements associated with 48 weeks of combi- nation therapy for all patients (Comb III) outweighed the long-term quality-of-life bene®ts in the additional responders. Thus, 24 weeks of combination therapy in (d) Genotype non-1 with four or ®ve favourable response factors G non-1 patients, and 48 weeks of combination therapy in 24-week responders in G1 patients, maximized As with any modelling approach, our analysis is limited by the scarce availability of data on the natural history of hepatitis C. Our conclusions, however, remain stable even when varying assumptions over a wide range.
Thus, decreasing yearly progression rate to cirrhosis by as much as 33%, increasing/decreasing long-term dis- IFN, interferon; DOM: dominated by other strategies; Comb I, combination therapy I; Comb II, combination therapy II; Comb III, combination therapy III.
ease-related costs by as much as 25%, or varying discount rates between 0% and 5% did not affect our conclusion that Comb I for G non-1 and Comb II for G1 are the most favourable treatment strategies.
QALY gained and thus within accepted CE limits (e.g.
Collectively, our data demonstrate that IFN-R therapy 31 577 euros per QALY gained for coronary stenting for non-cirrhotic patients with chronic hepatitis C is with single-vessel disease 2371±17 097 euros per most cost effective when genotype (1 versus non-1) life year gained for cholesterol-lowering therapy with a and, in G1, virological response at treatment week 24, are taken into account for determining duration of Our study is based on the original data of the two large randomized trials comparing the ef®cacy of IFN and IFN-R for treatment-naive patients. This allows for The authors thank the International Hepatitis Interven- a more detailed CE analysis than previous studies tional Therapy Group for providing their data for this We analysed ®ve different strategies separately for both G1 and G non-1. Cirrhotic patients were not included, as the antiviral treatment of cirrhotic patients is still under clinical evaluation. Our data suggest that a 1 Esteban JI, Lopez-Talavera JC, Genesca J, Madoz P, Viladomiu L, Muniz E, et al. High rate of infectivity and liver disease in blood donors with combination therapy according to genotype only is the antibodies to hepatitis C virus. Ann Intern Med 1991; 115:443±449.
488 European Journal of Gastroenterology & Hepatology 2001, Vol 13 No 5 2 Seeff LB, Buskell-Bales Z, Wright EC, Durako SJ, Alter HJ, Iber FL, et al.
29 Read JL, Quinn RJ, Berwick DM, Fineberg HV, Weinstein MC. Preferences Long term mortality after transfusion-associated non-A, non-B hepatitis. N for health outcomes: comparison of assessment methods. Med Decis 3 Di Bisceglie AM, Goodman ZD, Ishak KG, Hoofnagle JH, Melpolder JJ, Alter 30 Cohen DJ, Breall JA, Ho KK, Kuntz RE, Goldman L, Baim DS, et al.
HJ. Long term clinical and histopathological follow-up of chronic post- Evaluating the potential cost-effectiveness of stenting as a treatment for transfusion hepatitis. Hepatology 1991; 14:669±674.
symptomatic single-vessel coronary disease. Use of a decision-analytic 4 Fattovich G, Giustina G, Degos F, Treolada F, Diodati G, Almasio P, et al.
model. Circulation 1994; 89:1859±1874.
Morbidity and mortality in compensated cirrhosis type C: a retrospective 31 Johannesson M, JoÈnnsson B, Kjekhus J, Olsson AG, Pedersen TR, Wedel follow-up study of 384 patients. Gastroenterology 1997; 112:463±472.
H, et al. Cost effectiveness of simvastatin treatment to lower cholesterol 5 Tremolada F, Casarin C, Alberti A, Drago C, Tagger A, Ribero ML, et al.
levels in patients with coronary heart disease. N Engl J Med 1997; Long-term follow-up of non-A, non-B (type C) post transfusion hepatitis.
32 Kim WR, Poterucha JJ, Dickson ER, Gross JB. Optimal treatment strategy 6 Kiyosawa K, Sodeyama T, Tanaka E, Gibo Y, Yoshizawa K, Nakano Y, et al.
for chronic hepatitis C in 1999. Gastroenterology 1999; 116:L220.
Interrelationship of blood transfusion, non-A, non-B hepatitis and hepatocel- lular carcinoma: analysis by detection of antibody to hepatitis C virus.
7 Di Bisceglie AM, Order SE, Klein JL, Waggoner JG, Sjogren MH, Kuo G, et al. The role of chronic viral hepatitis in hepatocellular carcinoma in the United States. Am J Gastroenterol 1991; 86:335±338.
8 Poynard T, Leroy V, Cohard M, Thevenot T, Mathurin P, Opolon P. Meta- analysis of interferon randomized trials in the treatment of viral hepatitis C: effects of dose and duration. Hepatology 1996; 24:778±789.
9 Bennett WG, Inoue Y, Beck JR, Wong JB, Pauker SG, Davis GL. Estimates of the cost-effectiveness of a single course of interferon-á2b in patients with histology mild chronic hepatitis C. Ann Intern Med 1997; 127:855±865.
10 Dusheiko GM, Roberts JA. Treatment of chronic type B and C hepatitis with interferon alfa: an economic appraisal. Hepatology 1995; 22:1863±1873.
11 Kim WR, Poterucha JJ, Hermans JE, Therneau TM, Dickson ER, Evans RW.
Cost-effectiveness of 6 and 12 months of interferon-alfa therapy for chronic hepatitis C. Ann Intern Med 1997; 127:866±874.
12 Reichard O, Norkrans G, Fryden A, Braconier JH, Sonnerborg A, Weiland O. Randomised, double-blind, placebo-controlled trial of interferon á2b with and without ribavirin for chronic hepatitis C. Lancet 1998; 351:83±87.
13 Poynard T, Marcellin P, Lee SS, Niederau C, Minuk GS, Ideo G, et al.
Randomised trial of interferon á2b plus ribavirin for 48 weeks or for 24 weeks versus interferon á2b plus placebo for 48 weeks for treatment of chronic infection with hepatitis C virus. Lancet 1998; 352:1426±1432.
14 McHutchison JG, Gordoni SC, Schiff ER, Shiffmann ML, Lee WM, Rustgi VK, et al. Interferon alfa-2b alone or in combination with ribavirin as initial treatment for chronic hepatitis C. N Engl J Med 1998; 339:1485±1492.
15 Younossi ZM, Singer E, McHutchison JG, Shermock KM. Cost effectiveness of interferon á-2b combined with ribavirin for the treatment of chronic hepatitis. Hepatology 1999; 30:1318±1324.
16 Wong JB, Poynard T, Ling MH, Albrecht JK, Pauker SG. Cost-effectiveness of 24 or 48 weeks of interferon á-2b alone or with ribavirin as initial treatment of chronic hepatitis C. Am J Gastroenterol 2000; 95: 17 Poynard T, McHutchison J, Goodman Z, Ling MH, Albrecht J. Is an `a la carte' combination interferon alfa-2b plus ribavirin regimen possible for the ®rst line treatment in patients with hepatitis C? Hepatology 2000; 31: 18 Sonnenberg FA, Pauker SG. Decision Maker 7.0. Boston: New England 19 Wong JB, Bennett WB, Koff RS, Pauker SG. Pretreatment evaluation of hepatitis C: risks, bene®ts and costs. JAMA 1998; 280:2088±2093.
20 Sagmeister M, Renner EL, MuÈllhaupt B, Wong JB. Modeling of hepatitis C in Switzerland: prevalence, future epidemic and costs. Schweiz Med 21 European Association for the Study of the Liver. Consensus statement, EASL International Consensus Conference on Hepatitis C, 26±28 February, 1999, Paris. J Hepatol 1999; 30:956±961.
22 Schweizerische Expertengruppe fuer Virale Hepatitis (SEVHEP). Empfehlun- gen fuer die Therapie der chronischen Hepatitis C mit Interferon. Schwei- zerische Aerztezeitung 1996; 77:1744±1745.
23 Swiss Association for the Study of the Liver (SASL), Schweizerische Expertengruppe fuer Virale Hepatitis (SEVHEP), Schweizerische Gesellschaft fuer Gastroenterologie und Hepatologie (SGGH), Fachgesellschaft der Schweizerischen Gastroenterologen (FAGAS): Emp- fehlungen zur Therapie der Hepatitis C. Unpublished paper 1999.
24 National Institutes of Health Consensus. Management of hepatitis C.
25 Verbindung der Schweizer Aerzte, Versicherer gemaess UVG, Militaerversi- cherung, Invalidenversicherung: Aerztetarif. Luzern 1998.
26 VESKA. VESKA Diagnosenstatistik 1996. Aarau 1997.
27 Arbeitsgruppe Schweizerisches Institut fuer Gesundheits- und Krankenhaus- wesen (SKI). Lebertransplantation in der Schweiz. Aarau 1989.
28 Bundesamt fuÈr Statistik. Statistisches Jahrbuch der Schweiz 1997. Zurich:

Source: http://praxis-sagmeister.at/media/publikationen/euro2.pdf