- Email: [email protected]
Real-life Treatment Paradigms Show Infliximab Is Cost-effective for Management of Ulcerative Colitis
Clinical Gastroenterology and Hepatology 2014;-:-–-
Real-life Treatment Paradigms Show Infliximab Is Cost-effective for Management of Ulcerative Colitis Victoria Ung,* Nguyen Xuan Thanh,‡ Karen Wong,* Karen I. Kroeker,* Thomas Lee,* Haili Wang,* Arto Ohinmaa,‡ Philip Jacobs,‡ and Richard N. Fedorak* *Division of Gastroenterology, University of Alberta, Edmonton; and ‡Institute of Health Economics, Edmonton, Alberta, Canada BACKGROUND & AIMS:
Infliximab is effective for induction and maintenance of response in patients with moderate to moderately severe ulcerative colitis. Previous cost analyses of infliximab treatment for ulcerative colitis used models of colectomy vs infliximab and response rates derived from early clinical trials. In real life, therapeutic options are more complex; patients frequently choose to remain in an unwell state rather than undergo colectomy, and rates of response to infliximab are generally higher than those reported from clinical trials. We evaluate the real-life costeffectiveness of infliximab for treatment of ulcerative colitis where infliximab was readily available compared with not available, causing patients to remain in unwell states.
METHODS:
We constructed a Markov model to simulate disease progression of patients with moderate or moderately severe ulcerative colitis who depended on corticosteroids and/or did not respond to thiopurine therapy. Utility scores and transition probabilities between health states were determined by using data from randomized controlled trials and real-life rates published by expert inflammatory bowel disease centers. Health care costs were obtained from the Ontario Case Costing Initiative and the Alberta Health Schedule of Medical Benefits documents.
RESULTS:
The incremental cost-effectiveness ratios for infliximab treatment of ulcerative colitis were $79,000 and $64,000 per quality-adjusted life year, compared with ongoing medical therapy, at 5-year and 10-year treatment time horizons, respectively.
CONCLUSIONS:
By using real-life response rates and patients’ preference to avoid colectomy, infliximab therapy is a cost-effective strategy at a willingness-to-pay threshold of $80,000 for treatment of ulcerative colitis.
Keywords: Inflammatory Bowel Disease; Cost-Utility Analysis; Proctocolectomy; Corticosteroid; ICER.
lcerative colitis (UC) is a chronic inflammatory disease of the colon characterized by periods of quiescent disease and flares. In Canada, there are currently more than 155,000 patients, and 4000 new cases are added each year, representing a direct and indirect cost of nearly $1 billion yearly.1 The majority of patients with UC can expect to experience a relapsing-remitting course necessitating induction of remission and maintenance treatments. Approximately 15% of patients with UC will have an indolent disease course 5 years after diagnosis, and the remaining 85% will have at least 2 or more flares requiring corticosteroid therapy despite maintenance treatment with mesalamine.2 Furthermore, among patients who require a course of corticosteroids, half can be expected to become corticosteroid-dependent or require surgery.3 With the advent of immunosuppressives and biologics for induction and maintenance of UC response, improvements in health-related quality of life
U
and reductions in hospitalization and colectomy rates have been demonstrated.4–6 Furthermore, health care resource use and costs for treatment of inflammatory bowel disease (IBD) patients fall after infliximab therapy is started.7,8 The use of the anti–tumor-necrosis factor alpha (TNFa), infliximab, for treatment of outpatient UC generally follows a step-up approach advancing through mesalamine, corticosteroids, and thiopurines. For patients who are corticosteroid-dependent and fail or are intolerant to thiopurines, long-term treatment options are limited to continuous or intermittent corticosteroid Abbreviations used in this paper: IBD, inflammatory bowel disease; ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life year; TNF, tumor necrosis factor; UC, ulcerative colitis. © 2014 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2014.03.012
2
Ung et al
use, colectomy, or treatment with TNFa inhibitors. Longterm corticosteroid therapy is associated with significant complication rates and leaves the patient in a chronically unwell state.9 Colectomy with a permanent ileostomy or an ileoanal pouch procedure has morbidity and can have psychological effects for patients from this young population.10–12 Infliximab has demonstrated efficacy in the induction and maintenance of response in UC in both clinical trials and real-world settings.13,14 Nevertheless, the cost per patient treated with infliximab is significant.15 This has led payers to consider infliximab’s incremental costeffectiveness ratios (ICERs) per quality-adjusted life years (QALYs). In some jurisdictions, reports of high costs per QALY have limited formulary availability of infliximab for the management of UC. There are conflicting publications of the costeffectiveness of infliximab for the treatment of UC. Two studies report infliximab is cost-effective for UC treatment. The first examined acutely ill hospitalized UC patients (£19,545/QALY compared with cyclosporine),16 and the second examined those in scheduled infliximab maintenance during a period of 10 years (£86,322/QALY compared with standard therapy).17 However, for both these studies, at least one of the authors reported being an employee of the company that markets infliximab. In contrast, 3 studies did not identify infliximab to be cost-effective in moderate to moderately severe UC. The first, from Canada, reported an ICER of $CDN358,088/ QALY,18 the second, from the United Kingdom, reported an ICER of >£50,000/QALY,19 and the third, from the Common Drug Review, reported an ICER of $CDN125,000/QALY.20 All 3 of these studies simply compared infliximab therapy with immediate total colectomy. However, in a real-life scenario, the management choices are not this simple. Outpatients with moderate or moderately severe UC rarely opt for immediate colectomy as the first line of therapy, even when their disease is corticosteroid-dependent or refractory to thiopurines. Instead, because of the irreversible nature of colectomy, patients will opt for infliximab, but if infliximab is not available, they will generally select a chronically unwell state and a reduced quality of life, with or without corticosteroids, often for many years before agreeing to total colectomy.21–23 The aim of the present study was therefore to evaluate the cost utility of infliximab in the case of patients with moderate or moderately severe UC by using real-life patient preferences. These preferences include patient reluctance toward immediate colectomy but instead to remain in a chronically unwell state, with or without the use of corticosteroids, before considering surgery. Taking patients with corticosteroid-dependent and thiopurine refractory/intolerant disease and using the real-life patient preferences, we developed a cost-utility model for infliximab where the infliximab treatment was made readily available. We then compared this approach with a situation in which infliximab treatment was not available.
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Materials and Methods Target Population Our base-case analysis to determine infliximab cost used a theoretical cohort of patients with moderate to moderately severe UC that is corticosteroid-dependent and either refractory or intolerant to thiopurine treatment. The average age of this cohort is 40 years, with approximately 60% being male. For this base-case analysis, the estimated average weight of the cohort was considered to be 70 kg.
Type of Study and Outcome We used a cost-utility analysis, which uses the difference in costs divided by the difference in utilities between the study option and alternative intervention. The outcome is the ICER.
Markov Model Model structure, inputs, transition probabilities, costs of health states, utilities, and sensitivity analysis methods are detailed in Supplementary Methods, in Figure 1, and in Supplementary Tables 1 and 2.
Probabilities of Response Probabilities of response in moderate to moderately severe UC to infliximab in the first cycle was varied from 70.63% (randomized controlled studies used for regulatory filing)5,6,24,25 to 80.09% (open-label studies from IBD centers)13,14,17,24,26–29 among UC patients and from 71% to 84% among chronic pouchitis patients, with decreasing response rates in subsequent cycles as shown in Supplementary Table 3. There are no studies detailing 3-year and 5-year response rates of UC patients on infliximab. Thus in collaboration with the gastroenterologist/surgeon panel, it was agreed that in real-life practice, loss of response rates in UC outpatients on maintenance infliximab is similar to loss of response rates in Crohn’s disease patients on maintenance infliximab. Keshavarzian et al27 demonstrated infliximab response rates of 89% in the first year, 73% in the second, 58% in the fourth, and 54% in the fifth year. These data are comparable with those published by Teshima et al,28 where the response rate was 88% in the first year, 64% in the second, and 35% in the fifth year. Thus, we used the data from these studies and estimated an average rate of loss of response for patients on maintenance infliximab for each cycle, which was then applied to our data to determine the long-term rate of response of patients on infliximab (Supplementary Table 3). Dose escalation occurs almost universally after secondary loss of response to infliximab. From the ACT 1
-
2014
trial we estimated that dose escalation could have occurred in approximately 45% of subjects at 1 year.5 Rostholder et al30 and Ma et al31 in real-life open-label studies reported dose escalation rates at 1 year of 54% and 50%, respectively. Thus we used these data for the subanalysis determining the ICER associated with dose escalation. The assumptions were dose escalation to 10 mg/kg every 8 weeks in all patients with loss of response, a dose-escalated response rate of 60%, a utility score of the response to infliximab of 0.79, and an escalated loss of response rate into the unwell state as outlined in Supplementary Table 4. Those patients who did not respond or who lost response a second time entered the unwell state with a utility score of 0.32. To determine utility scores for the response-toinfliximab health state, we used 0.79 by time tradeoff and 0.82 by visual rating scale.32 A probabilistic sensitivity analysis for costs and utility scores was performed. A normal distribution was used for costs and a beta distribution for utility scores that are far from 0; utility scores close to 0 were transformed to utility decrement (¼1-U), and a gamma distribution was used.
Discounting Costs and utility scores were discounted annually at the rate of 5%.
Cost-Utility Analysis of Infliximab for UC
3
when comparing infliximab treatment with ongoing medical (steroid) therapy, was $US152,000/V117,000 per QALY gained when using a utility score of response to infliximab of 0.79 measured by time tradeoff (Table 1). For a utility score of response to infliximab of 0.82, measured by visual rating scale, the ICER was $US131,000/V100,000 per QALY gained (Table 1). Sensitivity analysis of probabilities, costs, and utility scores showed that the ICER at the 10-year horizon varied from $US104,000/V80,000 to $US289,000/ V218,000 (utility score 0.79) or from $US92,000/ V71,000 to $US217,000/V167,000 (utility score 0.82) (Table 1). The most sensitive variables were the cost of infliximab and the utility scores of unwell states, whereas the least sensitive variables were the probabilities of surgical complications and those of steroid complications. Sensitivity analyses were also performed with varying the time horizons. Assuming a utility score of the response to infliximab to be equal to 0.79, the ICERs were calculated to range from $US84,000/V64,000 to $US227,000/V174,000 and $US112,000/V86,000 to $US326,000/V250,000 per QALY gained at 5 and 15 years, respectively. For a utility score of 0.82, the ICERs decreased slightly and ranged from $US72,000/V55,000 to $US177,000/V136,000 and $US100,000/V77,000 to $US243,000/V186,000 per QALY gained at 5 and 15 years, respectively (Table 1).
Results Infliximab Not Available and Patients Opt for an Ongoing Unwell State (Steroid Therapy) With a utility score of the response to ongoing steroids for the management of UC equal to 0.79, the costutility analysis yielded a cost of $US86,000/V66,000 with 3.204 QALYs for a patient during a 10-year period.
Infliximab Available and Patients Opt for Infliximab Treatment to Induce and Maintain Response The cost-utility analysis with a utility score of response to infliximab equal to 0.79 yielded a cost of $US98,000/V75,600 with 3.284 QALYs for a patient during a 10-year period.
Incremental Cost-effective Ratio and Sensitivity Analysis With the Probability of Induction Response to Infliximab of 70.63% as Derived From Randomized Controlled Trials The ICER was calculated for infliximab induction response rate of 70.63% that was based on weighted average of published randomized controlled trials that were used for regulatory filings.5,6 The ICER at 10 years,
Incremental Cost-effective Ratio and Sensitivity Analysis With the Probability of Response to Infliximab of 80.09% as Derived From Real-life Open-label Studies Sensitivity analysis and ICER were also calculated for infliximab induction response rate of 80.09% that was based on the weighted average of real-life open-label studies from highly respectable IBD centers.13,14,26,33 The ICER for a 10-year time horizon, when comparing infliximab treatment with no infliximab treatment and ongoing medical (steroid), with a utility of 0.79 and 0.82 was found to be $US88,000/V68,000 and $US79,000/ V60,000 per QALY gained, respectively (Table 1). Sensitivity analysis of probabilities, costs, and utility scores showed that the ICER at the 10-year horizon ranged from $US62,000/V48,000 to $US114,000/ V87,000 (utility score 0.79) or from $US55,000/ V42,000 to $US102,000/V78,000 (utility score 0.82) (Table 1). The most sensitive variables were the cost of infliximab and the utility scores of unwell states, whereas the least sensitive variables were the probabilities of surgical complications and those of steroid complications. Sensitivity analyses were also performed with varying the time horizons. Assuming a utility score of the response to infliximab to be equal to 0.79, the ICERs were calculated to range from $US45,000/V35,000 to
4
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
-
2014
Cost-Utility Analysis of Infliximab for UC
5
Table 1. ICER Between Infliximab Treatment and No Infliximab Treatment but Instead Ongoing Medical Treatment
Utility score of response with infliximab measured by time tradeoff (u ¼ 0.79)c 5y 10 y 15 y Utility score of response with infliximab measured by visual rating scale (u ¼ 0.82)c 5y 10 y 15 y
ICER (range) Probability of initial response rate to infliximab derived from randomized controlled trials ¼ 70.63%a
ICER (range) Probability of initial response rate to infliximab derived from real-life studies ¼ 80.09%b
$US126,000 ($US84,000–$US 227,000) V97,000 (V64,000–V174,000) $US152,000 ($US104,000–$US 284,000) V117,000 (V80,000–V218,000) $US167,000 ($US112,000–$US326,000) V128,000 (V86,000–V250,000)
$US71,000 ($US45,000–$US 97,000 V54,000 (V35,000–V75,000) $US88,000 ($US62,000–$US114,000) V68,000 (V48,000–V87,000) $US96,000 ($US69,000–$US124,000) V74,000 (V53,000–V95,000)
$US108,000 ($US72,000–$US177,000) V83,000 (V55,000–V136,000) $US131,000 ($US92,000–$US217,000) V100,000 (V71,000–V167,000) $US142,000 ($US100,000–$US 243,000) V109,000 (V77,000–V186,000)
$US64,000 ($US41,000–$US86,000) V49,000 (V32,000–V66,000) $US79,000 ($US55,000–$US102,000) V60,000 (V42,000–V78,000) $US86,000 ($US62,000–$US111,000) V66,000 (V48,000–V85,000)
a
Sensitivity analysis by using infliximab induction response rate of 70.63% that is based on weighted average of published randomized controlled studies.5,6,24,25 Sensitivity analysis by using infliximab induction response rate of 80.09% that is based on weighted average of published open-label real-life studies.13,14,17,24,26–29 c u ¼ utility score for response to infliximab health state. b
$US97,000/V75,000 and $US69,000/V53,000 to $US124,000/V95,000 per QALY gained at 5 and 15 years, respectively. For a utility score of 0.82, the ICERs decreased slightly and ranged from $US41,000/V32,000 to $US86,000/V66,000 and $US62,000/V48,000 to $US110,000/V85,000 per QALY gained at 5 and 15 years, respectively (Table 1).
Incremental Cost-effective Ratio Associated With Infliximab Dose Escalation Exploratory analysis of dose escalation using response probabilities over time (Supplementary Table 5) revealed ICERs of $US114,000/V87,000 to $US306,000/V235,000 at 5 years, $US138,000/ V106,000 to $US371,000/V285,000 at 10 years, and $US152,000/V117,000 to $US424,000/V325,000 at 15 years for pre-escalation response rates of 70.63% to 80.09%, respectively.
Supplementary Figures 1 and 2. Assuming time horizon to be 10 years, the utility score of the response to infliximab health state to be equal to 0.79, and the probability of response to infliximab a real-life 80.09%, the graph shows 32% chance that infliximab treatment will be cost-effective if the willingness to pay for an extra QALY is $US50,000/V38,000. The percentage will be 47% and 54% if the willingness to pay is $US98,000/ V75,000 and $US147,000/V113,000, respectively (Supplementary Figure 1). With the same time horizon (10 years), if the utility score of the response to infliximab health state is 0.82 and the probability of response to infliximab a real-life 80.09%, the percentages of chances the infliximab treatment becomes cost-effective at the willingness-topay levels of $US50,000/V38,000, $US98,000/V75,000, and $US147,000/V113,000 are 34%, 50%, and 55%, respectively (Supplementary Figure 2).
Discussion Cost-effectiveness Acceptability Probabilistic sensitivity analysis results regarding cost-effective acceptability curves are shown in
Although studies suggest no impairment in quality of life after colectomy and ileal pouch-anal anastomosis as compared with pre-colectomy state, this does not seem
= Figure 1. Markov model simulating the progression of a real-life cohort of patients with moderate or moderately severe UC, corticosteroid-dependent or refractory to thiopurines. (A) Infliximab Not Available and Ongoing Unwell State—where infliximab therapy is not accessible for UC treatment and patients opt for an ongoing unwell state to delay colectomy. (B) Infliximab Available and Infliximab Therapy—where patients have access to infliximab for treatment of their UC and opt for infliximab to induce and maintain response. 5ASA, mesalamine; IFX, infliximab.
6
Ung et al
congruent with 17% overall rate of daytime incontinence and 40% infertility rate among women.34–36 Infliximab has been shown to be an effective treatment for steroid-dependent moderately severe UC by 2 large randomized controlled trials and also in the realworld setting, and economic analyses from the United Kingdom have demonstrated cost-effectiveness for both short-term and long-term use.5,6,13,14,16,17,24–26,28,29 In the current study, with pooled response data from the ACT I and ACT2 regulatory clinical trials, infliximab, within a 10-year time horizon, was not cost-effective, with ICER of $US131,000/V100,000 to $US152,000/ V117,000 depending on the utility. Nevertheless, by using the response data from real-life expert IBD center-reported studies, infliximab demonstrated costeffectiveness if the threshold of willingness to pay was set at $US80,000, with ICER of $US64,000/V49,000 after 5 years and $US79,000/V60,000 after 10 years. This is comparable with the cost-effective ICER of £27,424 for a 10-year time horizon seen in the United Kingdom study.17 Similar to the United Kingdom study, we also compared infliximab with standard therapy of ongoing intermittent corticosteroids that leave patients in a chronically unwell state with persistently active disease. Patients will often select this health state for extended periods of time before advancing to colectomy. Goals of therapy have to be balanced against the costs to society of achieving these goals. To allocate health spending efficiently, a willingness-to-pay threshold has been developed by the National Institute for Health and Clinical Excellence as the threshold at which the public is willing to pay for a given medical intervention. The currently accepted threshold, introduced in 1973, is $US50,000. However, there has been much debate as to whether this $US50,000 threshold is too low, resulting in good treatment interventions being denied funding.37 To date, many health care systems have not set a costeffectiveness threshold. Rocchi et al38,39 examined the recommendations made by Canadian Drug Expert Committee between 2003 and 2007 and showed that medicines up to $CDN80,000 were frequently accepted. Furthermore, a range of medications with ICERs between $CDN31,000 to $CDN137,000 were rejected on the basis of other criteria. One such criterion is the seriousness of the medical condition, with oncologic and medications for severe diseases having a higher threshold for acceptance.40 Patients with UC failing medical management and pending irreversible colectomy could be considered part of this group in which a higher ICER threshold is appropriate. In our study, we choose to use $CDN80,000 (approximately $US80,000/£60,000) as our ICER threshold, the threshold commonly used by Canadian Drug Expert Committee.38 Infliximab has been shown superior to alternate therapies in attaining the goals of mucosal healing, clinical remission, colectomy avoidance, and improved quality of life for steroid-dependent UC patients.24,25,41,42 Studies following the introduction of infliximab for the
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
treatment of Crohn’s disease demonstrated a decrease in health care resource utilization by these patients.7 It thus seems reasonable to suppose such a reduction might follow for infliximab treatment of UC.43 Indeed, our study found infliximab at both 5 and 10 years to be costeffective at a willingness-to-pay threshold of $US80,000 for patients with steroid-dependent moderate-to-severe UC when the infliximab induction response rate was set at 80.09% by using published results of real-life IBD expert centers. In contrast, infliximab was not costeffective at the $US80,000 threshold with a response rate (70.63%) derived from the randomized controlled clinical trials used for regulatory filing. This experience, given adequate compliance, of therapeutic outcomes in real-life clinical care being superior to those seen in the regulatory-directed randomized placebo-controlled clinical trials is well-described. There are several elements that may have led this study to underestimate the cost of infliximab therapy. Only infliximab complications requiring hospitalization were assessed; thus, costs related to complications treated as an outpatient are not included. In addition, patients who fail infliximab at the standard 5 mg/kg dose are often trialed on a dose-escalation regimen. We thus performed an exploratory analysis of ICERs associated with dose escalation. The ICER associated with dose escalation ranged from a low of $US114,000/V87,000 to a high of $US424,000/V325,000 and was highly dependent on the patient’s initial response to infliximab. In summary, this study demonstrates that using response rates from published expert IBD centers and real-life patient preference to avoid colectomy, infliximab is cost-effective at a willingness-to-pay threshold of $US80,000/V60,000 per QALY for treatment of UC compared with the chronic unwell state patients select in their avoidance of colectomy and ileostomy. Infliximab treatment of UC is less cost-effective at response rates reported in the randomized clinical trials used for regulatory filing. Dose escalation, if it occurs, increases these costs.
Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org, and at http://dx.doi.org/10.1016/j.cgh.2014.03.012.
References 1. Fedorak R, Wong K, Bridges R. Canadian Digestive Health foundation public impact series: inflammatory bowel disease in Canada—incidence, prevalence and direct and indirect economic impact. Can J Gastroenterol 2010;24:651–655. 2. Jess T, Riis L, Vind I, et al. Changes in clinical characteristics, course, and prognosis of inflammatory bowel disease during the last 5 decades: a population-based study from Copenhagen, Denmark. Inflamm Bowel Dis 2007;13:481–489.
-
2014
3. Faubion WA Jr, Loftus EV Jr, Harmsen WS, et al. The natural history of corticosteroid therapy for inflammatory bowel disease: a population-based study. Gastroenterology 2001;121: 255–260. 4. Feagan BG, Reinisch W, Rutgeerts P, et al. The effects of infliximab therapy on health-related quality of life in ulcerative colitis patients. Am J Gastroenterol 2007;102:794–802. 5. Rutgeerts P, Sandborn WJ, Feagan BG, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005;353:2462–2476. 6. Sandborn WJ, Rutgeerts P, Feagan BG, et al. Colectomy rate comparison after treatment of ulcerative colitis with placebo or infliximab. Gastroenterology 2009;137:1250–1260. 7. Loomes DE, Teshima C, Jacobs P, et al. Health care resource use and costs in Crohn’s disease before and after infliximab therapy. Can J Gastroenterol 2011;25:497–502. 8. Waters HC, Vanderpoel JE, Nejadnik B, et al. Resource utilization before and during infliximab therapy in patients with inflammatory bowel disease. J Med Econ 2012;15:45–52. 9. Manson SC, Brown RE, Cerulli A, et al. The cumulative burden of oral corticosteroid side effects and the economic implications of steroid use. Respir Med 2009;103:975–994. 10. Berndtsson IE, Carlsson EK, Persson EI, et al. Long-term adjustment to living with an ileal pouch-anal anastomosis. Dis Colon Rectum 2011;54:193–199.
Cost-Utility Analysis of Infliximab for UC
7
21. Heikens JT, de Vries J, van Laarhoven CJ. Quality of life, health related quality of life and health status in patients having restorative proctocolectomy with ileal pouch-anal anastomosis for ulcerative colitis: a systematic review. Colorectal Dis 2012; 14:536–544. 22. Lix LM, Graff LA, Walker JR, et al. Longitudinal study of quality of life and psychological functioning for active, fluctuating, and inactive disease patterns in inflammatory bowel disease. Inflamm Bowel Dis 2008;14:1575–1584. 23. Waljee A, Waljee J, Morris AM, et al. Threefold increased risk of infertility: a meta-analysis of infertility after ileal pouch anal anastomosis in ulcerative colitis. Gut 2006;55:1575–1580. 24. Ford AC, Sandborn WJ, Khan KJ, et al. Efficacy of biological therapies in inflammatory bowel disease: systematic review and meta-analysis. Am J Gastroenterol 2011;106:644–659. 25. Reinisch W, Sandborn WJ, Rutgeerts P, et al. Long-term infliximab maintenance therapy for ulcerative colitis: the ACT-1 and -2 extension studies. Inflamm Bowel Dis 2012;18:201–211. 26. Su C, Salzberg BA, Lewis JD, et al. Efficacy of anti-tumor necrosis factor therapy in patients with ulcerative colitis. Am J Gastroenterol 2002;97:2577–2584. 27. Keshavarzian A, Mayer L, Salzberg B, et al. A multicenter retrospective experience of infliximab in Crohn’s disease patients: infusion reaction rates and treatment persistency. Gastroenterol Hepatol 2007;3:381–390.
12. Carlsson E, Bosaeus I, Nordgren S. What concerns subjects with inflammatory bowel disease and an ileostomy? Scand J Gastroenterol 2003;38:978–984.
28. Teshima CW, Thompson A, Dhanoa L, et al. Long-term response rates to infliximab therapy for Crohn’s disease in an outpatient cohort. Can J Gastroenterol 2009;23:348–352. 29. Gornet JM, Couve S, Hassani Z, et al. Infliximab for refractory ulcerative colitis or indeterminate colitis: an open-label multicentre study. Aliment Pharmacol Ther 2003;18:175–181.
13. Gies N, Kroeker KI, Wong K, et al. Treatment of ulcerative colitis with adalimumab or infliximab: long-term follow-up of a single-centre cohort. Aliment Pharmacol Ther 2010;32: 522–528.
30. Rostholder E, Ahmed A, Cheifetz AS, et al. Outcomes after escalation of infliximab therapy in ambulatory patients with moderately active ulcerative colitis. Aliment Pharmacol Ther 2012;35:562–567.
14. Oussalah A, Evesque L, Laharie D, et al. A multicenter experience with infliximab for ulcerative colitis: outcomes and predictors of response, optimization, colectomy, and hospitalization. Am J Gastroenterol 2010;105:2617–2625.
31. Ma C, Huang V, Fedorak D, et al. Outpatients with ulcerative colitis being treated with adalimuamb and infliximab have similar rates of loss of response. Can J Gastroenterol Hepatol 2014;28:A141.
11. Sinclair LG. Young adults with permanent ileostomies: experiences during the first 4 years after surgery. J Wound Ostomy Continence Nurs 2009;36:306–316.
15. Schabert VF, Watson C, Gandra SR, et al. Annual costs of tumor necrosis factor inhibitors using real-world data in a commercially insured population in the United States. J Med Econ 2012; 15:264–275. 16. Punekar YS, Hawkins N. Cost-effectiveness of infliximab for the treatment of acute exacerbations of ulcerative colitis. Eur J Health Econ 2010;11:67–76. 17. Tsai HH, Punekar YS, Morris J, et al. A model of the long-term cost effectiveness of scheduled maintenance treatment with infliximab for moderate-to-severe ulcerative colitis. Aliment Pharmacol Ther 2008;28:1230–1239.
32. Arseneau KO, Sultan S, Provenzale DT, et al. Do patient preferences influence decisions on treatment for patients with steroid-refractory ulcerative colitis? Clin Gastroenterol Hepatol 2006;4:1135–1142. 33. Tursi A, Giorgetti GM, Brandimarte G, et al. Safety and effectiveness of long-term budesonide treatment in maintaining remission in patients with mild-to-moderate Crohn’s disease. Inflamm Bowel Dis 2007;13:1184–1186. 34. Hueting WE, Buskens E, van der Tweel I, et al. Results and complications after ileal pouch anal anastomosis: a metaanalysis of 43 observational studies comprising 9,317 patients. Dig Surg 2005;22:69–79.
18. Xie F, Blackhouse G, Assasi N, et al. Cost-utility analysis of infliximab and adalimumab for refractory ulcerative colitis. Cost Eff Resour Alloc 2009;7:20.
35. Rajaratnam SG, Eglinton TW, Hider P, et al. Impact of ileal pouch-anal anastomosis on female fertility: meta-analysis and systematic review. Int J Colorectal Dis 2011;26:1365–1374.
19. Dretzke J, Edlin R, Round J, et al. A systematic review and economic evaluation of the use of tumour necrosis factor-alpha (TNF-alpha) inhibitors, adalimumab and infliximab, for Crohn’s disease. Health Technol Assess 2011;15:1–244. 20. CEDAC final recommendation and reasons for recommendation: infliximab (Remicade - Centocor Inc) indication—ulcerative colitis. Ottawa, ON: Canada: Canadian Agency for Drugs and Technologies in Health (CADTH); 2009.
36. Jess T, Loftus EV Jr, Velayos FS, et al. Risk of intestinal cancer in inflammatory bowel disease: a population-based study from Olmsted county, Minnesota. Gastroenterology 2006;130:1039–1046. 37. Towse A. Should NICE’s threshold range for cost per QALY be raised? Yes. BMJ 2009;338:b181. 38. Rocchi A, Menon D, Verma S, et al. The role of economic evidence in Canadian oncology reimbursement decision-making: to lambda and beyond. Value Health 2008;11:771–783.
8
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
39. Rocchi A, Miller E, Hopkins RB, et al. Common Drug Review recommendations: an evidence base for expectations? Pharmacoeconomics 2012;30:229–246.
43. Rubenstein JH, Chong RY, Cohen RD. Infliximab decreases resource use among patients with Crohn’s disease. J Clin Gastroenterol 2002;35:151–156.
40. Ternouth AM, Chapman M, Modha R. An assessment of the variation in accepted ICERs by disease type: results from four HTAS (poster). Health Technology Assessment: A European Collaboration. ISPOR 13th Annual European Congress, Prague, Czech Republic. November 6–9, 2010.
Reprint requests Address requests for reprints to: Richard N. Fedorak, MD, 2-14A Zeidler Building, Division of Gastroenterology, University of Alberta, Edmonton, Alberta T6G 2X8, Canada. e-mail: [email protected]; fax: (780) 492-8121.
41. Carter CT, Leher H, Smith P, et al. Impact of persistence with infliximab on hospitalizations in ulcerative colitis. Am J Manag Care 2011;17:385–392. 42. Colombel JF, Rutgeerts P, Reinisch W, et al. Early mucosal healing with infliximab is associated with improved long-term clinical outcomes in ulcerative colitis. Gastroenterology 2011; 141:1194–1201.
Conflicts of interest This author discloses the following: Richard N. Fedorak has served as a consultant and speaker for Janssen Canada Inc. The remaining authors disclose no conflicts. Funding Supported by Centre of Excellence for Gastrointestinal Inflammation and Immunity Research (CEGIIR) and the Alberta Innovates - Health Solutions supported Alberta IBD Consortium.
-
2014
Supplementary Methods Markov Model A Markov model was constructed to compare the ICER of 2 management strategies (Figure 1). (1) Infliximab is not available, and patients opt for the ongoing unwell state to avoid colectomy; we termed this scenario Steroid Therapy, because patients were generally unwell on conventional medical treatments or were corticosteroid-dependent. (2) Infliximab is available, and patients opt for Infliximab Therapy to induce and maintain response; patients were modeled as being treated with 5 mg/kg infliximab induction therapy at 0, 2, and 6 weeks and then 5 mg/kg maintenance therapy every 8 weeks. The health states used in the model were defined and verified by 4 gastroenterologists and 1 surgeon with experience in the treatment of IBD (Supplementary Table 5). Patients were assigned to initial health states of 3 months and were evaluated every 3 months during a 10-year (40-cycle) time horizon; at the end of every 3-month cycle, patients were assigned probabilities of moving on to subsequent health states. Probabilities were as follows. Patients treated with infliximab either responded to the induction therapy or became non-responders. Patients whose disease responded to induction therapy could experience a response-to-treatment state over time, or they could subsequently lose the response. Along with those who did not respond to infliximab induction, patients who lost response were treated with ongoing steroid therapy; a proportion of these patients eventually underwent colectomy. Patients who experienced a complication associated with infliximab treatment could be treated for the complication; however, if the complication could not be treated, they were either taken off infliximab and returned to ongoing steroid therapy or offered colectomy. Patients who underwent surgery could develop surgical complications or remain in a response state. Patients who had a pouch constructed as part of the surgery could develop chronic pouchitis, which may or may not respond to infliximab or require further medical attention in the form of steroids or surgical intervention. The possibility of patient mortality was considered for each health node.
Model Inputs Our analysis follows the 2006 economic evaluation guidelines as set out by the Canadian Agency for Drugs and Health Technologies.1
Cost-Utility Analysis of Infliximab for UC
8.e1
Transition Probabilities The probabilities for patients moving between health states were derived by a review of the literature with prioritization of randomized controlled trials. In the absence of randomized controlled trials, observational studies were included. A meta-analysis to pool the probabilities was conducted if the state was reported by more than 1 study. All the values were subsequently assessed by the panel of gastroenterologists (R.N.F., K.I.K., K.W., T.L.) and 1 colorectal surgeon (H.W.) for face validity. All the probabilities are shown in Supplementary Tables 1 and 2. In this study, rates were used as proxies for probabilities.
Costs of Health States Services and costs per cycle were assessed for each health state on the basis of a review of literature. For those cycles where service-use data were missing, the gastroenterologist/surgeon panel estimated resource use. We took a health-system perspective, where we included hospital, physician, and outpatient drug costs. Physicians’ fees were obtained from the Alberta provincial fee schedules of Alberta Health and Wellness (2006).2 Hospital costs for all hospitalization episodes came from the Ontario Case Costing Initiative (2009).3 The costs of corticosteroid, infliximab, surgical complication, or chronic pouchitis health states were estimated by averaging the cost of each condition complication weighted by the likelihood of occurrence. Cost of death was counted once and equal to the cost of the health state that led to the death. The costs of drugs were obtained from the Alberta provincial formulary.4 Supplementary Table 1 outlines the cost of each health state per 3-month cycle. Costs are reported in $US and Euros at an exchange rate calculated on April 26, 2013 where $1US ¼ V0.767.
Utility To calculate the QALYs for each treatment regimen, utilities for each health state were determined through a review of literature, with adjustments based on expert opinion (Supplementary Tables 1 and 2).5
Sensitivity Analysis A multi-way sensitivity analysis with tornado diagrams (available on request) was conducted on all the key input parameters, including time horizon, probabilities, utility scores, and costs. The time horizon was varied from 5 to 15 years, the probabilities and utility scores were varied between the lower and upper ends of 95% confidence interval, and the costs were varied by 25%, as shown in Supplementary Tables 1 and 2.
8.e2
Ung et al
Supplementary References 1. Canadian Agency for Drugs and Technologies in Health. Guidelines for the economic evaluation of health technologies: Canada. 2006. Available at: http://www.cadth.ca/media/pdf/ 186_EconomicGuidelines_e.pdf. Accessed February 1, 2014. 2. Alberta Health Care insurance plan. 2011. Available at: http:// www.health.alberta.ca/documents/SOMB-Medical-Prices2012-04.pdf. Accessed March 23, 2012. 3. OCCI (Ontario Case Costing Initiative). Costing analysis tool (CAT) [OCCI website]. 2011. Available at: http://www.occp. com/. Accessed February 1, 2014. 4. Alberta Health and Wellness. Alberta Health and Wellness drug benefit list. 2011. Available at: https://www.ab.bluecross.ca/dbl/ pdfs/dbl_full_list.pdf. Accessed March 23, 2012. 5. Arseneau KO, Sultan S, Provenzale DT, et al. Do patient preferences influence decisions on treatment for patients with steroid-refractory ulcerative colitis? Clin Gastroenterol Hepatol 2006;4:1135–1142. 6. Sandborn WJ, Rutgeerts P, Feagan BG, et al. Colectomy rate comparison after treatment of ulcerative colitis with placebo or infliximab. Gastroenterology 2009;137:1250–1260.
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
12. Su C, Salzberg BA, Lewis JD, et al. Efficacy of anti-tumor necrosis factor therapy in patients with ulcerative colitis. Am J Gastroenterol 2002;97:2577–2584. 13. Gornet JM, Couve S, Hassani Z, et al. Infliximab for refractory ulcerative colitis or indeterminate colitis: an open-label multicentre study. Aliment Pharmacol Ther 2003;18:175–181. 14. Rutgeerts P, Sandborn WJ, Feagan BG, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005;353:2462–2476. 15. Saag KG, Koehnke R, Caldwell JR, et al. Low dose long-term corticosteroid therapy in rheumatoid arthritis: an analysis of serious adverse events. Am J Med 1994;96:115–123. 16. Arai K, Koganei K, Kimura H, et al. Incidence and outcome of complications following restorative proctocolectomy. Am J Surg 2005;190:39–42. 17. Holubar SD, Long KH, Loftus EV Jr, et al. Long-term direct costs before and after proctocolectomy for ulcerative colitis: a population-based study in Olmsted County, Minnesota. Dis Colon Rectum 2009;52:1815–1823. 18. McMullen K, Hicks TC, Ray JE, et al. Complications associated with ileal pouch-anal anastomosis. World J Surg 1991;15:763–766.
7. Gies N, Kroeker KI, Wong K, et al. Treatment of ulcerative colitis with adalimumab or infliximab: long-term follow-up of a singlecentre cohort. Aliment Pharmacol Ther 2010;32:522–528.
19. Barreiro-de Acosta M, Garcia-Bosch O, Souto R, et al. Efficacy of infliximab rescue therapy in patients with chronic refractory pouchitis: a multicenter study. Inflamm Bowel Dis 2012;18:812–817.
8. Ford AC, Sandborn WJ, Khan KJ, et al. Efficacy of biological therapies in inflammatory bowel disease: systematic review and meta-analysis. Am J Gastroenterol 2011;106:644–659.
20. Keshavarzian A, Mayer L, Salzberg B, et al. A multicenter retrospective experience of infliximab in Crohn’s disease patients: infusion reaction rates and treatment persistency. Gastroenterol Hepatol 2007;3:381–390.
9. Reinisch W, Sandborn WJ, Rutgeerts P, et al. Long-term infliximab maintenance therapy for ulcerative colitis: the ACT-1 and -2 extension studies. Inflamm Bowel Dis 2012;18:201–211. 10. Oussalah A, Evesque L, Laharie D, et al. A multicenter experience with infliximab for ulcerative colitis: outcomes and predictors of response, optimization, colectomy, and hospitalization. Am J Gastroenterol 2010;105:2617–2625. 11. Tsai HH, Punekar YS, Morris J, et al. A model of the long-term cost effectiveness of scheduled maintenance treatment with infliximab for moderate-to-severe ulcerative colitis. Aliment Pharmacol Ther 2008;28:1230–1239.
21. Teshima CW, Thompson A, Dhanoa L, et al. Long-term response rates to infliximab therapy for Crohn’s disease in an outpatient cohort. Can J Gastroenterol 2009;23:348–352. 22. Rostholder E, Ahmed A, Cheifetz AS, et al. Outcomes after escalation of infliximab therapy in ambulatory patients with moderately active ulcerative colitis. Aliment Pharmacol Ther 2012;35:562–567. 23. Ma C, Huang V, Fedorak D, et al. Outpatients with ulcerative colitis being treated with adalimuamband infliximab have similar rates of loss of response. Can J Gastroenterol Hepatol 2014;28:A141.
-
2014
Cost-Utility Analysis of Infliximab for UC
8.e3
Supplementary Figure 1. Cost-effectiveness acceptability curves by group (utility ¼ 0.79). *u ¼ utility score for the response to infliximab health state. **Sensitivity analysis by using induction rate of 70.63% for patients on infliximab that is based on weighted average of published randomized controlled studies.6–9 ***Sensitivity analysis by using induction rate of 80.09% for patients on infliximab that is based on actual weighted average of published open-label studies.8,10–13
8.e4
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Supplementary Figure 2. Cost-effectiveness acceptability curves by group (utility ¼ 0.82). *u ¼ utility score for the response to infliximab health state. **Sensitivity analysis by using induction rate of 70.63% for patients on infliximab that is based on weighted average of published randomized controlled studies.6–9,14 ***Sensitivity analysis by using induction rate of 80.09% for patients on infliximab that is based on actual weighted average of published open-label studies.8,10–13
-
2014
Cost-Utility Analysis of Infliximab for UC
8.e5
Supplementary Table 1. Markov Model Input Parameters for Chronic UC Transition probabilities (%)/cycle A. Ongoing steroids14,15 1. Response 2. Unwell 3. Complication 4. Surgery B. Response to steroids14 1. Response 2. Loss of response C. Unwell on steroids14,15 1. Unwell 2. Complication 3. Surgery D. Steroid complication15 1. Surgery 2. Death E. Surgery16,17 1. Early response 2. Complication 3. Death F. Response to surgery16,18 1. Response to surgery 2. Surgical complication 3. CP G. Chronic pouchitis19 1. Response to IFX (CP) 2. Nonresponse (unwell-CP) 3. IFX complication (CP) H. Surgical complication16 1. Hospitalization 2. Death I. Infliximab6,8,9,14 1. Response to IFX 2. IFX complication 3. Nonresponse (unwell) J. Response to IFX6–14,20–23 1. Response to IFX 2. IFX complication 3. Loss of response (unwell) K. IFX complications6,12,14 1. Response to IFX 2. Unwell on steroids 3. Surgery 4. Death L. Death
CP, chronic pouchitis; IFX, infliximab. a Complement probability.
Costs ($CAN)/cycle 864 (25%)
Utility scores/y 0.32 (0.31)
33.92 (28.09–40.33) 57.11 (50.5–63.27) 2.80 (0.56–7.63) a
0
0.79 (0.21)
53.30 (46.81–59.67) a
864 (25%)
0.32 (0.31)
22,527 (25%)
0.16 (0.16)
34,996 (25%)
0.16 (0.16)
0
0.58 (0.15)
a
2.80 (0.56–7.63) 10.00 (6.40–14.28) 98.00 (93.70–99.78) a
a
12.8 (8.76–17.91) 2.50 (0.98–5.69) 85.80 a
11.70 5446 (25%)
0.32 (0.31)
16,562 (25%)
0.49 (0.32)
5446 (25%)
0.32 (0.31)
See Supplementary Table 2 a
7.75 99.50 (97.22–99.99) a
70.63–80.09 1.49 a
5446 (25%)
0.79–0.82
See Supplementary Table 3 5.90 a
14,973 (25%)
0.16 (0.16)
Equal to cost of corresponding health state
0
70.00 14.00 14.00 a
1
8.e6
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Supplementary Table 2. Markov Model Input Parameters for Chronic Pouchitis Health states
Transition probabilities (%)/cycle
M. Response to IFX1 1. Response to IFX1 See Supplementary Table 3 a 2. Lost response (unwell1) 3. IFX complication1 7.75 N. Unwell1 a 1. Unwell1 2. Surgery1 10.00 3. Steroid complication1 2.80 O1. IFX complication1 in the IFX not available arm 1. Response to IFX1 64.60 2. Unwell1 16.70 3. Surgery1 16.70 a 4. Death O2. IFX complication1 in the IFX available arm 1. Response to IFX1 55 2. Unwell1 21.50 3. Surgery1 21.50 a 4. Death P. Surgery1 (permanent ileostomy) a 1. Response1 2. Surgery complication1 12.80 3. Death 2.50 Q. Steroid complication1 1. Surgery1 98.00 a 2. Death R. Response to surgery1 a 1. Response to surgery1 2. Surgical complication1 2.50 S. Surgical complication1 a 1. Response to surgery 2. Death 0.50
IFX, infliximab. a Complement probability.
Costs ($CAN)/cycle
Utility scores/y
5446 (25%)
0.58 (0.15)
864 (25%)
0.32 (0.31)
14,973 (25%)
0.16 (0.16)
14,973 (25%)
0.16 (0.16)
34,996 (25%)
0.16 (0.16)
22,527 (25%)
0.16 (0.16)
0
0.44 (0.11)
16,562 (25%)
0.37 (0.24)
-
2014
Cost-Utility Analysis of Infliximab for UC
8.e7
Supplementary Table 3. Maintenance Probabilities of Patients on Infliximab Over Time UC patientsb Cycle no. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
UC patientsa
6–8,10–13,20–23
6,8,9,14
(sensitivity analysis)
Chronic pouchitis patients (infliximab not available arm)
Chronic pouchitis patients (infliximab available arm)c
70.63 64.45 58.27 52.08 45.9 42.90 39.89 36.89 33.89 33.01 32.13 31.25 30.37 29.49 28.62 27.74 26.86 26.40 25.94 25.48 25.02
80.09 77.55 75.01 72.46 69.92 65.35 60.77 56.20 51.62 50.28 48.94 47.61 46.27 44.93 43.59 42.25 40.91 40.21 39.51 38.82 38.11
84 75 66 55.5 45 41.25 35.5 29.75 24 23.13 22.25 21.38 20.50 19.63 18.75 17.88 17 16 15 14 13
71 63.75 56 47.18 38.25 35.06 30.18 25.29 20.4 19.66 18.91 18.17 17.43 16.69 15.94 15.2 14.45 13.6 12.75 11.9 11.05
a These maintenance probabilities are based on a weighted average of clinical trials, including randomized control trials.6–9,14 The rate at which patients lose response to infliximab was calculated on the basis of data extrapolated from patients treated on infliximab.20–23 b Probabilities were derived from a weighted average of open-label studies reporting infliximab efficacy.8,10–13 The decrease in response over time of patients on infliximab was calculated on the basis of data derived from patients treated with infliximab.20–23 c To calculate the response probability for patients with chronic pouchitis who had been previously exposed to and failed by infliximab, a 15% discount was taken from the probability of response of patients with chronic pouchitis who had never been exposed to infliximab.
Supplementary Table 4. Response Probabilities of Patients Dose Escalated on Infliximab Over Time Cycle no. 0 1 2 3 4 (1 y) 5 6 7 8 (2 y) 9 10 11 12 (3 y) 13
a
UC patients rate of responsea 60 54.74 49.49 44.24 38.99 36.44 33.88 31.33 28.78 28.04 27.29 26.54 25.80 25.0
Decrease in response over time of patients on infliximab was calculated on the basis of data derived from patients treated with infliximab.22,23
8.e8
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Supplementary Table 5. Health State Definitions Health state
Definition
Response to medical treatment (steroid/IFX)
On their respective treatment regimens, patients have decreased symptoms: reduced bowel frequency as compared with active disease, may intermittently have blood with defecation, and are able to taper off steroid treatment. A disease pattern in which the patient is not acutely ill with unmistakable indication for resection or not in remission and symptom-free, despite being treated with medical therapy (oral steroids, mesalamine, azathioprine). Patients experience symptoms of chronic intractable disease (recurrent acute colitis, steroid dependence, chronic fecal urgency, persistent active disease, or complications of medical therapy). Patients in this cohort would have a UC Disease Activity Index score of 3–8 (out of 12) or a partial Mayo score of 2–6 (out of 9), corresponding to a mild-moderate disease activity state. Patients may develop mild anemia and have 5–8 bowel movements a day with some rectal bleeding. A common long-term complication after restorative proctocolectomy with ileal pouchanal anastomosis for patients with UC. This disease state is characterized by a nonspecific inflammation of the ileal pouch after surgery. Signs and symptoms may include increased stool frequency, urgency, incontinence, and dehydration. Majority of patients who develop chronic pouchitis are often treated with courses of antibiotics (metronidazole or ciprofloxacin); however, there is a subset of patients whose disease is refractory to antibiotic treatment. In this study, the chronic pouchitis health state refers to chronic pouchitis patients who are refractory to antibiotic therapy to attain remission. Any complication that occurred as result of a medical treatment (ongoing steroid or IFX) that necessitated a change in treatment or health state (steroid / surgery/ death, IFX / steroid/surgery/death). Patient whose disease never responded or at first responded but then lost response to IFX in subsequent cycles. Patients whose UC did not respond to medical treatment undergo 1 of 2 possible surgeries to manage their disease. Patients will undergo a 2-step or 3-step procedure as indicated by their physicians. Any complication that arises as a result of the surgery that requires the patient to be hospitalized or undergo further surgery to correct the complication.
Unwell
Chronic pouchitis
Steroid/IFX complication
Non-/loss response (IFX) Surgery
Surgical complication
IFX, infliximab.
Real-life Treatment Paradigms Show Infliximab Is Cost-effective for Management of Ulcerative Colitis Victoria Ung,* Nguyen Xuan Thanh,‡ Karen Wong,* Karen I. Kroeker,* Thomas Lee,* Haili Wang,* Arto Ohinmaa,‡ Philip Jacobs,‡ and Richard N. Fedorak* *Division of Gastroenterology, University of Alberta, Edmonton; and ‡Institute of Health Economics, Edmonton, Alberta, Canada BACKGROUND & AIMS:
Infliximab is effective for induction and maintenance of response in patients with moderate to moderately severe ulcerative colitis. Previous cost analyses of infliximab treatment for ulcerative colitis used models of colectomy vs infliximab and response rates derived from early clinical trials. In real life, therapeutic options are more complex; patients frequently choose to remain in an unwell state rather than undergo colectomy, and rates of response to infliximab are generally higher than those reported from clinical trials. We evaluate the real-life costeffectiveness of infliximab for treatment of ulcerative colitis where infliximab was readily available compared with not available, causing patients to remain in unwell states.
METHODS:
We constructed a Markov model to simulate disease progression of patients with moderate or moderately severe ulcerative colitis who depended on corticosteroids and/or did not respond to thiopurine therapy. Utility scores and transition probabilities between health states were determined by using data from randomized controlled trials and real-life rates published by expert inflammatory bowel disease centers. Health care costs were obtained from the Ontario Case Costing Initiative and the Alberta Health Schedule of Medical Benefits documents.
RESULTS:
The incremental cost-effectiveness ratios for infliximab treatment of ulcerative colitis were $79,000 and $64,000 per quality-adjusted life year, compared with ongoing medical therapy, at 5-year and 10-year treatment time horizons, respectively.
CONCLUSIONS:
By using real-life response rates and patients’ preference to avoid colectomy, infliximab therapy is a cost-effective strategy at a willingness-to-pay threshold of $80,000 for treatment of ulcerative colitis.
Keywords: Inflammatory Bowel Disease; Cost-Utility Analysis; Proctocolectomy; Corticosteroid; ICER.
lcerative colitis (UC) is a chronic inflammatory disease of the colon characterized by periods of quiescent disease and flares. In Canada, there are currently more than 155,000 patients, and 4000 new cases are added each year, representing a direct and indirect cost of nearly $1 billion yearly.1 The majority of patients with UC can expect to experience a relapsing-remitting course necessitating induction of remission and maintenance treatments. Approximately 15% of patients with UC will have an indolent disease course 5 years after diagnosis, and the remaining 85% will have at least 2 or more flares requiring corticosteroid therapy despite maintenance treatment with mesalamine.2 Furthermore, among patients who require a course of corticosteroids, half can be expected to become corticosteroid-dependent or require surgery.3 With the advent of immunosuppressives and biologics for induction and maintenance of UC response, improvements in health-related quality of life
U
and reductions in hospitalization and colectomy rates have been demonstrated.4–6 Furthermore, health care resource use and costs for treatment of inflammatory bowel disease (IBD) patients fall after infliximab therapy is started.7,8 The use of the anti–tumor-necrosis factor alpha (TNFa), infliximab, for treatment of outpatient UC generally follows a step-up approach advancing through mesalamine, corticosteroids, and thiopurines. For patients who are corticosteroid-dependent and fail or are intolerant to thiopurines, long-term treatment options are limited to continuous or intermittent corticosteroid Abbreviations used in this paper: IBD, inflammatory bowel disease; ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life year; TNF, tumor necrosis factor; UC, ulcerative colitis. © 2014 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2014.03.012
2
Ung et al
use, colectomy, or treatment with TNFa inhibitors. Longterm corticosteroid therapy is associated with significant complication rates and leaves the patient in a chronically unwell state.9 Colectomy with a permanent ileostomy or an ileoanal pouch procedure has morbidity and can have psychological effects for patients from this young population.10–12 Infliximab has demonstrated efficacy in the induction and maintenance of response in UC in both clinical trials and real-world settings.13,14 Nevertheless, the cost per patient treated with infliximab is significant.15 This has led payers to consider infliximab’s incremental costeffectiveness ratios (ICERs) per quality-adjusted life years (QALYs). In some jurisdictions, reports of high costs per QALY have limited formulary availability of infliximab for the management of UC. There are conflicting publications of the costeffectiveness of infliximab for the treatment of UC. Two studies report infliximab is cost-effective for UC treatment. The first examined acutely ill hospitalized UC patients (£19,545/QALY compared with cyclosporine),16 and the second examined those in scheduled infliximab maintenance during a period of 10 years (£86,322/QALY compared with standard therapy).17 However, for both these studies, at least one of the authors reported being an employee of the company that markets infliximab. In contrast, 3 studies did not identify infliximab to be cost-effective in moderate to moderately severe UC. The first, from Canada, reported an ICER of $CDN358,088/ QALY,18 the second, from the United Kingdom, reported an ICER of >£50,000/QALY,19 and the third, from the Common Drug Review, reported an ICER of $CDN125,000/QALY.20 All 3 of these studies simply compared infliximab therapy with immediate total colectomy. However, in a real-life scenario, the management choices are not this simple. Outpatients with moderate or moderately severe UC rarely opt for immediate colectomy as the first line of therapy, even when their disease is corticosteroid-dependent or refractory to thiopurines. Instead, because of the irreversible nature of colectomy, patients will opt for infliximab, but if infliximab is not available, they will generally select a chronically unwell state and a reduced quality of life, with or without corticosteroids, often for many years before agreeing to total colectomy.21–23 The aim of the present study was therefore to evaluate the cost utility of infliximab in the case of patients with moderate or moderately severe UC by using real-life patient preferences. These preferences include patient reluctance toward immediate colectomy but instead to remain in a chronically unwell state, with or without the use of corticosteroids, before considering surgery. Taking patients with corticosteroid-dependent and thiopurine refractory/intolerant disease and using the real-life patient preferences, we developed a cost-utility model for infliximab where the infliximab treatment was made readily available. We then compared this approach with a situation in which infliximab treatment was not available.
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Materials and Methods Target Population Our base-case analysis to determine infliximab cost used a theoretical cohort of patients with moderate to moderately severe UC that is corticosteroid-dependent and either refractory or intolerant to thiopurine treatment. The average age of this cohort is 40 years, with approximately 60% being male. For this base-case analysis, the estimated average weight of the cohort was considered to be 70 kg.
Type of Study and Outcome We used a cost-utility analysis, which uses the difference in costs divided by the difference in utilities between the study option and alternative intervention. The outcome is the ICER.
Markov Model Model structure, inputs, transition probabilities, costs of health states, utilities, and sensitivity analysis methods are detailed in Supplementary Methods, in Figure 1, and in Supplementary Tables 1 and 2.
Probabilities of Response Probabilities of response in moderate to moderately severe UC to infliximab in the first cycle was varied from 70.63% (randomized controlled studies used for regulatory filing)5,6,24,25 to 80.09% (open-label studies from IBD centers)13,14,17,24,26–29 among UC patients and from 71% to 84% among chronic pouchitis patients, with decreasing response rates in subsequent cycles as shown in Supplementary Table 3. There are no studies detailing 3-year and 5-year response rates of UC patients on infliximab. Thus in collaboration with the gastroenterologist/surgeon panel, it was agreed that in real-life practice, loss of response rates in UC outpatients on maintenance infliximab is similar to loss of response rates in Crohn’s disease patients on maintenance infliximab. Keshavarzian et al27 demonstrated infliximab response rates of 89% in the first year, 73% in the second, 58% in the fourth, and 54% in the fifth year. These data are comparable with those published by Teshima et al,28 where the response rate was 88% in the first year, 64% in the second, and 35% in the fifth year. Thus, we used the data from these studies and estimated an average rate of loss of response for patients on maintenance infliximab for each cycle, which was then applied to our data to determine the long-term rate of response of patients on infliximab (Supplementary Table 3). Dose escalation occurs almost universally after secondary loss of response to infliximab. From the ACT 1
-
2014
trial we estimated that dose escalation could have occurred in approximately 45% of subjects at 1 year.5 Rostholder et al30 and Ma et al31 in real-life open-label studies reported dose escalation rates at 1 year of 54% and 50%, respectively. Thus we used these data for the subanalysis determining the ICER associated with dose escalation. The assumptions were dose escalation to 10 mg/kg every 8 weeks in all patients with loss of response, a dose-escalated response rate of 60%, a utility score of the response to infliximab of 0.79, and an escalated loss of response rate into the unwell state as outlined in Supplementary Table 4. Those patients who did not respond or who lost response a second time entered the unwell state with a utility score of 0.32. To determine utility scores for the response-toinfliximab health state, we used 0.79 by time tradeoff and 0.82 by visual rating scale.32 A probabilistic sensitivity analysis for costs and utility scores was performed. A normal distribution was used for costs and a beta distribution for utility scores that are far from 0; utility scores close to 0 were transformed to utility decrement (¼1-U), and a gamma distribution was used.
Discounting Costs and utility scores were discounted annually at the rate of 5%.
Cost-Utility Analysis of Infliximab for UC
3
when comparing infliximab treatment with ongoing medical (steroid) therapy, was $US152,000/V117,000 per QALY gained when using a utility score of response to infliximab of 0.79 measured by time tradeoff (Table 1). For a utility score of response to infliximab of 0.82, measured by visual rating scale, the ICER was $US131,000/V100,000 per QALY gained (Table 1). Sensitivity analysis of probabilities, costs, and utility scores showed that the ICER at the 10-year horizon varied from $US104,000/V80,000 to $US289,000/ V218,000 (utility score 0.79) or from $US92,000/ V71,000 to $US217,000/V167,000 (utility score 0.82) (Table 1). The most sensitive variables were the cost of infliximab and the utility scores of unwell states, whereas the least sensitive variables were the probabilities of surgical complications and those of steroid complications. Sensitivity analyses were also performed with varying the time horizons. Assuming a utility score of the response to infliximab to be equal to 0.79, the ICERs were calculated to range from $US84,000/V64,000 to $US227,000/V174,000 and $US112,000/V86,000 to $US326,000/V250,000 per QALY gained at 5 and 15 years, respectively. For a utility score of 0.82, the ICERs decreased slightly and ranged from $US72,000/V55,000 to $US177,000/V136,000 and $US100,000/V77,000 to $US243,000/V186,000 per QALY gained at 5 and 15 years, respectively (Table 1).
Results Infliximab Not Available and Patients Opt for an Ongoing Unwell State (Steroid Therapy) With a utility score of the response to ongoing steroids for the management of UC equal to 0.79, the costutility analysis yielded a cost of $US86,000/V66,000 with 3.204 QALYs for a patient during a 10-year period.
Infliximab Available and Patients Opt for Infliximab Treatment to Induce and Maintain Response The cost-utility analysis with a utility score of response to infliximab equal to 0.79 yielded a cost of $US98,000/V75,600 with 3.284 QALYs for a patient during a 10-year period.
Incremental Cost-effective Ratio and Sensitivity Analysis With the Probability of Induction Response to Infliximab of 70.63% as Derived From Randomized Controlled Trials The ICER was calculated for infliximab induction response rate of 70.63% that was based on weighted average of published randomized controlled trials that were used for regulatory filings.5,6 The ICER at 10 years,
Incremental Cost-effective Ratio and Sensitivity Analysis With the Probability of Response to Infliximab of 80.09% as Derived From Real-life Open-label Studies Sensitivity analysis and ICER were also calculated for infliximab induction response rate of 80.09% that was based on the weighted average of real-life open-label studies from highly respectable IBD centers.13,14,26,33 The ICER for a 10-year time horizon, when comparing infliximab treatment with no infliximab treatment and ongoing medical (steroid), with a utility of 0.79 and 0.82 was found to be $US88,000/V68,000 and $US79,000/ V60,000 per QALY gained, respectively (Table 1). Sensitivity analysis of probabilities, costs, and utility scores showed that the ICER at the 10-year horizon ranged from $US62,000/V48,000 to $US114,000/ V87,000 (utility score 0.79) or from $US55,000/ V42,000 to $US102,000/V78,000 (utility score 0.82) (Table 1). The most sensitive variables were the cost of infliximab and the utility scores of unwell states, whereas the least sensitive variables were the probabilities of surgical complications and those of steroid complications. Sensitivity analyses were also performed with varying the time horizons. Assuming a utility score of the response to infliximab to be equal to 0.79, the ICERs were calculated to range from $US45,000/V35,000 to
4
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
-
2014
Cost-Utility Analysis of Infliximab for UC
5
Table 1. ICER Between Infliximab Treatment and No Infliximab Treatment but Instead Ongoing Medical Treatment
Utility score of response with infliximab measured by time tradeoff (u ¼ 0.79)c 5y 10 y 15 y Utility score of response with infliximab measured by visual rating scale (u ¼ 0.82)c 5y 10 y 15 y
ICER (range) Probability of initial response rate to infliximab derived from randomized controlled trials ¼ 70.63%a
ICER (range) Probability of initial response rate to infliximab derived from real-life studies ¼ 80.09%b
$US126,000 ($US84,000–$US 227,000) V97,000 (V64,000–V174,000) $US152,000 ($US104,000–$US 284,000) V117,000 (V80,000–V218,000) $US167,000 ($US112,000–$US326,000) V128,000 (V86,000–V250,000)
$US71,000 ($US45,000–$US 97,000 V54,000 (V35,000–V75,000) $US88,000 ($US62,000–$US114,000) V68,000 (V48,000–V87,000) $US96,000 ($US69,000–$US124,000) V74,000 (V53,000–V95,000)
$US108,000 ($US72,000–$US177,000) V83,000 (V55,000–V136,000) $US131,000 ($US92,000–$US217,000) V100,000 (V71,000–V167,000) $US142,000 ($US100,000–$US 243,000) V109,000 (V77,000–V186,000)
$US64,000 ($US41,000–$US86,000) V49,000 (V32,000–V66,000) $US79,000 ($US55,000–$US102,000) V60,000 (V42,000–V78,000) $US86,000 ($US62,000–$US111,000) V66,000 (V48,000–V85,000)
a
Sensitivity analysis by using infliximab induction response rate of 70.63% that is based on weighted average of published randomized controlled studies.5,6,24,25 Sensitivity analysis by using infliximab induction response rate of 80.09% that is based on weighted average of published open-label real-life studies.13,14,17,24,26–29 c u ¼ utility score for response to infliximab health state. b
$US97,000/V75,000 and $US69,000/V53,000 to $US124,000/V95,000 per QALY gained at 5 and 15 years, respectively. For a utility score of 0.82, the ICERs decreased slightly and ranged from $US41,000/V32,000 to $US86,000/V66,000 and $US62,000/V48,000 to $US110,000/V85,000 per QALY gained at 5 and 15 years, respectively (Table 1).
Incremental Cost-effective Ratio Associated With Infliximab Dose Escalation Exploratory analysis of dose escalation using response probabilities over time (Supplementary Table 5) revealed ICERs of $US114,000/V87,000 to $US306,000/V235,000 at 5 years, $US138,000/ V106,000 to $US371,000/V285,000 at 10 years, and $US152,000/V117,000 to $US424,000/V325,000 at 15 years for pre-escalation response rates of 70.63% to 80.09%, respectively.
Supplementary Figures 1 and 2. Assuming time horizon to be 10 years, the utility score of the response to infliximab health state to be equal to 0.79, and the probability of response to infliximab a real-life 80.09%, the graph shows 32% chance that infliximab treatment will be cost-effective if the willingness to pay for an extra QALY is $US50,000/V38,000. The percentage will be 47% and 54% if the willingness to pay is $US98,000/ V75,000 and $US147,000/V113,000, respectively (Supplementary Figure 1). With the same time horizon (10 years), if the utility score of the response to infliximab health state is 0.82 and the probability of response to infliximab a real-life 80.09%, the percentages of chances the infliximab treatment becomes cost-effective at the willingness-topay levels of $US50,000/V38,000, $US98,000/V75,000, and $US147,000/V113,000 are 34%, 50%, and 55%, respectively (Supplementary Figure 2).
Discussion Cost-effectiveness Acceptability Probabilistic sensitivity analysis results regarding cost-effective acceptability curves are shown in
Although studies suggest no impairment in quality of life after colectomy and ileal pouch-anal anastomosis as compared with pre-colectomy state, this does not seem
= Figure 1. Markov model simulating the progression of a real-life cohort of patients with moderate or moderately severe UC, corticosteroid-dependent or refractory to thiopurines. (A) Infliximab Not Available and Ongoing Unwell State—where infliximab therapy is not accessible for UC treatment and patients opt for an ongoing unwell state to delay colectomy. (B) Infliximab Available and Infliximab Therapy—where patients have access to infliximab for treatment of their UC and opt for infliximab to induce and maintain response. 5ASA, mesalamine; IFX, infliximab.
6
Ung et al
congruent with 17% overall rate of daytime incontinence and 40% infertility rate among women.34–36 Infliximab has been shown to be an effective treatment for steroid-dependent moderately severe UC by 2 large randomized controlled trials and also in the realworld setting, and economic analyses from the United Kingdom have demonstrated cost-effectiveness for both short-term and long-term use.5,6,13,14,16,17,24–26,28,29 In the current study, with pooled response data from the ACT I and ACT2 regulatory clinical trials, infliximab, within a 10-year time horizon, was not cost-effective, with ICER of $US131,000/V100,000 to $US152,000/ V117,000 depending on the utility. Nevertheless, by using the response data from real-life expert IBD center-reported studies, infliximab demonstrated costeffectiveness if the threshold of willingness to pay was set at $US80,000, with ICER of $US64,000/V49,000 after 5 years and $US79,000/V60,000 after 10 years. This is comparable with the cost-effective ICER of £27,424 for a 10-year time horizon seen in the United Kingdom study.17 Similar to the United Kingdom study, we also compared infliximab with standard therapy of ongoing intermittent corticosteroids that leave patients in a chronically unwell state with persistently active disease. Patients will often select this health state for extended periods of time before advancing to colectomy. Goals of therapy have to be balanced against the costs to society of achieving these goals. To allocate health spending efficiently, a willingness-to-pay threshold has been developed by the National Institute for Health and Clinical Excellence as the threshold at which the public is willing to pay for a given medical intervention. The currently accepted threshold, introduced in 1973, is $US50,000. However, there has been much debate as to whether this $US50,000 threshold is too low, resulting in good treatment interventions being denied funding.37 To date, many health care systems have not set a costeffectiveness threshold. Rocchi et al38,39 examined the recommendations made by Canadian Drug Expert Committee between 2003 and 2007 and showed that medicines up to $CDN80,000 were frequently accepted. Furthermore, a range of medications with ICERs between $CDN31,000 to $CDN137,000 were rejected on the basis of other criteria. One such criterion is the seriousness of the medical condition, with oncologic and medications for severe diseases having a higher threshold for acceptance.40 Patients with UC failing medical management and pending irreversible colectomy could be considered part of this group in which a higher ICER threshold is appropriate. In our study, we choose to use $CDN80,000 (approximately $US80,000/£60,000) as our ICER threshold, the threshold commonly used by Canadian Drug Expert Committee.38 Infliximab has been shown superior to alternate therapies in attaining the goals of mucosal healing, clinical remission, colectomy avoidance, and improved quality of life for steroid-dependent UC patients.24,25,41,42 Studies following the introduction of infliximab for the
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
treatment of Crohn’s disease demonstrated a decrease in health care resource utilization by these patients.7 It thus seems reasonable to suppose such a reduction might follow for infliximab treatment of UC.43 Indeed, our study found infliximab at both 5 and 10 years to be costeffective at a willingness-to-pay threshold of $US80,000 for patients with steroid-dependent moderate-to-severe UC when the infliximab induction response rate was set at 80.09% by using published results of real-life IBD expert centers. In contrast, infliximab was not costeffective at the $US80,000 threshold with a response rate (70.63%) derived from the randomized controlled clinical trials used for regulatory filing. This experience, given adequate compliance, of therapeutic outcomes in real-life clinical care being superior to those seen in the regulatory-directed randomized placebo-controlled clinical trials is well-described. There are several elements that may have led this study to underestimate the cost of infliximab therapy. Only infliximab complications requiring hospitalization were assessed; thus, costs related to complications treated as an outpatient are not included. In addition, patients who fail infliximab at the standard 5 mg/kg dose are often trialed on a dose-escalation regimen. We thus performed an exploratory analysis of ICERs associated with dose escalation. The ICER associated with dose escalation ranged from a low of $US114,000/V87,000 to a high of $US424,000/V325,000 and was highly dependent on the patient’s initial response to infliximab. In summary, this study demonstrates that using response rates from published expert IBD centers and real-life patient preference to avoid colectomy, infliximab is cost-effective at a willingness-to-pay threshold of $US80,000/V60,000 per QALY for treatment of UC compared with the chronic unwell state patients select in their avoidance of colectomy and ileostomy. Infliximab treatment of UC is less cost-effective at response rates reported in the randomized clinical trials used for regulatory filing. Dose escalation, if it occurs, increases these costs.
Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org, and at http://dx.doi.org/10.1016/j.cgh.2014.03.012.
References 1. Fedorak R, Wong K, Bridges R. Canadian Digestive Health foundation public impact series: inflammatory bowel disease in Canada—incidence, prevalence and direct and indirect economic impact. Can J Gastroenterol 2010;24:651–655. 2. Jess T, Riis L, Vind I, et al. Changes in clinical characteristics, course, and prognosis of inflammatory bowel disease during the last 5 decades: a population-based study from Copenhagen, Denmark. Inflamm Bowel Dis 2007;13:481–489.
-
2014
3. Faubion WA Jr, Loftus EV Jr, Harmsen WS, et al. The natural history of corticosteroid therapy for inflammatory bowel disease: a population-based study. Gastroenterology 2001;121: 255–260. 4. Feagan BG, Reinisch W, Rutgeerts P, et al. The effects of infliximab therapy on health-related quality of life in ulcerative colitis patients. Am J Gastroenterol 2007;102:794–802. 5. Rutgeerts P, Sandborn WJ, Feagan BG, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005;353:2462–2476. 6. Sandborn WJ, Rutgeerts P, Feagan BG, et al. Colectomy rate comparison after treatment of ulcerative colitis with placebo or infliximab. Gastroenterology 2009;137:1250–1260. 7. Loomes DE, Teshima C, Jacobs P, et al. Health care resource use and costs in Crohn’s disease before and after infliximab therapy. Can J Gastroenterol 2011;25:497–502. 8. Waters HC, Vanderpoel JE, Nejadnik B, et al. Resource utilization before and during infliximab therapy in patients with inflammatory bowel disease. J Med Econ 2012;15:45–52. 9. Manson SC, Brown RE, Cerulli A, et al. The cumulative burden of oral corticosteroid side effects and the economic implications of steroid use. Respir Med 2009;103:975–994. 10. Berndtsson IE, Carlsson EK, Persson EI, et al. Long-term adjustment to living with an ileal pouch-anal anastomosis. Dis Colon Rectum 2011;54:193–199.
Cost-Utility Analysis of Infliximab for UC
7
21. Heikens JT, de Vries J, van Laarhoven CJ. Quality of life, health related quality of life and health status in patients having restorative proctocolectomy with ileal pouch-anal anastomosis for ulcerative colitis: a systematic review. Colorectal Dis 2012; 14:536–544. 22. Lix LM, Graff LA, Walker JR, et al. Longitudinal study of quality of life and psychological functioning for active, fluctuating, and inactive disease patterns in inflammatory bowel disease. Inflamm Bowel Dis 2008;14:1575–1584. 23. Waljee A, Waljee J, Morris AM, et al. Threefold increased risk of infertility: a meta-analysis of infertility after ileal pouch anal anastomosis in ulcerative colitis. Gut 2006;55:1575–1580. 24. Ford AC, Sandborn WJ, Khan KJ, et al. Efficacy of biological therapies in inflammatory bowel disease: systematic review and meta-analysis. Am J Gastroenterol 2011;106:644–659. 25. Reinisch W, Sandborn WJ, Rutgeerts P, et al. Long-term infliximab maintenance therapy for ulcerative colitis: the ACT-1 and -2 extension studies. Inflamm Bowel Dis 2012;18:201–211. 26. Su C, Salzberg BA, Lewis JD, et al. Efficacy of anti-tumor necrosis factor therapy in patients with ulcerative colitis. Am J Gastroenterol 2002;97:2577–2584. 27. Keshavarzian A, Mayer L, Salzberg B, et al. A multicenter retrospective experience of infliximab in Crohn’s disease patients: infusion reaction rates and treatment persistency. Gastroenterol Hepatol 2007;3:381–390.
12. Carlsson E, Bosaeus I, Nordgren S. What concerns subjects with inflammatory bowel disease and an ileostomy? Scand J Gastroenterol 2003;38:978–984.
28. Teshima CW, Thompson A, Dhanoa L, et al. Long-term response rates to infliximab therapy for Crohn’s disease in an outpatient cohort. Can J Gastroenterol 2009;23:348–352. 29. Gornet JM, Couve S, Hassani Z, et al. Infliximab for refractory ulcerative colitis or indeterminate colitis: an open-label multicentre study. Aliment Pharmacol Ther 2003;18:175–181.
13. Gies N, Kroeker KI, Wong K, et al. Treatment of ulcerative colitis with adalimumab or infliximab: long-term follow-up of a single-centre cohort. Aliment Pharmacol Ther 2010;32: 522–528.
30. Rostholder E, Ahmed A, Cheifetz AS, et al. Outcomes after escalation of infliximab therapy in ambulatory patients with moderately active ulcerative colitis. Aliment Pharmacol Ther 2012;35:562–567.
14. Oussalah A, Evesque L, Laharie D, et al. A multicenter experience with infliximab for ulcerative colitis: outcomes and predictors of response, optimization, colectomy, and hospitalization. Am J Gastroenterol 2010;105:2617–2625.
31. Ma C, Huang V, Fedorak D, et al. Outpatients with ulcerative colitis being treated with adalimuamb and infliximab have similar rates of loss of response. Can J Gastroenterol Hepatol 2014;28:A141.
11. Sinclair LG. Young adults with permanent ileostomies: experiences during the first 4 years after surgery. J Wound Ostomy Continence Nurs 2009;36:306–316.
15. Schabert VF, Watson C, Gandra SR, et al. Annual costs of tumor necrosis factor inhibitors using real-world data in a commercially insured population in the United States. J Med Econ 2012; 15:264–275. 16. Punekar YS, Hawkins N. Cost-effectiveness of infliximab for the treatment of acute exacerbations of ulcerative colitis. Eur J Health Econ 2010;11:67–76. 17. Tsai HH, Punekar YS, Morris J, et al. A model of the long-term cost effectiveness of scheduled maintenance treatment with infliximab for moderate-to-severe ulcerative colitis. Aliment Pharmacol Ther 2008;28:1230–1239.
32. Arseneau KO, Sultan S, Provenzale DT, et al. Do patient preferences influence decisions on treatment for patients with steroid-refractory ulcerative colitis? Clin Gastroenterol Hepatol 2006;4:1135–1142. 33. Tursi A, Giorgetti GM, Brandimarte G, et al. Safety and effectiveness of long-term budesonide treatment in maintaining remission in patients with mild-to-moderate Crohn’s disease. Inflamm Bowel Dis 2007;13:1184–1186. 34. Hueting WE, Buskens E, van der Tweel I, et al. Results and complications after ileal pouch anal anastomosis: a metaanalysis of 43 observational studies comprising 9,317 patients. Dig Surg 2005;22:69–79.
18. Xie F, Blackhouse G, Assasi N, et al. Cost-utility analysis of infliximab and adalimumab for refractory ulcerative colitis. Cost Eff Resour Alloc 2009;7:20.
35. Rajaratnam SG, Eglinton TW, Hider P, et al. Impact of ileal pouch-anal anastomosis on female fertility: meta-analysis and systematic review. Int J Colorectal Dis 2011;26:1365–1374.
19. Dretzke J, Edlin R, Round J, et al. A systematic review and economic evaluation of the use of tumour necrosis factor-alpha (TNF-alpha) inhibitors, adalimumab and infliximab, for Crohn’s disease. Health Technol Assess 2011;15:1–244. 20. CEDAC final recommendation and reasons for recommendation: infliximab (Remicade - Centocor Inc) indication—ulcerative colitis. Ottawa, ON: Canada: Canadian Agency for Drugs and Technologies in Health (CADTH); 2009.
36. Jess T, Loftus EV Jr, Velayos FS, et al. Risk of intestinal cancer in inflammatory bowel disease: a population-based study from Olmsted county, Minnesota. Gastroenterology 2006;130:1039–1046. 37. Towse A. Should NICE’s threshold range for cost per QALY be raised? Yes. BMJ 2009;338:b181. 38. Rocchi A, Menon D, Verma S, et al. The role of economic evidence in Canadian oncology reimbursement decision-making: to lambda and beyond. Value Health 2008;11:771–783.
8
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
39. Rocchi A, Miller E, Hopkins RB, et al. Common Drug Review recommendations: an evidence base for expectations? Pharmacoeconomics 2012;30:229–246.
43. Rubenstein JH, Chong RY, Cohen RD. Infliximab decreases resource use among patients with Crohn’s disease. J Clin Gastroenterol 2002;35:151–156.
40. Ternouth AM, Chapman M, Modha R. An assessment of the variation in accepted ICERs by disease type: results from four HTAS (poster). Health Technology Assessment: A European Collaboration. ISPOR 13th Annual European Congress, Prague, Czech Republic. November 6–9, 2010.
Reprint requests Address requests for reprints to: Richard N. Fedorak, MD, 2-14A Zeidler Building, Division of Gastroenterology, University of Alberta, Edmonton, Alberta T6G 2X8, Canada. e-mail: [email protected]; fax: (780) 492-8121.
41. Carter CT, Leher H, Smith P, et al. Impact of persistence with infliximab on hospitalizations in ulcerative colitis. Am J Manag Care 2011;17:385–392. 42. Colombel JF, Rutgeerts P, Reinisch W, et al. Early mucosal healing with infliximab is associated with improved long-term clinical outcomes in ulcerative colitis. Gastroenterology 2011; 141:1194–1201.
Conflicts of interest This author discloses the following: Richard N. Fedorak has served as a consultant and speaker for Janssen Canada Inc. The remaining authors disclose no conflicts. Funding Supported by Centre of Excellence for Gastrointestinal Inflammation and Immunity Research (CEGIIR) and the Alberta Innovates - Health Solutions supported Alberta IBD Consortium.
-
2014
Supplementary Methods Markov Model A Markov model was constructed to compare the ICER of 2 management strategies (Figure 1). (1) Infliximab is not available, and patients opt for the ongoing unwell state to avoid colectomy; we termed this scenario Steroid Therapy, because patients were generally unwell on conventional medical treatments or were corticosteroid-dependent. (2) Infliximab is available, and patients opt for Infliximab Therapy to induce and maintain response; patients were modeled as being treated with 5 mg/kg infliximab induction therapy at 0, 2, and 6 weeks and then 5 mg/kg maintenance therapy every 8 weeks. The health states used in the model were defined and verified by 4 gastroenterologists and 1 surgeon with experience in the treatment of IBD (Supplementary Table 5). Patients were assigned to initial health states of 3 months and were evaluated every 3 months during a 10-year (40-cycle) time horizon; at the end of every 3-month cycle, patients were assigned probabilities of moving on to subsequent health states. Probabilities were as follows. Patients treated with infliximab either responded to the induction therapy or became non-responders. Patients whose disease responded to induction therapy could experience a response-to-treatment state over time, or they could subsequently lose the response. Along with those who did not respond to infliximab induction, patients who lost response were treated with ongoing steroid therapy; a proportion of these patients eventually underwent colectomy. Patients who experienced a complication associated with infliximab treatment could be treated for the complication; however, if the complication could not be treated, they were either taken off infliximab and returned to ongoing steroid therapy or offered colectomy. Patients who underwent surgery could develop surgical complications or remain in a response state. Patients who had a pouch constructed as part of the surgery could develop chronic pouchitis, which may or may not respond to infliximab or require further medical attention in the form of steroids or surgical intervention. The possibility of patient mortality was considered for each health node.
Model Inputs Our analysis follows the 2006 economic evaluation guidelines as set out by the Canadian Agency for Drugs and Health Technologies.1
Cost-Utility Analysis of Infliximab for UC
8.e1
Transition Probabilities The probabilities for patients moving between health states were derived by a review of the literature with prioritization of randomized controlled trials. In the absence of randomized controlled trials, observational studies were included. A meta-analysis to pool the probabilities was conducted if the state was reported by more than 1 study. All the values were subsequently assessed by the panel of gastroenterologists (R.N.F., K.I.K., K.W., T.L.) and 1 colorectal surgeon (H.W.) for face validity. All the probabilities are shown in Supplementary Tables 1 and 2. In this study, rates were used as proxies for probabilities.
Costs of Health States Services and costs per cycle were assessed for each health state on the basis of a review of literature. For those cycles where service-use data were missing, the gastroenterologist/surgeon panel estimated resource use. We took a health-system perspective, where we included hospital, physician, and outpatient drug costs. Physicians’ fees were obtained from the Alberta provincial fee schedules of Alberta Health and Wellness (2006).2 Hospital costs for all hospitalization episodes came from the Ontario Case Costing Initiative (2009).3 The costs of corticosteroid, infliximab, surgical complication, or chronic pouchitis health states were estimated by averaging the cost of each condition complication weighted by the likelihood of occurrence. Cost of death was counted once and equal to the cost of the health state that led to the death. The costs of drugs were obtained from the Alberta provincial formulary.4 Supplementary Table 1 outlines the cost of each health state per 3-month cycle. Costs are reported in $US and Euros at an exchange rate calculated on April 26, 2013 where $1US ¼ V0.767.
Utility To calculate the QALYs for each treatment regimen, utilities for each health state were determined through a review of literature, with adjustments based on expert opinion (Supplementary Tables 1 and 2).5
Sensitivity Analysis A multi-way sensitivity analysis with tornado diagrams (available on request) was conducted on all the key input parameters, including time horizon, probabilities, utility scores, and costs. The time horizon was varied from 5 to 15 years, the probabilities and utility scores were varied between the lower and upper ends of 95% confidence interval, and the costs were varied by 25%, as shown in Supplementary Tables 1 and 2.
8.e2
Ung et al
Supplementary References 1. Canadian Agency for Drugs and Technologies in Health. Guidelines for the economic evaluation of health technologies: Canada. 2006. Available at: http://www.cadth.ca/media/pdf/ 186_EconomicGuidelines_e.pdf. Accessed February 1, 2014. 2. Alberta Health Care insurance plan. 2011. Available at: http:// www.health.alberta.ca/documents/SOMB-Medical-Prices2012-04.pdf. Accessed March 23, 2012. 3. OCCI (Ontario Case Costing Initiative). Costing analysis tool (CAT) [OCCI website]. 2011. Available at: http://www.occp. com/. Accessed February 1, 2014. 4. Alberta Health and Wellness. Alberta Health and Wellness drug benefit list. 2011. Available at: https://www.ab.bluecross.ca/dbl/ pdfs/dbl_full_list.pdf. Accessed March 23, 2012. 5. Arseneau KO, Sultan S, Provenzale DT, et al. Do patient preferences influence decisions on treatment for patients with steroid-refractory ulcerative colitis? Clin Gastroenterol Hepatol 2006;4:1135–1142. 6. Sandborn WJ, Rutgeerts P, Feagan BG, et al. Colectomy rate comparison after treatment of ulcerative colitis with placebo or infliximab. Gastroenterology 2009;137:1250–1260.
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
12. Su C, Salzberg BA, Lewis JD, et al. Efficacy of anti-tumor necrosis factor therapy in patients with ulcerative colitis. Am J Gastroenterol 2002;97:2577–2584. 13. Gornet JM, Couve S, Hassani Z, et al. Infliximab for refractory ulcerative colitis or indeterminate colitis: an open-label multicentre study. Aliment Pharmacol Ther 2003;18:175–181. 14. Rutgeerts P, Sandborn WJ, Feagan BG, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005;353:2462–2476. 15. Saag KG, Koehnke R, Caldwell JR, et al. Low dose long-term corticosteroid therapy in rheumatoid arthritis: an analysis of serious adverse events. Am J Med 1994;96:115–123. 16. Arai K, Koganei K, Kimura H, et al. Incidence and outcome of complications following restorative proctocolectomy. Am J Surg 2005;190:39–42. 17. Holubar SD, Long KH, Loftus EV Jr, et al. Long-term direct costs before and after proctocolectomy for ulcerative colitis: a population-based study in Olmsted County, Minnesota. Dis Colon Rectum 2009;52:1815–1823. 18. McMullen K, Hicks TC, Ray JE, et al. Complications associated with ileal pouch-anal anastomosis. World J Surg 1991;15:763–766.
7. Gies N, Kroeker KI, Wong K, et al. Treatment of ulcerative colitis with adalimumab or infliximab: long-term follow-up of a singlecentre cohort. Aliment Pharmacol Ther 2010;32:522–528.
19. Barreiro-de Acosta M, Garcia-Bosch O, Souto R, et al. Efficacy of infliximab rescue therapy in patients with chronic refractory pouchitis: a multicenter study. Inflamm Bowel Dis 2012;18:812–817.
8. Ford AC, Sandborn WJ, Khan KJ, et al. Efficacy of biological therapies in inflammatory bowel disease: systematic review and meta-analysis. Am J Gastroenterol 2011;106:644–659.
20. Keshavarzian A, Mayer L, Salzberg B, et al. A multicenter retrospective experience of infliximab in Crohn’s disease patients: infusion reaction rates and treatment persistency. Gastroenterol Hepatol 2007;3:381–390.
9. Reinisch W, Sandborn WJ, Rutgeerts P, et al. Long-term infliximab maintenance therapy for ulcerative colitis: the ACT-1 and -2 extension studies. Inflamm Bowel Dis 2012;18:201–211. 10. Oussalah A, Evesque L, Laharie D, et al. A multicenter experience with infliximab for ulcerative colitis: outcomes and predictors of response, optimization, colectomy, and hospitalization. Am J Gastroenterol 2010;105:2617–2625. 11. Tsai HH, Punekar YS, Morris J, et al. A model of the long-term cost effectiveness of scheduled maintenance treatment with infliximab for moderate-to-severe ulcerative colitis. Aliment Pharmacol Ther 2008;28:1230–1239.
21. Teshima CW, Thompson A, Dhanoa L, et al. Long-term response rates to infliximab therapy for Crohn’s disease in an outpatient cohort. Can J Gastroenterol 2009;23:348–352. 22. Rostholder E, Ahmed A, Cheifetz AS, et al. Outcomes after escalation of infliximab therapy in ambulatory patients with moderately active ulcerative colitis. Aliment Pharmacol Ther 2012;35:562–567. 23. Ma C, Huang V, Fedorak D, et al. Outpatients with ulcerative colitis being treated with adalimuamband infliximab have similar rates of loss of response. Can J Gastroenterol Hepatol 2014;28:A141.
-
2014
Cost-Utility Analysis of Infliximab for UC
8.e3
Supplementary Figure 1. Cost-effectiveness acceptability curves by group (utility ¼ 0.79). *u ¼ utility score for the response to infliximab health state. **Sensitivity analysis by using induction rate of 70.63% for patients on infliximab that is based on weighted average of published randomized controlled studies.6–9 ***Sensitivity analysis by using induction rate of 80.09% for patients on infliximab that is based on actual weighted average of published open-label studies.8,10–13
8.e4
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Supplementary Figure 2. Cost-effectiveness acceptability curves by group (utility ¼ 0.82). *u ¼ utility score for the response to infliximab health state. **Sensitivity analysis by using induction rate of 70.63% for patients on infliximab that is based on weighted average of published randomized controlled studies.6–9,14 ***Sensitivity analysis by using induction rate of 80.09% for patients on infliximab that is based on actual weighted average of published open-label studies.8,10–13
-
2014
Cost-Utility Analysis of Infliximab for UC
8.e5
Supplementary Table 1. Markov Model Input Parameters for Chronic UC Transition probabilities (%)/cycle A. Ongoing steroids14,15 1. Response 2. Unwell 3. Complication 4. Surgery B. Response to steroids14 1. Response 2. Loss of response C. Unwell on steroids14,15 1. Unwell 2. Complication 3. Surgery D. Steroid complication15 1. Surgery 2. Death E. Surgery16,17 1. Early response 2. Complication 3. Death F. Response to surgery16,18 1. Response to surgery 2. Surgical complication 3. CP G. Chronic pouchitis19 1. Response to IFX (CP) 2. Nonresponse (unwell-CP) 3. IFX complication (CP) H. Surgical complication16 1. Hospitalization 2. Death I. Infliximab6,8,9,14 1. Response to IFX 2. IFX complication 3. Nonresponse (unwell) J. Response to IFX6–14,20–23 1. Response to IFX 2. IFX complication 3. Loss of response (unwell) K. IFX complications6,12,14 1. Response to IFX 2. Unwell on steroids 3. Surgery 4. Death L. Death
CP, chronic pouchitis; IFX, infliximab. a Complement probability.
Costs ($CAN)/cycle 864 (25%)
Utility scores/y 0.32 (0.31)
33.92 (28.09–40.33) 57.11 (50.5–63.27) 2.80 (0.56–7.63) a
0
0.79 (0.21)
53.30 (46.81–59.67) a
864 (25%)
0.32 (0.31)
22,527 (25%)
0.16 (0.16)
34,996 (25%)
0.16 (0.16)
0
0.58 (0.15)
a
2.80 (0.56–7.63) 10.00 (6.40–14.28) 98.00 (93.70–99.78) a
a
12.8 (8.76–17.91) 2.50 (0.98–5.69) 85.80 a
11.70 5446 (25%)
0.32 (0.31)
16,562 (25%)
0.49 (0.32)
5446 (25%)
0.32 (0.31)
See Supplementary Table 2 a
7.75 99.50 (97.22–99.99) a
70.63–80.09 1.49 a
5446 (25%)
0.79–0.82
See Supplementary Table 3 5.90 a
14,973 (25%)
0.16 (0.16)
Equal to cost of corresponding health state
0
70.00 14.00 14.00 a
1
8.e6
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Supplementary Table 2. Markov Model Input Parameters for Chronic Pouchitis Health states
Transition probabilities (%)/cycle
M. Response to IFX1 1. Response to IFX1 See Supplementary Table 3 a 2. Lost response (unwell1) 3. IFX complication1 7.75 N. Unwell1 a 1. Unwell1 2. Surgery1 10.00 3. Steroid complication1 2.80 O1. IFX complication1 in the IFX not available arm 1. Response to IFX1 64.60 2. Unwell1 16.70 3. Surgery1 16.70 a 4. Death O2. IFX complication1 in the IFX available arm 1. Response to IFX1 55 2. Unwell1 21.50 3. Surgery1 21.50 a 4. Death P. Surgery1 (permanent ileostomy) a 1. Response1 2. Surgery complication1 12.80 3. Death 2.50 Q. Steroid complication1 1. Surgery1 98.00 a 2. Death R. Response to surgery1 a 1. Response to surgery1 2. Surgical complication1 2.50 S. Surgical complication1 a 1. Response to surgery 2. Death 0.50
IFX, infliximab. a Complement probability.
Costs ($CAN)/cycle
Utility scores/y
5446 (25%)
0.58 (0.15)
864 (25%)
0.32 (0.31)
14,973 (25%)
0.16 (0.16)
14,973 (25%)
0.16 (0.16)
34,996 (25%)
0.16 (0.16)
22,527 (25%)
0.16 (0.16)
0
0.44 (0.11)
16,562 (25%)
0.37 (0.24)
-
2014
Cost-Utility Analysis of Infliximab for UC
8.e7
Supplementary Table 3. Maintenance Probabilities of Patients on Infliximab Over Time UC patientsb Cycle no. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
UC patientsa
6–8,10–13,20–23
6,8,9,14
(sensitivity analysis)
Chronic pouchitis patients (infliximab not available arm)
Chronic pouchitis patients (infliximab available arm)c
70.63 64.45 58.27 52.08 45.9 42.90 39.89 36.89 33.89 33.01 32.13 31.25 30.37 29.49 28.62 27.74 26.86 26.40 25.94 25.48 25.02
80.09 77.55 75.01 72.46 69.92 65.35 60.77 56.20 51.62 50.28 48.94 47.61 46.27 44.93 43.59 42.25 40.91 40.21 39.51 38.82 38.11
84 75 66 55.5 45 41.25 35.5 29.75 24 23.13 22.25 21.38 20.50 19.63 18.75 17.88 17 16 15 14 13
71 63.75 56 47.18 38.25 35.06 30.18 25.29 20.4 19.66 18.91 18.17 17.43 16.69 15.94 15.2 14.45 13.6 12.75 11.9 11.05
a These maintenance probabilities are based on a weighted average of clinical trials, including randomized control trials.6–9,14 The rate at which patients lose response to infliximab was calculated on the basis of data extrapolated from patients treated on infliximab.20–23 b Probabilities were derived from a weighted average of open-label studies reporting infliximab efficacy.8,10–13 The decrease in response over time of patients on infliximab was calculated on the basis of data derived from patients treated with infliximab.20–23 c To calculate the response probability for patients with chronic pouchitis who had been previously exposed to and failed by infliximab, a 15% discount was taken from the probability of response of patients with chronic pouchitis who had never been exposed to infliximab.
Supplementary Table 4. Response Probabilities of Patients Dose Escalated on Infliximab Over Time Cycle no. 0 1 2 3 4 (1 y) 5 6 7 8 (2 y) 9 10 11 12 (3 y) 13
a
UC patients rate of responsea 60 54.74 49.49 44.24 38.99 36.44 33.88 31.33 28.78 28.04 27.29 26.54 25.80 25.0
Decrease in response over time of patients on infliximab was calculated on the basis of data derived from patients treated with infliximab.22,23
8.e8
Ung et al
Clinical Gastroenterology and Hepatology Vol.
-,
No.
-
Supplementary Table 5. Health State Definitions Health state
Definition
Response to medical treatment (steroid/IFX)
On their respective treatment regimens, patients have decreased symptoms: reduced bowel frequency as compared with active disease, may intermittently have blood with defecation, and are able to taper off steroid treatment. A disease pattern in which the patient is not acutely ill with unmistakable indication for resection or not in remission and symptom-free, despite being treated with medical therapy (oral steroids, mesalamine, azathioprine). Patients experience symptoms of chronic intractable disease (recurrent acute colitis, steroid dependence, chronic fecal urgency, persistent active disease, or complications of medical therapy). Patients in this cohort would have a UC Disease Activity Index score of 3–8 (out of 12) or a partial Mayo score of 2–6 (out of 9), corresponding to a mild-moderate disease activity state. Patients may develop mild anemia and have 5–8 bowel movements a day with some rectal bleeding. A common long-term complication after restorative proctocolectomy with ileal pouchanal anastomosis for patients with UC. This disease state is characterized by a nonspecific inflammation of the ileal pouch after surgery. Signs and symptoms may include increased stool frequency, urgency, incontinence, and dehydration. Majority of patients who develop chronic pouchitis are often treated with courses of antibiotics (metronidazole or ciprofloxacin); however, there is a subset of patients whose disease is refractory to antibiotic treatment. In this study, the chronic pouchitis health state refers to chronic pouchitis patients who are refractory to antibiotic therapy to attain remission. Any complication that occurred as result of a medical treatment (ongoing steroid or IFX) that necessitated a change in treatment or health state (steroid / surgery/ death, IFX / steroid/surgery/death). Patient whose disease never responded or at first responded but then lost response to IFX in subsequent cycles. Patients whose UC did not respond to medical treatment undergo 1 of 2 possible surgeries to manage their disease. Patients will undergo a 2-step or 3-step procedure as indicated by their physicians. Any complication that arises as a result of the surgery that requires the patient to be hospitalized or undergo further surgery to correct the complication.
Unwell
Chronic pouchitis
Steroid/IFX complication
Non-/loss response (IFX) Surgery
Surgical complication
IFX, infliximab.