SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Background  Infliximab has been shown to be efficacious in moderate-to-severe Crohn’s disease (CD).

Aim  To evaluate the cost-effectiveness of scheduled maintenance treatment with infliximab in luminal and fistulizing CD patients.

Methods  Markov models were constructed to simulate the progression of adult CD patients with and without fistulae during treatment with infliximab (5 mg/kg). Transitions were estimated from published clinical trials of infliximab. Standard care, comprising immunomodulators and/or corticosteroids was used as a comparator. An average weight of 60 kg was used to estimate the dose of infliximab. The costs and outcomes were discounted at 3.5% over 5 years. The primary effectiveness measurement was quality-adjusted life years (QALYs) estimated using EQ-5D. One-way and probabilistic sensitivity analyses were performed by varying the infliximab efficacy estimates, costs and utilities.

Results  The incremental cost per QALY gained was £26 128 in luminal CD and £29 752 in fistulizing CD at 5 years. Results were robust and remained in the range of £23 752–£38 848 for luminal CD and £27 047–£44 206 for fistulizing CD. Patient body weight was the most important factor affecting cost-effectiveness.

Conclusion  Eight-week scheduled maintenance treatment with infliximab is a cost-effective treatment for adult patients suffering from active luminal or fistulizing CD.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Crohn’s disease (CD) is a chronic relapsing intestinal inflammatory condition requiring multiple hospital visits and in-patient stays. Observational studies have demonstrated that patients with CD progress from inflammatory to complex stricturing or penetrating disease over time.1 This results in a cumulative risk of intestinal resection of 70% during the first 15 years after diagnosis.2 Despite a significant increase in the use of immunosuppressive agents such as azathioprine over the last 30 years, the requirement for surgical resection has remained constant.3 In addition, CD has a significant impact on an individual’s productivity and quality of life, with 15–24% being considered disabled 5 years and 15% of patients being unable to work 15 years after diagnosis.4 Therefore, given the high prevalence, CD results in a significant use of health service resources5 and is a considerable economic burden to society.

The financial impact of CD has been reported in a single study from the UK.6 This suggested that hospital admissions account for 75% of the financial burden, with mean costs over 6 months nearing £7000 for patients, compared to £516 for patients not requiring hospitalization. The findings of this study were consistent with research from Sweden and the US, which also suggested that hospitalization accounts for a large proportion of patient treatment costs in CD.7, 8 Furthermore, these studies do not include the considerable financial implications of the reduction in productivity in patients with active disease.4 Therefore, achieving symptom-free remission and reducing hospital episodes remain important goals in CD treatment.

Over the last decade, the management of CD has been revolutionized initially by the introduction of infliximab and more recently by other anti-TNF-α therapies that have been shown to induce and maintain remission in patients resistant to standard medical therapies.9–11 Clinical guidelines issued by the British Society for Gastroenterology recommend that patients with severe active CD [Harvey Bradshaw Index >8, CD activity index (CDAI) >300], who are refractory to or intolerant of steroids and immunosuppression, and for whom surgery is inappropriate, the TNF-α inhibitor infliximab (Remicade; Schering-Plough Ltd, Welwyn Garden City, UK) should be the treatment of choice.12–14 This recommendation is in concordance with the guidance issued by the National Institute for Health and Clinical Excellence (NICE) for infliximab in CD,15 which is due to be updated shortly. Other clinical societies such as the European Crohn’s and Colitis Organisation recommend use of infliximab earlier in the disease management; for steroid-dependent Crohn’s patients and for patients with a single immunomodulator failure.16

The ACCENT I trial showed that scheduled maintenance therapy with infliximab was significantly more effective than placebo at maintaining clinical remission in patients who had responded to an initial infusion.9 In addition, patients receiving scheduled maintenance therapy with infliximab were more likely to achieve mucosal healing,17 and were less likely to be hospitalized and less likely to require surgery than those receiving episodic treatment.18 The ACCENT II trial reported a significant benefit of scheduled maintenance infliximab over placebo in the treatment of fistulizing CD.19 Scheduled maintenance infliximab therapy also improved health-related quality of life,20 and reduced the requirement for both hospitalization and surgery.21, 22 This suggests that maintenance infliximab therapy may bring significant quality-adjusted life-year (QALY) gains in adult patients with either severe active luminal CD or fistulizing CD.

A retrospective study conducted on 205 patients who received infliximab across seven centres in the UK demonstrated that the mean costs over the 6 months preceding the initial infusion exceeded the mean costs over the 6 months following the infusion by an average of £138 per patient.23 Therefore, it is possible that a substantial part of the acquisition cost of infliximab could be offset by savings on other disease-related costs in patients not responding to, or intolerant to steroids and immunomodulators.

The present economic evaluation was performed to assess the cost-effectiveness of infliximab scheduled maintenance treatment at the licensed dose of 5 mg/kg, in comparison with standard care without infliximab, for the treatment of patients with active luminal CD or fistulizing CD.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Model overview

Markov models were used to simulate the disease progression and track the associated costs and outcomes as QALYs over 5 years of treatment in hypothetical cohorts of patients with active luminal and fistulizing CD. Figure 1a illustrates the Markov model used in the analysis of active luminal CD. For patients receiving treatment in the model, the disease severity was characterized by two discrete ‘on-treatment’ health states, namely remission (CDAI ≤150) and treatment response but not achieving remission (CDAI >150; from here on referred to as the active state). All patients started in the active health state and remained in this health state for the first model cycle. At the end of the first and each subsequent model cycle, patients either remained in the active health state, or moved to a different health state. Patients responding to treatment and achieving a CDAI ≤150 moved to the remission and remained on treatment. Patients demonstrating a clinical response (as defined in ACCENT I) but not achieving remission remained in the active state and continued to receive treatment.9 Patients classified as nonresponders or patients discontinuing treatment stopped treatment and moved to the nonresponding active state. Once patients had failed treatment and moved to this ‘off-treatment’ state, they could not restart infliximab treatment and return to the ‘on-treatment’ states. However, these nonresponders and treatment failures were followed throughout the model to capture their costs and outcomes.

image

Figure 1.  (a) Markov model for active luminal CD. (b) Markov model modification used in fistulizing CD. Two separate numbers denote different transitions in standard care and infliximab (*) treatment arm. A single number denotes identical transitions in both groups.

Download figure to PowerPoint

Patients in both the ‘on-treatment’ and ‘off-treatment’ health states could transition to surgery. In the subsequent model cycle, patients undergoing surgery could either undergo repeat surgery due to immediate postsurgery complications and remain in the surgery state or move to a postsurgery health state (postsurgery remission or postsurgery complications). Patients in postsurgery remission could continue in the same health state, enter surgery to undergo repeat surgery, suffer from a complication to enter postsurgery complications or have recurrence of CD to enter the nonresponding active state. In the absence of any evidence to estimate the effectiveness of re-treatment with infliximab for infliximab failures, it was assumed that patients with a recurrence of their CD would not be offered re-treatment with infliximab. Similarly, patients experiencing postsurgery complications could continue in the same health state, enter surgery to undergo repeat surgery, respond to the treatment for their complications and enter postsurgery remission, or have recurrence of CD and enter the nonresponding active state.

The model used for fistulizing CD was similar to that used in the active luminal CD patients except that the ‘on-treatment’ health states of remission and active were further classified as with or without fistulae. This resulted in four ‘on-treatment’ health states as shown in Figure 1b. The remaining health states and the patient progression within these health states were identical to the active luminal CD model.

The patient and treatment parameters were based on standard treatment protocols and ACCENT trial evidence as detailed in Table 1.9, 19 Scheduled maintenance as opposed to episodic therapy was chosen as the comparator, as scheduled maintenance now represents standard medical care in expert centres within the UK. There is clear evidence that scheduled therapy achieves higher levels of mucosal healing, with lower immunogenicity than episodic therapy.16

Table 1.   Patient and treatment parameters in the Markov models
Model parameterLuminal active CDFistulizing active CD
  1. CD, Crohn’s disease; CDAI, CD activity index.

  2. * Treatment strategy in accordance with the UK-marketing authorization for infliximab.

  3. † Based on the response criteria defined in ACCENT I and II, respectively.

  4.  The cycle lengths were selected to match the assessment visits in the ACCENT I and II trials.

Patient populationCDAI score between 220 and 400 History of CD for at least 3 months Patients with a history of CD with single or multiple draining fistulae, including perianal fistulae and enterocutaneous fistulae, for at least 3 months
Sources of efficacy estimatesTargan et al.26 ACCENT I9Present et al.27 ACCENT II19
TreatmentInfliximab 5 mg/kg at weeks 0, 2, 6 and 8 weeks thereafterInfliximab 5 mg/kg at weeks 0, 2, 6 and 8 weeks thereafter
ComparatorStandard care without infliximabStandard care without infliximab
Treatment strategy*Identify responders at week 2† Continue treatment to the point of failureIdentify responders at week 14† Continue treatment to the point of failure
Time horizon5 years5 years
Cycle lengths‡Weeks 0–2, 2–6, 6–10, 10–14 and every 8 weeks thereafterWeeks 0–14, 14–30 and every 24 weeks thereafter

Transition probabilities

Probability of death.  There is conflicting evidence on the impact of CD on mortality.23 However, some recent studies have suggested that the survival rates in patients with active luminal CD and fistulizing CD were similar to those of the general population, after adjustment for age and gender.24, 25 Therefore, the current analysis assumes mortality in CD patients to be equivalent to general population in England and Wales.

Transition probabilities for different health states.  Results from the published, randomized placebo-controlled induction studies were used to estimate the initial response to infliximab,26, 27 whereas infliximab efficacy in maintaining this remission was obtained from the ACCENT I and II maintenance trials.9, 19 All patients in ACCENT I received infliximab at week 0 and all patients in ACCENT II received the full induction dose of infliximab (week 0, 2 and 6). Hence, a ‘true placebo effect’ could not be estimated from ACCENT trials. To overcome this bias, the transition probabilities for the ‘on-treatment’ health states were estimated from the Targan study for weeks 0–2 in the active luminal CD analysis.26 The transition probabilities for subsequent model cycles were derived from patients randomized as responders at week 2 in ACCENT I.9 Patients in the standard care treatment arm of ACCENT I received a single infusion of infliximab at week 0 followed by placebo infusions at weeks 2 and 6, and every 8 weeks thereafter. Therefore, it was assumed that by the end of next model cycle (week 10), the effect of this single infusion would be washed away and the patients in standard care arm would demonstrate the appropriate placebo effect.

Likewise, the transition probabilities for the induction phase (week 0–14) in the fistulizing CD analysis were estimated using the data from the present study,27 and subsequent transitions were estimated using patient level data from responders to the induction regimen in ACCENT II.19 The transition probabilities for the first 54 weeks were derived from the patient level data in ACCENT trials as explained above. The transitions observed in the last assessment cycle in ACCENT trials (weeks 46–54) were then used to extrapolate the analysis up to 5 years. Figures 1a and 1b display the average transition probabilities over the first 54 weeks between the health states.

Probability of surgery, postsurgical states and CD recurrence.  Transition probabilities for surgery and postsurgical states were obtained from the published literature24, 28, 29, 52 and are displayed in Table 2. The probability of a postsurgery complication was based on a cohort control study, which compared the postsurgical complications between groups of patients treated with or without infliximab before surgery.28 As this study concluded that use of infliximab did not have any significant impact on the frequency of postsurgery complications, a weighted average of complications observed in infliximab and non-infliximab treatment arms was used in our analysis.

Table 2.   Surgery, postsurgery complications and recurrence rates in Crohn’s disease patients
 Observed incidence (%)Rate/8-week cycle (%)Ref.
  1. CD, Crohn’s disease.

Surgery (10 years)640.1624
Early complications (within 10 days)15.210.428
Late complications (within 3 months)10.110.128
CD recurrence (luminal CD)15.6 per year2.629
CD recurrence (fistulizing CD)59 per 2 years6.652
Repeat surgery16.7 per year2.829

Costs

Perspective and drug costs.  The perspective adopted on costs was that of the National Health Service (NHS) in England and Wales using a reference year 2005–2006. Productivity costs are significant in CD, but were omitted because of this choice of perspective. The total cost associated with infliximab treatment was broken down into acquisition costs (£419.73 per 100 mg vial) and administration costs (£96.00 per infusion30) leading to a total cost per infusion of £1355.19 for an adult patient. The cost of drug acquisition was calculated assuming an average body weight of 60 kg based on the previous NICE guidance.15

The use of concomitant medications such as immunomodulators, aminosalisylates and corticosteroids was obtained from the baseline data in the ACCENT trials.9, 19 This baseline medication use was assigned to patients in nonresponding active state in our model. In the absence of any data to suggest a change in use of immunomodulators and aminosalisylates with improvement in health states, it was assumed that patients would continue their use even after entering remission or active health states. As patients responded to treatment and achieved remission, the use of corticosteroids was assumed to reduce gradually such that patients in remission were free of any corticosteroid use.

The model also did not explicitly account for the costs of adverse events. Potential costs associated with infusion reactions, such as headache, dizziness, nausea, injection-site reaction, flushing chest pain, dyspnoea and pruritus, were considered to be incorporated in the drug administration cost. Serious adverse events, related either to infliximab or other medications for CD such as corticosteroids or immunomodulators, were assumed to be included as part of the hospitalization costs.

Surgery, hospitalization and other assessments.  The estimated cost of surgical interventions used was £5277 on the basis of data published in the NHS National Schedule of Reference Cost (NSRC) for 2005/06.31 This was an average cost, taking into account elective and non-elective admissions, with or without complications or comorbidities. The cost of hospitalization and other assessments was adapted from the Jewell study by adjusting the reported resource use with published estimates from NSRC 2005/06 (Table 3).23, 31 The study did not distinguish resource use for individual health states; therefore, all three presurgery health states (remission, active and nonresponding active) were assumed to have identical resource use per cycle. The impact of this counter intuitive assumption was further explored in one-way sensitivity analysis (scenario A). A study by Lichtenstein et al.22 demonstrated that patients who spent 0–25% of their time in remission have a 5.57-fold increase in hospitalizations and patients who spend 50% of their time in remission have a threefold increase in hospitalizations compared to patients who spend 75–100% of their time in remission. Therefore, based on the Lichtenstein study, it was assumed that patients in active health state would have threefold increase and patients in nonresponding active health state would have 5.57-fold increases in resource use compared to patients in remission.

Table 3.   Hospitalization and assessments
ResourcePublished estimate of resource use (n = 205)*Cost per patient per cycleUnit costSource
PreinfliximabPostinfliximabStandard careInfliximab
  1. TELIP, Elective In-patient HRG data; TNELIP, Non-elective In-patient HRG data; TOPS FAA, Out-patient Adult First Attendance Data; TOPS FUA, Out-patient Adult Follow-up Attendance Data.

  2. * Resource use of 205 patients for 6 months23.

Diagnostic procedures16263£354.92£138.98£1424.50TELIP, TNELIP31
Examination under anaesthetic5017£139.62£47.47£1860.50TELIP, TNELIP31
In-patient days1435342£337.58£80.46£156.7347
Out-patient visits555534£90.24£86.83£108.33TOPS FAA, TOPS FUA31
Total  £922.37£353.73  

The Jewell study did not estimate the cost associated with postsurgery health states. Therefore, the costs associated with postsurgery remission was assumed to be equivalent to medical remission and the cost of postsurgery complications (£1460.40/cycle) was calculated based on the resource use by a hypothetical patient in these health states as estimated by a panel of UK gastroenterologists.32 Each panel member estimated the resource use independently and values used in the economic model were averages of individual estimates, costed using NSRC 2005/06.31

Outcomes

The primary effectiveness measure used in these analyses was the QALY. Evidence relating to a survival benefit associated with infliximab in CD was not available. Gains in QALYs are therefore driven exclusively by quality of life benefits estimated using published data on health state preferences.33 Casellas et al.33 in their study estimated health state preferences of Spanish CD patients (n = 200) using EQ-5D. The EQ-5D responses of these patients (obtained by personal communication with Dr Casellas) were then converted into utilities using UK tariffs.34 Health state preferences for surgery and postsurgery health states were unavailable in Casellas study. Therefore, the corresponding health state preference values were estimated by a panel of UK gastroenterologists.32Table 4 provides a summary of all health state preference estimates employed in the economic evaluation.

Table 4.   Utility estimates33
Markov model health stateCorresponding state from the sourceSourceUtility estimate
  1. HBI, Harvey Bradshaw Index.14

RemissionRemission (HBI <3)330.83
ActiveActive (HBI >3)330.55
Nonresponding activeNot available; assigned by panel of UK gastroenterologists320.4
SurgeryPostsurgery (<2 months)470.73
Postsurgery remissionPostsurgery (>2 months)470.67
Postsurgery complicationsNot available; assigned by panel of UK gastroenterologists320.5
Remission + fistula closureRemission330.83
Remission + fistulaNot available; assigned by panel of UK gastroenterologists320.68
Active + fistula closureActive330.55
Active + fistulaNot available; assigned by panel of UK gastroenterologists320.4
Death (D)

Cost-effectiveness analyses

The cost-effectiveness analysis has been presented as incremental cost per QALY gained. Costs and outcomes were calculated separately for the entire cohort of patients starting on both infliximab and standard care irrespective of whether they were responders or nonresponders and were discounted to present values at 3.5% per annum, in accordance with NICE guidelines.35 Multiple one-way sensitivity analyses were conducted varying the parameters such as patient age and weight, time horizon, health state preferences, cost of healthcare resources, discounting rate and administration cost of infliximab to assess the variability surrounding the model results.

To estimate the uncertainty around the important parameters, the transition probabilities associated with presurgery health states, costs of healthcare resources and health state preferences were subjected to probabilistic sensitivity analysis (PSA). The parameters of transition probabilities were set by using observed patient numbers in the corresponding trials. As no treatment effect was assumed beyond surgery, postsurgery transition probabilities were not subjected to PSA.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Cost-effectiveness analyses (base cases)

In active luminal CD, scheduled maintenance therapy with infliximab derived a mean additional 0.19 QALYs at an additional cost of £4873 compared with standard care without infliximab. Therefore, the incremental cost per QALY gained for infliximab against standard care, was £26 128 (Table 5). In fistulizing CD, scheduled maintenance treatment with infliximab derived an additional 0.20 QALYs at an additional cost of £5998 compared to standard care without infliximab. The estimated incremental cost per QALY gained was £29 752 in fistulizing CD (Table 5).

Table 5.   Cost-effectiveness results for infliximab scheduled maintenance vs. standard care in severe active luminal CD and fistulizing CD at 5 years
Patient populationMean costsDifferenceMean QALYsDifferenceICERs
  1. CD, Crohn’s disease; QALYs, quality-adjusted life years; ICER, incremental cost-effectiveness ratio.

Severe active luminal CD
 Infliximab £31 499£48732.1450.186£26 128
 Standard care£26 6271.959
Fistulizing CD
 Infliximab £37 488£59982.4490.202£29 752
 Standard care£31 4902.247

Sensitivity analyses

One-way sensitivity analyses showed that results remained in the range of £23 752–£38 848 at 5 years for active luminal CD and £27 047–£44 206 at 5 years for fistulizing CD (Table 6). Because of the weight-based dosing of infliximab, patient weight had the most impact on the incremental cost-effectiveness ratio (ICER) with the ICER increasing to £38 848 in luminal CD and £44 206 in fistulizing CD for an 80-kg patient. The change in health state preferences had the most significant impact on improving the ICERs with a 10% increase in utilities resulting in an ICER of £23 752 in luminal CD and £27 047 in the fistulizing CD. The impact of the other parameters was less prominent with patient age being the least significant. The cost-effectiveness planes resulting from the PSA are displayed in Figures 2a and 2b. The PSA showed that the results were robust with majority of simulations clustered together. In both severe active CD and fistulizing CD, infliximab scheduled maintenance resulted in additional QALYs at an additional cost compared to standard care.

Table 6.   One-way sensitivity analysis*– infliximab scheduled maintenance vs. standard care in severe active luminal and fistulizing Crohn’s disease
ParameterBase case estimateSensitivity estimateCost/QALY in severe active luminal CD†Cost/QALY in fistulizing CD‡
  1. CD, Crohn’s disease; QALYs, quality-adjusted life years.

  2. * All results except ‘time horizon’ assumes the time horizon of 5 years used in the base case.

  3. † Base case: cost per QALY = £26 128.

  4. ‡ Base case: cost per QALY = £29 752.

  5. § Cost (nonresponding active) = 5.57 × cost (remission) and cost (active) = 3 × cost (remission).

  6. ¶ Cost (nonresponding active or A + F) = 5.57 × cost (R + FC), cost (R + F or A + FC) = 3 × cost (R + FC).

Patient weight60 kg80 kg£38 848£44 206
70 kg (with vial sharing)£32 488£36 979
Time horizon5 years1 year£29 060£33 318
Lifetime£28 864£30 006
Discount rateCosts – 3.5% QALYs – 3.5%Cost and QALYs – 1.5%£26 112£29 574
Cost – 1.5% and QALYs – 6%£27 540£31 234
Cost – 6% and QALYs – 1.5%£24 793£28 380
Cost and QALYs – 6%£26 149£29 973
Baseline age45 years30 years£26 126£29 739
60 years£26 133£29 817
Healthcare costs Scenario A£27 332§£30 214¶
Health state utilities Increase by 10%£23 752£27 047
Decrease by 10%£29 031£33 058
Infliximab administration cost£96.00£124.00£26 976£30 717
image

Figure 2.  (a) Cost-effectiveness plane in active luminal CD. (b) Cost-effectiveness plane in fistulizing CD.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Infliximab is an effective induction and maintenance strategy for patients with moderate-to-severe inflammatory luminal and fistulizing CD.9, 19 Placebo-controlled trials have demonstrated that scheduled maintenance therapy results in a reduced requirement for surgery and hospitalization compared to both placebo and episodic ‘on-demand’ infliximab therapy.18, 21, 22 The purpose of this analysis was to assess the cost-effectiveness of infliximab scheduled maintenance treatment at the licensed dose of 5 mg/kg compared with standard care without infliximab, for the treatment of patients with active luminal CD or fistulizing CD over a 5-year time frame.

Several published studies have attempted to estimate the cost-effectiveness of infliximab in CD.36–39 Three of these studies37–39 used CD progression documented by Silverstein et al.40 as a basis for their modelling approach, whereas the fourth study by Kaplan et al.36 used response to anti-TNF therapies as a basis for health state classification. The Silverstein study and the cost-effectiveness analyses based on it were conducted prior to ACCENT trial publications and included all CD patients. Our analysis focussed on moderate-to-severe CD patients eligible to receive anti-TNF therapy and hence only comprised of a subpopulation of Silverstein cohort. In addition, Silverstein study classified patients into different health states based on their medication use instead of CDAI score.13 Therefore, rather than modifying the transitions in Silverstein cohort, we used the outcome measurements used in ACCENT trials (response and remission) to build our model framework. Kaplan et al.36 also used a similar approach in their evaluation of anti-TNF therapies.

Our analysis used two separate infliximab treatment strategies for active luminal and fistulizing CD. In active luminal CD, the responders were identified at week 2 and the infliximab treatment was only continued for these week 2 responders until failure. In contrast, fistulizing CD patients received full induction dose of infliximab and only week 14 responders continued with infliximab to the point of failure. This approach was adopted to comply with infliximab license, which recommends different responder identification strategies in active luminal and fistulizing CD. The other important consideration is the choice of time horizon. The published cost-effectiveness analyses have used time horizons ranging from 1 year36 to lifetime.37 In our analysis, we extrapolated the efficacy of the treatments up to 5 years. Studies published in the literature have demonstrated long-term efficacy of scheduled maintenance treatment with infliximab in maintaining remission up to 4 years.41–43 Therefore, a 5-year time horizon was deemed to be appropriate to capture all relevant costs and outcomes associated with infliximab treatment. The model predictions for CD outcomes from our analysis were similar to those observed in clinical practice. Of all patients in remission at the end of first year in our analysis, 11% maintained their remission at the end of fifth year compared to 12.2% in Casellas study.41 This assumption was also reinforced in sensitivity analyses wherein the duration of therapy did not have a significant impact on ICER.

The other important parameters affecting ICER were health state preferences and costs. The PSA demonstrated that health state preferences had a significant impact on ICERs. A proportion of PSA simulations resulted in negative incremental QALY gain for patients treated with infliximab. This, however, is contrary to the trial and clinical practice evidence, both of whom suggest significant quality of life gain for patients treated with infliximab.20, 22, 41 This may be attributed to the uncertainty around the health state preference values used in this analysis. We used health state preference values adopted from Casellas study,33 as this information was not available in the ACCENT trials. Similar health state preferences also were available from other published studies.36, 39, 44 However, none of them conformed with the NICE’s analysis framework within which our analysis was conducted.35

The resource use estimates in our model were derived from Jewell study,23 which compared the resource use before and after the use of infliximab in CD patients in the UK. Resource use estimates with and without the use of infliximab were also available from two other studies.45, 46 We estimated resource use based on Jewell study, as it represented the clinical practice specific to UK, but explored the uncertainty around the estimates in the sensitivity analysis. Both one-way sensitivity analysis where the resource use associated with a particular health state was varied (scenario A) and the PSA where the costs attributable to a particular resource were varied did not affect the ICERs significantly. The resource use predicted by our model also was comparable to other published estimates. In the study conducted by Saro et al.,45 the rate of major surgeries was 9.8% before infliximab and 4% after infliximab. The corresponding surgery rate at the end of first year was 9% for standard care and 7% for infliximab in our analysis. The hospitalization rate in Saro study reduced 5.68-fold after introduction of infliximab, whereas we assumed a 5.57-fold reduction by achieving remission in our analysis. The benefit of infliximab was less pronounced in Rubenstein study,46 which demonstrated a 38% reduction in surgeries, but no significant change in hospitalizations or hospital stay.

Another important factor affecting the ICERs was the patient weight. The baseline analysis used patient weight of 60 kg, as this was what had been used in the initial infliximab appraisal by NICE.15 However, this was substantially lower than the mean weight of patients entering the ACCENT I (71.1 kg) and II (73.4 kg) studies, which may be biased towards a North American population.9, 19 A review of patients with severe CD in the UK reports a mean weight of 67.8 kg.47 This would still increase the ICER for both luminal and fistulizing CD above the threshold used by NICE. However, most UK infusion units treat several patients at once and utilize ‘vial sharing’ to minimize drug costs.48 The exact ICER for a patient weighing 67.8 kg treated in a unit that uses vial sharing is £31 089 for severe active luminal disease and £36 979 for fistulizing disease.

Overall, the base case ICERs were £26 128 in active luminal CD and £29 752 in fistulizing CD. An unofficial ICER threshold of £30 000 is used by NICE when considering whether a therapeutic intervention is cost-effective.49 Therefore, using the baseline parameters of this model, scheduled maintenance infliximab is cost-effective in the treatment of both active luminal and fistulizing CD. Extensive one-way sensitivity analyses demonstrated a range of ICER of £23 752–£38 848 for active luminal CD, and £27 047–£44 206 for fistulizing CD at 5 years. The PSA demonstrated the robustness of these estimates. In spite of this shortcoming, the overall analysis confirms the superiority of scheduled maintenance infliximab over standard care in both luminal and fistulizing CD.

A recent open randomized trial found benefit of infliximab in inducing and maintaining remission in newly diagnosed CD patients.50 One of the major deterrents for early use of biologics in CD treatment pathway is the cost of treatment associated with the biologics. However, our lifetime cost-effectiveness analysis demonstrated that the cost of long-term treatment with infliximab can be justified by the clinical and quality of life benefit achieved through remission in CD patients. Although infliximab is an effective medication for patients with moderate-to-severe CD, loss of response to the 5 mg/kg dose occurs. In our analysis, patients who lost response after having initially responded stopped infliximab treatment and were switched back to standard care without infliximab. However, in clinical practice, these patients are treated with a dose escalation of infliximab to 10 mg/kg or a reduction in the dose interval to 4–6 weekly. An alternative strategy is to swap to an alternative anti-TNF-α such as adalimumab. The GAIN trial demonstrated that adalimumab given at a dose of 160 mg, 80 mg 2 weeks later and then 40 mg every other week is an effective and safe option for infliximab failures.51 These strategies were not assessed in our analysis and may alter the ICER, given the significant cost of surgery. A recently published decision analysis found that for patients losing response to infliximab, dose-escalating infliximab to 10 mg/kg 8 weekly instead of starting adalimumab yielded an excess of 0.03 QALY per patient.36 However, the cost to achieve this benefit was considerable.

In conclusion, infliximab is a highly effective and well-tolerated therapy for the management of moderate-to-severe and fistulizing adult CD patients, and provides a significant clinical benefit over standard care. This economic analysis demonstrated that the incremental costs associated with achieving these clinical benefits are reasonable and that scheduled maintenance therapy with infliximab represents a cost-effective treatment option.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Declaration of personal interests: James Lindsay has served as a speaker, a consultant and an advisory board member for Schering-Plough Ltd, Abbott Laboratories Ltd and Shire Pharmaceutical Ltd. Guy Chung-Faye has served as a speaker, a consultant and an advisory board member for Schering-Plough Ltd. Yogesh Punekar and James Morris are employees of Schering-Plough Ltd. Declaration of funding interests: James Lindsay received research funding from Schering Plough Ltd, Abbott Laboratories Ltd and UCB Pharma Ltd. This study was funded in full by Schering-Plough Ltd, and the writing of this paper was funded in part by Schering-Plough Ltd. Writing support was provided by Maria Hansson and funded by Schering-Plough Ltd.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Cosnes J, Cattan S, Blain A, et al. Long-term evolution of disease behavior of Crohn’s disease. Inflamm Bowel Dis 2002; 8: 24450.
  • 2
    Munkholm P, Langholz E, Davidsen M, Binder V. Intestinal cancer risk and mortality in patients with Crohn’s disease. Gastroenterology 1993; 105: 171623.
  • 3
    Cosnes J, Nion-Larmurier I, Beaugerie L, et al. Impact of the increasing use of immunosuppressants in Crohn’s disease on the need for intestinal surgery. Gut 2005; 54: 23741.
  • 4
    Munkholm P, Langholz E, Davidsen M, Binder V. Disease activity courses in a regional cohort of Crohn’s disease patients. Scand J Gastroenterol 1995; 30: 699706.
  • 5
    Hanauer SB, Cohen RD, Becker RV, Larson LR, Vreeland MG. Advances in the management of Crohn’s disease: economic and clinical potential of infliximab. Clin Ther 1998; 20: 100928.
  • 6
    Bassi A, Dodd S, Williamson P, Bodger K. Cost of illness of inflammatory bowel disease in the UK: a single centre retrospective study. Gut 2004; 53: 14718.
  • 7
    Blomqvist P, Ekbom A. Inflammatory bowel diseases: health care and costs in Sweden in 1994. Scand J Gastroenterol 1997; 32: 11349.
  • 8
    Feagan BG, Fedorak RN, Irvine EJ, et al. A comparison of methotrexate with placebo for the maintenance of remission in Crohn’s disease. North American Crohn’s Study Group Investigators. N Engl J Med 2000; 342: 162732.
  • 9
    Hanauer SB, Feagan BG, Lichtenstein GR, et al. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. Lancet 2002; 359: 15419.
  • 10
    Colombel JF, Sandborn WJ, Rutgeerts P, et al. Adalimumab for maintenance of clinical response and remission in patients with Crohn’s disease: the CHARM trial. Gastroenterology 2007; 132: 5265.
  • 11
    Schreiber S, Khaliq-Kareemi M, Lawrance IC, et al. Maintenance therapy with certolizumab pegol for Crohn’s disease. N Engl J Med 2007; 357: 23950.
  • 12
    Carter MJ, Lobo AJ, Travis SP. Guidelines for the management of inflammatory bowel disease in adults. Gut 2004; 53(Suppl 5.): V116.
  • 13
    Best WR, Becktel JM, Singleton JW, Kern F. Development of a Crohn’s disease activity index: National Cooperative Crohn’s Disease Study. Gastroenterology 1976; 70: 43944.
  • 14
    Harvey R, Bradshaw J. A simple index for Crohn’s disease activity. Lancet 1980; 1: 514.
  • 15
    National Institute of Health and Clinical Excellence (NICE). Guidance on the Use of Infliximab for Crohn’s Disease. London: Technology Appraisal Guidance 40, April 2002.
  • 16
    Travis S, Trange E, Lemann A, et al. European evidence based consensus on the diagnosis and management of Crohn’s disease: current management. Gut 2006; 55(Suppl. 1): i1635.
  • 17
    Rutgeerts P, Diamond R, Bala M, et al. Scheduled maintenance treatment with infliximab is superior to episodic treatment for the healing of mucosal ulceration associated with Crohn’s disease. Gastrointest Endosc 2006; 63: 43342.
  • 18
    Rutgeerts P, Feagan BG, Lichtenstein GR, et al. Comparison of scheduled and episodic treatment strategies of infliximab in Crohn’s disease. Gastroenterology 2004; 126: 40213.
  • 19
    Sands BE, Anderson FH, Bernstein CN, et al. Infliximab maintenance therapy for fistulizing Crohn’s disease. N Engl J Med 2004; 350: 87685.
  • 20
    Feagan BG, Yan S, Bala M, Bao W, Lichtenstein GR. The effects of infliximab maintenance therapy on health-related quality of life. Am J Gastroenterol 2003; 98: 22328.
    Direct Link:
  • 21
    Lichtenstein G, Yan S, Bala M, et al. Infliximab maintenance treatment reduces hospitalizations, surgeries and procedures in fistulizing Crohn’s disease. Gastroenterology 2005; 128: 8629.
  • 22
    Lichtenstein G, Yan S, Bala M, et al. Remission in patients with Crohn’s disease is associated with improvement in employment and quality of life and a decrease in hospitalisations and surgery. Am J Gastroneterol 2004; 99: 916.
    Direct Link:
  • 23
    Jewell DP, Satsangi J, Lobo A, et al. Infliximab use in Crohn’s disease: impact on health care resources in the UK. Eur J Gastroenterol Hepatol 2005; 17: 104752.
  • 24
    Jess T, Loftus E, Scott Harmsen W, et al. Survival and cause specific mortality in patients with inflammatory bowel disease: a long term outcome study in Olmsted County, Minnesota, 1940–2004. Gut 2006; 55: 124854.
  • 25
    Lichtenstein G, Feagan B, Cohen R, et al. Serious infections and mortality in association with therapies for Crohn’s Disease: TREAT registry. Clin Gastroenterol Hepatol 2006; 4: 62130.
  • 26
    Targan SR, Hanauer SB, Van Deventer SJ, et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn’s disease. Crohn’s Disease cA2 Study Group. N Engl J Med 1997; 337: 102935.
  • 27
    Present DH, Rutgeerts P, Targan S, et al. Infliximab for the treatment of fistulas in patients with Crohn’s disease. N Engl J Med 1999; 340: 1398405.
  • 28
    Marchal L, D’Haens G, Van Assche G, et al. The risk of post-operative complications associated with infliximab therapy for Crohn’s disease: a controlled cohort study. Aliment Pharmacol Ther 2004; 19: 74954.
  • 29
    Wolters FL, Russel MG, Sijbrandij J, et al. Phenotype at diagnosis predicts recurrence rates in Crohn’s disease. Gut 2006; 55: 112430.
  • 30
    National Institute of Health and Clinical Excellence (NICE). Guidance on the Use of Infliximab for Psoriasis. London: Technology Appraisal Guidance 134, January 2008.
  • 31
    Department of Health. NHS Reference Costs 2005–06. Published 7 December 2006. Available at http://www.dh.gov.uk/en/publicationsandstatistics/publications/PublicationsPolicyAndGuidance/DH_062884.
  • 32
    Schering-Plough Ltd. UK Advisory Panel on Resource Use for CD Patients in Individual Post-Surgery Health States. Data on File. Available on request: Schering-Plough Ltd, 2007.
  • 33
    Casellas F, Arenas J, Baudet J. Impairment of health-related quality of life in patients with inflammatory bowel disease: a Spanish multicenter study. Inflamm Bowel Dis 2005; 11: 48896.
  • 34
    Dolan P. Modeling valuations for EuroQol health states. Med Care 1997; 35: 1095108.
  • 35
    NICE. Guide to the Methods of Technology Appraisal. National Institute of Health and Clinical Excellence, April 2004:26.
  • 36
    Kaplan GG, Hur C, Korzenik J, Sands BE. Infliximab dose escalation vs. initiation of adalimumab for loss of response in Crohn’s disease: a cost-effectiveness analysis. Aliment Pharmacol Ther 2007;26:150920.
  • 37
    Jaisson-Hot I, Flourié B, Descos L, Colin C. Management for severe Crohn’s disease: a lifetime cost-utility analysis. Int J Technol Assess Health Care 2004; 20: 2749.
  • 38
    Clark W, Raftery J, Song F, et al. Systematic review and economic evaluation of the effectiveness of infliximab for the treatment of Crohn’s disease. Health Technol Assess 2003; 7: 167.
  • 39
    Arseneau KO, Cohn SM, Cominelli F, Connors AF. Cost-utility of initial medical management for Crohn’s disease perianal fistulae. Gastroenterology 2001; 120: 164056.
  • 40
    Silverstein MD, Loftus EV, Sandborn WJ, et al. Clinical course and costs of care for Crohn’s disease: Markov model analysis of a population-based cohort. Gastroenterology 1999; 117: 4957.
  • 41
    Casellas F, Rodrigo L, Niño P, Pantiga C, Riestra S, Malagelada JR. Sustained improvement of health-related quality of life in Crohn’s disease patients treated with infliximab and azathioprine for 4 years. Inflamm Bowel Dis 2007; 13: 1395400.
  • 42
    Rudolph SJ, Weinberg DI, McCabe RP. Long-term durability of Crohn’s disease treatment with infliximab. Dig Dis Sci 2008; 53: 103341.
  • 43
    Caviglia R, Ribolsi M, Rizzi M, Emerenziani S, Annunziata ML, Cicala M. Maintenance of remission with infliximab in inflammatory bowel disease: efficacy and safety long-term follow-up. World J Gastroenterol 2007; 13: 523844.
  • 44
    Gregor JC, McDonald JWD, Klar N, et al. An evaluation of utility measurement in Crohn’s disease. Inflamm Bowel Dis 1997; 3: 26576.
  • 45
    Saro C, Da La Coba C, Casado MA, et al. Resource use in patients with Crohn’s disease treated with infliximab. Aliment Pharmacol Ther 2007; 10: 131323.
  • 46
    Rubenstein JH, Chong RY, Cohen RD. Infliximab decreases resource use among patients with Crohn’s disease. J Clin Gastroenterol 2002; 35: 1516.
  • 47
    Woehl A, Hawthorne B, Morgan C, Punekar Y, McEwan P. The epidemiology and healthcare resource use in patients with Crohn’s disease: a population based UK study. Value Health 2007; 10: A355.
  • 48
    Somerville M, Brooksby A, Scott DG. Maximizing the use of scarce resources: vial optimization. Rheumatology (Oxford) 2006; 45: 3534.
  • 49
    Towse A. What is NICE’s threshold? An external view. In: DevlinN, TowseA, eds. Cost Effectiveness Threshold: Economic and Ethical Issues. London: King’s Fund/Office of Health Economics, 2002.
  • 50
    D’Haens G, Baert F, Assche GV, et al. Early combined immunosuppression or conventional management in patients with newly diagnosed Crohn’s disease: an open randomised trial. Lancet 2008; 371: 6607.
  • 51
    Sandborn W, Rutgeerts P, Enns R, et al. Adalimumab induction therapy for Crohn’s disease previously treated with infliximab: a randomized trial. Ann Intern Med 2007; 146: 82938.
  • 52
    Makowiec F, Jehle EC, Starlinger M. Clinical course of perianal fistulas in Crohn’s disease. Gut 1995; 37: 696701.