Cite as: Marshall J, Blackhouse G, Goeree R, Brazier N, Irvine E, Faulkner L, Dipchand C, O'Brien B. Infliximab for the Treatment of Crohn's Disease: A Systematic Review and Cost-Utility Analysis. Ottawa: Canadian Coordinating Office for Health Technology Assessment (CCOHTA); 2002. Technology report no 24.
Crohn's disease (CD) is a chronic, inflammatory disorder of the gastrointestinal tract of uncertain etiology. The morbidity and clinical manifestations of CD are variable, and reflect the distribution and severity of the disease. CD often follows a relapsing and remitting course, and can be complicated by intestinal strictures, fistulas and abscesses. Although effective treatments are available, a minority of patients with CD develop refractory disease, suffer poor quality of life and consume considerable healthcare resources. Novel treatments to attenuate inflammation, improve symptoms, and avoid hospitalization are needed. Infliximab (RemicadeTM) is a chimeric human-murine monoclonal antibody to the pro-inflammatory cytokine tumour necrosis factor-alpha (TNFa), and the first biological therapy to win regulatory approval for the treatment of CD. Clinical trials have evaluated infliximab for treatment of patients with fistulizing and active CD resistant to conventional therapy. The costs and potential toxicity of infliximab must be weighed against its effectiveness in improving health outcomes and reducing healthcare resource utilization in a challenging patient population.
(1) To review available data supporting the efficacy, effectiveness and adverse effects of infliximab in the treatment of patients with CD. (2) To review available data evaluating the economic impact of infliximab used to treat patients with CD. (3) To conduct a primary cost-utility analysis of infliximab treatment for patients with active CD resistant to conventional therapy.
Clinical Effectiveness Review
Because a limited number of clinical trials have evaluated infliximab for treatment of CD, no attempt was made to pool literature quantitatively. Rather, a qualitative summary of the available clinical data was undertaken.
For the treatment of fistulizing CD, one controlled clinical trial has been reported and another is underway. Three infusions of infliximab (5 or 10 mg/kg) at Weeks 0, 2 and 6 were superior to placebo in achieving partial (62% vs. 26%, p=0.002) and complete (46% vs. 13%, p=0.001) closure of fistulas over 18 weeks. No significant dose response was observed, although numerically higher closure rates were seen with 5mg/kg.
For the treatment of active CD resistant to conventional medical therapy, only one acute treatment trial and its extension to maintenance therapy have been fully published. The results show that a single intravenous infusion of infliximab is superior to placebo in inducing clinical response (65% vs. 16%, p<0.001) and clinical remission (33% vs. 4%, p=0.005) at four weeks. The gains in response (41% vs. 12%, p=0.008) and remission (24% vs. 8%, p=0.31) were attenuated by 12 weeks. Again, no dose-response was observed, with numerically greater treatment effects at 5 mg/kg than 10 or 20 mg/kg. Among subjects who responded to a blinded infliximab infusion or an open-label re-infusion (10 mg/kg), re-infusions of 10 mg/kg at eight-week intervals yielded significantly higher rates of clinical remission (44% vs. 20%, p=0.013) and numerically higher rates of clinical response (62% vs. 37%, p=0.16) at Week 44. Preliminary results from a larger trial evaluating maintenance strategies for subjects who achieve clinical response two weeks after infliximab infusion also suggest that repeat infusions of infliximab (5 mg/kg or 10 mg/kg) every eight weeks are numerically superior to placebo in providing clinical response (55% vs. 27%) or remission (42% vs. 21%) at Week 30 (significance testing not reported). Full results of this trial are awaited.
No clinical subgroups in which infliximab consistently offers preferential benefit have been identified. In controlled clinical trials of CD, treatment with infliximab has been tolerated well, with mild and self-limited infusion reactions in three to seven percent of patients. Increased rates of acute respiratory infection were observed. With rare cases of reactivated tuberculosis reported in post-marketing surveillance, screening for tuberculosis is now recommended among candidates for infliximab treatment. Patients treated with infliximab have been noted to develop de novo autoimmune markers and human anti-chimeric antibodies (HACA), although their clinical significance remains uncertain. The long-term risks of infliximab, including malignancy and autoimmune disease, are currently unknown.
Economic Analysis and Review
Six previous economic analyses of infliximab and two observational studies of infliximab-associated resource utilization were identified. Four evaluations, submitted by industry, generated favourable results for treatment of fistulizing and active CD, and suggested it to be cost-saving. A cost-utility analysis of single-dose infliximab for the treatment of active CD estimated its incremental cost-utility ratio (ICUR) to range from US $14,200/QALY to US $40,000/QALY, but it has been published only in abstract form with limited information on methods and assumptions. The only analysis to be published in a peer-reviewed journal concluded that the ICUR of primary treatment with infliximab for fistulizing CD, relative to usual care, was US $355,450/QALY. Two pre-post observational studies noted lower resource utilization after infliximab infusion, than in a matched period before treatment.
The authors undertook a cost-utility analysis of infliximab for active CD resistant to conventional therapy. Its use for fistulizing CD was not evaluated. A Markov model was used to compare three infliximab treatment strategies to usual care, using transition data from a published natural history study of CD and estimates of treatment effects from clinical trials. From the perspective of a Canadian provincial ministry of health, no strategy was dominant in the base-case analysis. Usual care yielded the fewest QALY and incurred the lowest costs over one year. A single-infusion of infliximab was estimated to yield 0.01524 additional QALY for incremental direct medical costs of C $2,762 (ICUR C $181,201/QALY). Re-treatment of responders further improved outcomes (ICUR C $480,111/QALY) while adding maintenance therapy for responders provided the best outcome (ICUR C $696,078/QALY). In one-way sensitivity analyses, the results were sensitive to extreme reductions in the cost of infliximab and increases in the rate of medical admission for drug-refractory disease. In a probabilistic sensitivity analysis, usual care was the strategy most likely to be cost effective for QALY values less than approximately C $180,000.
This economic analysis is methodologically rigorous and relevant to the Canadian practice setting, and used advanced probabilistic sensitivity analysis to explore the impact of parameter uncertainty. In its base-case, the economic impact of infliximab exceeded what is generally considered good value for money. However, several limitations warrant recognition. First, explicit assumptions regarding natural history, resource utilization and drug dosing were required with reliance upon expert opinion. Second, the analysis used a limited time horizon in keeping with the limited available efficacy data, and did not address long-term costs, outcomes and adverse effects of the alternative strategies. Third, indirect costs and benefits were not assessed. Fourth, the economic impact of infliximab may continue to evolve, with changes in drug delivery, dose and cost. Finally, decision makers must recognize that infliximab offers a potential treatment to selected patients with refractory CD for whom few, if any, other alternatives are available.
Infliximab appears to be clinically effective for the treatment of fistulizing CD and active CD resistant to conventional therapy. While more information on the long-term consequences of infliximab therapy is needed, its short-term safety profile is acceptable. A cost-utility analysis of infliximab in treatment-resistant active CD suggests the incremental costs per additional quality-adjusted life year exceed traditional benchmarks for cost per QALY.