4 Consideration of the evidence
4.1 The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of infliximab, having considered evidence on the nature of acute exacerbations of severely active ulcerative colitis and the value placed on the benefits of infliximab by people with the condition, those who represent them, and clinical specialists. It was also mindful of the need to take account of the effective use of NHS resources.
4.2 The Committee understood that the symptoms of ulcerative colitis may fluctuate in severity and often the disease can remain inactive for some time. In patients who present with severe exacerbations, hospital admission is often necessary.
4.3 The Committee heard from the clinical specialists that if acute severe ulcerative colitis does not respond to treatment with intravenous corticosteroids within 72 hours, patients are very likely to require urgent colectomy. In these circumstances, infliximab or ciclosporin may be used in an attempt to avoid the need for surgery. The Committee understood that although these strategies may reduce the need for urgent colectomy, the likelihood of colectomy as an elective procedure may still remain in the long term. The Committee further understood from the patient experts that it is important psychologically for patients to delay colectomy in order to have time to consider and come to terms with the implications of major surgery and the possibility of the provision of a stoma.
4.4 The clinical specialists told the Committee that a single dose of infliximab is often used in clinical practice to treat acute severe ulcerative colitis in order to delay or avoid urgent colectomy after failure of intravenous corticosteroids. The Committee appreciated that this approach clearly differs from the regimen specified in the SPC for infliximab for ulcerative colitis, which involves at least three doses for induction.
4.5 The clinical specialists described to the Committee the current management of acute ulcerative colitis using ciclosporin. They expressed the view that there was widespread concern among clinicians treating acute ulcerative colitis about the adverse effects of ciclosporin in this indication, particularly the risk of serious infections and the associated risk of mortality. The clinical specialists also indicated that ciclosporin is not used routinely in all centres because of these concerns. However, the Committee noted that the evidence on the relative safety of infliximab in this setting was not adequately researched given the small sizes of the available studies. The Committee was mindful of the results of the indirect comparison presented in the manufacturer's submission. Specifically, the Committee discussed whether the estimated colectomy rates following the use of ciclosporin were in line with those observed in clinical practice. The Committee considered that the manufacturer's submission overestimated the clinical effectiveness of infliximab compared with ciclosporin based on the probability of colectomy because:
the clinical benefit of ciclosporin in the appropriate population (that is, those resistant to intravenous steroids) was underestimated in the manufacturer's submission, and
the effectiveness of placebo in the clinical trials may have been overestimated, because of possible confounding due to the use of ciclosporin as a 'rescue therapy' in patients whose disease failed to respond to placebo treatment.
The Committee noted the critical importance of the comparative colectomy rates for infliximab and ciclosporin in the cost-effectiveness analysis. Although noting the concerns of the clinical specialists, the Committee concluded that ciclosporin was clinically effective and an appropriate comparator for infliximab in this setting. The Committee also considered the views of patients about the potential adverse effects associated with drug treatments for acute severe ulcerative colitis, and their need to receive full information about these options.
4.6 The clinical specialists advised the Committee that, of those people with acute severe ulcerative colitis that was refractory to intravenous corticosteroids, approximately half would, with medical management (infliximab or ciclosporin), avoid colectomy in the short term (that is, during the course of the initial hospital admission) and approximately one quarter would avoid it in the long term. The Committee also discussed how appropriate it was to include the comparison of ciclosporin with corticosteroids in newly treated acute ulcerative colitis in the manufacturer's submission in the synthesis of relative treatment effects; both the population (people in whom a course of corticosteroids had not yet been tried) and the comparator (intravenous corticosteroids) studied in this trial differed from those in the other clinical trials included in the synthesis. The Committee agreed with the ERG that the inclusion of the comparison of ciclosporin with corticosteroids was inappropriate. Therefore, the Committee accepted the ERG's corrected predicted probabilities of colectomy for patients treated with ciclosporin for the first 3 months as the best available estimate. Because of the lack of any direct comparison between infliximab and ciclosporin, the Committee considered that the comparative effectiveness of ciclosporin and infliximab in acute ulcerative colitis was subject to considerable uncertainty and that it could not estimate the true clinical effectiveness of infliximab relative to ciclosporin based on the evidence available. The Committee concluded that there was insufficient evidence to assume a clinical benefit of infliximab over ciclosporin and was persuaded by the clinical specialists that the colectomy rates with the two drugs were likely to be similar.
4.7 The Committee considered the economic analysis presented by the manufacturer and noted that the manufacturer's base-case analysis suggested that the ICER for infliximab relative to ciclosporin was £18,400 per QALY gained. However, the Committee also noted that when the ERG made corrections to the model and excluded the study they considered inappropriate from the analysis of the effectiveness of ciclosporin for the first 3 months, the ICER rose to £48,400 per QALY gained. The Committee observed that in this analysis, the estimate of the colectomy rate during months 4–12 following ciclosporin treatment had been taken from the study that had been excluded. The Committee was aware that there was no alternative estimate of this parameter and therefore considered the effect that varying this parameter had on the estimate of effectiveness. They noted sensitivity analyses presented by the manufacturer showing that when a high value of the colectomy rate was assumed (0.48), the ICER for infliximab versus ciclosporin was reduced (£9300 per QALY) and conversely when a low value was assumed (0.143) the ICER increased (£52,000 per QALY). The Committee considered that it was extremely unlikely that the colectomy rate during months 4–12 would be as high as 0.48, recalling their earlier discussions with the clinical experts that suggested that those patients who did not respond to ciclosporin would be likely to have a colectomy within the first few weeks following presentation. The Committee agreed that the lower estimates of colectomy rate during months 4–12 were more likely, and that the ERG reanalysis giving an ICER of £48,400 per QALY represented the best available reflection of the cost effectiveness of infliximab relative to ciclosporin.
4.8 The Committee appreciated that the cost effectiveness of infliximab would be sensitive to the number of doses given. It noted that the economic analysis presented by the manufacturer assumed that all patients received the full induction course of three doses even if they underwent urgent colectomy. The Committee believed that the average course would be fewer than three doses because those who underwent colectomy would not receive further infliximab, and some people would discontinue treatment for other reasons. The Committee noted that if the average number of doses of infliximab was reduced from three to two and a half, while keeping the initial colectomy rates used in the model constant, then the ICER for the comparison with ciclosporin would fall from over £48,000 to approximately £33,000 per QALY gained; if the number of doses was assumed to be two then the ICER for infliximab compared with ciclosporin was approximately £20,000 per QALY gained. However, the Committee considered that on the evidence available to it the actual average number of doses used in clinical practice was uncertain.
4.9 The Committee also noted that there was further uncertainty around the ICERs, in that the univariate sensitivity analyses for the comparison of infliximab with ciclosporin resulted in ICERs ranging from £1400 to £64,500. The former value occurred when all responders were assumed to continue in remission with no colectomy after the first year, and the latter when all patients were assumed to undergo colectomy within the first cycle of the economic model (0–3 months) after the first year. In addition, the Committee noted that not including adverse events and mortality in the manufacturer's economic analyses increased the uncertainty around the relative cost-effectiveness estimates. Because the available evidence did not show infliximab to be more clinically and cost effective than ciclosporin, the Committee felt unable to recommend it as a treatment option in people for whom ciclosporin is suitable. The Committee stated that further research is needed in order to establish a more accurate representation of the relative clinical and cost effectiveness of infliximab and ciclosporin. The Committee therefore concluded that using infliximab in research would reduce this uncertainty.
4.10 The Committee heard from the clinical specialists that some patients requiring treatment for acute ulcerative colitis would have other conditions that may mean that ciclosporin was contraindicated or inappropriate. The Committee noted that ciclosporin was not licensed for the treatment of ulcerative colitis and that the contraindications listed in the SPCs for ciclosporin related to its use in the other conditions for which it has been approved (in transplantation and in inflammatory conditions including skin conditions, rheumatoid arthritis and nephrotic syndrome). The Committee was aware that the contraindications to short-term use of ciclosporin in the treatment of acute ulcerative colitis may differ from those to the longer-term use of ciclosporin in conditions listed in the SPC. It was also aware that the decision about whether ciclosporin was contraindicated or inappropriate would have to be a matter for clinical judgement, based on a careful assessment of the risks and benefits of treatment in the individual patient. In addition, the Committee appreciated that the nature of the patient's acute presentation, their history of ulcerative colitis and their current use of immunomodulators may also have a bearing on the appropriateness of ciclosporin. The Committee was persuaded that there could be situations in which clinicians and patients judged that the potential risks from using ciclosporin in acute ulcerative colitis outweighed the likely clinical benefits. The Committee noted that in those situations, the ICER for infliximab relative to standard care from the manufacturer's submission was £11,600 per QALY gained and the ICER relative to immediate surgery was £13,400 per QALY gained. Even after the ERG corrections to the model, these ICERs were similar: £12,300 and £14,400 per QALY gained, respectively. The Committee concluded that infliximab would be a cost-effective use of NHS resources if ciclosporin is contraindicated or clinically inappropriate based on an assessment of the risks and benefits.