Introduction The objective of this analysis was to examine the clinical and economic impact of vedolizumab (VDZ) compared with conventional therapy (CT) in the treatment of moderately to severely active ulcerative colitis (UC) in the UK (UK).
Methods A Markov decision analytic model in Microsoft Excel was used to compare VDZ with CT (aminosalicylates, corticosteroids, immunomodulators) for the treatment of UC patients in the UK. We considered three populations: the overall intent-to-treat (ITT) population; anti-tumour necrosis factor (TNF)-naïve patients; and patients who previously failed anti-TNF therapy from the GEMINI 1 trial.
Population characteristics, efficacy data, and UC health-state utility data were obtained from the GEMINI 1 trial. Other inputs (e.g., unit costs, adverse event disutilities, probability of surgery, mortality) were obtained from published literature. Costs are presented in 2014 British pounds. Outcomes included quality-adjusted life-years (QALY), time spent in clinical response, and time spent in clinical remission. Time horizons included 10 year (base-case) and lifetime (scenario) horizons, with costs and outcomes discounted by 3.5% per year. Incremental cost-effectiveness ratios (ICER) were presented for VDZ compared with CT. Univariate and multivariate probabilistic sensitivity analyses were conducted to assess model robustness to parameter uncertainty, and a scenario analysis was explored using efficacy data from a network meta-analysis.
Results Over the base-case (10 year) time horizon, the model predicted that patients on VDZ accrued more QALY than patients on CT: 5.551 QALY vs 5.397 QALY in the ITT population (ICER=£33,297/QALY); 5.597 vs 5.403 QALY for anti-TNF-naïve patients (ICER=£24,657/QALY; network meta-analysis results: utilities 5.898 for VDZ vs 5.555 for CT and ICER=£4,862/QALY); and 5.463 vs 5.373 QALY for anti-TNF-failure patients (£64,999/QALY). Patients on VDZ spent more time in clinical response (0.99 years vs 0.27 years for the ITT population) and clinical remission (0.64 years vs 0.13 years) than patients on CT. Scenario analyses with a lifetime horizon showed VDZ to be even more cost-effective (ITT population ICER=£20,599/QALY). Sensitivity analyses suggest that results are most sensitive to treatment response and transition probabilities.
Conclusion Our model predicted that treatment with VDZ improves QALY, increases time in remission and response, and is a cost-effective treatment option for both anti-TNF naïve and anti-TNF failure patients with moderately to severely active UC compared with CT over 10 year and lifetime horizons.
Disclosure of Interest M. Wilson Conflict with: Employee of RTI Health Solutions, a company hired to conduct the study by Takeda Pharmaceuticals, M. Kerrigan Conflict with: Employee of PHMR Ltd, a company hired to conduct the study by Takeda Pharmaceuticals, M. Smyth Employee of: Takeda Development Centre Europe Ltd, London, UK, H. Chevrou-Severac Employee of: Takeda Pharmaceuticals GmbH, Zurich, Switzerland, A. Bergman Employee of: Takeda Pharmaceuticals GmbH, Zurich, Switzerland, R. Selby Employee of: Takeda UK Ltd., Bucks, UK