A Microsimulation Model to Determine the Cost-Effectiveness of Treat-to-Target Strategies for Crohn's Disease

INTRODUCTION:Cost-effectiveness of biomarker- vs endoscopy-based treat-to-target monitoring in Crohn's disease (CD) is unknown.METHODS:A microsimulation model for CD was built to simulate biomarker (fecal calprotectin) vs endoscopy-based monitoring in a treat-to-target fashion. Published literature in combination with patient-level data from phase 3 clinical trials and population estimates for therapeutic drug monitoring were used to generate transition probabilities, costs, and utilities. Tracker variables were used to modify downstream probabilities and outcomes based on previous exposures, response patterns, and disease-related complications or surgery history. The primary outcome was cost-effectiveness over a 5-year horizon at a willingness-to-pay threshold of $100,000/quality-adjusted life-year. Probabilistic sensitivity analyses in addition to multiple 1-, 2-, and 3-way microsimulation sensitivity analyses were performed.RESULTS:In the base-case model, the endoscopy-based monitoring strategy dominated the biomarker-based monitoring strategy over a 5-year horizon. Over shorter periods of observation, the biomarker-based monitoring strategy became progressively more cost-effective, with cost-effectiveness achieved for this strategy over a 1-year horizon. Therapeutic drug monitoring did not influence short-term cost-effectiveness of biomarker-based monitoring. Once in endoscopic remission, continued biomarker-based vs endoscopy-based monitoring was more cost-effective. A hybrid biomarker-endoscopy-based monitoring strategy dominated the endoscopy-based monitoring strategy over a 5-year horizon. The strongest determinants for cost-effectiveness were cost of colonoscopy and diagnostic performance of fecal calprotectin.DISCUSSION:The most cost-effective approach for treat-to-target monitoring in CD is up-front biomarker-based monitoring followed by endoscopy-based monitoring if not in endoscopic remission by 1 year and then returning to biomarker-based monitoring once in endoscopic remission.