Development of the natural history component of an early economic model for primary sclerosing cholangitis

Primary sclerosing cholangitis (PSC) is a rare, chronic cholestatic disease that can progress to cirrhosis and liver failure. The natural history of PSC is variable as liver enzymes and liver symptoms fluctuate over time. Several drugs for PSC are under investigation, but there are currently no economic models to evaluate the cost-effectiveness and value of new treatments. The objective of this study was to develop an early economic model for PSC and validate the natural history component.

A lifetime horizon Markov cohort model was developed to track the progression of adults with PSC with or without inflammatory bowel disease. Based on relevant literature and clinical expert advice, fibrosis staging was used to model disease progression. Evidence on disease progression, mortality, PSC-related complications, and secondary cancers was identified by literature searches and validated by interviews with clinical and cost-effectiveness modelling experts. Model outcomes were overall survival and transplant-free survival years, and the proportions of patients receiving liver transplants, 2nd liver transplants after recurrent PSC (rPSC), and developing rPSC after liver transplantation during their lifetime. Cumulative incidence of secondary cancers and quality-adjusted life-years (QALYs) were also tracked.

Model outcomes are in line with estimates reported in literature recommended by clinical experts. Overall survival (95% uncertainty interval [UI]) was estimated to be 25.0 (23.2-26.3) years and transplant-free survival was estimated to be 22.0 (20.2-23.6) years. The estimated proportion (95% UI) of patients receiving first liver transplants was 14.5% (11.6-17.1%), while the proportion of patients developing rPSC and receiving 2nd liver transplants after rPSC was 24.2% (20.4-28.0%) and 21.6% (12.9-29.7%), respectively. The cumulative incidence (95% UI) of cholangiocarcinoma, colorectal cancer, and gallbladder cancer were estimated at 5.2% (2.1-10.0%), 3.6% (1.4-5.4%), and 3.3% (1.2-7.6%), respectively. Discounted lifetime QALYs per patient (95% UI) were estimated at 16.4 (15.6-17.1).

We have developed a model framework to simulate the progression of PSC with estimates of overall and transplant-free survival. This model, which calibrates well with existing estimates of disease progression, may be useful to evaluate the clinical and economic benefits of future treatments.