Systematic Review of Prognostic Models Compared to the Mayo Risk Score for Primary Sclerosing Cholangitis

Background: Primary sclerosing cholangitis (PSC) is a cholestatic liver disease with a variable clinical course that can ultimately lead to end-stage liver disease, cholangiocarcinoma, and the need for liver transplantation. Several prognostic models have been developed to predict clinical outcomes and have been compared to the revised Mayo Risk Score (rMRS). Aim: To conduct a systematic review comparing the rMRS to other non-invasive prognostic tests for PSC. Methods: A systematic review of studies from 2000 to 2020 was performed that compared non-invasive biochemical prognostic models to the rMRS in predicting outcomes in patients with PSC. Results: Thirty-seven studies were identified, of which five studies that collectively included 3230 patients were reviewed. Outcomes included transplant-free survival or composite clinical outcomes. The rMRS was better than the Amsterdam–Oxford model for predicting 1-year transplant-free survival, c-statistics 0.75 and 0.70, respectively. The UK-PSC score outperformed the rMRS for 10-year transplant-free survival, c-statistics 0.85 and 0.69, respectively. An enhanced liver fibrosis score was independently associated with transplant-free survival after adjusting for rMRS. PREsTo predicts 5-year hepatic decompensation with a c-statistic modestly higher than rMRS; 0.90 and 0.85, respectively. Conclusion: Newer prognostic models, including the UK-PSC score and PREsTo, are more accurate at predicting clinical endpoints in PSC compared to the rMRS. Time frames and clinical endpoints are not standard among studies.