Systematic review: non-invasive prognostic tests for primary sclerosing cholangitis

Background Several prognostic tests for primary sclerosing cholangitis (PSC) have been developed, including biochemical models, elastography and magnetic resonance imaging scores. Aim To conduct a systematic review of non-invasive prognostic tests for PSC. Methods A systematic review was conducted from 1987 to 2020 of blood tests, biochemical models, elastography and imaging scores associated with outcomes in PSC. Results Forty studies of prognostic tests that collectively included 16 094 subjects with PSC were reviewed, of which 26 studies of non-invasive tests including 13 759 subjects with PSC were included. Normalisation or reduction of alkaline phosphatase with or without therapy was associated with transplant-free survival and reduced risk of hepatobiliary cancers but cut-off values for alkaline phosphatase were not consistent among studies. The most studied prognostic biochemical model was the Mayo Risk Score (MRS) evaluated in 18 studies with a c-statistic from 0.63 to 0.85 for clinical outcomes. One study demonstrated that the UK-PSC score outperforms MRS for predicting clinical outcomes with a c-statistics of 0.81and 0.75 respectively. A transient elastography score greater than 11.1 kPa is associated with survival and liver-related complications. The Anali score, derived from specific MRI and MRCP features, is associated with the development of cholangiocarcinoma and decompensated cirrhosis. Promising prognostic models include the enhanced liver fibrosis (ELF) score, ELF test and PREsTo scores. Conclusion MRS is the most studied prognostic score for clinical outcomes in PSC but the UK-PSC score and PREsTo have better test performance. Further studies comparing MRS to UK-PSC score, PREsTo or ELF with elastography or imaging-based scores are warranted.

Automated summary

This systematic review analyzed 40 studies on prognostic tests for PSC, comprising a total of 16,094 subjects. Non-invasive tests were associated with transplant-free survival and reduced risk of hepatobiliary cancers, but the consistency of cut-off values for alkaline phosphatase varied among studies. The Mayo Risk Score (MRS) is the most studied prognostic model, although the UK-PSC score and PREsTo showed better test performance. Further studies are needed to compare these models.