Clinical Application Of New Methods For Risk Stratification In Primary Biliary Cholangitis

Background:Primary biliary cholangitis(PBC,formerly known as primary biliary cirrhosis)is an autoimmune-mediated chronic cholestatic liver disease characterized by non-suppurative granulomatous cholangitis,ongoing bile-duct destruction,and portal fibrosis that progresses slowly to biliary cirrhosis and end-stage liver disease.The disease progression is quite heterogeneous in PBC,especially the distinct mode of portal hypertensive progression which could presented in the early stage without proof of cirrhosis.Ursodeoxycholic acid(UDCA)is the only FDA-approved first-line drug for PBC,while the biochemical responsiveness is variable,and strongly predicts the long-term outcome.Vitamin D,as a kind of fat-soluble vitamin,is at risk of malabsorption and deficiency in PBC due to the loss of functioning intrahepatic bile ducts and inadequate bile salt secretion.Increasing evidence reveals that vitamin D pathway is crucial for the regulation of autoimmunity,cholestasis,and fibrosis.Our previous study found that vitamin D level is associated with disease severity and response to UDCA in PBC,indicating that measuting vitamin D status is of important clinical significance and should be further investigated,especially as a tool of risk stratigication for responsiveness and prognosis after UDCA treatment.Liver stiffness measurement(LSM)using transient elastography could assess the severity of portal hypertension indirectly.But the application for portal hypertension risk stratification in PBC,which characterized by distinct portal hypertensive progression type,has not yet been reported.Aims:1.To assess the predictive value of vitamin D level for biochemical responses and prognostic risk in patients with PBC after UDCA treatment,and investigate the related factors affecting vitamin D level to optimize the application of serum vitamin D level measurement.2.To evaluate the value of LSM in early warning and screening of esophageal varices in PBC,and explore the optimization effects by combination of other indexes such as platelet.Methods:1.A retrospective cohort study was conducted with 217 PBC patients from our center who have been treated by UDCA and consecutively followed-up for more than 1 year.Baseline and short-term follow-up data of vitamin D levels were collected.DNA samples were extracted to detect the SNP sites associated with vitamin D metabolism.Receiver operating characteristic(ROC)curves were generated for baseline and follow-up vitamin D level respectively.The area under the ROC curve(AUROC)and performance characteristic were compared to each other.2.PBC patients who had data of liver stiffness measurement and underwent endoscopic gastroscopy before or after half a year were retrospectively reviewed.AUROC and performance characteristics of LSM were compared to those proposed previously to identify patients with suspected esophagogastric varices.Results：(1-1)The AUROC from baseline vitamin D levels for the prediction of UDCA responsiveness after 12 months was 0.69.The spearman correlation index of baseline vitamin D levels and GLOBE scores after 12 months treatment was 0.215.(1-2)Serum vitamin D level was sharply increased from 16.5ng/ml at baseline to 20.9ng/ml after vitamin D supplementation for 1 month,and sustained within 21-22ng/ng to the third month.(1-3)The AUROC from highest follow-up vitamin D levels for the prediction of UDCA responsiveness after 12 months was 0.76,significantly greater than baseline vitamin D levels.The spearman correlation index of follow-up vitamin D levels and GLOBE scores after 12 months treatment was 0.39.The classical cut-off value of vitamin D deficiency was also an optimal cut-off value for the prediction of UDCA responsiveness,with a specialty of 78.6 and NPV of 62%.(1-4)Two SNPs of Vitamin D Binding Protein(VDBP,GC)are risk alleles of persistent vitamin D deficiency in 3monthes.OR for rs7041-A is 1.4 and rs2282679-G is 1.6.The UDCA responsiveness in patients with persistent vitamin D deficiency and low-risk genotype(TT-CC/TT-CA/TT-AA/GT-CA)is 32.4%.While 91.3% patients wirh adequate vitamin D level and high-risk genotype(GG-CA/GG-AA/GT-AA),and the response rate was much higher than the patients with adequate vitamin D level and low-risk genotype.(2-1)LSMs were higher in patients with esophagogastric varices,and were closely correlate to previously proposed risk scores of esophagogastric varices in PBC,including NVP scores,NVP-S scores,Mayo scores and plateles.The AUROCs from simple LSMs for the prediction of any esophagogastric varices and high-risk esophagogastric varices were 0.889 and 0.899 respectively,similar to NVP scores and NVP-S scores while significantly greater than Mayo risk scores.And LSMs retained high discriminating power and performance characteristics of LSM were compared to those proposed previously to identify patients with suspected esophagogastric varices.Results:(AUROC of 0.90)in patients with early PBC.(2-2)47.7% patients met PLT>150x109/L & LSM<20k Pa(Baveno VI criteria),and recognized as low risk ones who could avoid an endoscopy examination,and 1.2% of them diagnosed with high-risk varices by endoscopy.(2-3)The percent of patients who could avoid an endoscopy examination increased to 64.5% when expanding the above criteria to PLT>110x109/L & LSM<25k Pa,with 1.8% of them had high-risk varices.(2-4)The AUROCs from LSPSs which calculated by LSMs for the prediction of high-risk esophagogastric varices was 0.879.64.9% could avoid an endoscopy examination if cut-off applied as LSPS>3.5.The percentage was greater than the PLT>110x109/L & LSM<25k Pa(47.4%)in the included patients(n=97).3.2% of the patients who could avoid endoscopy were missed high-risk varices.The AUROC from LSPSs(0.916)was significantly greater than simple LSMs for the prediction of fibrosis.Conclusions:Serum vitamin D level could be used as a risk stratification marker in PBC patients for UDCA responsiveness prediction,and the clinical significance of vitamin D level during 3 monthes follow-up is higher than baselinbe vitamin D level.Follow-up vitamin D level was influenced by SNPs of VDBP,and combination of particular genotype could improve the discrimination power of vitamin D level in UDCA responsiveness prediction.LSM was an ideal risk stratification marker of esophagogastric varices in PBC.Patients met PLT>110x109/L & LSM<25k Pa could avoid endoscopy with very low possibility of missing high-risk varices.LSPS,which calculating based on LSM,may had greater power for early warning and screening of esophageal varices in PBC,and warranted further invesgation in larger cohort.