INTRODUCTION: A straightforward, reliable, and noninvasive method for predicting the development of liver-related events (LRE) in primary biliary cholangitis (PBC) has not been attained thus far. This study investigated whether serum autotaxin (ATX) could predict LRE in patients with PBC. METHODS: This retrospective multicenter investigation included 190 biopsy-proven untreated patients with PBC. All subjects were followed for at least 1 year, during which time the prevalence of LRE, including newly developing hepatocellular carcinoma, esophagogastric varices, ascites, and hepatic encephalopathy, was investigated in relation with ATX levels at the time of liver biopsy. RESULTS: During the median follow-up period of 9.7 years, LRE were observed in 22 patients (11.6%). The area under the receiver operating characteristic curve and cutoff value of serum ATX for predicting LRE were 0.80 and 1.086 mg/L, respectively. Patients with serum ATX ≥1.086 had a significantly higher cumulative incidence of LRE compared with patients with ATX < 1.086 (33.3% vs 3.6%, P < 0.00001). Notably, the predictive capability of ATX for LRE in patients with PBC surpassed that of FIB-4, ALBI, APRI, and Mac-2-binding protein glycan isomer. A multivariate Cox proportional hazards model revealed ATX as an independent associated factor for LRE (hazard ratio 6.24, 95% confidence interval 1.87–20.80, P = 0.003) along with Nakanuma stage (hazard ratio 2.75, 95% confidence interval 1.52–4.99, P < 0.001). These results were closely replicated in a serologically diagnosed PBC validation cohort (n = 32). DISCUSSION: Serum ATX levels may serve as a predictive marker for LRE in patients with PBC.
Background/Objectives: Adrenomedullin (AM) is a circulating peptide involved in tumor progression, angiogenesis, and fibrosis. However, its clinical utility as a biomarker in gastrointestinal cancers remains unclear. Methods: This single-center prospective observational study included patients with unresectable advanced gastrointestinal cancers (esophageal, gastric, colorectal, biliary tract, and pancreatic cancers) and healthy volunteers. Plasma AM levels were measured before the initiation of systemic chemotherapy in patients with cancer and at the time of enrollment in healthy volunteers. Results: In the unadjusted analyses, plasma mature and total AM levels were significantly higher in the cancer group than in healthy volunteers. Age was significantly associated with both mature and total AM levels, whereas sex did not show a significant association. After adjustment for age, between-group differences in mature and total AM levels were not statistically significant. Among patients with pancreatic cancer, plasma mature and total AM levels were significantly higher in stage IV than in stage III disease. Conclusions: Higher plasma AM levels were observed in the cancer group than in healthy controls in the unadjusted analyses; however, these differences were not statistically significant after adjustment for age. A stage-related difference was observed in pancreatic cancer. These findings should be considered hypothesis-generating and require validation in larger studies with appropriately matched control groups.