Purpose <p>Bile acids (BAs) are important molecules in the carcinogenesis of pancreatic cancer (PC). The BA profile in pancreatic diseases and its role in PC remain unclear.</p> Methods <p>The concentrations of 63 BAs in serum were measured in a total of 421 participants from three medical centers using targeted metabolomics. The concentrations of seven common types of BAs in serum were determined and compared. The least absolute shrinkage and selection operator algorithm analysis was used to confirm the independent predictors and build a metabolite panel for prediction and discrimination.</p> Results <p>A panel of two BA metabolites DCA and ω-MCA showed high diagnostic performance for PC versus healthy controls with an AUC of 0.914 (95% CI 0.873–0.956) in the validation cohort and achieved a 94.20% positive rate in the validation cohort. This diagnostic panel correlated with advanced clinical stage (<i>p</i> = 0.006) and poorer overall survival (HR = 3.11, 95% CI 0.79–12.23, <i>p</i> = 0.006). Importantly, biliary obstruction altered total BA concentrations (2.4-fold increase, <i>p</i> &lt; 0.001) but did not change BA compositional proportions (<i>p</i> &gt; 0.05). Additionally, ω-MCA, GCA-3S, GCDCA-3S, UDCA, and secondary BAs may distinguish PC from pancreatic benign tumors. GLCA was able to differentiate PC from pancreatic premalignant lesions. TLCA, 3-oxoLCA, and NorCA can effectively separate PC from chronic pancreatitis.</p> Conclusion <p>BAs are potential diagnostic and prognostic biomarkers for PC, which may also be helpful in differentiating pancreatic diseases.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Prognostic and Diagnostic Value of Serum Bile Acid Profiles in Pancreatic Cancer: a Retrospective Multicenter Cohort Study

  • Chang Liu,
  • Henan Qin,
  • Zeming Wu,
  • Jiwei Liu,
  • Minjie Wang,
  • Jinpeng Yao,
  • Dong Shang,
  • Peiyuan Yin

摘要

Purpose

Bile acids (BAs) are important molecules in the carcinogenesis of pancreatic cancer (PC). The BA profile in pancreatic diseases and its role in PC remain unclear.

Methods

The concentrations of 63 BAs in serum were measured in a total of 421 participants from three medical centers using targeted metabolomics. The concentrations of seven common types of BAs in serum were determined and compared. The least absolute shrinkage and selection operator algorithm analysis was used to confirm the independent predictors and build a metabolite panel for prediction and discrimination.

Results

A panel of two BA metabolites DCA and ω-MCA showed high diagnostic performance for PC versus healthy controls with an AUC of 0.914 (95% CI 0.873–0.956) in the validation cohort and achieved a 94.20% positive rate in the validation cohort. This diagnostic panel correlated with advanced clinical stage (p = 0.006) and poorer overall survival (HR = 3.11, 95% CI 0.79–12.23, p = 0.006). Importantly, biliary obstruction altered total BA concentrations (2.4-fold increase, p < 0.001) but did not change BA compositional proportions (p > 0.05). Additionally, ω-MCA, GCA-3S, GCDCA-3S, UDCA, and secondary BAs may distinguish PC from pancreatic benign tumors. GLCA was able to differentiate PC from pancreatic premalignant lesions. TLCA, 3-oxoLCA, and NorCA can effectively separate PC from chronic pancreatitis.

Conclusion

BAs are potential diagnostic and prognostic biomarkers for PC, which may also be helpful in differentiating pancreatic diseases.