CA19-9 capability as predictor of pancreatic cancer resectability in a Spanish cohort

Mol Biol Rep. 2020 Mar;47(3):1583-1588. doi: 10.1007/s11033-020-05245-5. Epub 2020 Jan 8.

Abstract

CA19-9 serum has been suggested as a marker of unresectability but different cut-off levels have been published. A cut-off of 500 U/ml is currently considered in an international consensus as biological criteria of borderline resectable pancreatic adenocarcinoma. To evaluate whether serum CA19-9 threshold of 500 U/ml could be adequate predictor of resectability in pancreatic adenocarcinoma. Multicenter, observational, prospective study performed in Spain including 203 patients diagnosed with pancreatic adenocarcinoma. 43 (21.2%) cases were resectable and 160 (78.8%) unresectable. Among the 176 preoperative CA19-9 available values, 98 (58.3%) were ≤ 500 U/ml and 73 (42.7%) > 500 U/ml. Resectability rate in those patients with CA19-9 ≤ 500 U/ml was 60% while it was found to be 18% when CA19-9 > 500 U/ml. Statistical model to predict resectability based on CA19-9 provide an AUC of 0.6618 (95% CI 0.53-0.83) when only CA19-9 values > 500 U/ml are studied. Serum levels of CA19-9 higher than 500 U/ml are indicative of unresectable disease, however reduced sensitivity and specificity lead to a limited clinical applicability for resectability.

Keywords: Biomarkers; CA19-9; Pancreatic cancer; Resectability; Sensitivity; Specificity.

Publication types

  • Multicenter Study
  • Observational Study

MeSH terms

  • Adenocarcinoma / blood*
  • Adenocarcinoma / diagnosis
  • Adenocarcinoma / surgery
  • Aged
  • Aged, 80 and over
  • Biomarkers, Tumor / blood*
  • CA-19-9 Antigen / blood*
  • Female
  • Humans
  • Male
  • Middle Aged
  • Pancreatic Neoplasms / blood*
  • Pancreatic Neoplasms / diagnosis
  • Pancreatic Neoplasms / surgery
  • Prognosis
  • Prospective Studies
  • ROC Curve
  • Spain

Substances

  • Biomarkers, Tumor
  • CA-19-9 Antigen