Improved Pancreatic Adenocarcinoma Diagnosis in Jaundiced and Non-Jaundiced Pancreatic Adenocarcinoma Patients through the Combination of Routine Clinical Markers Associated to Pancreatic Adenocarcinoma Pathophysiology

BACKGROUND: There is still no reliable biomarker for the diagnosis of pancreatic adenocarcinoma. Carbohydrate antigen 19-9 (CA 19-9) is a tumor marker only recommended for pancreatic adenocarcinoma follow-up. One of the clinical problems lies in distinguishing between this cancer and other benign pancreatic diseases such as chronic pancreatitis. In this study we will assess the value of panels of serum molecules related to pancreatic cancer physiopathology to determine whether alone or in combination could help to discriminate between these two pathologies. METHODS: CA 19-9, carcinoembryonic antigen (CEA), C-reactive protein, albumin, insulin growth factor-1 (IGF-1) and IGF binding protein-3 were measured using routine clinical analyzers in a cohort of 47 pancreatic adenocarcinoma, 20 chronic pancreatitis and 15 healthy controls. RESULTS: The combination of CA 19-9, IGF-1 and albumin resulted in a combined area under the curve (AUC) of 0.959 with 93.6% sensitivity and 95% specificity, much higher than CA 19-9 alone. An algorithm was defined to classify the patients as chronic pancreatitis or pancreatic cancer with the above specificity and sensitivity. In an independent validation group of 20 pancreatic adenocarcinoma and 13 chronic pancreatitis patients, the combination of the four molecules classified correctly all pancreatic adenocarcinoma and 12 out of 13 chronic pancreatitis patients. CONCLUSIONS: Although this panel of markers should be validated in larger cohorts, the high sensitivity and specificity values and the convenience to measure these parameters in clinical laboratories shows great promise for improving pancreatic adenocarcinoma diagnosis.