Metabolic biomarkers for the differential diagnosis of pancreatic ductal adenocarcinoma vs. chronic pancreatitis

Julia Mayerle,Volker Liebenberg,Christian Pilarsky,Beate Kamlage,Robert Grützmann,Bodo Schniewind,Sandra González-Maldonado,Holger Kalthoff,Ulrich F Weiss,Philipp Schatz,Regina C Reszka,Erik Peter,Jonas A Schreiber

doi:10.1186/2049-3002-2-s1-p60

Abstract

Background The incidence of chronic pancreatitis (CP) varies between 4 and 23/100.000 in different populations and a tenfold higher prevalence. Current diagnostic tests such as transabdominal ultrasound and CA 19-9 can distinguish between pancreatic cancer (PDAC) and chronic pancreatitis (CP) in only about two thirds of patients. CA19-9 has been reported to discriminate between pancreatic cancer patients and healthy controls with a sensitivity of 0.80 (95 % CI 0.787-0.83) and a specificity of 0.80 (95 % CI 0.78-0.82). Therefore more sensitive biomarkers for the early detection of pancreatic cancer would be urgently needed. Our aim was to identify a panel of plasma metabolite biomarkers for this diagnostic purpose.

Highlights

The incidence of chronic pancreatitis (CP) varies between 4 and 23/100.000 in different populations and a tenfold higher prevalence. Current diagnostic tests such as transabdominal ultrasound and CA 19-9 can distinguish between pancreatic cancer (PDAC) and chronic pancreatitis (CP) in only about two thirds of patients
A consistent multimarker 10-metabolite panel was identified by the Elastic Net algorithm providing an area under the curve (AUC) of up to 0.96 [95% CI 0.93-0.98]
With a fixed sensitivity of 88% and an estimated incidence of 1.95 for pancreatic cancer a test specificity of 93.8% [95% CI84-98%] for the training set and of 91.3% [95% CI 76-94%] for the validation set was reached with a negative predictive value (NPV) of 99.75% [95% CI 99.7-99.8%]

Summary

Introduction

The incidence of chronic pancreatitis (CP) varies between 4 and 23/100.000 in different populations and a tenfold higher prevalence. Current diagnostic tests such as transabdominal ultrasound and CA 19-9 can distinguish between pancreatic cancer (PDAC) and chronic pancreatitis (CP) in only about two thirds of patients. CA19-9 has been reported to discriminate between pancreatic cancer patients and healthy controls with a sensitivity of 0.80 (95 % CI 0.787-0.83) and a specificity of 0.80 (95 % CI 0.78-0.82). More sensitive biomarkers for the early detection of pancreatic cancer would be urgently needed. Our aim was to identify a panel of plasma metabolite biomarkers for this diagnostic purpose

Objectives

Methods

Results

Conclusion