Pancreatic cancer (PC) is the third leading cause of cancer death because of its unobvious clinical symptoms in the early stage, making the urgency to discover particular metabolites as potential diagnostic markers for early diagnosis of PC. The study aims to identify differential metabolites distinguishing PC, benign diseases (BD) and healthy individuals (normal control, NC) using liquid chromatography-mass spectrometry (LC-MS).We combined both reversed-phase (C18) and hydrophilic interaction (HILIC) liquid chromatography separation and a high-resolution quadrupole time-of-flight mass spectrometer detection together which was operated in negative (ESI–) and positive (ESI+) electrospray ionization mode separately, in order to comprehensively analyze serum metabolites obtained from 20 patients with PC, 10 patients with BD and 10 NC. Differential metabolites were selected by univariate (Student’s t test) and multivariate (orthogonal partial least squares-discriminant analysis (OPLS-DA)) statistics. Bile acids, amino acids and choline-containing glycerophospholipids were identified as differential metabolites for discriminating PC, BD and NC. These altered metabolites corresponding to 3 most important pathways, including primary bile acid biosynthesis, biosynthesis of amino acids, and glycerophospholipid metabolism. The newly found metabolites by using LC-MS-based metabolomics may be potential biomarkers for the early diagnosis of PC.