Plasma Metabolome Profiling Identifies Metabolic Subtypes of Pancreatic Ductal Adenocarcinoma
Ujjwal Mukund Mahajan, Ahmed Alnatsha, Qi Li, Bettina Oehrle, Frank Ulrich Weiß, Matthias Sendler, Marius Distler, Waldemar Uhl, Tim Fahlbusch, Elisabetta Goni, Georg Beyer, Ansgar M. Chromik, Markus Bahra, Fritz Klein, Christian Pilarsky, Robert Grützmann, Markus M. Lerch, Kirsten Lauber, Nicole Christiansen, Beate Kamlage, Ivonne Regel, Julia Mayerle
Abstract
Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest cancers. Developing biomarkers for early detection and chemotherapeutic response prediction is crucial to improve the dismal prognosis of PDAC patients. However, molecular cancer signatures based on transcriptome analysis do not reflect intratumoral heterogeneity. To explore a more accurate stratification of PDAC phenotypes in an easily accessible matrix, plasma metabolome analysis using MxP® Global Profiling and MxP® Lipidomics was performed in 361 PDAC patients. We identified three metabolic PDAC subtypes associated with distinct complex lipid patterns. Subtype 1 was associated with reduced ceramide levels and a strong enrichment of triacylglycerols. Subtype 2 demonstrated increased abundance of ceramides, sphingomyelin and other complex sphingolipids, whereas subtype 3 showed decreased levels of sphingolipid metabolites in plasma. Pathway enrichment analysis revealed that sphingolipid-related pathways differ most among subtypes. Weighted correlation network analysis (WGCNA) implied PDAC subtypes differed in their metabolic programs. Interestingly, a reduced expression among related pathway genes in tumor tissue was associated with the lowest survival rate. However, our metabolic PDAC subtypes did not show any correlation to the described molecular PDAC subtypes. Our findings pave the way for further studies investigating sphingolipids metabolisms in PDAC.