Metabolic Profiling for Pancreatic Cancer Predicts Patient Outcomes

Metabolic Profiling for Pancreatic Cancer Predicts Patient Outcomes
Credit: Dr_Microbe/Getty Images

Pancreatic ductal adenocarcinoma (PDAC), is an aggressive cancer with a high mortality rate due to many factors including the high heterogeneity of tumors, late diagnosis, and high resistance to chemotherapies. To better understand the biology of the disease, a team of researchers from Centre de Recherche en Cancerologie de Marseille (CRCM), INSERM in France have developed a new method for capturing the metabolic signature of the disease to predict clinical outcomes.

“In this study, we establish another facet to characterize PDAC tumors at the metabolic level, as a promising area of investigation to offer new insights into the association between PDAC metabolic profiles and their transcriptomic profiles, as well as their resistance to standard chemotherapeutic drugs,” the authors write in their paper published in The Lancet.

Metabolomics profiling of 77 PDAC patient-derived tumor samples engrafted in mice (PDTX) were studied to investigate the relationship of metabolic profiles with overall survival in PDAC patients, tumor phenotypes and resistance to five anticancer drugs (gemcitabine, oxaliplatin, docetaxel, SN-38 and 5- Fluorouracil).

These PDTX samples revealed varying metabolic profiles after mass spectrometry coupled to ultra-performance liquid chromatography. A total of 502 were chosen for further analysis. The vast majority of these metabolites were from the lipid class suggesting that the heterogeneity in human PDAC tumors is comprised largely of the lipid class of metabolites. Glycerophospholipids were the most represented class with 45% metabolites followed by glycerolipids (17.1%), fatty acids (10%), sphingolipids (8.2%), amino acids (5.8%), nucleotides (2.8%) and sterols (2.2%). The other metabolites, representing 9% of all metabolites, included carbohydrates, such as monosaccharides and disaccharides, as well as alkylamines.

Next, the team created cell cultures of 35 of these PDTX samples with metabolomics data and exposed each to increasing doses of gemcitabine, oxaliplatin, docetaxel, the active metabolite of Irinotecan SN-38, and 5-fluorouracil (5-FU).

The team was able to identify a metabolic signature based on this metabolic profiling that was able to predict the clinical outcome of PDAC patients. They found that triacylglycerols, such as TG30 and TG56 along with several oxidized fatty acids, had a negative association suggesting that these metabolites could be associated with poor prognosis. However, cholesteryl esters (ChoE_11 and ChoE_8) and glycerophospholipids, especially lysophospholipids, were positively associated indicating a possible link with improved prognosis in patients.

However, glycerophospholipids seem to be the most altered metabolites between resistant and sensitive tumors. In particular, glycerophospholipid content was linked with multi-drug resistance in three out of five drugs, gemcitabine, oxaliplatin, and F-FU. They theorize that increased amounts of glycerophospholipids, which are made up of very long chains of fatty acid and are major components of the cell membrane, could limit drug diffusion into the cell.

Further, they showed that targeting this lipid profile improved chemosensitivity to standard anticancer drugs in PDAC models. Inhibiting glycerophospholipid synthesis, by acting on the GPAT1 enzyme, improved sensitivity to the three drugs in primary cultures, indicating it may be a promising therapeutic target to overcome challenges related to drug resistance in pancreatic cancer cells.