Lipid Aging Clocks as predictive and prognostic biomarker in cancer and inflammaging

Lipids are a heterogenous class of molecules involved in signaling, cell structure and energy storage. Lipid metabolism is dysregulated in aging and aging-related diseases such as cancer, metabolic disorders, and neurodegeneration. In this study, we developed a biological age predictor - a Lipid Aging Clock - based on human serum lipidome data of pancreatic ductal adenocarcinoma (PDAC) patients, that has a Pearson correlation coefficient of 0.81 to chronological age with a median absolute error of 4.5 years. This shows that it is possible to build aging clocks measuring aging from pathological cohorts. We find that LipidAgeAcceleration is increased in both PDAC and pancreatitis, indicating that these pancreatic conditions accelerate aging or that individuals with age acceleration or more likely to acquire them (or both). Furthermore, the lipid age clock is predictive of PDAC survival, where positive accelerated Lipid Age is associated with an 86% higher mortality risk. Among the lipid species associated with LipidAgeAcceleration, Ceramides, Sphingomyelins and Glycerophosphocholines, have statistically significant hazard ratios, and directly impact increased mortality. Pathway analysis of lipid species selected by the lipid clock further identifies age-dependent dysregulation of specific lipid pathways, including Sphingolipid, Glycerolipid, Glycerophospholipid metabolism, and steroid biosynthesis. Sphingolipid metabolism is significantly dysregulated in both aging and PDAC, connecting aging dynamics and cancer mortality. Moreover, sphingolipids are involved in inflammatory processes, and therefore the lipid aging clock could be, at least in part, reflecting inflammaging and is likely influenced by age-related alterations to the immune system. In summary, our work shows that lipid alterations are a robust biological age predictor with utility in cancer and aging research, as well as in predicting disease-associated outcomes.