Abstract 6610: Metabolic patterns of pancreatic cancer cachexia: Cross-tissue lipid networks predict cachexia progression

Abstract Pancreatic ductal adenocarcinoma (PDAC) is among the most lethal of all cancer types. A key — yet poorly understood — facet in the disease state is cachexia, a multi-organ pathological state characterized by physical wasting and tissue catabolism. It occurs in 80% of PDAC patients, and to date there are no preventative or early detection methods. Cachexia leads to limited tolerance to anti-cancer therapy and contributes to disease lethality. Here, we present the first-of-its-kind systemic metabolomic analysis across cachectic stages to better understand disease progression. We used the well-established mutant KRASG12D (LSL/+) mutant p53 inducible mouse model of PDAC. Model physiology faithfully recaptures human cachexia: we observe progressive overall weight loss as well as loss of skeletal muscle and adipose tissue. As with human disease, weight loss occurs prior to loss of appetite in the animals, suggesting physiological changes independent of nutrition. Importantly, our model captures the period prior to weight loss (pre-CAC), as well as early-CAC (10% weight loss) and late-CAC (10% weight loss). We profiled the metabolomes of the pancreas, interstitial fluid, plasma, liver, 3 different adipose tissues, and 3 skeletal muscles, from male and female mice, control and tumor-bearing, across all 3 stages of cachexia. Each tissue has a unique metabolic trajectory across cachexia stages; pathway and correlation analyses showed a particular emphasis on lipid alterations in peripheral tissues. Strikingly, we find systemic metabolic changes prior to tissue wasting, including rewiring in the liver before the presence of metastases. This indicates systemic alterations as early as pre-CAC. We therefore used computational modeling to identify metabolites that may be participating in cross-tissue networks. A key finding, confirmed by 13C tracing, was the circulation of lactate and glucose and their uptake by the liver and pancreas. Network maps also suggest lipid rewiring in distal tissues, changes in abundances of these same lipid species in plasma, and their consequent alterations in the tumor and liver. A fundamental question remains: can we predict cachexia development before the manifestation of symptoms? We used feature selection algorithms based on statistical learning, training the models only on a subset of data. Specific sphingolipids and triglycerides were the key lipid species that distinguished control and pre-CAC. We find that the model — trained only on pre-CAC — is able to determine early-CAC with an 85% accuracy and late-CAC with a 90% accuracy. Thus, a limited set of metabolites that are altered in pre-CAC could have predictive value for future cachexia development. Overall, our work provides a resource for the field and advances our understanding of systemic metabolism in PDAC cachexia. We hope this will lay the foundation for cachexia prevention and treatment. Citation Format: Deepti Mathur, Blanca Majem, Courtney Beaulieu, Lucas Dailey, Sarah Jeanfavre, Joao Xavier, Clary Clish, Nada Kalaany. Metabolic patterns of pancreatic cancer cachexia: Cross-tissue lipid networks predict cachexia progression abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 6610.