Abstract 5968: Prognostic value of cachexia consensus criteria in pancreatic cancer

Abstract Background: Pancreatic ductal adenocarcinoma (PDAC) carries a poor prognosis due in part to cachexia diagnosed in up to 87% of patients. Cancer-associated cachexia is a muscle wasting syndrome unresponsive to nutritional support and associated with reduced treatment efficacy and higher patient mortality. Defining the relationship between cachexia and patient prognosis remains an unmet need, particularly in settings outside of clinical trials. Here, we applied weight-based markers of cachexia to real-world electronic health record (EHR) data to assess the prognostic ability of current consensus criteria toward PDAC patient survival. Methods: We utilized EHR data from the Northwestern Medicine Enterprise Data Warehouse to design a PDAC-associated cachexia phenotype from weight-based markers of cachexia. To form our patient cohort, we identified patients by a PDAC diagnostic code billed by an oncology provider from records backdated to 2012. Of 1469 patients, 1099 patients had sufficient weight data for cachexia classification at the time of diagnosis, defined by a pathology note indicating the presence of PDAC. We used the classification method from Fearon et al. 2011 to stratify patients by no cachexia, pre-cachexia and cachexia before constructing Kaplan-Meier curves. Our model, validated by manual adjudication of 300 patient charts, had an F1 score of 0.961. Results: There was no significant difference in PDAC patient median overall survival (OS) comparing patients with no cachexia, pre-cachexia and cachexia, which was 19.04 months, 19.27 months and 16.96 months, respectively. However, there were differences in survival rate at one year, which were 60%, 67% and 59%, respectively (log-rank p=0.04). The pre-cachexia associated 1-year survival benefit was most pronounced in Stage IV patients. With pre-cachexia as the reference, the HRs for no cachexia and cachexia were 1.18 (0.97-1.44) and 1.23 (1.02-1.47), respectively. As compared to patients with no cachexia, those with pre-cachexia and cachexia experienced greater survival benefit with resection and chemotherapy. Discussion: Cachexia status as defined by the Fearon criteria demonstrated a limited prognostic ability, as patient OS did not vary significantly with stratification. However, pre-cachexia was associated with a statistically significant improved 1-year survival and enhanced response to treatment. The unexpected pre-cachexia associated protective benefit suggests cachexia may serve as a double-edged sword in cancer, an observation deserving of further investigation. Overall, our results indicate that while cachexia status should be considered during PDAC management, better measures of cachexia are needed. Prognostication may be improved in future models by combining weight-based measures with factors that inform patient care, such as cachexia biomarkers and patient functionality. Citation Format: April Bell, Noah Forrest, Steven Tran, Amber Willbanks, Vijeeth Guggilla, Akhil Chawla, Jodi Johnson, Ishan Roy, Theresa Walunas. Prognostic value of cachexia consensus criteria in pancreatic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5968.