Abstract Background: Pancreatic ductal adenocarcinoma (PDAC) is lethal in part due to high frequency and severity of cachexia. Cachexia comprises unintentional weight loss, adipose and muscle wasting, inflammation, and dysmetabolism. It reduces quality of life, treatment response, and survival. Validated biomarkers for PDAC cachexia are lacking. Methods: We queried clinical data (age, sex, weight loss, diabetes, cancer stage, BMI, tobacco and alcohol use) from 1599 patients in BIOPAC (NCT03311776), which includes all PDAC stages, and 121 in the PACTO trial (NCT02767557) for advanced/metastatic PDAC. Olink immuno-oncology proteins were measured in pre-treatment plasma from 572 patients in BIOPAC and 131 in PACTO. Cachexia was binarized, either by Fearon criteria (weight loss ≥5% from habitual for BMI≥20, weight loss ≥2% with BMI≤20) or as BMI-adjusted weight loss grade (WLG) ≥3 (0-4 scale). In PACTO, muscularity was assessed by CT-measured skeletal muscle Z-score. Logistic regression (cachexia, weight loss grade outcomes) or linear regression (skeletal muscle Z-score) multivariable models were constructed using the purposeful selection method. Assumptions were checked and multicollinearity was limited to moderate levels (condition index 30, variance inflation factor 1. 5). Results: In BIOPAC, cachexia by Fearon criteria associated to male sex OR 1. 46; 95% CI 1. 15–1. 84, diabetes 1. 46; 1. 11–1. 94, cancer stage 3 or 4 1. 44; 1. 10–1. 87, overweight 1. 62; 1. 25–2. 09, obesity 2. 65; 1. 87–3. 81, and smoking 1. 77; 1. 30–2. 44; plasma biomarkers were IL-4 0. 71; 0. 57–0. 88, TNFSF14 0. 77; 0. 62–0. 97 and MICA/MICB 1. 14; 1. 02–1. 27. The combined model retained IL-4 0. 73; 0. 58–0. 91, overweight 1. 94; 1. 24–3. 06, obesity 2. 79; 1. 55–5. 23, prior smoking 1. 9; 1. 19–3. 05, and smoking 2. 64; 1. 56–4. 54. In PACTO, cachexia by Fearon criteria associated to overweight 3. 78; 1. 42–10. 99 and obesity 17. 23; 3. 17–321. 93, and with biomarkers IL-12 0. 45; 0. 24–0. 79 and GZMH 1. 84; 1. 05–3. 44. The combined model retained overweight 4. 54; 1. 62–14. 05, obesity 22. 02; 3. 83–421. 96 and IL-12 0. 49; 0. 24–0. 89. For WLG≥3, BIOPAC revealed associations with diabetes 1. 59; 1. 25–2. 03, PDAC stage 3 or 4 OR 1. 37; 1. 07–1. 74, obesity 0. 55; 0. 41–0. 73, prior smoking 1. 27; 1. 00–1. 61, and smoking 2. 3; 1. 74–3. 07; plasma biomarkers were IL-12 0. 81; 0. 67–0. 98, PTN 1. 16; 1. 02–1. 32, and IFN-beta 0. 57; 0. 39–0. 81. The combined model revealed associations with smoking 2. 97; 1. 83–4. 88, IFN-beta 0. 54; 0. 35–0. 79, NOS3 0. 69; 0. 50–0. 96, and CXCL12 2. 00; 1. 13–3. 60. In PACTO, WLG≥3 associated only with plasma biomarkers ADGRG1 1. 74; 1. 21–2. 60, IL-12 0. 51; 0. 31–0. 78, and FGF2 0. 54; 0. 31–0. 93. Skeletal muscle Z-score associated with CXCL9 0. 20; 0. 09–0. 32, IL-4 0. 24; 0. 07–0. 41, CD4 -0. 46; -0. 79– -0. 13, and IL-8 0. 07; 0. 00–0. 14. Conclusions: Overweight, obesity, low IL-12, and low IL-4 associated with at least one cachexia outcome in each study. Our results highlight at-risk populations and suggest immune dysfunction and exhaustion in cachexia. Citation Format: Sidsel C. Lindgaard, Shauna K. Rakshe, Tetiana Korzun, Astrid Z. Johansen, Susann Theile, Inna M. Chen, Julia S. Johansen, Teresa A. Zimmers. Biomarkers of cachexia due to pancreatic ductal adenocarcinoma: Results from two Danish studies abstract. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research—Emerging Science Driving Transformative Solutions; Boston, MA; 2025 Sep 28-Oct 1; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2025;85 (18Suppl₃): Abstract nr A008.
Lindgaard et al. (Sun,) studied this question.