Background: Pancreatic ductal adenocarcinoma (PDAC) is a leading cause of death in the United States. Treatment is difficult and often complicated by the presence of cachexia, a systemic wasting condition affecting most PDAC patients at diagnosis. Skeletal muscle atrophy is a hallmark of cachexia and predicts outcomes after surgery and chemotherapy. In this study, we sought to identify possible serum biomarkers of cachexia, which correlates with changes to lean body mass and survival, using an aptamer-based platform in patients with PDAC. We hypothesize that unique proteins associate with cachexia and survival measures. Methods: Using SomaScan we measured the serum levels of 4,006 proteins in 21 patients with PDAC undergoing curative surgery. Clinical data and anthropometric measurements derived from pre-operative CT imaging were compared to protein levels with Spearman correlational analyses and logistic regression. We then queried DAVID Bioinformatics database to identify enriched functional protein categories. Results: Mean age was 66.9 ± 8.91 years, and mean body-weight loss in the six months preceding diagnosis was 12.6 ± 8.80%. Using consensus guidelines considering percent weight loss and skeletal muscle mass and radiation attenuation, 10 subjects (46.7%) were defined as “cachectic.” We found 241 proteins significantly correlated to Cancer Weight Loss Grade (a composite measure of percent body weight loss and body mass index). High IDUA, CTLA4 and USE1, in particular, predicted the worst weight loss grade with 75% sensitivity and 92% specificity (AUC = 0.8894, odds ratio: 36, p = 0.0068). Four additional proteins significantly correlated with skeletal muscle index (CK-MM, CK-MB, PCOC2 and ADH4) while 13 proteins significantly correlated with muscle radiation attenuation. Lastly, we identified 18 proteins that significantly correlated with survival quartile. Of these, we found that elevated HS6ST2 and DEPP significantly predicted early recurrence and death compared to those with lower levels (medial survival: 177 days v. 850 days, p = 0.0049). Of the correlative proteins, enriched gene ontology terms included signal peptide (28.2), cytokine (6.86), negative regulation of endopeptidase activity (3.39), and immunoglobulin domain (3.24). Conclusions: Cancer cachexia in PDAC and various gastrointestinal malignancies remains a key clinical issue. While there are no definitive biomarkers currently in use to diagnose and manage cachexia preemptively, we present several potential candidates in a small cohort of patients. Ultimately, such assays may better elucidate common cachexia-inducing pathways, a shared spectrum of biomarkers and allow for the development of more specific therapies targeting cachexia.