Abstract 1539P
Background
Chronic inflammation is a known risk factor for pancreatic ductal adenocarcinoma (PDAC). Inflammation-related proteins are therefore of interest in the acquisition of novel biomarkers. The aim of this study was to explore a panel of serum proteins for the diagnosis of PDAC.
Methods
Serum samples from 701 patients with PDAC (Stage I n=22; Stage II n=250; Stage III n=110; and stage IV n=319) and 102 patients with benign pancreatic diseases (chronic pancreatitis n=26; intraductal papillary mucinous neoplasm n=38; and other n=38) were collected from the prospective BIOPAC study (NCT03311776). Furthermore, we included serum samples from 180 healthy blood donors. The samples were divided in discovery (n=645) and replication (n=338) cohorts. The samples were analyzed at BioXpedia using the Proximity Extension Assay from Olink (www.olink.com) for 92 proteins in the Immuno-oncology panel. Cancer antigen 19-9 (CA19-9) levels were included in the analyses as the 93rd protein. Logistic LASSO and Ridge regression models were used to explore the performance and stability of the complete set and to select serum proteins for the diagnostic signatures. Findings were tested in the replication cohort.
Results
19 signatures were identified. The best performing signature included 10 proteins: T-cell surface glycoprotein CD4, Cytotoxic and regulatory T-cell molecule, Fas Ligand, Interleukin-8, Interleukin-10, Monocyte chemotactic protein 3, Matrix metalloproteinase-7, TNF-related apoptosis-inducing ligand, Vascular endothelial growth factor C and CA19-9. This signature was able to discriminate PDAC patients from patients with benign pancreatic diseases and healthy individuals with a receiver operating characteristic area under the curve (AUC) of 0.92 (95% CI 0.89-0.92) in the discovery cohort and 0.92 (95% CI 0.87-0.92) in the replication cohort. When splitting the PDAC patients in two groups according to stage (I-II and III-IV) and comparing these to benign and healthy controls the protein signature gave AUCs of 0.90 (stage I-II) and 0.94 (stage III-IV) in the replication cohort.
Conclusions
We identified a 10-serum protein signature which could contribute to PDAC diagnosis. Further validation is needed.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Julia Sidenius Johansen.
Funding
Celgene; The Harboe Foundation; The Beckett Foundation; The Foundation of M. Kristian Kjær and wife Margrethe Kjær, born la Cour-Holmen; The Holm Memorial Fund.
Disclosure
All authors have declared no conflicts of interest.