Background: Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies due to late diagnosis at an advanced stage. Early detection of PDAC remains a critical challenge, underscoring the need for novel biomarkers. In this study, we ...
Background: Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies due to late diagnosis at an advanced stage. Early detection of PDAC remains a critical challenge, underscoring the need for novel biomarkers. In this study, we aimed to identify novel mRNA biomarkers for diagnosing PDAC, focusing on early-stage tumorigenesis and associated immunological changes.
Methods: Buffy coat samples and clinical information were obtained from the Seoul National University Bundang Hospital between 2015 and 2021. Candidate mRNA biomarkers were identified through literature review, and their relative expression levels in buffy coat samples were measured using reverse transcription quantitative polymerase chain reaction. Machine learning-based feature selection was used to construct the final biomarker panel, which was tested using an independent dataset for diagnostic performance.
Results: In total, 1,504 individuals (417 PDAC patients and 1,087 non-disease controls) were eligible for the study. Among the 19 candidate biomarkers identified, 15 mRNAs (ARG1, CCL2, CLEC7A, CXCL8, CXCR2, CXCR4, FOXP3, IFNA1, IFNL1, PTGES, PTGES2, PTGS2, SLC27A2, TNF, and VEGFA) were selected based on their diagnostic performance in distinguishing resectable PDAC (RPC) from control groups. The final model, SVM-15, exhibited higher AUC of 0.976 in distinguishing RPC from control groups in the test set compared to CA19-9 alone (AUC = 0.910). The combination of SVM-15 and CA19-9 demonstrated an AUC of 0.965. For patients with normal CA19-9 levels, the AUC of the SVM-15 was 0.967, compared to 0.716 for CA19-9 alone and 0.896 for the combination.
Conclusion: Compared to CA19-9, the mRNA biomarker panel, SVM-15, improved diagnostic performance in patients with PDAC especially in patients with RPC and normal CA19-9 levels.