Diagnostic model for pancreatic cancer using a multi-biomarker panel

Area of Science:

• Biochemistry
• Oncology
• Medical Diagnostics

Background:

• Pancreatic ductal adenocarcinoma (PDAC) has a poor prognosis, necessitating early detection biomarkers.
• Current clinically feasible biomarkers for PDAC are limited.
• Previous research identified leucine-rich alpha-2-glycoprotein (LRG1), transthyretin (TTR), and CA 19-9 as potential PDAC biomarkers.

Purpose of the Study:

• To develop an automated multi-marker ELISA kit combining LRG1, TTR, and CA 19-9 for PDAC diagnosis.
• To propose a diagnostic model based on this kit for clinical application.
• To evaluate the diagnostic performance of the developed triple-biomarker assay.

Main Methods:

• An automated ELISA kit was created by combining individual panels of LRG1, TTR, and CA 19-9.
• The kit was tested on 728 plasma samples (381 PDAC, 347 normal).
• A logistic regression model was developed to predict PDAC risk (high, intermediate, low).

Main Results:

• The automated triple-marker ELISA panel showed high consistency with individual marker assays (Pearson correlation coefficient = 0.865).
• The diagnostic model achieved high performance metrics: positive predictive value (92.05%), negative predictive value (90.69%), specificity (90.69%), and sensitivity (92.05%).
• All performance metrics simultaneously exceeded the 90% cutoff value.

Conclusions:

• The diagnostic model utilizing the triple ELISA kit demonstrates superior diagnostic performance for PDAC compared to previous markers.
• This automated assay offers a promising tool for PDAC early detection and risk stratification.
• External validation is required for future clinical implementation.