Pancreatic Cancer Detection Consortium Biomarker Bakeoff: A Phase II Blinded Biomarker Validation and Panel Discovery...

Area of Science:

• Biomarker discovery and validation
• Pancreatic cancer diagnostics
• Early detection research

Background:

• Pancreatic ductal adenocarcinoma (PDAC) detection remains challenging.
• Biomarker panels are being investigated to improve early detection rates.
• Cancer antigen 19-9 (CA19-9) is a commonly used but limited biomarker for PDAC.

Purpose of the Study:

• To evaluate blood-based biomarker panels for pancreatic cancer detection.
• To compare panel performance against CA19-9 alone.
• To explore novel biomarker combinations for improved diagnostic accuracy.

Main Methods:

• A phase 2 biomarker bakeoff study involving ten biomarkers in eight panels and CA19-9.
• Evaluation using plasma, serum, and germline DNA from 140 PDAC cases and 140 controls.
• LASSO regression used for biomarker combination exploration; Area Under the ROC Curve (AUC) was the primary metric.

Main Results:

• Biomarker panel AUCs ranged from 89.9% to 96.3%.
• The CA19-9/FUT2/3 panel (AUC=96.3%) and TFPI/TNC-FNIII-C panel (AUC=95%) significantly outperformed serum CA19-9 alone (AUC=91.7%).
• Exploratory models combining biomarkers achieved an optimism-corrected AUC of 96.4%.

Conclusions:

• The CA19-9/FUT2/3 panel demonstrated superior performance in this biomarker bakeoff.
• This panel warrants evaluation in larger, prospective studies for pancreatic cancer early detection.
• Biomarker bakeoffs are effective for comparing diagnostic candidates, with PCDC well-positioned for future studies.