Nonlinear Modeling Reveals Novel Associations Between Genetically Predicted Protein Levels and Pancreatic Cancer Risk

Area of Science:

• Genetics
• Oncology
• Biomarker Discovery

Background:

• Pancreatic ductal adenocarcinoma (PDAC) is a deadly cancer with a significant public health impact.
• Understanding PDAC etiology is crucial for developing effective prevention and treatment strategies.
• Previous studies identified PDAC risk-associated proteins using linear models, but nonlinear associations remain underexplored.

Purpose of the Study:

• To investigate nonlinear associations between genetically predicted plasma protein concentrations and PDAC risk.
• To identify novel protein biomarkers for PDAC using advanced statistical methods.

Main Methods:

• Employed a nonlinear modeling approach combining two-stage sliced inverse regression (2SIR) and adjusted inverse regression (AIR).
• Integrated blood proteome and genome data from the INTERVAL study (n=3301) with a large PDAC genome-wide association study (8275 cases, 6723 controls).

Main Results:

• Identified 25 genetically predicted proteins associated with PDAC risk after multiple comparison correction.
• Confirmed 22 previously reported proteins and discovered three novel proteins: APOF, CCL15, and CHIT1.
• Found literature support for the potential roles of novel proteins in PDAC development.

Conclusions:

• Accounting for nonlinear relationships is vital for uncovering novel PDAC risk proteins.
• The identified novel proteins (APOF, CCL15, CHIT1) warrant further investigation for their role in PDAC pathogenesis.
• Findings may enhance understanding of PDAC and inform future therapeutic and risk assessment strategies.