Evaluating a Mendelian Risk Prediction Model That Aggregates Across Genes and Cancers.

Area of Science:

• Genetics
• Computational Biology
• Oncology

Background:

• Mendelian risk prediction models identify individuals at high risk for hereditary cancer susceptibility variants.
• Existing models like Fam3PRO are effective but face challenges with rare gene-cancer associations and obtaining detailed family histories.
• Pre-screening for broad hereditary cancer gene panels necessitates simplified yet accurate risk assessment tools.

Purpose of the Study:

• To develop and evaluate an aggregate Mendelian model for hereditary cancer risk prediction.
• To simplify risk assessment by aggregating information across multiple genes and cancers.
• To reduce the need for extensive patient family history data and robust parameter estimation for rare genetic factors.

Main Methods:

• Developed a novel Mendelian model that aggregates genetic and cancer information.
• Evaluated the aggregate model's performance using computational simulations.
• Applied the model to two independent clinical cohorts for validation.

Main Results:

• The aggregate Mendelian model demonstrated comparable results to individual gene-cancer models for assessing the risk of carrying any cancer susceptibility variant.
• The proposed model significantly simplifies model assumptions and user input requirements.
• Performance was validated across simulations and clinical datasets.

Conclusions:

• An aggregate Mendelian model offers a simplified and efficient approach to hereditary cancer risk prediction.
• This model is suitable for pre-screening germline testing for broad cancer gene panels.
• The approach reduces clinical burden and improves feasibility in real-world settings.