Pathway polygenic risk scores (pPRS) for the analysis of gene-environment interaction

• Novel Method: The paper introduces Pathway Polygenic Risk Scores (pPRS), a targeted method for GxE analysis that restricts genetic variants to specific, biologically informed genomic pathways.
• Increased Power: Simulations and empirical analysis demonstrated that pPRS yields substantially greater statistical power to detect true GxE interactions compared to using a standard, overall PRS.
• Empirical Example: The method identified a significant interaction between pPRS based on the TGF-\(\beta\)/GRHR pathway and NSAID use for colorectal cancer (CRC) risk, suggesting the strongest protective effect of NSAIDs in those with high pathway-specific genetic risk.

PubMed: 40763299 DOI: 10.1371/journal.pgen.1011543 Overview generated by: Gemini 2.5 Flash, 28/11/2025

Key Findings: Enhancing Power for Gene-Environment Interaction (GxE)

The paper introduces pathway polygenic risk scores (pPRS) as a novel and more powerful method for detecting Polygenic Risk Score by Environment (PRS x E) interactions in complex human traits. The central finding is that standard PRS often include too many genetic variants that affect disease independently of the environment, which dilutes the true GxE signal and reduces statistical power. By focusing the PRS on biologically relevant pathways, pPRS substantially improves the ability to detect these crucial interactions.

Study Design and Motivation

The Problem with Standard PRS x E Analysis

A standard Polygenic Risk Score (PRS) is a comprehensive aggregate of many genetic variants. When testing for PRS x E interaction, the inclusion of a large number of variants that do not interact with the environmental factor (E) “waters down” the underlying signal from the few truly interacting variants. This leads to reduced statistical power to identify genuine GxE effects, potentially masking important biological insights and opportunities for targeted prevention.

Introducing Pathway Polygenic Risk Scores (pPRS)

The authors propose the use of pPRS scores, which are constructed by annotating subsets of genetic variants (SNPs) to specific genomic pathways using state-of-the-art annotation tools. This approach integrates existing biological knowledge to create a more targeted genetic risk measure, hypothesized to be more sensitive to GxE effects mediated through those pathways.

Methods and Results

Simulation Studies

Through extensive simulation studies, the researchers demonstrated that testing a targeted pPRS x E interaction yields substantially greater statistical power compared to testing the interaction using a broad, overall PRS.

Empirical Application: Colorectal Cancer (CRC) and NSAIDs

The pPRS method was applied to a large case-control study (N = 78,253) of colorectal cancer (CRC), using non-steroidal anti-inflammatory drugs (NSAIDs) as the environmental factor, a known protective exposure for CRC.

• Overall PRS Result: No evidence of an overall PRS x NSAIDs interaction was observed (\(p = 0.41\)).
• pPRS Result: A highly significant pPRS x NSAIDs interaction (\(p = 0.0003\)) was identified based on SNPs restricted to the TGF-\(\beta\)/gonadotropin releasing hormone receptor (GRHR) pathway.

Interpretation of Empirical Results

The interaction analysis showed that the protective effect of NSAIDs against CRC was significantly stronger among individuals with higher genetic risk captured by the TGF-\(\beta\)/GRHR pPRS. For example, the odds ratio (OR) for NSAIDs protecting against CRC was 0.84 for individuals at the 5th percentile of the pPRS (low risk) but significantly better at 0.70 for those at the 95th percentile (high risk).

Conclusions and Recommendations

The pPRS approach successfully addresses the power limitations of standard PRS in GxE analysis. From a biological perspective, this suggests that NSAIDs may act to reduce CRC risk specifically through genes within the identified pathways. From a population health perspective, the findings support a precision prevention approach, suggesting that focusing on genetic susceptibility within biologically informed pathways may be more sensitive for identifying individuals who would benefit most from specific prevention efforts. The authors recommend the use of pPRS to integrate biological insight and maximize statistical power in future GxE studies.