A general approach to detect gene (G)-environment (E) additive interaction leveraging G-E independence in...

Area of Science:

• Statistical genetics
• Epidemiology
• Biostatistics

Background:

• Testing for additive interactions between genetic (G) and environmental (E) risk factors is crucial in statistical genetics.
• Traditional tests for no additive G×E interaction, like the relative excess risk due to interaction (RERI), have limitations.
• Existing methods, such as the RERI-LRT, rely on strict model assumptions that can be problematic with non-categorical exposures or covariates.

Purpose of the Study:

• To develop a general statistical approach for testing additive G×E interaction that exploits G-E independence.
• To overcome limitations of existing methods, particularly when dealing with non-categorical exposures or auxiliary covariates.
• To allow for unrestricted regression models for the binary outcome while enabling covariate adjustment.

Main Methods:

• A novel statistical approach is presented to test for additive G×E interaction under the G-E independence assumption.
• The method allows for flexible, unrestricted regression models for the binary outcome.
• Covariate adjustment is incorporated to ensure G-E independence and address potential confounding.

Main Results:

• The proposed method offers a more general framework for testing additive G×E interaction compared to the RERI-LRT.
• It accommodates non-categorical exposures and auxiliary covariates without a priori ruling out the null hypothesis.
• Extensive simulation studies and an ovarian cancer study demonstrate the method's utility.

Conclusions:

• The developed approach provides a robust and flexible tool for investigating additive genetic-environmental interactions in epidemiological studies.
• It enhances the ability to accurately test for G×E interactions, especially in complex scenarios with covariates.
• This method has significant implications for understanding disease etiology and risk factor contributions.