Genome-based prediction of common diseases: methodological considerations for future research

Area of Science:

• Genomics
• Public Health
• Clinical Medicine

Background:

• Translating genomic knowledge into clinical practice remains a significant challenge.
• Current genome-based predictions for common diseases like diabetes, heart disease, and cancer lack informativeness.
• Limited contribution of identified genetic susceptibility variants to multifactorial disease development.

Purpose of the Study:

• Critically appraise the usefulness of genome-based applications in clinical and public health.
• Propose a framework for evaluating the clinical validity and utility of genetic tests.
• Review studies on the predictive value of genetic profiling from a methodological standpoint.

Main Methods:

• Review of recent studies on genetic profiling predictive value.
• Methodological analysis focusing on study population, genetic profile construction, and prediction model validation.
• Assessment of clinical utility and calibration of prediction models.

Main Results:

• Discrepancy exists between limited predictive value of genetic profiles and their commercial availability.
• Identified susceptibility variants currently contribute marginally to disease development.
• Need for robust methodologies to evaluate genome-based applications.

Conclusions:

• A framework is necessary for designing and analyzing studies to evaluate genetic tests.
• Careful consideration of methodological issues is crucial for identifying useful genome-based applications.
• Implementation of validated genome-based applications in clinical and public health requires further research.