Developing and validating continuous genomic signatures in randomized clinical trials for predictive medicine.

Area of Science:

• Genomic medicine
• Translational oncology
• Biostatistics

Background:

• Predicting patient response to anticancer therapies before treatment is challenging.
• Genomic signatures offer a promising approach for personalized treatment selection.
• Randomized trials are crucial for validating predictive biomarkers.

Purpose of the Study:

• To develop and validate a framework for creating genomic signatures within randomized clinical trials.
• To enable quantitative prediction of treatment effect heterogeneity.
• To facilitate the development of diagnostic tools for individual patient treatment.

Main Methods:

• Proposed a framework for co-developing predictive and prognostic genomic signatures.
• Applied the framework to gene-expression microarray data from a multiple myeloma randomized trial.
• Utilized cross-validation to develop patient-level survival curves.

Main Results:

• Identified that approximately 50% of multiple myeloma patients responded to thalidomide.
• Demonstrated a statistically significant improvement in survival for responsive patients.
• Generated cross-validated patient-level survival curves for predicting individual outcomes.

Conclusions:

• The framework advances predictive medicine by integrating genomic data into randomized trials.
• It provides a validated method for assessing treatment efficacy considering patient heterogeneity.
• Offers tools for personalized treatment selection based on predicted patient response.