Integrative Pharmacogenomics Analysis of Patient-Derived Xenografts

Area of Science:

• Oncology
• Bioinformatics
• Pharmacogenomics

Background:

• Precision medicine requires robust biomarkers for predicting drug response.
• Patient-derived tumor xenografts (PDX) are valuable preclinical models but lack adequate computational analysis tools for high-throughput data.
• Analyzing PDX molecular and pharmacologic profiles is crucial for advancing cancer therapy.

Purpose of the Study:

• To develop an open-source software package, Xenograft Visualization & Analysis (Xeva), for analyzing in vivo pharmacogenomic datasets from PDX models.
• To quantify gene expression and pathway activity variability across PDX passages.
• To identify reliable biomarkers predictive of drug response in cancer.

Main Methods:

• Developed Xeva, an open-source software package for in vivo pharmacogenomic data analysis.
• Quantified gene expression and pathway activity variability across PDX passages.
• Utilized the largest PDX pharmacogenomic dataset to identify drug response biomarkers.

Main Results:

• Identified few passage-specific gene and pathway alterations, indicating model stability for biomarker discovery.
• Discovered 87 pathways significantly associated with response to 51 drugs (FDR < 0.05).
• Found novel biomarkers based on gene expression, copy number aberrations, and mutations with high predictive power (concordance index > 0.60; FDR < 0.05).

Conclusions:

• Xeva offers a flexible platform for integrative analysis of preclinical in vivo pharmacogenomics data.
• The study successfully identified biomarkers predictive of drug response, advancing precision oncology.
• Xeva facilitates the discovery of consistent gene and pathway activity across PDX passages and validates drug response biomarkers.