Candidate biomarker assessment for pharmacological response

Area of Science:

• * Genomics and Molecular Biology
• * Cancer Research
• * Bioinformatics

Background:

• * The NCI-60 human cancer cell line panel is a valuable resource for cancer drug discovery.
• * Understanding molecular events linked to drug response is crucial for developing targeted therapies.

Purpose of the Study:

• * To identify molecular events significantly correlated with pharmacological response across diverse gene-drug pairings.
• * To compare the predictive power of different molecular data types (DNA, RNA, protein) for drug response.

Main Methods:

• * Utilized CellMiner and CellMinerCDB web applications to analyze molecular data from the NCI-60 cell lines.
• * Investigated associations between 3978 molecular events (copy number, methylation, mutation, transcript, protein expression) and drug response.
• * Assessed correlations between various molecular parameters and pharmacological outcomes.

Main Results:

• * Identified 3978 molecular events significantly linked to drug response.
• * Phosphoprotein levels (31%) showed the highest correlation with drug response, followed by transcript levels (16%) and total protein levels (14%).
• * The type of molecular event significantly linked to response varied widely depending on the specific drug and gene target.

Conclusions:

• * Molecular profiling, particularly phosphoprotein and transcript levels, can effectively predict drug response in cancer cell lines.
• * CellMiner and CellMinerCDB are powerful tools for integrating multi-omics data to uncover biologically relevant molecular-pharmacological relationships.
• * Findings support the use of specific molecular biomarkers for guiding personalized cancer treatment strategies.