|
?Systems Biology Approach to Identifying Biomarkers for Predict Sensitivity to Cancer Drugs
|
|
By Kendric, Section Biology
Posted on Fri Apr 30th, 2010 at 09:44:25 PM PST
|
 |
|
Background
Although heterogeneity between tumours of the same cancer type can affect its sensitivity to drug treatments, we still typically treat patients with the same series of drugs. Alternatively, we should aim to tailor specific treatments to individual patients based on the tumour's profile. For this goal, I propose using a systems biology approach integrating genomic, transcriptomic, and proteomic data to identify biomarkers that can predict a tumour's sensitivity to cancer drug.
|
Experiment
For the experiment, we will selecting a set of cell lines that is representative of the genomic heterogeneity that is observed in primary tumours of the cancer of interest. For each cell line, we will use aCGH to identify copy number aberrations in the genome, microarray to obtain RNA expressions, as well as measure proteomic concentrations. Then, we will treat each cell line with different drugs and measure cell line response to treatment quantified by GI50, which is the dose of treatment required for 50% relative growth inhibition. Using a statistical approach (i.e. non-parametric regression) or supervised learning approach (i.e. SVMs), we can correlate the GI50 values for cell lines with their pretreatment genomic, transcriptional, and proteomic profiles in order to identify genes associated with cellular response to each drug. In other words, genes associated with drug response or resistance when present at elevated levels are possible response predictors. Then, we can map a subset of the top ranked response predictors to interaction maps or signaling pathways to look for specific mechanisms related to drug sensitivity.
In this manner, we can perform large-scale screening for drugs used in treatment of specific cancers to see its effectiveness in other cancers.
References
Kuo, Wei-Lin et al. A systems analysis of the chemosensitivity of breast cancer cells to the polyamine analogue PG-11047 (2009). BMC Medicine 7:77. |
|
|
|
|
|
|
