In This Section

A Comprehensive Cancer Center Designated by the National Cancer Institute

Breast Cancer SPORE Project 4

Project 4:  Pharmacogenomics of Breast Cancer Therapies

Co-Leaders:
M. Eileen Dolan, PhD
(basic science)
Peter O’Donnell, MD (clinical)
Gini Fleming, MD
(clinical)


SPECIFIC AIMS

Chemotherapy remains a mainstay of treatment for breast cancer, particularly for women with triple-negative breast cancer. However, it is generally viewed as a toxic treatment modality. We believe therapeutic decisions in both the adjuvant and palliative scenario could be greatly improved by incorporating patient-specific knowledge regarding toxicity risk, part of the ideal process of delivering precision medicine. The contribution of genetic variation to toxicities associated with chemotherapy has generally focused on a single candidate gene or pathway that ignores the often multigenic and complex nature of drug effects. During the past funding cycle, we led the field by taking on the challenge of developing cell-based models with known genetic information and gene expression, evaluating sensitivity to drugs to identify “predictive genetic markers” of chemotherapy induced toxicity. We used Genome Wide Association Studies (GWAS) for two breast cancer drugs, capecitabine and carboplatin, and have extended our work to paclitaxel.

We translated our fundamental basic research using these models into the design of a hypothesis-based clinical trial within the Translational Breast Cancer Research Consortium (TBCRC) using capecitabine. In the next funding period, we plan to further characterize the function of single nucleotide polymorphisms (SNPs) identified in preclinical and clinical studies governing toxicity to paclitaxel and capecitabine. Importantly, we will extend these pharmacogenomic discovery studies to include analysis of the now completed capecitabine TBCRC clinical trial, and additional clinical trials evaluating paclitaxel (CALGB studies 40101 and 40502).

A culminating, important new aim of our work in the next funding period will be to routinely test for well-validated genetic variants and develop polygenic risk scores associated with these breast cancer agents in patients being considered for breast cancer therapy in order to study the effect of offering preemptive pharmacogenomic testing on chemotherapy decision-making, including choice of agent to be administered and choice of starting dosage. We hypothesize that germline genetic variation accounts for a significant proportion of the inter-individual variability in toxicity of chemotherapy agents used in breast cancer therapy, and that provision of information regarding genetic risks for toxicity to clinicians will alter chemotherapy treatment choices for patients with this disease.

Our long-term goal is to provide a framework for discovery, validation, and implementation of predictive genetic markers/models for breast cancer treatment that can be used in routine clinical practice, particularly among those at high genetic risk for toxicity, thereby decreasing the suffering of patients undergoing treatment for breast cancer. Specifically, we plan to:

1. Identify a set of SNPs associated with capecitabine and paclitaxel toxicity in preclinical models, including characterization of these variants using pre-clinical functional studies. We have evaluated cellular sensitivity to capecitabine and paclitaxel in LCLs and will identify an optimal set of commonly (and rarely)-occurring SNPs associated with drug sensitivity for evaluation in the available clinical trials in specific Aim 2. We will utilize iCell neurons to functionally validate prioritized genes/genetic variants that will be identified in LCLs or the clinical trials described in aim 2.

2. To use five large sets of clinical samples from breast cancer patients treated with paclitaxel or capecitabine in order a) to validate LCL toxicity associated SNPs from aim 1; b) to identify novel genetic variants associated with paclitaxel or capecitabine related toxicity and; c) to build multi-genic predictive models associated with sensitivity to paclitaxel and capecitabine. Five clinical trial datasets available to us include: 1) a cohort of 859 breast cancer patients treated with paclitaxel alone (CALGB 40101); 2) a cohort of 143 breast and lung cancer patients from Spain treated with paclitaxel/carboplatin; 3) a cohort of 790 breast cancer patients treated with paclitaxel, nab-paclitaxel or ixabepilone in CALGB 40502; 4) the 250-patient prospective TBCRC clinical trial of single-agent capecitabine and 5) a  cohort of 163 patients in Spain treated with capecitabine. In addition, DNA from patients at the University of Chicago with extreme paclitaxelinduced neuropathy will be collected to help validate SNPs and statistical modeling.

3. To translate well-characterized variants from the above clinical studies of breast cancer pharmacogenomics into a clinical, preemptive testing treatment approach for breast cancer patients. This clinical implementation study will examine the utility of providing patients’ comprehensive breast cancer pharmacogenomic results to breast cancer oncologists in routine practice. We expect that breast cancer patients and oncologists will find that pharmacogenomics adds information to the clinical decisionmaking algorithm in the adjuvant and metastatic treatment settings, and that oncologists will change agents (e.g. from paclitaxel to docetaxel in adjuvant setting) or starting dosage (capecitabine therapy in palliative setting) for high risk patients. We aim to demonstrate that paclitaxel and capecitabine are less likely to be prescribed at full dosage to patients highly likely to have toxicity based on germline SNPs when pharmacogenomic results are preemptively ascertained and delivered to treating oncologists.

Project Publications

 

 

Events