Using pharmacoepidemiology to test medications associated with altered breast cancer risk identified from connectivity mapping

Study type
Protocol
Date of Approval
Study reference ID
15_212
Lay Summary

Connectivity mapping is a new technique that allows medications to be screened to assess their ability to cause cancer or have anti-cancer properties that are unrelated to the established therapeutic effects of these drugs. This connectivity map process has been implemented and has identified 10 prescription medications (used for non-cancer purposes) that potentially reduce or increase breast cancer risk.

The aim of this study is to determine the association between the 10 identified medications (from connectivity mapping) and breast cancer risk.

The study could identify (previously unrecognized) medications which increase the risk of breast cancer development; current licensing and use of such medications may need to be reconsidered. This study also has the potential to identify licensed medications that reduce breast cancer risk and warrant further study in clinical trials.

Technical Summary

Background: Connectivity mapping is a novel bioinformatics technique which links gene expression data with expression events induced by a wide range of compounds, including medications. It therefore allows medications to be screened to assess carcinogenicity and anti-cancer properties. This connectivity map process has been implemented to identify 10 prescription medications (used for non-cancer purposes) that may alter breast cancer risk.

Aims: To determine the association between the 10 identified medications (from connectivity mapping) and breast cancer risk.

Methods: A nested case-control study will be conducted. Breast cancer patients (identified from cancer registries) will be matched to five controls on age, calendar year and GP practice. Prescriptions for candidate medications will be determined prior to breast cancer diagnosis or index date in cases and controls, respectively. Conditional logistic regression will be used to calculate odds ratios (ORs), and 95% confidence intervals (95%CIs), for breast cancer in candidate medication users compared with non-users and adjust for potential confounders.

Potential: The study could identify a medication (from connectivity mapping) which increase the risk of breast cancer development; current licensing and use of such medications may need to be reconsidered. This study also could identify licensed medications that reduce breast cancer risk and warrant further study.

Collaborators

Chris Cardwell - Chief Investigator - Queen's University Belfast
Fabio Liberante - Collaborator - Queen's University Belfast
Gayathri Thillaiyampalam - Collaborator - Queen's University Belfast
Ken Mills - Collaborator - Queen's University Belfast
Liam Murray - Collaborator - Queen's University Belfast
Shu-Dong Zhang - Collaborator - Queen's University Belfast

Linkages

NCRAS Cancer Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation