Long-term Use of CCB and Breast Cancer Risk

Aims: To examine the association between long-term calcium channel blocker (CCB) use and the development of breast cancer. Data sources: the Australian Longitudinal Study on Women's Health , 45 and Up Study, Rotterdam study and linked administrative data. Study cohort: Women with self-reported hypertension (HTN) without prior breast cancer. Harmonisation: Relevant variables will be checked for harmonisation in which variables available in all three cohort will be used in the harmonised analysis. The variables will be checked on the definition, measurement and harmonisation rules. Variables that cannot be harmonised across the cohorts will be used in the cohort specific analyses. Exposure measurement: The primary exposure is the use of CCB, which will be identified by the anatomical therapeutic chemical code (C08) in the medicine data. Exposure to CCB as well as other antihypertensive (AHT) medicines will be captured separately as the cumulative dose-duration of exposure during follow up. Participants will be stratified in four groups: women with HTN with no AHT use, women with HTN exposed to CCB only, women with HTN exposed to non-CCB and women with HTN exposed to both CCB and non-CCB. Outcome measurement: Data on a diagnosis of invasive breast cancer will be obtained from cancer registries and hospital admission data using ICD-10 code of C50.x. Potential confounders: age, education, marital status, socioeconomic status, body mass index, diabetes, heart disease, stroke, age when had first child, parity, history of hysterectomy or oophorectomy, use of hormonal contraception, use of hormonal replacement therapy. Statistical analysis: The association between CCB use and breast cancer risk will be estimated by the Fine and Gray competing risk regression model in which a first diagnosis of invasive breast cancer will be treated as the principle event and death and bilateral mastectomy (without a diagnosis of breast cancer) will be competing risks. The nonlinear threshold models will be used to capture any differential effect of the cumulative dose-duration of CCB while simultaneously accounting for the cumulative dose-duration of other AHT exposure on breast cancer risk. Other confounders will be accounted for in the models using propensity scores. Implications: Results from this study will contribute to addressing the concerns about using CCB for hypertension treatment in women, particularly in those who have high risk of breast cancer.