Use of Calcium Channel Blockers and Breast Cancer Risk in the Women's Health Initiative Free

The use of calcium channel blockers (CCBs) recently has been found to be associated with increased risk of breast cancer (1, 2), although inconsistently. The inconsistency may be due to differences inherent in study design, or inadequately controlled confounding, including factors related to prescription for CCBs. Indeed, a number of important risk factors are shared between hypertension and breast cancer, making interpretation of results from studies that do not restrict to hypertensive women challenging. Given the high prevalence of CCB use and their hypothesized potential to disrupt apoptotic pathways, additional high-quality prospective data are needed.

Here, we examine the association between CCB use and breast cancer risk in the Women's Health Initiative (WHI), a large cohort of postmenopausal women. To further control for potential confounding, we compared CCB exposure with use of any other antihypertensive medication (AHM).

Information about the WHI methods has been published previously (3). From 1993 to 1998, 161,808 postmenopausal women, ages 50 to 79 years, were recruited into an observational study (OS) and one or more clinical trials (CT). Women were followed up to 2005 and, via an extension study, up to 2010. For this analysis, we excluded at baseline women who had prevalent breast cancer (n = 5,551); did not self-report a history of hypertension (n = 95,530), were nonusers of CCBs or other AHM (n = 26,840), or who used CCBs in combination with other AHM (n = 5,325) or were missing these data (n = 1); and leaving n = 28,561 for analysis.

Participants attended baseline screening visits, during which they completed extensive baseline questionnaires. Height and weight were measured by clinical staff. In-person medication inventories were obtained by review of participants' pill containers at baseline and year 3 in the OS and additionally in years 1, 6, and 9 in the CT. CCBs were subclassified into dihydropyridines or non-dihydropyridines and short acting or long acting. Duration of medication use was categorized as <5 years, 5 to 9.9 years, and ≥10 years. Other AHM data (including diuretics, ACE inhibitors, adrenergic receptor antagonists, angiotensin II receptor antagonists, renin inhibitors, and vasodilators) were obtained in an identical manner.

Incident, first-primary, invasive breast cancers were self-reported annually in the OS and semiannually in the CT until 2005 and annually thereafter. Cases were confirmed by medical record review by physician-adjudicators. After a median follow-up of 12.7 years, 1,402 invasive breast cancers were identified. Breast cancer subtypes, defined here as joint expressions of estrogen receptor (ER), progesterone receptor (PR), and HER2, were abstracted from medical records.

Cox proportional hazards models were used to estimate HRs and 95% confidence intervals (CI) for the associations between baseline CCB and breast cancer risk relative to other AHM. Categories of CCB duration were compared with the same categories of other AHM in regression models. Regression models were adjusted a priori for breast cancer risk factors thought to potentially confound associations and CT randomization. We performed several sensitivity analyses: (i) in the WHI-CT, CCB and other AHM use was treated as time-varying in regression models; (ii) we additionally examined associations of CCB use versus nonuse (n = 156,255) in the larger WHI cohort (including women without hypertension; n = 156,255) to compare our findings with others that did not account for confounding by shared risk factors.

Differences by medication for participants' baseline characteristics were small (Table 1). Compared with other AHM use, CCB use was not associated with breast cancer risk (HR, 1.06; 95% CI, 0.94–1.20; Table 2). No associations were observed when CCB use was stratified by length of action or drug class. When cancers were stratified on molecular subtype defined by ER, PR, and HER2, CCB use was associated with elevated risk of triple-negative breast cancers (HR, 1.60; 95% CI, 1.04–2.48). In the sensitivity analysis, time-varying CCB use was also not associated with breast cancer risk (HR, 0.99; 95% CI, 0.78–1.26). When CCB use was contrasted against nonuse (thus insufficiently controlling for shared hypertension/breast cancer risk factors; n = 156,255), associations were elevated (HR, 1.30; 95% CI, 0.84–2.02). When we restricted the comparison to women who reported prevalent hypertension (n = 60,726; HR, 1.08; 95% CI, 0.98–1.18) and who used ≥1 AHM (n = 33,886; HR, 1.08; 95% CI, 0.98–1.20), the association was attenuated.

We observed no association between CCB use and breast cancer risk in the WHI. Although these results contrast with recent case–control analyses (1, 2) and an early (ref. 4; but not later; ref. 5) report from a prospective study, our findings of no association are compatible with recent data from several prospective cohorts (6–8). Although a recent case–control study among Spanish women reported higher postmenopausal breast cancer risk associated with CCB use (OR, 1.72; 95% CI, 1.05–2.80; ref. 1), it was neither restricted to hypertensive women nor were CCBs compared with users of other antihypertensive medications, leaving a strong possibility for confounding. A >2-fold higher risk reported by Li and colleagues (2) persisted after restriction of the analysis to hypertensive women; however, the referent group included women with untreated hypertension. No study has examined associations with breast cancers characterized by molecular subtypes; the elevated association observed here may be due to chance but warrants consideration.

The advantages of this study include its comprehensive collection of medication use and its strong control of confounding by restriction of the analysis to women with hypertension and comparing CCB use with that of other AHMs. Furthermore, attrition bias was minimized with near-complete follow-up in the WHI.

We provide here additional evidence that CCBs do not broadly influence breast cancer risk in postmenopausal women.

R.T. Chlebowski is a consultant for AstraZeneca, Novartis, Amgen, and Genomic Health and serves as a speaker for Novartis and Genentech. No potential conflicts of interest were disclosed by the other authors.

Conception and design: T.M. Brasky, J.L. Krok-Schoen, J. Wactawski-Wende, A. Wang, J.E. Manson

Development of methodology: L. Qi, T.M. Brasky, K.L. Margolis, J. Wactawski-Wende, A. Wang, J.E. Manson

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): R.T. Chlebowski, K.L. Margolis, M.S. Simon, J. Wactawski-Wende, J.E. Manson

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): T.M. Brasky, J.L. Krok-Schoen, J. Liu, R.T. Chlebowski, J.L. Freudenheim, S. Lavasani, K.L. Margolis, L. Qi, K.W. Reding, P.G. Shields, M.S. Simon, J. Wactawski-Wende, A. Wang, C. Womack, J.E. Manson

Writing, review, and/or revision of the manuscript: T.M. Brasky, J.L. Krok-Schoen, J. Liu, R.T. Chlebowski, J.L. Freudenheim, S. Lavasani, K.L. Margolis, L. Qi, K.W. Reding, P.G. Shields, M.S. Simon, J. Wactawski-Wende, A. Wang, C. Womack, J.E. Manson

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): R.T. Chlebowski, J. Wactawski-Wende, J.E. Manson

Study supervision: J. Wactawski-Wende, J.E. Manson

This work was supported by the National Heart, Lung, and Blood Institute, NIH, and U.S. Department of Health and Human Services grants HHSN2682011000046C, HHSN268201100001C, HHSN268201100002C, HHSN268201100003C, and HHSN268201100004C.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.