2003 Calcium Channel Blockers, Cancer Incidence, and Cancer Mortality in a Cohort of U.S. Women The Nurses’ Health Study Karin B. Michels, Sc.D.1,2 Bernard A. Rosner, Ph.D.1,3 Alexander M. Walker, M.D.2 Meir J. Stampfer, M.D.1,2,4 JoAnn E. Manson, M.D.1,2,5 Graham A. Colditz, M.D.1,2 Charles H. Hennekens, M.D.2,5 Walter C. Willett, M.D.1,2,4 1 Channing Laboratory, Department of Medicine, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts. 2 Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts. 3 Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts. 4 Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts. 5 Division of Preventive Medicine, Department of Medicine, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts. Supported by research grant CA 40356 from the National Cancer Institute, National Institutes of Health, Bethesda, Maryland. The authors thank Professor Frank E. Speizer, M.D., principal investigator of the Nurses’ Health Study, for helpful advice regarding the analysis conducted in this study and regarding the manuscript. They also thank the participants in the Nurses’ Health Study for providing the relevant information necessary for this analysis. Address for reprints: Karin B. Michels, Sc.D., Channing Laboratory, 181 Longwood Avenue, Boston, MA 02146. Received January 26, 1998; accepted April 6, 1998. © 1998 American Cancer Society BACKGROUND. Some studies have suggested that the use of calcium channel blockers may increase the risk of cancer. A possible association of the use of calcium channel blockers with cancer incidence and cancer mortality was addressed using data from the Nurses’ Health Study. METHODS. In this study, a total of 18,635 female nurses reported regularly taking at least 1 of 4 cardiovascular medications in 1988: diuretics, beta-blockers, calcium channel blockers, and/or angiotensin-converting enzyme (ACE) inhibitors. Cancer incidence and cancer deaths were ascertained until 1994. RESULTS. During 6 years of follow-up, 852 women were newly diagnosed with cancer and 335 women died of cancer. Women who reported the use of calcium channel blockers had no increased risk of newly diagnosed cancer compared with those taking other cardiovascular drugs (relative risk ⫽ 1.02; 95% CI 0.83–1.26). The relative risk of dying from cancer associated with the self-reported use of calcium channel blockers was 1.25 (95% CI 0.91–1.72). Relative risks were adjusted for the following self-reported factors: age; weight; height; cholesterol level; systolic and diastolic blood pressure; smoking; alcohol intake; physical activity; menopausal status; postmenopausal hormone use; aspirin use; and history of diabetes, cancer, stroke, myocardial infarction, coronary artery bypass graft or percutaneous transluminal coronary angioplasty, angina, and hypertension. Regarding site specific cancer incidence and mortality, only lung cancer incidence was somewhat increased (RR ⫽ 1.61; 95% CI 0.88 –2.96). CONCLUSIONS. These data suggest no important increase in overall cancer incidence or cancer mortality related to the self-reported use of calcium channel blockers. Cancer 1998;83:2003–7. © 1998 American Cancer Society. KEYWORDS: epidemiology, cancer incidence, cancer mortality, calcium channel blockers, apoptosis. C alcium channel blockers have recently been associated with a possible increase in the incidence of cancer. An increased risk of cancer was first reported by Pahor et al.1,2 Among 750 hypertensives who were cancer free at baseline, the relative risk (RR) of developing cancer was found to be 2.02 (95% CI 1.16 –3.54) for those taking calcium channel blockers (27 events) as compared with those taking beta-blockers (28 events).1 In the full cohort (not restricted to hypertensives), relative risks for cancer were 1.72 (95% CI 1.27–2.34) for reported use of calcium antagonists compared with those not taking them; the highest risk was associated with verapamil (RR ⫽ 2.49; 95% CI 1.54 – 4.01).2 In a small Swedish case– control study, patients with colon carcinoma had taken verapamil more often than control participants, 2004 CANCER November 1, 1998 / Volume 83 / Number 9 but only 11 participants (10 cases, 1 control) reported the use of verapamil.3 In a British case– control study, Jick et al. reported an odds ratio for cancer of 1.27 (95% CI 0.98 –1.63) for users of calcium channel blockers (178 cases) compared with users of beta-blockers (183 cases).4 In a cohort in Germany, no increase in cancer incidence was observed with calcium channel blockers.5 An elevated risk of breast carcinoma was recently attributed to the use of calcium channel blockers based on a multisite cohort study.6 Fitzpatrick et al. reported a hazard ratio of 2.57 (95% CI 1.47– 4.49) for women who used calcium channel blockers (20 cases) compared with women who did not use them (55 cases).6 Most recently, a large casecontrol study including 9513 incident cancer cases yielded a relative risk of 1.1 (95% CI 0.9 –1.3) for the use of calcium channel blockers.6a It has been suggested that calcium channel blockers increase the risk of cancer by inhibiting apoptosis,7 the programmed cell death by DNA fragmentation of dysfunctional and old cells.8 Intracellular calcium ion balance regulates apoptosis in vitro9,10 and calcium antagonists inhibit apoptosis in vitro.11,12 Animal data are not available. In the Nurses’ Health Study, we investigated whether self-reported use of calcium channel blockers, beta-blockers, or angiotensin-converting enzyme (ACE) inhibitors, relative to the use of diuretics, was associated with incidence of or mortality from cancer. METHODS Study Population and Data Collection The Nurses’ Health Study was initiated in 1976 and includes 121,701 female registered U.S. nurses who were ages 30 –55 years at the time of entry. Participants are followed through biennial self-administered questionnaires. On baseline and follow-up questionnaires, nurses are asked to provide demographic and life-style information as well as their disease status. In 1988, information was requested on the regular use of cardiovascular medications, including thiazide diuretics, beta-blockers, calcium channel blockers, and ACE inhibitors. Participants were asked, “Are you currently taking any of the following medications at least once a week?” On the questionnaire, medications were identified only as classes of pharmaceutical drugs (with examples), as listed above. Information on the use of calcium channel blockers was not updated until 1994; therefore, medication use was not updated in the analysis. According to self-reported information on medications from 1994, 57% of women who reported calcium channel blocker use in 1988 were still using them in 1994. Of women who used other or no medications in 1988, 14% reported use of calcium channel blockers in 1994. In 1988, long-acting calcium channel blockers were rarely prescribed; it can therefore be assumed that the majority of calcium antagonists were short-acting formulations. The endpoints included in this analysis were cancer incidence and cancer mortality. Cancer mortality analyses included women with prevalent cancers in 1988. All women who reported cancer were contacted for permission to review the relevant hospital records and confirm the self-reported diagnosis. Study physicians blinded to the exposure information reviewed the medical records to extract information on the histologic type, the anatomic location, and the stage of the cancer. Deaths of study participants have been ascertained since the beginning of the Nurses’ Health Study in 1976. The mortality surveillance includes the National Death Index to identify any deaths among participants who did not respond during each questionnaire cycle. This surveillance supplements reports by the subjects’ next of kin and by postal authorities. Mortality follow-up has been estimated to be 98% complete for this cohort.13 When deaths were identified, written permission was requested from the next of kin to review the medical records and obtain pathology reports. Statistical Analysis Analyses were restricted to women who reported in 1988 that they took thiazide diuretics, beta-blockers, calcium channel blockers, ACE inhibitors, or any combination. Women who reported use of both calcium channel blockers and ACE inhibitors were excluded due to inadequate sample size (n ⫽ 209). Women were also excluded from the analysis if their year of birth (n ⫽ 5), smoking status in 1988 (n ⫽ 40), menopausal status in 1988 (n ⫽ 48), or weight or height (n ⫽ 29) was unknown, or if their date of death was prior to 1988 or not yet known (n ⫽ 110); these exclusions left a study population of 18,635 women. During the 6-year observation period, 684 women (3.7%) died and 502 (2.7%) were lost to follow-up. Person-time of follow-up was allocated to each participant starting with the return of the 1988 questionnaire and accumulated up to May 1, 1994, or the diagnosis of cancer or death, whichever occurred first. A total of 107,256 person-years were accumulated. Drug exposure was classified according to the 1988 responses throughout the follow-up period, without further updating. Cancer incidence or mortality rates were calculated for each pharmaceutical agent by dividing the number of events by the person-time of follow-up for that agent. Relative risks were estimated as ratios of incidence rates, comparing women who Calcium Channel Blockers and Cancer/Michels et al. reported regular use of calcium channel blockers with women who used other cardiovascular drugs. To control simultaneously for other potential risk factors, a pooled logistic regression model was used, in which risk sets were updated every 2 years.14 In estimating the cause specific hazard for cancer death, we censored deaths from other causes at each 2-year interval, thus conditioning on survival up to that age.15 We calculated 95% confidence intervals around the relative risks.16 Age-adjusted relative risks were obtained for all medications considered and for any combinations used. Covariate-adjusted relative risks were calculated, with adjustment for various self-reported risk factors: age; weight; height; cholesterol level; systolic and diastolic blood pressure; smoking; amount of current smoking; alcohol intake; regular physical activity, defined as working up a sweat at least once a week; menopausal status; current postmenopausal hormone (PMH) use for at least 5 years; use of aspirin on at least 5 days each month; and history of diabetes, cancer, stroke, myocardial infarction, coronary artery bypass graft (CABG) or percutaneous transluminal coronary angioplasty (PTCA), angina pectoris, or hypertension in or prior to 1988. RESULTS The distribution of risk factors and history of disease among women who reported use of calcium channel blockers and those who reported other medications is displayed in Table 1. Women prescribed calcium channel blockers were more likely to have a history of diabetes, pulmonary disease (asthma, chronic bronchitis, or emphysema), or ischemic heart disease (myocardial infarction, angina pectoris, or CABG/ PTCA). Women who reported use of other medications were more likely to be hypertensive. Of the 18,635 nurses included in this analysis, 852 were newly diagnosed with cancer between 1988 and 1994, and 335 died of cancer during this follow-up period. Total cancer incidence was not found to be associated with the use of calcium channel blockers (Table 2). No association was found between calcium channel blocker use and breast carcinoma incidence. Of the site specific cancers, only lung carcinoma incidence was elevated among women who reported calcium channel blocker use (RR ⫽ 1.96; 95% 1.15–3.33; adjusted for age, smoking status in 1988, and mean number of cigarettes smoked per day among women who smoked in 1988). When additional adjustment was made for age at the start of smoking, multiple drug use, weight, height, alcohol intake, physical activity, menopausal status, and postmenopausal hormone use, the relative risk was slightly higher. After 2005 TABLE 1 Risk Factor Profile and Concomitant Disease among Regular Users of CCB, Compared with Users of Cardiovascular Medications (thiazide diuretics, beta blockers, ACE inhibitors) Other than CCB, Self-Reported in 1988 among 18,635 Participants in the Nurses’ Health Study Risk factors/Concomitant Disease CCB No CCB Total no. of participants Mean age (yrs) Mean weight (kg) Mean heighta (cm) Mean cholesterolb (mg/dL) Mean systolic BPb (mmHg) Mean diastolic BPb (mmHg) Current smoker (%) Former smoker (%) Amount smokedb,c Alcohol intaked (gm/day) Regular physical activitye Postmenopausal Current PMH use for 5⫹ yrs Aspirin intakef Diabetes Pulmonary disease Cancer Myocardial infarctiong Angina pectorisg CABG/PTCAg Strokeg Hypertensiong 2361 58.0 71.9 163.3 220.5 135.8 84.2 396 (16.8) 1009 (42.7) 19.4 5.6 899 (38.1) 2131 (90.3) 340 (14.4) 1008 (42.7) 321 (13.6) 382 (16.2) 200 (8.5) 362 (15.3) 605 (25.6) 105 (4.5) 65 (2.8) 918 (38.9) 16,274 56.8 73.2 163.8 212.9 136.1 85.1 2686 (16.5) 6391 (39.3) 18.5 6.3 6391 (39.3) 13,787 (84.7) 2455 (15.1) 5834 (35.9) 1386 (8.5) 1721 (10.6) 1453 (8.9) 518 (3.2) 1003 (6.2) 112 (0.7) 229 (1.4) 11,482 (70.6) CCB: calcium channel blockers; ACE: angiotension-converting enzyme; PMH: postmenopausal hormone; CABG/PTCA: coronary artery bypass graft or percutaneous transluminal coronary angioplasty. a Data for 1976 is given. b Average values were calculated as mean values from prespecified categoric responses. c Mean no. of cigarettes smoked per day among current smokers in 1988. d In grams per day; includes women who do not drink alcohol. e Assessed in 1980; regular physical activity was defined as working up a sweat at least once a week. f Regular aspirin intake was defined as use on at least 5 days each month. g Categories are mutually exclusive with the following hierarchy: 1) stroke, 2) myocardial infarction, 3) CABG/PTCA, 4) angina pectoris, 5) hypertension. additional adjustment for cardiovascular risk factors as well as concomitant cardiovascular disease and diabetes, the relative risk was somewhat attenuated (RR ⫽ 1.78; 95% CI 0.98 –3.23) (Table 2). When pulmonary disease in or prior to 1988 and years since women stopped smoking were also controlled in the analysis, the relative risk for lung carcinoma was 1.61 (95% CI 0.88 –2.96). For women who had ever smoked, the covariate-adjusted relative risk was 1.55 (95% CI 0.80 – 3.02), additionally adjusted for a history of pulmonary disease and years since women stopped smoking. Calcium channel blocker users had a relative risk of smoking-related cancers (lung, pancreatic, kidney, or bladder carcinoma) of 1.68 (95% CI 1.06 –2.65, adjusted for all covariates, including prior pulmonary disease and years since women stopped smoking). 2006 CANCER November 1, 1998 / Volume 83 / Number 9 TABLE 2 Relative Risk (95% Confidence Interval) of Site Specific Cancer Incidence and Mortality between 1988 and 1994, According to Use of CCB SelfReported in 1988 among Women in the Nurses’ Health Study CCB Cancer site Total cancer Incidence (n ⫽ 16,954) Mortality Breast carcinoma Incidence (n ⫽ 16,482) Mortality Lung carcinoma Incidence (n ⫽ 16,746) Mortality Colon carcinoma Incidence (n ⫽ 16,981) Mortality Cancer of the lymphatic system Incidence (n ⫽ 16,978) Mortality No CCB Covariate-adjusted relative risk I (95% CI)a CCB vs. no CCB Covariate-adjusted relative risk II (95% CI)b CCB vs. no CCB Events Person-years Events Personyears Age-adjusted relative risk (95% CI) CCB vs. no CCB 122 54 12,206 13,404 730 281 84,787 93,852 1.09 (0.90–1.33) 1.23 (0.92–1.65) 1.10 (0.90–1.34) 1.27 (0.94–1.71) 1.02 (0.83–1.26) 1.25 (0.91–1.72) 51 13 11,807 13,404 304 77 82,524 93,852 1.12 (0.83–1.51) 1.12 (0.62–2.02) 1.16 (0.86–1.57)c 1.13 (0.61–2.07) 1.07 (0.78–1.48)c 1.33 (0.70–2.54) 18 14 12,170 13,404 56 56 84,588 93,852 1.96 (1.15–3.33)d 1.58 (0.88–2.85) 2.01 (1.16–3.49)e 1.61 (0.88–2.94) 1.78 (0.98–3.23)e 1.32 (0.69–2.53) 6 5 12,324 13,404 49 28 85,746 93,852 0.76 (0.33–1.77) 1.15 (0.44–2.98) 0.73 (0.31–1.74)f 1.10 (0.41–2.92) 0.87 (0.35–2.14)f 1.03 (0.37–2.88) 6 6 12,332 13,404 43 23 85,737 93,852 0.90 (0.38–2.13) 1.66 (0.67–4.07) 0.93 (0.39–2.21) 1.59 (0.63–4.01) 0.88 (0.35–2.18) 1.86 (0.71–4.88) CCB: calcium channel blockers; CI: confidence interval; CABG/PTCA: coronary artery bypass graft or percutaneous transluminal coronary angioplasty; Mortality: covariate-adjusted relative risks are adjusted for pre-existing cancer. a The relative risk and 95% CI estimates are adjusted for age, multiple drug use, self-reported weight, height, smoking status and mean number of cigarettes smoked per day among women who smoked in 1988, alcohol intake in 1988, physical activity, menopausal status in 1988, and postmenopausal hormone use. b Additional adjustment was made for cholesterol level; systolic and diastolic blood pressure in 1988; aspirin intake; diabetes; and history of stroke, myocardial infarction, CABG/PTCA, angina, and hypertension in or prior to 1988. c Additional adjustment was made for family history of breast carcinoma, history of benign breast disease, age at menarche, parity, age at first birth, and age at menopause. d Additional adjustment was made for smoking status and mean no. of cigarettes smoked per day among women who smoked in 1988. e Additional adjustment was made for age at the start of smoking. f Additional adjustment was made for history of colorectal polyps. Mortality from total or site specific incident and prevalent cancers combined did not differ significantly among nurses who reported the use of calcium channel blockers compared with those who did not report the use of this drug (Table 2). Of the 335 women who died of cancer among our study population, 154 were diagnosed with cancer prior to the time of assessment of medication use in 1988, and 181 women were diagnosed with cancer after 1988. DISCUSSION These data do not suggest an important association between calcium channel blocker use and cancer incidence or cancer mortality. However, we observed an increase in lung carcinoma incidence among women who reported the use of calcium channel blockers. This association remained after statistical adjustment for smoking status in 1988, amount of smoking in 1988, and age at the start of smoking. Women who took calcium channel blockers had more pulmonary disease than women who took other medications. Since we did not know when these women began taking calcium channel blockers, pulmonary disease may or may not have preceded the use of this medication. Controlling for pulmonary disease in the analysis weakened the association somewhat. Because beta-blockers are contraindicated for patients with pulmonary disease, other medications are more commonly prescribed for hypertensive patients with such conditions. In this population, women who reported regular use of calcium channel blockers in 1988 also reported pulmonary disease prior to 1988 more often than women who used other medications. It is likely that calcium channel blockers were preferentially prescribed for hypertensive women who also had pulmonary disease. We have previously described an excess in the incidence of myocardial infarction among women who were smokers and reported the use of calcium channel blockers.17 A possible interaction between calcium channel blockers and cigarette smoking could also underlie the increased incidence of lung carcinoma observed; however, further data are needed to address this issue. Calcium Channel Blockers and Cancer/Michels et al. A number of epidemiologic studies have suggested that hypertension itself may be associated with cancer mortality and incidence.18,19 In our study population, physician-diagnosed hypertension was not an important predictor of cancer mortality (RR ⫽ 0.92; 95% CI 0.66 –1.28) or cancer incidence (RR ⫽ 0.94; 95% CI 0.78 –1.13). We could not confirm an increased risk of breast carcinoma for users of calcium channel blockers as described by Fitzpatrick et al.6 These authors reported a hazard ratio of 2.57 (95% CI 1.47– 4.49) based on 20 cases of breast carcinoma among women who took calcium channel blockers and 55 cases among women who did not take them. Neither Pahor et al.2 nor Jick et al.4 found significantly elevated relative risks for breast carcinoma in their site specific cancer analyses. The current study has a number of limitations. We had to rely on self-reported medication use, but the reliability of such data can be assumed to be high because the study population consists of health professionals. As use of calcium channel blockers was first assessed in 1988 in this cohort, new and preexisting prescriptions before 1988 could not be separated. We further assumed that exposure in 1988 was relevant, although we had no information on duration of use. If an extended duration of use was crucial to any association with cancer, exposure may have sometimes been misclassified and any true effect underestimated. Since cancer has a considerable latent period, medication use in 1988 or before might be more relevant than later use. Because we had no information about the start of medication use, we could not determine what proportion of women were diagnosed with cancer prior to taking any of the medications. Similarly, doses of prescriptions were not known. We had no information on the type of drug within a particular class. Therefore, we could not study the association of calcium channel blockers of the dihydropyridine type (e.g., nifedipine) separately from that of the nondihydropyridines (diltiazem or verapamil). In conclusion, these prospective data suggest that a major association between the use of calcium channel blockers of the short-acting formulation and cancer incidence or cancer mortality is unlikely. The elevated risk found for lung carcinoma incidence deserves further investigation. Further follow-up is needed to rule out late effects of calcium channel blockers on cancer risk. The weight of the current evidence does not give reason for concern regarding an association between the use of calcium channel blockers and overall or site specific cancer incidence or mortality. Although the incidence of one or the other site specific cancer has been found to be increased among calcium antagonist 2007 users in some studies, no two studies have found significant elevations in the incidence of the same type of cancer. REFERENCES 1. 2. 3. 4. 5. 6. 6a. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Pahor M, Guralnik JM, Salive ME, Corti MC, Carbonin P, Havlik RJ. Do calcium channel blockers increase the risk of cancer? Am J Hypertens 1996;9:695–9. Pahor M, Guralnik JM, Ferrucci L, Corti MC, Salive ME, Cerhan JR, et al. Calcium-channel blockade and incidence of cancer in aged populations. Lancet 1996;348:493–7. Hardell L, Fredrikson M, Axelson O. Case-control study on colon cancer regarding previous diseases and drug intake. Int J Oncol 1996;8:439 – 44. Jick H, Jick S, Derby LE, Vasilakis C, Myers MW, Meier CR. Calcium-channel blockers and risk of cancer. Lancet 1997; 349:525– 8. Trenkwalder PRA, Hendricks P, Hense H-W. Calcium-channel blockers do not increase the risk of fatal or non-fatal cancer in an elderly European population: results from STEPHY II. Circulation 1997;96:I–763. Fitzpatrick AL, Daling JR, Furberg CD, Kronmal RA, Weissfeld JL. Use of calcium channel blockers and breast carcinoma risk in postmenopausal women. Cancer 1997;80:1438 – 47. Rosenberg L, Rao S, Palmer JR, Strom BL, Stolley PD, Zauber AG, et al. Calcium channel blockers and the risk of cancer. JAMA 1998;279:1000 – 4. Daling J. Calcium channel blockers and cancer: is an association biologically plausible? Am J Hypertens 1996;9:713– 4. Kerr JFR, Wyllie AH, Currie AR. Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer 1972;26:239 –57. McConkey DJ, Orrenius S. The role of calcium in the regulation of apoptosis. J Leukoc Biol 1996;59:775– 83. McConkey DJ, Hartzell P, Nicotera P, Orrenius S. Calciumactivated DNA fragmentation kills immature thymocytes. FASEB J 1989;3:1843–9. Connor J, Sawczuk IS, Benson MC, Tomashefsky P, O’Toole KM, Olsson CA, et al. Calcium channel antagonists delay regression of androgen-dependent tissues and suppress gene activity associated with cell death. Prostate 1988;13:119 –30. Ray SD, Kamendulis LM, Gurule MW, Yorkin RD, Corcoran GB. Ca2⫹ antagonists inhibit DNA fragmentation and toxic cell death induced by acetaminophen. FASEB J 1993;7:453– 63. Stampfer MJ, Willett WC, Speizer FE, Dysert DC, Lipnick R, Rosner B, et al. Test of the National Death Index. Am J Epidemiol 1984;119:837–9. D’Agostino RB, Lee ML, Belanger AJ, Cupples LA, Anderson K, Kannel WB. Relation of pooled logistic regression to time dependent Cox regression analysis: the Framingham Heart Study. Stat Med 1990;9:1501–15. Cox DR. Regression models and life tables. J R Stat Soc 1972;32;187–220. Rosner B. Fundamentals of biostatistics. 4th edition. Belmont, CA: Duxbury Press, 1995. Michels KB, Rosner BA, Manson JE, Stampfer MJ, Walker AM, Willett WC, et al. Calcium channel blockers in relation to incidence of cardiovascular disease and total mortality among hypertensive women. Circulation 1998;97:1540 – 8. Hamet P. Cancer and hypertension. Hypertension 1996; 28:321– 4. Dyer AR, Stamler J, Berkson DM, Lindberg HA, Stevens E. High blood pressure: a risk factor for cancer mortality? Lancet 1975;1:1051– 6.