Cimetidine and Other Histamine₂-Receptor Antagonist Use in Relation to Risk of Breast Cancer

Cimetidine, a member of the histamine₂-receptor antagonist (H₂ blocker) class, is used for treatment of the symptoms of peptic and duodenal ulcer. Introduced in 1977, the drug was awarded over-the-counter status in 1995 and is commonly used. The possible influence of cimetidine on the risk of cancer has garnered interest because cimetidine influences both hormonal and immune pathways. Cimetidine has been shown to reduce estradiol 2-hydroxylation and increase serum estradiol levels in men (1) and women (2). It increases the serum concentrations of prolactin, which may be a risk factor for breast cancer among postmenopausal women (3). Cimetidine also inhibits histamine-induced immune suppression (4), an effect that has been tested in the treatment of various malignancies (5, 6), and may also be relevant to tumor development (7). Other members of the H₂ blocker class, including ranitidine, famotidine, and nizatidine, are thought to have little or no hormonal (8) or immunologic effects (4).

In two large pharmacy databases (9, 10) and one cohort study (11), cimetidine use did not influence the risk of breast cancer. We investigated the influence of cimetidine use on the risk of breast cancer in our hospital-based case control surveillance study. This study includes information on frequency and duration of both prescription and over-the-counter medication use.

Data used in this analysis were collected from patients admitted to participating hospitals in New York, Philadelphia, Boston, and Baltimore from 1977 to 2002. The population base for the study comprised people living in zip codes within 50 miles of a participating hospital. Nurse-interviewers visited the hospitals on a rotating basis, from once to thrice per week, depending on the size of the institution. On visit days, all patients meeting study criteria were identified through examination of admission lists and ward logs. Eligible patients were aged 18 to 79, were under the care of a physician participating in the study, did not have certain excluded diagnoses (e.g., psychiatric diagnoses), were able to complete the interview (e.g., not deaf), and lived in an eligible zip code. Patients were approached for an interview if they did not have visitors or were not too ill. The vast majority of interviews were conducted when the patient was an inpatient in the hospital. A small proportion of cancer patients were interviewed when visiting outpatient cancer clinics. Thus, women with breast cancer who did not require surgery or who did not visit a clinic for chemotherapy were not included in the study. Methods are described in detail elsewhere (12).

Nurse-interviewers gave standard questionnaires to obtain information on demographic factors, medical and reproductive history, and habits such as smoking and alcohol consumption. Histories of drug use were elicited by asking about 43 indications for use that included those for which H₂ blockers are used (e.g., stomach/ulcers/bowel). For each episode of use, the drug name and the duration, timing, and frequency of use were recorded. Details of the diagnosis were abstracted from discharge summaries and pathology reports. From 1977 through 1997, 95% of patients approached for an interview participated. Since 1998, 88% of patients have participated. The study was approved by the Institutional Review Boards of all participating institutions.

The breast cancer cases comprised 6,994 women aged 30 through 79 with a first occurrence of primary breast cancer diagnosed within the previous year, and no concurrent or previous cancer other than non–melanoma cancer of the skin.

We selected two control groups, cancer controls and noncancer controls, that comprised patients admitted to the hospital with diagnoses that we judged to be unrelated to cimetidine use. Thus, diagnoses involving the esophagus or the gastrointestinal tract were excluded as were diagnoses that are contraindications for H₂ blocker use (e.g., diseases of the kidney). Controls had no prior history of cancer other than non–melanoma cancer of the skin and were diagnosed no more than 1 year prior to hospital admission. Controls were frequency-matched to the cases in a ratio of up to 4:1 on age in 5-year intervals, study center, and year of interview. The noncancer controls included 6,004 patients admitted for trauma (24%), uterine fibroids (23%), orthopedic conditions (15%), acute infection (12%), endometriosis (5%), and other conditions (e.g., cataract, uterine prolapse, sciatica; 21%). The 2,478 cancer controls included patients with colorectal cancer (50%), leukemia and lymphoma (25%), cancer of the bone and connective tissue (8%), bladder cancer (6%), lung and other respiratory cancers (5%), and other cancers (e.g., gallbladder, liver, thyroid; 6%).

We defined regular H₂ blocker use to be use of at least 4 days per week for at least 3 continuous months. All other use was considered sporadic. Regular use was categorized according to when it began (within the previous year or at least 1 year prior to admission); only use that began a year or more before admission was considered etiologically relevant. Regular use was further categorized according to when it ended; discontinued use was use that ended 12 or more months prior to hospital admission, and continuing use was use that continued into the year prior to admission. Subjects who used cimetidine in the year prior to hospital admission only were excluded from the analyses. Two categories of histamine blocker use were assessed: use of cimetidine (with or without use of another H₂ blocker) and use of other H₂ blockers (without use of cimetidine).

The age-adjusted prevalence of regular cimetidine use that began at least 1 year prior to hospital admission was 1.3% among cancer controls and 1.2% among noncancer controls; corresponding figures for use of other H₂ blockers was 1.2% among cancer controls, and 1.1% among noncancer controls. The most commonly used of the other H₂ blockers was ranitidine.

Conditional logistic regression models which accounted for the matching on age (5-year age intervals), year of interview (1977-1980, 1981-1984, 1985-1990, and 1991-2002), and study center (Boston, New York, Philadelphia, and Baltimore) were used to estimate the odds ratios for breast cancer among regular users of H₂ blockers relative to never users of any drug in the H₂ blocker class. Estimates from models that included family history of breast cancer, race, education, parity, menopausal status, body mass index, and use of conjugated estrogens or other female hormones varied little from models that accounted only for the matching variables. Likewise, adjusting for H₂ blocker use in the cimetidine analysis, and for cimetidine use in the H₂ blocker analysis, did not change the odds ratios. Therefore, we present unadjusted odds ratios. A test for trend was done by including duration of use among regular users in the model as a continuous variable.

Characteristics of cases, noncancer controls, and cancer controls are shown in Table 1. Cancer controls were more similar to the cases in terms of age, race, and body mass index compared with the noncancer controls. Noncancer controls were more likely to be younger than age 50, black, and premenopausal compared with cases or cancer controls.

Results for cimetidine and breast cancer risk were similar regardless of whether cancer controls or noncancer controls were used: the odds ratio for breast cancer among regular users of cimetidine was 0.9 [95% confidence interval (CI), 0.6-1.5] using cancer controls and 0.9 (95% CI, 0.6-1.3) using noncancer controls. Among women who used the drug for 4 or more years, the odds ratios were 0.6 (95% CI, 0.2-1.9) with cancer controls and 0.5 (95% CI, 0.2-1.4) with noncancer controls. Because the prevalence of cimetidine use was similar in cancer and noncancer controls, and results with cancer controls were similar to those with noncancer controls, we combined the two control groups. This resulted in a case/control ratio of up to 1:8. As shown in Table 2, the odds ratios for regular and sporadic use were close to null. Among regular users of cimetidine for 4 or more years, the odds ratio (OR) was reduced (OR, 0.5; 95% CI, 0.2-1.3). There was no trend across the duration of use among regular users (P = 0.16). The odds ratios for continuing and discontinued regular use were close to null (Table 2). Odds ratios for regular users did not vary from 1.0 among premenopausal and postmenopausal women (data not shown).

The odds ratio for breast cancer among users of H₂ blockers other than cimetidine was 0.9 (95% CI, 0.6-1.3) using the combined control group (Table 3). The odds ratio for use of 2 to 3 years' duration was significantly reduced (OR, 0.4; 95% CI, 0.2-0.9), but the odds ratio for 4 or more years was compatible with 1.0 (OR, 0.9; 95% CI, 0.4-2.0; Table 3). Odds ratios for continuing and discontinued regular use were near null. Odds ratios were similar among premenopausal and postmenopausal women (data not shown).

In these data, the regular use of cimetidine did not affect the risk of breast cancer. The majority of other H₂ blocker use was of ranitidine. The odds ratios for regular use of these drugs did not differ significantly from 1.0. There was a significant reduction in risk among women who had used other H₂ blockers for 2 to 3 years, but given the lack of an effect with either longer or shorter term use, this is likely to be due to chance. Our duration analysis was limited by small numbers, and chance is a likely explanation for the reduced odds ratio for women who had used cimetidine for 4 or more years.

Our results agree with those of three prior cohort studies which failed to find an association between use of cimetidine and breast cancer risk. Moller et al. (11) assembled a cohort of 16,739 cimetidine users who had applied for reimbursement for the drug from the Danish National Board of Health from 1977 to 1981. Using the Danish population as the standard, the relative risk for breast cancer was 0.9, based on 43 case users. Duration of use among these subjects was not known. In a study using several Northern California Kaiser Permanente pharmacy databases, cimetidine use between 1982 and 1987 was not associated with breast cancer during follow-up through 1995 (relative risk, 1.09; 95% CI, 0.83-1.43, based on 58 case users) and estimates did not vary according to the length of time since the prescription was filled (9). This study also reported a relative risk of 0.92 (95% CI, 0.49-1.71) for the use of ranitidine, based on 10 case users. In a cohort of 54,239 people enrolled in the Group Health Cooperative of Puget Sound who had filled two or more prescriptions for an H₂ blocker within a 6-month period between 1987 and 1995, the risk of breast cancer among cimetidine users was similar to the risk among users of other H₂ blockers (standardized incidence ratio, 1.0; 95% CI, 0.9-1.1) based on 507 case users (10). Risk did not vary by the number of prescriptions filled or by the time since the first prescription was filled. There was no association between use of other H₂ blockers and breast cancer in this data set.

Although recall bias is of concern in case-control studies, it is not likely to have played a role here, because study subjects and interviewers were blind to this particular hypothesis, and gastric ulcer as an indication for drug use was just one among a long list of indications about which subjects were queried. In addition, results were similar using both cancer and noncancer controls. Selection bias is a theoretical concern. However, we chose patients with diagnoses that were unrelated to cimetidine use as controls, and the prevalence of regular cimetidine use was the same in both subgroups of controls, suggesting that the control groups were appropriate. A limitation of this study is that drug use information was self-reported. However, we focused on regular use of at least 3 months' duration, which is more likely to be remembered than sporadic use. There is no alternative to self-reporting in the case of an over-the-counter drug, which cimetidine has been since 1995. The strength of our study compared with the studies that used pharmacy databases was our ability to focus on regular use and to capture over-the-counter use in addition to prescription drug use.

In conclusion, our data support the results of three other studies that indicate that cimetidine is not associated with the risk of breast cancer.

We thank the many physicians who allowed their patients to be interviewed; the nurse-interviewers who collected the data; Jane Sheehan and Marguerite Angeloni, who coordinated data collection; Michael Stanley and Leonard Gaetano, who were responsible for data management; and Dong Hua, M.S., who conducted the data analysis.