Mindfully.org
Home | Air | Energy | Farm | Food | Genetic Engineering | Health | Industry | Nuclear | Pesticides | Plastic
Political | Sustainability | Technology | Water



Antidepressant Medication Use and Breast Cancer Risk 

American Journal of Epidemiology Vol. 152, No. 11 : 1104-1016 1dec00

Debbie A. Lawlor

Department of Social Medicine University of Bristol Bristol, United Kingdom

The issue of a possible association between antidepressant medication and increased risk of breast cancer is important. In the developed world, breast cancer is very common, and large numbers of women are given prescriptions for and take antidepressants. The need for quality research in this area is therefore paramount. I am concerned about the conclusions reached in the paper by Cotterchio et al. (1). From their results, they draw three main conclusions: 1) that there is no association between "ever" use of antidepressant medication and increased breast cancer risk; 2) that the use of tricyclic antidepressants for 2 years or more is associated with a twofold increase in risk; and 3) that ever use of the specific selective serotonine reuptake inhibitor (SSRI), paroxetine, is associated with a sevenfold increase in risk.

The first of these conclusions seems justified on the basis of the results and the a priori hypothesis. However, the next two conclusions are not justified. It is not surprising that if multiple subgroup analyses are undertaken to find positive results, this will occur by chance. In this case, the positive results claimed by the authors are not actually statistically significant. In multivariate analysis, the odds ratio for risk of breast cancer associated with use of tricyclic antidepressants for more than 25 months is 2.1 (95 percent CI: 0.9, 5.0); p values were not presented in the paper but when calculated from the log of the odds ratio and confidence interval in this case, p = 0.09. For the association between paraxetine and breast cancer, the odds ratio is 7.2 (95 percent CI: 0.9, 58.3, p = 0.06).

Given the low number of cases, the lack of statistical significance, the problem of multiple analyses, and recall bias, these conclusions cannot be accepted. Interestingly (and unusually), in the introduction, one specific SSRI (i.e., Prozac, Eli Lilly and Company, Indianapolis, Indiana), along with its manufacturer, is used to define an SSRI. It is unknown whether any potential conflicts of interest from associations with this manufacturer exist, but if they did this would add further to my concerns about the bias of this paper.

REFERENCES

  1. Cotterchio M, Kreiger N, Darlington G, et al. Antidepressant medication use and breast cancer risk. Am J Epidemiol 2000;151:951–7.
    [Abstract Below]

Katherine L. Beebe, Rocco Zaninelli and Bruce Trock

SmithKline Beecham Pharmaceuticals Research and Development, North America Collegeville, PA 19426–0989
Lombardi Cancer Center Georgetown University Medical Center Washington, DC 20007

In a recent population-based case-control study, Cotterchio et al. (1) reported on the association between breast cancer risk and use of antidepressants. They found no increase in risk with "ever use" of antidepressants in general or with specific classes of drugs, an increased risk with prolonged use only for tricyclic medication, and a potentially large, but very imprecisely estimated, risk associated with the selective serotonin reuptake inhibitor (SSRI) paroxetine (odds ratio 5 7.2, 95% confidence interval: 0.9, 58.3) (1). However, their abstract concluded that "use of paroxetine may be associated with a substantial increase in breast cancer risk" (1, p. 951). This observation was based on only nine cases and one control out of 1,403 women in the study. Given that associations based on such small numbers are highly susceptible to biases other than sampling error, greater caution should have been used in describing this association in the abstract.

Moreover, the authors' proposed mechanism to explain this finding lacks biologic plausibility. Paroxetine was introduced in 1993, so no one in the study could have had more than 3 years of exposure. No known breast carcinogens have such a short latency period. The strongest known breast carcinogen, exposure to ionizing radiation from the atomic bomb explosion in Hiroshima and Nagasaki, produced increases in breast cancer incidence only 10 or more years after the blasts and very little excess risk in those aged 40 years or older at the time of the initial exposure (2, 3). Theoretically, exposure to an agent that affects a late stage of carcinogenesis, i.e., promotion or progression, could reduce the latency period. Beyond that, mathematical models that provide a good fit to breast cancer incidence data still assume a period of 2.5–7 years from the time of malignant transformation until clinical detection (4).

The authors state that one possible mechanism of their finding is stimulation of prolactin secretion by paroxetine (1). It should be noted that the product information for fluoxetine, sertraline, and paroxetine mentions voluntary reports of adverse events involving hyperprolactinemia and galactorrhea that are temporally associated with each of these agents. Nevertheless, there was no evidence from chronic toxicology or 2-year carcinogenicity studies in laboratory animals that would indicate a paroxetine-related increase in the spectrum of toxicologic pathology lesions that are consistent with hyperprolactinemia (SmithKline Beecham, data on file). Furthermore, even if such stimulation occurs, it is unlikely to produce an increase in risk as large as that observed by Cotterchio et al. Human studies of prolactin as a breast cancer risk factor are inconsistent, and much smaller risks are observed (5). Even exposure to exogenous estrogens produces only a twofold or smaller risk and only after prolonged exposure (6, 7).

Another potential mechanism proposed by Cotterchio et al. is inhibition of CYP2D6 (1). Most studies that have evaluated the risk associated with this enzyme system, however, have considered genetic polymorphisms rather than epigenetic phenomena. A recent review and meta-analysis found only a weak, nonsignificant risk of breast cancer associated with poor metabolizer genotype or phenotype (pooled odds ratio = 1.36, 95 percent confidence interval: 0.96, 1.91) (8). Moreover, other studies have found that the SSRI fluoxetine, which was associated with breast cancer risk in the study by Cotterchio et al. (1), is a potent inhibitor of CYP2D6 (9).

Reasons other than biologic plausibility suggest that the purported association with paroxetine is spurious. Among controls, the rate of exposure to paroxetine (0.1 percent) is much lower than the rate of exposure to sertraline (1.1 percent), which is surprising given that both agents were introduced to the market within a year of each other and had a similar volume of use in Canada during the time the study was conducted (SmithKline Beecham, data on file). The apparent increase in risk could therefore represent an artificially low rate of self-reported paroxetine use among controls. A similar bias has been observed previously in epidemiologic studies, in which it may have contributed to an apparent increase in reported risk of congenital anomalies associated with vaginal spermicide use (10).

Because paroxetine was a fairly new drug during the time period when cases were diagnosed, postdiagnosis prescriptions may have switched patients to paroxetine from another drug. Because the total number of paroxetine users in this study is so small, only a few such cases would produce a large bias. This scenario seems probable, given the fact that 21 of the 22 cases who took an SSRI (1, table 1) were exposed to fluoxetine (1, table 3), indicating that the majority of cases exposed to paroxetine had prior exposure to fluoxetine. We feel that the tenuous conclusions of the paper by Cotterchio et al. regarding the relation of paroxetine to breast cancer are overstated in their abstract.

REFERENCES

  1. Cotterchio M, Kreiger N, Darlington G, et al. Antidepressant medication use and breast cancer risk. Am J Epidemiol 2000;151:951–7. [Medline]

  2. Boice JD Jr. Ionizing radiation. In: Schottenfeld D, Fraumeni JF, eds. Cancer epidemiology and prevention. New York, NY: Oxford University Press, 1996:319–53.

  3. Tokunaga M. Malignant breast tumors among atomic bomb survivors, Hiroshima and Nagasaki, 1950–74. J Natl Cancer Inst 1979;62:1347–59. [Medline]

  4. Moolgavkar SH. Two-stage model for carcinogenesis: epidemiology of breast cancer in females. J Natl Cancer Inst 1980;65:559–69. [Medline]

  5. Bernstein L, Ross RK. Endogenous hormones and breast cancer risk. Epidemiol Rev 1993;15:48–65. [Medline]

  6. Brinton LA, Schairer C. Estrogen replacement therapy and breast cancer risk. Epidemiol Rev 1993;15:66–79. [Medline]

  7. Malone KE, Daling JR, Weiss NS. Oral contraceptives and breast cancer risk. Epidemiol Rev 1993;15:80–97. [Medline]

  8. Dunning Am, Healey CS, Pharaoh PDP, et al. A systematic review of genetic polymorphisms and breast cancer risk. Cancer Epidemiol Biomarkers Prev 1999;8:843–54. [Medline]

  9. Alfaro CL, Lam YW, Simpson J, et al. CYP2D6 status of extensive metabolizers after multiple-dose fluoxetine, fluvoxamine, paroxetine, or sertraline. J Clin Psychopharmacol 1999;19:155–63. [Medline]

  10. Oakley GP. Spermicides and birth defects. (Editorial). JAMA 1982;247:2405. [Medline]

TWO OF THE AUTHORS REPLY

Michelle Cotterchio and Nancy Kreiger

Research Unit, Division of Preventive Oncology, Cancer Care Ontario Toronto, Ontario, M5G 2L7 Canada

We thank Drs. Lawlor (1) and Beebe et al. (2) for their letters regarding our recently published article on antidepressant use and breast cancer risk (3). They express concern that we have overstated the association between paroxetine and breast cancer risk. Our abstract concluded, however, that "use of paroxetine may be associated with a substantial increase in breast cancer risk" (italics added by author) (3, p. 951). We deliberately chose the word "may" because the results of our multivariate analysis were of borderline significance (i.e., the lower 95 percent confidence interval was 0.9). In addition, we indicated in our paper that whether or not our finding of increased risk among the few paroxetine users is due to chance can be determined only by future studies. These studies should help to determine whether our finding is spurious, and we are currently conducting such a study.

Beebe et al. (2) argue that the association between paroxetine and breast cancer risk may be due to the lower-than-expected rate of exposure among the controls. Assuming their unpublished "volume of use" data (2, p. 1105) are correct and assuming that "volume of use" is analogous to expected exposure rates, this suggestion is a plausible explanation for our finding. We believe it is unlikely, however, that the controls underreported antidepressant use, since the proportion of the controls who reported use (8 percent) is actually slightly higher than proportions from earlier studies that collected data predominantly during the 1970s and 1980s (3–6 percent) (4–6). Furthermore, it is hard to imagine why only one particular antidepressant would be underreported by controls. (Given that some of the data cited by Beebe et al. (2) are unpublished, we cannot comment on, for example, the prevalence of paroxetine exposure and hyperprolactinemia).

While we did not suggest that paroxetine is a carcinogen, we believe that it may be a promoter. Animal studies support the hypothesis that antidepressants may be tumor promoters (7). Finally, we clarify that all reports of postdiagnosis anti-depressant medication use were appropriately excluded from our analysis.

REFERENCES

  1. Lawlor DA. Re: "Antidepressant medication use and breast cancer risk." (Letter). Am J Epidemiol 2000;152:1104. [Full Text]

  2. Beebe KL, Zaninelli R, Trock B. Re: "Antidepressant medication use and breast cancer risk." (Letter). Am J Epidemiol 2000;152:1104–5. [Full Text]

  3. Cotterchio M, Kreiger N, Darlington G, et al. Antidepressant medication use and breast cancer risk. Am J Epidemiol 2000;151:951–7. [Medline]

  4. Kelly JP, Rosenberg L, Rao RS, et al. Is use of antidepressants associated with the occurrence of breast cancer? (Abstract). Am J Epidemiol 1998;147:S69.

  5. Harlow BL, Cramer DW, Baron JA, et al. Psychotropic medication use and risk of epithelial ovarian cancer. Cancer Epidemiol Biomarkers Prev 1998;7:697–702. [Medline]

  6. Wallace RB, Sherman, BA, Bean JA. A case-control study of breast cancer and psychotropic drug use. Oncology 1982;39:279–83. [Medline]

  7. Brandes LJ, Arron RJ, Bogdanovic RP, et al. Stimulation of malignant growth in rodents by antidepressant drugs at clinically relevant doses. Cancer Res 1992;52:3796–800. [Medline]

Antidepressant medication use and breast cancer risk.

Am J Epidemiol 2000 May 15;151(10):951-7

Cotterchio M, Kreiger N, Darlington G, Steingart A

Division of Preventive Oncology, Cancer Care Ontario, Toronto, Canada.

Experimental and epidemiologic studies suggest that antidepressant medication use may be associated with breast cancer risk. This hypothesis was investigated using a population-based case-control study; cases diagnosed in 1995-1996 were identified using the Ontario Cancer Registry, and controls were randomly sampled from an Ontario Ministry of Finance database. Data were collected using a self-administered questionnaire, and multivariate logistic regression was used to estimate odds ratios and 95% confidence intervals. Adjusted odds ratio estimates ranged from 0.7 to 0.8 and were not statistically significant for "ever" use of antidepressants, tricyclics, and selective serotonin reuptake inhibitors. Compared with no antidepressant use, use of tricyclic antidepressants for greater than 2 years' duration was associated with an elevated risk of breast cancer (odds ratio (OR) = 2.1, 95% confidence interval (CI): 0.9, 5.0). Of the six most commonly reported antidepressant medications, only paroxetine use was associated with an increase in breast cancer risk (OR = 7.2, 95% CI: 0.9, 58.3). Results from this study do not support the hypothesis that "ever" use of any antidepressant medications is associated with breast cancer risk. Use of tricyclic medications for greater than 2 years, however, may be associated with a twofold elevation, and use of paroxetine may be associated with a substantial increase in breast cancer risk.

If you have come to this page from an outside location click here to get back to mindfully.org