Skip Navigation


American Journal of Epidemiology Advance Access originally published online on July 27, 2007
American Journal of Epidemiology 2007 166(6):709-716; doi:10.1093/aje/kwm216
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
166/6/709    most recent
kwm216v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kirsh, V. A.
Right arrow Articles by Theis, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kirsh, V. A.
Right arrow Articles by Theis, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology © The Author 2007. Published by the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.

ORIGINAL CONTRIBUTIONS

Nonsteroidal Antiinflammatory Drug Use and Breast Cancer Risk: Subgroup Findings

Victoria A. Kirsh1,2, Nancy Kreiger1,2,3, Michelle Cotterchio1,2, Margaret Sloan1 and Beth Theis1

1 Division of Preventive Oncology, Research Unit, Cancer Care Ontario, Toronto, Ontario, Canada
2 Department of Public Health Sciences, University of Toronto, Toronto, Ontario, Canada
3 Department of Nutritional Sciences, University of Toronto, Toronto, Ontario, Canada

Correspondence to Dr. Victoria A. Kirsh, Division of Preventive Oncology, Cancer Care Ontario, Toronto, Ontario, Canada M5G 2L6 (e-mail: vicki.kirsh{at}cancercare.on.ca).

Received for publication August 22, 2006. Accepted for publication January 31, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Nonsteroidal antiinflammatory drugs (NSAIDs) may play a role in breast cancer prevention; however, breast cancer subtypes and lifestyle/host factors may influence their impact. During 1996–1998 in Canada, the authors examined the association between regular NSAID use (defined as daily use for at least 2 months) and breast cancer risk by estrogen receptor (ER) and progesterone receptor (PR) status, cigarette smoking exposure, and history of arthritis. Breast cancer cases (n = 3,125, including 1,600 ER+PR+ and 591 ER–PR–) and an age-matched, random sample of controls (n = 3,062) completed a general risk factor questionnaire, including detailed questions on prescription and nonprescription NSAID use. NSAID use was associated with reduced risk of breast cancer (odds ratio = 0.76, 95% confidence interval: 0.66, 0.88). The association was not significantly different for ER+PR+ (odds ratio = 0.71, 95% confidence interval: 0.60, 0.84) and ER–PR– cancers (odds ratio = 0.80, 95% confidence interval: 0.62, 1.03) (pheterogeneity = 0.66). The magnitude of the NSAID inverse association was similar for women with and without arthritis and across smoking strata (risk estimates ranged from 0.74 to 0.84). Breast cancer risk tended to decrease with increasing duration of NSAID use and was generally lowest for ≥7 years of use, and both acetylsalicylic acid and non–acetylsalicylic acid use were associated with reduced risks.

anti-inflammatory agents, non-steroidal; arthritis; breast neoplasms; receptors, estrogen; receptors, progesterone; smoking

Abbreviations: CI, confidence interval; ER, estrogen receptor; NSAID, nonsteroidal antiinflammatory drug; OR, odds ratio; PR, progesterone receptor


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Nonsteroidal antiinflammatory drugs (NSAIDs) such as ibuprofen and aspirin are widely used to treat inflammation, pain, and fever (1), and they may have utility as anticancer agents (2). Inhibition of one or both of the two isoforms of the cyclooxygenase enzymes is thought to be the basis by which NSAID use may lower cancer risk; cyclooxygenase-1 is expressed in nearly all cells, and cyclooxygenase-2 is induced during inflammation or by tumor oncogenes and promoters and is overexpressed in breast cancers (3). Inhibition of the cyclooxygenase enzymes by NSAIDs leads to reduced prostaglandin production, which, in addition to blocking inflammatory reactions, may inhibit mutagenesis and angiogenesis and promote apoptosis (3). The experimental and epidemiologic evidence for the efficacy of long-term NSAID prophylaxis is strongest for colorectal cancer (4); the evidence for breast cancer remains inconclusive (5–10).

Recently, the Women's Health Study, a randomized 2 x 2 factorial chemoprevention trial including nearly 40,000 US women (11), refuted the accumulating observational evidence (10, 12–15) regarding a role for NSAIDs in the chemoprevention of breast cancer. Women who were randomly assigned to receive low-dose (100-mg) aspirin every other day and were followed for an average of 10 years did not experience a reduction in breast cancer risk compared with those receiving placebo (11). Unexpectedly, nonsmokers experienced a higher incidence of breast cancer associated with NSAID use, and past smokers had a reduced risk.

Findings from the Women's Health Study trial raise important issues that require further investigation to reconcile the discordance between the observational studies and the randomized controlled trial. While it is possible that NSAIDs may function as an antiinflammatory agent only when given in higher doses (16), other factors may be relevant. First, proinflammatory processes could increase the effectiveness of NSAIDs in chemoprevention; accordingly, women who elect to use NSAIDs for underlying conditions such as arthritis or chronic pain may be more likely to benefit from NSAIDs than women without such chronic illnesses. Similarly, proinflammatory tobacco carcinogens may modify the efficacy of NSAIDs (11). In fact, emerging data from several epidemiologic studies provide evidence of a chemopreventive agent–smoking interaction in cancer etiology (17).

In addition to these host characteristics, heterogeneity in disease subtype may lead to differential effects of chemopreventive agents. An inverse association between long-term, regular aspirin use and breast cancer, restricted to tumors that were estrogen receptor (ER) or progesterone receptor (PR) positive, was recently reported in one cohort (18) and one case-control study (19); the Women's Health Study did not replicate this subgroup finding (11) nor did two additional investigations (14, 20).

In an earlier report, we observed an inverse association between NSAID use and breast cancer risk (21). Here, we examined the association between NSAID use and breast cancer risk by smoking status, history of arthritis, and ER and PR status.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Cases and controls
Cases were women aged 25–74 years with incident breast cancer (International Classification of Diseases, Ninth Revision, code 174), sampled from pathology reports received by the population-based Ontario Cancer Registry between July 1996 and September 1998. Controls were a random sample of women identified from the assessment rolls of the Ministry of Finance, which documents all residents (owners and tenants) of Ontario; controls were frequency matched to cases within 5-year age groups.

The Ontario Cancer Registry records all new cases of cancer in Ontario, most within 3 months of diagnosis, and is estimated to be more than 98 percent complete (22). The system relies predominantly on passive collection of pathology reports with any mention of cancer, electronic hospital discharge summaries and day surgery reports with cancer diagnoses, electronic records from regional cancer treatment centers, and electronic reports of deaths in which cancer is the underlying cause. Computerized probabilistic record linkage is used to reconcile information and to identify duplicates. The Ontario Ministry of Finance database is an accurate representation of the source population that gave rise to the cases; a 1994 reabstraction study (n = 1,192) showed that more than 95 percent of the Ontario Cancer Registry cases were included in the Ministry of Finance database (23).

Receptor status was determined by biochemical assays routinely conducted by four Ontario hospital laboratories, or from immunohistochemical assay findings recorded in pathology laboratory reports, on the basis of methods previously described (24). Briefly, hormone receptor status was defined as positive if the biochemical findings indicated a concentration of estrogen and progesterone binding protein of >10 fmol/mg or if the immunohistochemical assays were positive for antibody nuclear staining; if the binding protein concentration was ≤10 fmol/mg or if antibody nuclear staining was absent, hormone receptor status was defined as negative. Cases for whom ER and/or PR status was missing were excluded from the analyses by receptor status.

Complete ER/PR status was obtained for 84 percent of the 3,125 cases (n = 2,626) who provided epidemiologic data. For the remaining cases, the tissue sample was not sent, the tissue sample was too small, or there were problems with the assay (7 percent, n = 230); no result was available (6 percent, n = 189); or either ER or PR data were missing (3 percent, n = 80). Sixty-three percent of the ER/PR data came from the four hospital laboratories (biochemical assay) and 37 percent from pathology laboratory reports (immunohistochemical assay).

Data collection
As described in our previous report (21), physicians were identified from pathology reports and were asked to provide the vital status of the patient, to confirm the diagnosis, to supply or verify the address and telephone number, and to give consent to contact the patient (response rate, 89 percent). A questionnaire was mailed to cases and controls with a $5 incentive included in the mailing to controls. Postcard reminders were sent within 2 weeks, and nonresponding subjects became eligible for follow-up telephone calls within 4 weeks.

Cases (response rate, 73 percent) and controls (response rate, 61 percent) completed a self-administered questionnaire that solicited information on breast cancer risk factors and details about current and previous drug use, including use of prescription and over-the-counter medications to control pain or inflammation, taken daily for at least 2 months. Included in the questionnaire were color photographs of common NSAIDs at various doses: acetylsalicylic acid (i.e., aspirin, Entrophen (PendoPharm, Quebec, Canada), Novasen (Novopharm, Ontario, Canada)), ibuprofen (i.e., Motrin (McNeil Consumer Products Company, Fort Washington, Pennsylvania)), indomethacin (i.e., Novo-methacin (Novopharm)), piroxicam (i.e., Apo-piroxicam (Apotex, Ontario, Canada), Novo-pirocam (Novopharm)), diclofenac sodium (i.e., Voltaren, Voltaren SR novo-difenac (Novartis Pharmaceuticals Corporation, East Hanover, New Jersey)), and naproxen (i.e., Apo-naproxen (Apotex), Novo-naprox (Novopharm)). Participants were also asked to list medications used that were not specifically pictured, which were classified as acetylsalicylic acid or non–acetylsalicylic acid according to the main active ingredient. In addition to those using the acetylsalicylic acid trade names mentioned above, we included women taking acetylsalicylic acid (not otherwise specified) and Fiorinal (Novartis) in the subgroup of acetylsalicylic acid users. Those who used NSAIDs for less than 2 months were considered nonusers. Duration of use was reported by participants as the total length of time that the medication was taken throughout their lives. To determine the number of years since last NSAID use, we subtracted the age at which the participant reported having last taken the medication from her current age. We also solicited information on smoking history, history of migraine headaches, and physician-diagnosed arthritis.

Statistical analysis
Breast cancer cases were stratified by joint receptor expression as follows: ER+PR+, ER+PR–, ER–PR+, and ER–PR–. The total case series included 1,600 ER+PR+ (60.9 percent), 591 ER–PR– (22.5 percent), 289 ER+PR– (11.0 percent), and 146 ER–PR+ (5.6 percent) breast cancers. Odds ratios associating NSAID use with breast cancers of differing hormone receptor status were computed by using polytomous logistic regression, adjusting for age, history of arthritis, and benign breast disease; subgroup analyses by smoking status and history of arthritis were performed by using logistic regression. We evaluated the potential for confounding by hormone replacement therapy, oral contraceptive use, alcohol consumption, tobacco smoking, body mass index, physical activity, history of migraines, reproductive history (including age at menarche, age at menopause, childbearing, and breastfeeding), education, marital status, previous breast cysts, family history of breast cancer, other medication use, and dietary fat intake. None of these variables was associated with a change in the risk estimates of 10 percent or more. To test for differences in odds ratios across ER and PR categories, we calculated a p value for heterogeneity by using the –2-log likelihood statistic. Tests for multiplicative interaction were obtained by including a cross-product term of NSAID use and the stratification variable and by testing the significance using the Wald {chi}2 statistic or the –2-log likelihood statistic, as appropriate. Tests for trend were based on the continuous variable. All p values were two sided. Statistical analysis for polytomous logistic regression was performed by using STATA software (Stata Corporation, College Station, Texas); other analyses were performed with SAS software (SAS Institute, Inc., Cary, North Carolina).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Regular NSAID use (defined as daily use of any NSAID medication for at least 2 months) was reported by 26 percent of the controls (n = 786). Acetylsalicylic acid was the most commonly used NSAID, with 14 percent (n = 418) of the controls reporting regular use. Among regular acetylsalicylic acid users for whom data on dose were available (86 percent), the median daily dose was 325 mg, with almost 78 percent reporting an intake at that dosage level. Compared with nonusers, regular NSAID users were older, were more likely to have a body mass index of >27 kg/m2, were current smokers, were users of hormone replacement therapy, and were more likely to report a history of migraine headaches and arthritis (table 1, among controls). Findings were similar in the case group. Cases were more likely to have a family history of breast cancer, personal history of benign breast cysts or lumps, higher educational level, earlier age at menarche, and lower parity (data shown elsewhere (21)).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Description of selected characteristics of controls* by use of NSAIDs,{dagger} Canada, 1996–1998

 
As reported previously, women who reported regular NSAID use had a statistically significant 24 percent lower risk of breast cancer (odds ratio (OR) = 0.76, 95 percent confidence interval (CI): 0.66, 0.88) (21). The decreased risk associated with regular NSAID use did not differ significantly by hormone receptor status (pheterogeneity = 0.66) (table 2); however, the inverse association was slightly more pronounced for those with joint ER+ and PR+ breast cancer (ER+PR+) (OR = 0.71, 95 percent CI: 0.60, 0.84). A modest inverse association was also evident for those whose hormone receptor status was negative (ER–PR–) (OR = 0.80, 95 percent CI: 0.62, 1.03). The associations did not differ when acetylsalicylic acid and non–acetylsalicylic acid were considered separately. Increasing duration of NSAID use was associated with decreased risk of breast cancer in the unstratified analyses (21); no clear differences in this trend emerged when duration of, and time since last, NSAID use were compared across breast cancer subsets. The inclusion of benign breast cysts in the multivariate models did not consistently lead to attenuation in the odds ratio, and conclusions did not change.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Odds ratios (with 95% confidence intervals) of breast cancer by regular NSAID* use{dagger} and joint ER* and PR* status, Canada, 1996–1998

 
The odds ratios associating NSAID use with breast cancer risk were comparable between nonsmokers (50 percent), former smokers (30.6 percent), and current smokers (18.3 percent), with estimates within the range of 0.74–0.84 (pinteraction = 0.71) (table 3). Analyses of type of NSAID, duration of use, and time since last use showed similar risk estimates for non-, current, former, and heavy smokers.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Odds ratios (with 95% confidence intervals) of breast cancer by regular NSAID* use{dagger} and smoking status, Canada, 1996–1998

 
The inverse association between ever use of NSAIDs and breast cancer risk was evident for women who reported a history of physician-diagnosed arthritis (35.0 percent) (OR = 0.77, 95 percent CI: 0.64, 0.93) as well as for those who did not (OR = 0.75, 95 percent CI: 0.60, 0.93) (pinteraction = 0.77) (table 4). Analyses of type of NSAID, duration of use, and time since last use yielded similar risk estimates for women with and without arthritis.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Odds ratios (with 95% confidence intervals) of breast cancer by regular NSAID* use{dagger} and physician-diagnosed arthritis, Canada, 1996–1998

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
In our analysis of more than 3,000 breast cancer cases, the inverse association between breast cancer risk and regular, high-dose NSAID use was consistent across strata defined by hormone receptor status, smoking status, and history of arthritis. The various subgroup estimates were not markedly different from the overall analysis, which indicated a 24 percent lower risk of breast cancer for women who reported daily use of any NSAID medication for at least 2 months at some point in their lives (21). The reduction in risk was similar for acetylsalicylic acid NSAIDs—which are far more potent inhibitors of cyclooxygenase-1 than cyclooxygenase-2—and non–acetylsalicylic acid NSAIDs, which inhibit both cyclooxygenase-1 and cyclooxygenase-2 to varying degrees, depending on the specific medication (25).

Our findings—and those of others (10, 13, 26, 27)—of an inverse association between high-dose NSAID use and reduced risk of breast cancer are not refuted by the null findings from the Women's Health Study chemoprevention trial, which included nearly 40,000 women (including 1,230 breast cancer cases). While the Women's Health Study results were unexpected, the aspirin dose administered (100 mg every other day) was much lower than the median daily acetylsalicylic acid dose reported by the women in our study (325 mg). Correspondingly, the Women's Health Initiative Observational Study Cohort, which included more than 80,000 women (1,392 breast cancer cases), reported a 21 percent reduced risk of breast cancer for women with 5 or more years of NSAID use and a 28 percent reduced risk associated with 10 or more years of use; these associations were no longer apparent when analyses were restricted to low-dose aspirin (<100 mg/day) (10). A Canadian prescription database study found that frequent use of aspirin at a dose greater than 100 mg/day was associated with a decreased risk of breast cancer, whereas lower doses were not (26). In contrast, in a large, nested-case control study (including 3,708 breast cancer cases) (28), the authors noted a 33 percent reduced risk among women taking 75 mg of aspirin daily, but they did not find statistically significant reductions with higher doses (150 mg and 300 mg). Many studies focusing on low-dose NSAIDs have indeed reported a reduced risk of breast cancer (12); thus, the totality of the evidence regarding a dose-response effect remains inconclusive.

In light of the evidence suggesting that different molecularly defined subsets of breast cancer have different risk factor profiles (29–31), we evaluated the association between NSAID use and risk of breast cancer by ER and PR status. The possibility of a differential effect according to receptor status is supported biologically. Prostaglandin E2 has been shown to stimulate aromatase transcription, which leads to increased estrogen levels (32), which in turn could lead to progression of estrogen-dependent breast cancer (i.e., ER+ disease) (3). If the method of action of cyclooxygenase inhibitors is predominantly through inhibition of aromatase, this points to a reduction in risk for mainly receptive-positive tumors.

We observed an inverse association between acetylsalicylic acid and non–acetylsalicylic acid NSAID use and breast cancer risk for both hormone-receptor-positive and -negative breast tumors, results that do not reproduce the selective beneficial effect found in certain other studies. Among women in the California Teachers Study (n = 2,391 cases), long-term (≥5 years) daily aspirin use was associated with a nonstatistically significant decreased risk of ER/PR-positive breast cancer (includes tumors that were either ER+ or PR+) (OR = 0.80, 95 percent CI: 0.62, 1.03) but a statistically significant increased risk of ER–PR– breast cancer (OR = 1.81, 95 percent CI: 1.12, 2.92); results were not significant, nor did they differ by receptor status when ibuprofen use alone was considered (18). Similarly, the Long Island Breast Cancer Study Project (n = 1,442 cases and 1,420 controls) reported a reduction in risk of ER/PR-positive breast cancer (evident for subgroups in which either ER or PR was positive) associated with regular aspirin use (≥4 pills/week for ≥3 months) (OR = 0.71, 95 percent CI, 0.55, 0.93), but not for those with ER–PR– breast cancer (19). In contrast, results from the Cancer Prevention Study II Nutrition Cohort (n = 3,008 cases) indicated that long-duration, regular NSAID use (≥30 pills/month for ≥5 years) was not associated with breast cancer risk overall, nor was it associated with ER+PR+ breast cancer (relative risk = 1.15, 95 percent CI: 0.85, 1.56) (20). The Women's Health Study also did not find a protective effect of NSAIDs on risk of ER+PR+ breast cancer (11), and another study noted that the decreased risk of breast cancer they observed did not vary according to hormone receptor status (14).

Chronic inflammatory conditions induced by infectious and other etiologic agents have been associated with cancer development; examples include the well-established associations between Helicobacter pylori infection and risk of stomach cancer, hepatitis B and C viruses and risk of liver cancer, inflammatory bowel disease and risk of colorectal cancer, reflux esophagitis and risk of esophageal cancer, and smoking-associated bronchitis and risk of lung cancer (33). We reported results stratified by smoking status in view of this body of research and, to compare with findings from the Women's Health Study that indicated an inverse association between NSAID use and risk of breast cancer, restricted to former smokers. Evidence suggests that smoking might induce cyclooxygenase-2 expression (34, 35), so, in theory, smokers may have more to gain from a chemopreventive agent that inhibits cyclooxygenase-2 expression. We nevertheless did not find a differential effect by smoking status. Tobacco use itself is not unequivocally associated with breast cancer (36), so the Women's Health Study findings stratified by smoking status are not compelling and may be, as the authors note, simply a chance finding from multiple testing.

We assessed the potential for effect modification by history of arthritis, reasoning that NSAID use might be particularly beneficial for breast cancer prevention among a subgroup of individuals whose expression of cyclooxygenase enzymes is elevated (37), similar to smokers. Our results, however, were not modified by arthritis status. The interpretation of this result is limited; women in our study sample who reported regular use of NSAIDs may represent a homogenous group, all of whom may have underlying chronic inflammation, manifesting as arthritis or types of pain. A limitation of our study was that we could not evaluate the effect of low-dose NSAID use on breast cancer risk. We largely captured use of high-dose medication for chronic pain or inflammation. Accordingly, dose options listed in the questionnaire tended to cover the higher range of exposure (e.g., 325 mg of aspirin, 300–600 mg of ibuprofen, 125–500 mg of naproxen), and the number of users and variation in exposure within any particular type of NSAID were insufficient to examine dose response. Furthermore, in instances where participants listed medications used that were not specifically pictured, we did not solicit data on dose. It should be noted that because intermittent users of low-dose NSAIDs would have been included in the nonuser category, observed associations would, if anything, underestimate true association between NSAIDs and breast cancer.

Response rates were somewhat low among the cases (73 percent) and lower among the controls (61 percent), introducing the potential for selection bias in the data. If those controls who participate are likely to be more health conscious and use NSAIDs as heart disease prophylaxis, it could bias the odds ratio away from the null. Refuting this possibility, we noted that among controls, those with higher levels of education—an indicator of health-conscious behavior—were not more likely to use NSAIDs. Furthermore, classic breast cancer risk factors were found to be associated with risk of breast cancer in our data set (21), suggesting that the population-based sampling frame was not compromised.

Information bias is always a concern in case-control studies; however, missing data on NSAID use were similarly distributed between cases and controls (data on regular NSAID use were missing for 2.5 percent of cases and 3.1 percent of controls; duration of use was missing for 9.8 percent of cases and 9.0 percent of controls), and systematic overreporting of medication use by controls seems unlikely. Although several studies found concordance between medical record medication use and self-reported/interview data to be high (38–40), the validity of our exposure assessment remains a potential source of error. We would not, however, expect self-reported use to be biased by disease status, particularly because the purported association between NSAID use and reduced breast cancer risk was not widely known when the data were collected (1996–1998); nondifferential error would not account for our inverse findings.

A potential bias—confounding by indication—seems unlikely in our study because we collected data on the main conditions associated with regular NSAID use. Nearly two thirds of regular NSAID users reported a history of arthritis, and nearly one quarter reported a history of migraines; neither of these conditions was associated with risk of breast cancer in our study. Nevertheless, the possibility of uncontrolled confounding by duration of arthritis, by other potential confounders such as diet, or by other indications for NSAID use cannot be excluded.

The major strengths of our study include the large number of breast cancer cases, which allowed for stratification by ER/PR tumor subtype (although subgroup analyses by duration of use and time since last use were hampered by small numbers), and the use of a population-based cancer registry with nearly complete cancer ascertainment. In addition, detailed information on all types of NSAIDs, including prescription use, together with color photographs of 28 NSAID/dose combinations, likely reduced potential sources of exposure misclassification.

In summary, we found that high-dose NSAID use—for chronic pain or for inflammation—was associated with reduced risk of breast cancer and that this association was not modified by hormone receptor status, tobacco smoking, or arthritis.


   ACKNOWLEDGMENTS
 
Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Thompson PW, Tee L, McBride J, et al. Long-term NSAID use in primary care: changes over a decade and NICE risk factors for gastrointestinal adverse events. Rheumatology (Oxford) (2005) 44:1308–10.[CrossRef][Medline]
  2. Thun MJ, Henley SJ, Patrono C. Nonsteroidal anti-inflammatory drugs as anticancer agents: mechanistic, pharmacologic, and clinical issues. J Natl Cancer Inst (2002) 94:252–66.[Abstract/Free Full Text]
  3. Mazhar D, Ang R, Waxman J. COX inhibitors and breast cancer. Br J Cancer (2006) 94:346–50.[CrossRef][Web of Science][Medline]
  4. Asano TK, McLeod RS. Non steroidal anti-inflammatory drugs (NSAID) and aspirin for preventing colorectal adenomas and carcinomas. Cochrane Database Syst Rev (2004) CD004079.
  5. Schreinemachers DM, Everson RB. Aspirin use and lung, colon, and breast cancer incidence in a prospective study. Epidemiology (1994) 5:138–46.[Web of Science][Medline]
  6. Coogan PF, Rao SR, Rosenberg L, et al. The relationship of nonsteroidal anti-inflammatory drug use to the risk of breast cancer. Prev Med (1999) 29:72–6.[CrossRef][Web of Science][Medline]
  7. Harris RE, Namboodiri KK, Farrar WB. Nonsteroidal antiinflammatory drugs and breast cancer. Epidemiology (1996) 7:203–5.[Web of Science][Medline]
  8. Harris RE, Kasbari S, Farrar WB. Prospective study of nonsteroidal anti-inflammatory drugs and breast cancer. Oncol Rep (1999) 6:71–3.[Web of Science][Medline]
  9. Egan KM, Stampfer MJ, Giovannucci E, et al. Prospective study of regular aspirin use and the risk of breast cancer. J Natl Cancer Inst (1996) 88:988–93.[Abstract/Free Full Text]
  10. Harris RE, Chlebowski RT, Jackson RD, et al. Breast cancer and nonsteroidal anti-inflammatory drugs: prospective results from the Women's Health Initiative. Cancer Res (2003) 63:6096–101.[Abstract/Free Full Text]
  11. Cook NR, Lee IM, Gaziano JM, et al. Low-dose aspirin in the primary prevention of cancer: the Women's Health Study: a randomized controlled trial. JAMA (2005) 294:47–55.[Abstract/Free Full Text]
  12. Khuder SA, Mutgi AB, Johnson TW, et al. Breast cancer and NSAID use: a meta-analysis. Br J Cancer (2001) 84:1188–92.[CrossRef][Web of Science][Medline]
  13. Moorman PG, Grubber JM, Millikan RC, et al. Association between non-steroidal anti-inflammatory drugs (NSAIDs) and invasive breast cancer and carcinoma in situ of the breast. Cancer Causes Control (2003) 14:915–22.[CrossRef][Web of Science][Medline]
  14. Zhang Y, Coogan PF, Palmer JR, et al. Use of nonsteroidal antiinflammatory drugs and risk of breast cancer: the Case-Control Surveillance Study revisited. Am J Epidemiol (2005) 162:165–70.[Abstract/Free Full Text]
  15. Swede H, Mirand AL, Menezes RJ, et al. Association of regular aspirin use and breast cancer risk. Oncology (2005) 68:40–7.[CrossRef][Web of Science][Medline]
  16. Umar A, Viner JL, Anderson WF, et al. Development of COX inhibitors in cancer prevention and therapy. Am J Clin Oncol (2003) 26:S48–57.[Web of Science][Medline]
  17. Mayne ST, Lippman SM. Cigarettes: a smoking gun in cancer chemoprevention. J Natl Cancer Inst (2005) 97:1319–21.[Free Full Text]
  18. Marshall SF, Bernstein L, Anton-Culver H, et al. Nonsteroidal anti-inflammatory drug use and breast cancer risk by stage and hormone receptor status. J Natl Cancer Inst (2005) 97:805–12.[Abstract/Free Full Text]
  19. Terry MB, Gammon MD, Zhang FF, et al. Association of frequency and duration of aspirin use and hormone receptor status with breast cancer risk. JAMA (2004) 291:2433–40.[Abstract/Free Full Text]
  20. Jacobs EJ, Thun MJ, Connell CJ, et al. Aspirin and other nonsteroidal anti-inflammatory drugs and breast cancer incidence in a large U.S. cohort. Cancer Epidemiol Biomarkers Prev (2005) 14:261–4.[Abstract/Free Full Text]
  21. Cotterchio M, Kreiger N, Sloan M, et al. Nonsteroidal anti-inflammatory drug use and breast cancer risk. Cancer Epidemiol Biomarkers Prev (2001) 10:1213–17.[Abstract/Free Full Text]
  22. Holowaty EJ, Theriault ME, Bowler J. Estimating cancer incidence from mortality data: in search of the holy grail? Boston, Massachusetts: Presented at the North American Association of Central Cancer Registries Annual Conference. June 9, 2005.
  23. Holowaty EJ, Lee G, Dale D, et al. A reabstraction study to estimate the accuracy and completeness of data elements in the Ontario Cancer Registry (1994) Niagara-on-the-Lake, Ontario, Canada: American Association of Central Cancer Registries.
  24. Cotterchio M, Kreiger N, Theis B, et al. Hormonal factors and the risk of breast cancer according to estrogen- and progesterone-receptor subgroup. Cancer Epidemiol Biomarkers Prev (2003) 12:1053–60.[Abstract/Free Full Text]
  25. Vane JR, Bakhle YS, Botting RM. Cyclooxygenases 1 and 2. Annu Rev Pharmacol Toxicol (1998) 38:97–120.[CrossRef][Web of Science][Medline]
  26. Rahme E, Ghosn J, Dasgupta K, et al. Association between frequent use of nonsteroidal anti-inflammatory drugs and breast cancer. BMC Cancer (2005) 5:159.[CrossRef][Medline]
  27. Harris RE, Beebe-Donk J, Alshafie GA. Reduction in the risk of human breast cancer by selective cyclooxygenase-2 (COX-2) inhibitors. BMC Cancer (2006) 6:27.[CrossRef][Medline]
  28. Garcia Rodriguez LA, Gonzalez-Perez A. Risk of breast cancer among users of aspirin and other anti-inflammatory drugs. Br J Cancer (2004) 91:525–9.[CrossRef][Web of Science][Medline]
  29. Althuis MD, Fergenbaum JH, Garcia-Closas M, et al. Etiology of hormone receptor-defined breast cancer: a systematic review of the literature. Cancer Epidemiol Biomarkers Prev (2004) 13:1558–68.[Abstract/Free Full Text]
  30. Yasui Y, Potter JD. The shape of age-incidence curves of female breast cancer by hormone-receptor status. Cancer Causes Control (1999) 10:431–7.[CrossRef][Web of Science][Medline]
  31. Potter JD, Cerhan JR, Sellers TA, et al. Progesterone and estrogen receptors and mammary neoplasia in the Iowa Women's Health Study: how many kinds of breast cancer are there? Cancer Epidemiol Biomarkers Prev (1995) 4:319–26.[Abstract]
  32. Harris RE, Robertson FM, Abou-Issa HM, et al. Genetic induction and upregulation of cyclooxygenase (COX) and aromatase (CYP19): an extension of the dietary fat hypothesis of breast cancer. Med Hypotheses (1999) 52:291–2.[CrossRef][Web of Science][Medline]
  33. Shacter E, Weitzman SA. Chronic inflammation and cancer. Oncology (Williston Park) (2002) 16:217–26. 229; discussion 230–2.[Medline]
  34. Badawi AF, Habib SL, Mohammed MA, et al. Influence of cigarette smoking on prostaglandin synthesis and cyclooxygenase-2 gene expression in human urinary bladder cancer. Cancer Invest (2002) 20:651–6.[CrossRef][Web of Science][Medline]
  35. Martey CA, Pollock SJ, Turner CK, et al. Cigarette smoke induces cyclooxygenase-2 and microsomal prostaglandin E2 synthase in human lung fibroblasts: implications for lung inflammation and cancer. Am J Physiol Lung Cell Mol Physiol (2004) 287:L981–91.[Abstract/Free Full Text]
  36. Russo IH. Cigarette smoking and risk of breast cancer in women. Lancet (2002) 360:1033–4.[CrossRef][Web of Science][Medline]
  37. Korotkova M, Westman M, Gheorghe KR, et al. Effects of antirheumatic treatments on the prostaglandin E2 biosynthetic pathway. Arthritis Rheum (2005) 52:3439–47.[CrossRef][Web of Science][Medline]
  38. Kwon A, Bungay KM, Pei Y, et al. Antidepressant use: concordance between self-report and claims records. Med Care (2003) 41:368–74.[CrossRef][Web of Science][Medline]
  39. Norell SE, Boethius G, Persson I. Oral contraceptive use: interview data versus pharmacy records. Int J Epidemiol (1998) 27:1033–7.[Abstract/Free Full Text]
  40. Caskie GI, Willis SL. Congruence of self-reported medications with pharmacy prescription records in low-income older adults. Gerontologist (2004) 44:176–85.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
J. Fortuny, C. Sima, S. Bayuga, H. Wilcox, K. Pulick, S. Faulkner, A. G. Zauber, and S. H. Olson
Risk of Endometrial Cancer in Relation to Medical Conditions and Medication Use
Cancer Epidemiol. Biomarkers Prev., May 1, 2009; 18(5): 1448 - 1456.
[Abstract] [Full Text] [PDF]

Home page
 Arch Intern MedHome page
A. H. Eliassen, W. Y. Chen, D. Spiegelman, W. C. Willett, D. J. Hunter, and S. E. Hankinson
Use of Aspirin, Other Nonsteroidal Anti-inflammatory Drugs, and Acetaminophen and Risk of Breast Cancer Among Premenopausal Women in the Nurses' Health Study II
Arch Intern Med, January 26, 2009; 169(2): 115 - 121.
[Abstract] [Full Text] [PDF]

Home page
 JNCI J Natl Cancer InstHome page
L. R. Howe and S. M. Lippman
Modulation of Breast Cancer Risk by Nonsteroidal Anti-inflammatory Drugs
J Natl Cancer Inst, October 15, 2008; 100(20): 1420 - 1423.
[Full Text] [PDF]

Home page
 Cancer Epidemiol. Biomarkers Prev.Home page
A. G. Hudson, G. L. Gierach, F. Modugno, J. Simpson, J. W. Wilson, R. W. Evans, V. G. Vogel, and J. L. Weissfeld
Nonsteroidal Anti-inflammatory Drug Use and Serum Total Estradiol in Postmenopausal Women
Cancer Epidemiol. Biomarkers Prev., March 1, 2008; 17(3): 680 - 687.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
166/6/709    most recent
kwm216v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Kirsh, V. A.
Right arrow Articles by Theis, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kirsh, V. A.
Right arrow Articles by Theis, B.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?