Skip to main content

Advertisement

Regular use of aspirin and other non-steroidal anti-inflammatory drugs and breast cancer risk for women at familial or genetic risk: a cohort study

Article metrics

Abstract

Background

The use of aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs) has been associated with reduced breast cancer risk, but it is not known if this association extends to women at familial or genetic risk. We examined the association between regular NSAID use and breast cancer risk using a large cohort of women selected for breast cancer family history, including 1054 BRCA1 or BRCA2 mutation carriers.

Methods

We analyzed a prospective cohort (N = 5606) and a larger combined, retrospective and prospective, cohort (N = 8233) of women who were aged 18 to 79 years, enrolled before June 30, 2011, with follow-up questionnaire data on medication history. The prospective cohort was further restricted to women without breast cancer when medication history was asked by questionnaire. Women were recruited from seven study centers in the United States, Canada, and Australia. Associations were estimated using multivariable Cox proportional hazards regression models adjusted for demographics, lifestyle factors, family history, and other medication use. Women were classified as regular or non-regular users of aspirin, COX-2 inhibitors, ibuprofen and other NSAIDs, and acetaminophen (control) based on self-report at follow-up of ever using the medication for at least twice a week for ≥1 month prior to breast cancer diagnosis. The main outcome was incident invasive breast cancer, based on self- or relative-report (81% confirmed pathologically).

Results

From fully adjusted analyses, regular aspirin use was associated with a 39% and 37% reduced risk of breast cancer in the prospective (HR = 0.61; 95% CI = 0.33–1.14) and combined cohorts (HR = 0.63; 95% CI = 0.57–0.71), respectively. Regular use of COX-2 inhibitors was associated with a 61% and 71% reduced risk of breast cancer (prospective HR = 0.39; 95% CI = 0.15–0.97; combined HR = 0.29; 95% CI = 0.23–0.38). Other NSAIDs and acetaminophen were not associated with breast cancer risk in either cohort. Associations were not modified by familial risk, and consistent patterns were found by BRCA1 and BRCA2 carrier status, estrogen receptor status, and attained age.

Conclusion

Regular use of aspirin and COX-2 inhibitors might reduce breast cancer risk for women at familial or genetic risk.

Background

Women vary greatly in their underlying familial risk of breast cancer (BC). Those with an affected first-degree relative are on average at 2-fold increased risk of BC. [1] Women with a BRCA1 or BRCA2 mutation are at about a 10-fold increased risk of BC, depending on their age, family history, and location of mutation [2]. The two leading risk-reduction strategies for women at increased BC risk are risk-reducing mastectomy, which could reduce risk by over 90% [3], and use of medications such as the selective estrogen receptor modulators or aromatase inhibitors, which reduce risk of estrogen receptor (ER)-positive BC by about 30–65% [4,5,6]. Despite the proven efficacy of these options, uptake remains low- and high-risk women often inquire about alternative BC prevention strategies [7,8,9,10,11,12]. Regular use of aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs) including COX-2 inhibitors could be one such alternative. NSAIDs might impede tumor development and growth by modulating cellular proliferation and apoptosis, predominately by suppressing endogenous production of prostaglandin through the inhibition of cyclooxygenase (COX) enzyme activity, particularly COX-2, which is shown to be over-expressed in cancer cells [13, 14]. NSAIDs may also impede the development of ER positive BC through the inhibition of aromatase [13, 15]. The use of aspirin and other NSAIDs for BC prevention is an attractive strategy given that over-the-counter NSAIDs are inexpensive and widely available. However, even if regular NSAID use proves to be an effective BC prevention strategy, as with other risk-reducing options, the potential benefits of NSAIDs will need to be weighed against the potential harms of long-term use [16,17,18,19,20,21].

The cancer prevention effects of aspirin and other NSAIDs are well established for colon cancer [22, 23], and accumulating evidence from epidemiologic studies of women unselected for familial or genetic risk suggests that regular, long-term use of aspirin could reduce BC risk by about 14% [24, 25]. Comparable estimates have been reported for COX-2 inhibitors [26, 27]. However, the current body of evidence is far from conclusive [28], especially given that the only mature randomized controlled trial (RCT) of aspirin and primary prevention of BC did not find evidence for an effect, although no effect was found for colon cancer either [29]. While ongoing secondary prevention trials in women affected with breast cancer, such as the Aspirin for Breast Cancer (ABC) trial and Add-Aspirin trial [30, 31], will also inform this question, results from these trials have yet to be published. Recently published findings from the Aspirin in Reducing Events in the Elderly (ASPREE) found that cancer-related deaths, including BC, were higher in the aspirin group compared to those in the placebo group [21].

Little is known about whether aspirin and other NSAIDs reduce BC risk for women across the familial risk spectrum. For example, no study appears to have estimated the association for BRCA1 and BRCA2 mutation carriers. One study tested the association stratified by first-degree BC family history (12% of the overall sample) and found that regular aspirin use (≥ 6 times per week versus never) was associated with a reduced BC risk both for women with and without an affected first-degree relative (OR = 0.62, 95% CI = 0.41–0.93 and OR = 0.73, 95% CI = 0.61–0.88, respectively) [13]. The Sister Study, a prospective cohort study of women with a sister diagnosed with BC, also found a negative association between lifetime NSAID use (≥ 49 versus < 0.75 pill-years) and BC risk, although only for premenopausal women (HR = 0.66, 95% CI = 0.50–0.87; postmenopausal HR = 0.95, 95% CI = 0.82–1.09) [32]. However, both of these studies relied on a binary definition of family history, which discounts the fact that there is a strong gradient in risk due to underlying familial risk factors such as number of affected relatives and their age at diagnosis. Mathematical modeling demonstrates that in order to explain the average 2-fold increased risk of BC associated with having an affected first-degree relative, the risk of developing BC must vary by approximately 20-fold between people in the lowest quartile of familial risk versus the highest quartile of familial risk [33]. In our family cohort enriched with women with a family history of BC, remaining lifetime risk of BC ranges anywhere from < 1% to > 90% in women unaffected with BC at baseline [34]. It is possible to get a reliable estimate of this underlying familial risk, referred to as familial risk profile, from multi-generational breast and ovarian cancer history data using risk models such as the Breast Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm (BOADICEA), which includes consideration of BRCA1 and BRCA2 gene mutations [35,36,37]. In this study, we employed the BOADICEA model to evaluate associations of regular NSAID use and BC risk by familial risk profile using a large cohort of women enriched for family history, including 1054 women with a BRCA1 or BRCA2 mutation.

Methods

Study sample

This study was based on the Prospective Family Study Cohort (ProF-SC), which comprises baseline and follow-up data from the Breast Cancer Family Registry (BCFR) [38] and the Kathleen Cuningham Foundation Consortium for Research into Familial Breast Cancer (kConFab) [39]; additional details are available elsewhere [34]. These cohorts involved women affected with BC and included their affected and unaffected female relatives; ProF-SC is therefore enriched for familial risk of BC (82% with a first-degree relative and 95% with a first- or second-degree relative with BC). Baseline and follow-up questionnaires asked about personal and family history of BC, demographics, reproductive history, and lifestyle factors. Medication history was not asked about at baseline, but was included on follow-up questionnaires. Women were followed prospectively for cancer and other health outcomes for up to 20 years, and screening for germline BRCA1 and BRCA2 mutations has been conducted over time [39, 40]. The BCFR and kConFab received ethical approval by each participating study center’s institutional review board. All participants provided written informed consent.

Prospective cohort

The prospective cohort included women who were enrolled in the BCFR or kConFab before June 30, 2011, aged 18–79 years at follow-up, who self-reported medication history by follow-up questionnaire, and had not undergone a bilateral mastectomy or been diagnosed with BC prior to follow-up questionnaire (N = 5606). To ensure that we did not include prevalent cancers in the prospective cohort, person-years were calculated from age at 2 months after the questionnaire with medication history was completed to age at first invasive breast cancer diagnosis, based on self- or relative-report and confirmed pathologically for 81% of cases, or censoring (Fig. 1). Women were censored at the earliest of the following events: risk-reducing bilateral mastectomy, age 80 years, loss to follow-up, or death.

Fig. 1
figure1

Overview of the timeline of events in the Prospective Family Study Cohort. Legend: BCFR, Breast Cancer Family Registry; kConFab, Kathleen Cuningham Foundation Consortium for Research into Familial Breast Cancer. The prospective cohort includes women who were enrolled before June 30, 2011, aged 18–79 years at follow-up, with data on regular NSAID use, and with no personal history of breast cancer when regular NSAID use was asked by follow-up questionnaire (N = 5606). The combined cohort includes all women enrolled before June 30, 2011, aged 18–79 years at baseline, with data on regular NSAID use asked by follow-up questionnaire. In both cohorts, women were censored at the earliest of the following events: risk-reducing bilateral mastectomy, age 80 years, loss to follow-up, or death

Combined cohort

The combined, retrospective and prospective, cohort included all women who were enrolled in the BCFR or kConFab before 30 June 2011, aged 18–79 years at baseline, who self-reported medication history by follow-up questionnaire. In addition to women in the prospective cohort, this included women who were diagnosed with BC prior to baseline (N = 1973), women who were diagnosed with BC after baseline but before follow-up questionnaire (N = 232), and women who were censored prior to follow-up questionnaire (N = 422). We excluded 16 cases missing BC diagnosis date, resulting in a final sample of 8233 women in the combined cohort. We used a general modeling approach for the combined cohort that was similar to the prospective cohort, except that we calculated person-years from age 17 years, 1 year prior to earliest age at diagnosis (Fig. 1).

Exposure assessment

Medication history was asked by follow-up questionnaire including use of (1) aspirin, (2) COX-2 inhibitors, (3) ibuprofen and other non-selective NSAIDs, and (4) acetaminophen (paracetamol). Median time between baseline and follow-up questionnaire was 8.7 years. In the BCFR, women in the prospective cohort were asked if they had ever used each of the four types of medication for at least twice a week for 1 month or longer at any time in the past; kConFab participants were asked a slightly modified question that asked if medication use occurred for at least twice a week for more than 1 month. Participants were prompted with country-specific examples of each medication type to help with recall (e.g., Tylenol, Anacin-3, and Panadol were provided as examples of acetaminophen-based medications on the US-based questionnaire). Women in the combined cohort with a personal history of BC were asked if they had ever used (for at least twice a week for ≥ 1 month) the listed medications prior to diagnosis. Women who gave affirmative responses for a given medication were classified as regular users of that drug. Women who used these medications less frequently than twice per week for ≥ 1 month or never were classified as non-regular users. A subset of women (77% of the combined cohort) also reported total duration of regular medication use in either months or years.

Statistical analysis

We used multivariable Cox proportional hazards regression, with age as the time scale, to estimate associations of regular medication use with BC risk. The proportionality assumption was assessed by evaluating Schoenfeld residuals. We estimated associations separately for regular use of aspirin, COX-2 inhibitors, ibuprofen and other NSAIDs, and acetaminophen, the latter as a negative control to determine if any associations observed between NSAIDs and BC risk reflected a non-specific use of analgesics [41]. We used a robust variance estimator to account for the family structure of the cohort. We stratified models by birth cohort (in 10-year categories) and adjusted for baseline age (continuous), race/ethnicity (non-Hispanic white versus otherwise), and study center (Model 1). We adjusted for familial risk profile using the 1-year BC risk score predicted from the BOADICEA model (Model 2) [35]. We also tested models adjusted for baseline health behaviors (never, former, current) including cigarette smoking, alcohol consumption, hormone therapy use, and hormonal birth control use, which collectively altered parameter estimates of some NSAID variables by > 10% (Model 3). Further adjustment for parity, breastfeeding, age at menarche, and body mass index did not alter the parameter estimates by > 10% and were not included in the final parsimonious model. Lastly, we tested a model further adjusted for use of the other three types of medications (Model 4). We estimated cross-product terms to test for multiplicative interactions between regular medication use and familial risk profile (continuous). We also plotted the predicted age-specific absolute cumulative risk for women with different familial risk based on BOADICEA and underlying age-specific incidences from the Surveillance, Epidemiology, and End Results Program [42,43,44]. We chose three scenarios of familial risk: 12% (population average), 20–30% (moderate familial risk), and > 30% (high familial risk) full lifetime BC risk, and two scenarios of medication use: regular aspirin user and non-regular aspirin user.

Subgroup analyses

Using the combined cohort, we estimated associations stratified by gene mutation carrier status defined as non-carriers (either true negative or not tested), BRCA1 carriers, or BRCA2 carriers. We conducted a sensitivity analysis using the weighting approach of Antoniou et al. (2005) [45] to account for non-random selection of mutation carriers. Weighting did not substantively alter medication-associated risk estimates or their standard errors. We also estimated associations by tumor ER status (positive or negative); the alternative ER subtype was censored at diagnosis. For example, ER-negative BC cases were censored at age at diagnosis in the analysis of ER-positive BC. Finally, we fitted attained age models, truncating follow-up time at ages 45, 55, and 65 years, to assess associations for younger women.

Sensitivity analyses

Although our primary analysis was based on regular NSAID use, we did an additional analysis examining duration of use (categorized as ≥ 5 years versus < 5 years) for women who provided this information. We conducted a sensitivity analysis excluding women who reported tamoxifen use at baseline (N = 64) and found this did not alter estimates. We conducted another sensitivity analysis further adjusting for diabetes and other cancers to account for comorbidities and found that this also did not appreciably alter estimates. To account for missing data, we conducted a sensitivity analysis using multiple imputation by chained equations, which produced comparable findings (data not shown). Statistical significance was determined as p < 0.05 for a two-sided hypothesis test. Analyses were conducted using Stata 15.1 (College Station, TX) [46].

Results

There were 139 incident cases of BC in the prospective cohort over 27,923 person-years. Prevalences of medication use were similar in the prospective and combined cohorts (18% and 19% for aspirin, 9% and 8% for COX-2 inhibitors, 17% and 19% for ibuprofen, and 17% and 17% for acetaminophen, respectively). Tetrachoric correlations between medications were statistically significant, but all ≤ 0.47 (see Additional File 1). At study enrollment, regular aspirin users were older on average than non-regular users (53.1 versus 41.7 years). Regular aspirin users were also less likely to smoke cigarettes, consume alcohol, or use hormonal birth control than non-regular users, but were more likely to use hormone therapy and other types of medication (Table 1). On average, regular aspirin users had a higher 1-year BOADICEA risk score than non-regular users, which is influenced by baseline age; a smaller proportion of aspirin users had a known mutation in the BRCA1 or BRCA2 gene. The distribution of 1-year BOADICEA risk score by regular medication use is provided in the supplemental materials (see Additional File 2).

Table 1 Baseline characteristics of women in the Prospective Family Study Cohort by regular aspirin use

As shown in Table 2, from analysis of the prospective cohort, regular aspirin use was associated with a 39% reduced BC risk in the fully adjusted model (Model 4: prospective hazard ratio (HRp) = 0.61, 95% confidence interval (CI) = 0.33 to 1.14); a very similar, but more precise, estimate was obtained from analysis of the combined cohort (Model 4: combined hazard ratio (HRc) = 0.63, 95% CI = 0.57 to 0.71). When we considered duration of aspirin use in the combined cohort, < 5 years versus never use was associated with an estimated 32% reduced BC risk (HRc = 0.68, 95% CI = 0.58 to 0.80); ≥ 5 years versus never use was associated with an estimated 66% reduced BC risk (HRc = 0.34, 95% CI = 0.27 to 0.44). From fully adjusted models, regular use of COX-2 inhibitors was associated with a 61% reduced BC risk for the prospective cohort (Model 4: HRp = 0.39, 95% CI = 0.15 to 0.97) and a 71% reduced BC risk for the combined cohort (Model 4: HRc = 0.29, 95% CI = 0.23 to 0.38). After adjusting for regular use of other medications (Model 4), regular use of ibuprofen and other NSAIDs (or ibuprofen exclusively) was not associated with BC risk, nor was acetaminophen.

Table 2 Adjusted hazard ratios (HRs) and 95% confidence intervals (CI) of breast cancer risk comparing regular medication users with non-regular users in the Prospective Family Study Cohort

Associations of regular use of aspirin and Cox-2 inhibitors with reduced breast cancer risk were not modified by familial risk profile as estimated by the BOADICEA 1-year BC risk score (Fig. 2a). When we stratified the combined cohort by gene mutation carrier status (Fig. 2b), similar HR estimates were found for women not known to be mutation carriers (HRc = 0.71, 95% CI = 0.63 to 0.80), BRCA1 carriers (HRc = 0.73, 95% CI = 0.49 to 1.09), and BRCA2 carriers (HRc = 0.80, 95% CI = 0.53 to 1.21), although the latter two confidence intervals were wide. Consistent estimates were also found for the association between COX-2 inhibitors and BC risk for women not known to be mutation carriers (HRc = 0.34, 95% CI = 0.25 to 0.48), BRCA1 carriers (HRc = 0.46, 95% CI = 0.20 to 1.08), and BRCA2 carriers (HRc = 0.51, 95% CI = 0.26 to 1.03); confidence intervals were again wide for known mutation carriers. Similar associations were also found when we stratified by ER status (Fig. 2c) and when we assessed attained age models (Fig. 2d).

Fig. 2
figure2

Adjusted hazard ratios and 95% confidence intervals of breast cancer risk comparing regular users of aspirin and COX-2 inhibitors with non-regular users by subgroups from analysis of the combined cohort of the Prospective Family Study Cohort (N = 8233). Legend: Models are adjusted for race/ethnicity, study center, baseline age, familial risk profile, cigarette smoking, alcohol consumption, hormone therapy use, hormonal birth control use, and regular use of other medications; stratified by birth cohort. Sample sizes: non-carriers (includes true negatives and women who did not undergo genetic testing) N = 6395; BRCA1 mutation carriers N = 506; BRCA2 mutation carriers N = 418; ER status: N = 7319; attained age 45: N = 2222; attained age 55: N = 4401; attained age 65: N = 6325. Alternative ER subtypes were censored at diagnosis (e.g., ER negative and ER status missing breast cancers censored at age at diagnosis in the analysis of ER positive breast cancer). P values are for the Wald chi-square test statistic for the interaction between categories of familial risk profile or BRCA carrier status and regular medication use

No associations with BC risk were found for regular use of ibuprofen and other NSAIDs or acetaminophen from subgroup analyses by familial risk profile, BRCA1 or BRCA2 mutation carrier status, tumor ER status, or attained age (Fig. 3).

Fig. 3
figure3

Adjusted hazard ratios and 95% confidence intervals of breast cancer risk comparing regular users of ibuprofen and other NSAIDs and acetaminophen with non-regular users by subgroup in the combined cohort of the Breast Cancer Prospective Family Study Cohort (N = 8233). Legend: Models are adjusted for race/ethnicity, study center, baseline age, familial risk profile, cigarette smoking, alcohol consumption, hormone therapy use, hormonal birth control use, and regular use of other medications; stratified by birth cohort. Sample sizes: non-carriers (includes true negatives and women who did not undergo genetic testing) N = 6395; BRCA1 mutation carriers N = 506; BRCA2 mutation carriers N = 418; ER status: N = 7319; attained age 45: N = 2222; attained age 55: N = 4401; attained age 65: N = 6325. Alternative ER subtypes were censored at diagnosis (e.g., ER negative and ER status missing breast cancers censored at age at diagnosis in the analysis of ER-positive breast cancer). P values are for the Wald chi-square test statistic for the interaction between categories of familial risk profile or BRCA carrier status and regular medication use

Figure 4 shows the overall implications of the study estimates on the predicted age-specific BC cumulative risk for non-regular users of aspirin and regular users of aspirin with different familial risk profiles. In terms of absolute risk, the risk difference between regular users and non-regular users is greater for women with higher familial risk. For cumulative BC risk to age 80 years, the risk difference is 4.1%, 6.9%, and 9.8% for women at population average risk, moderate familial risk, and high familial risk, respectively.

Fig. 4
figure4

Cumulative risk of breast cancer by regular aspirin use and underlying familial risk. Legend: Predicted age-specific cumulative risk (from birth) for breast cancer, by regular aspirin use and familial risk, where 12% lifetime risk is approximately the population risk of breast cancer by age 80 years, where moderate familial risk (> 20–30% full lifetime BOADICEA) is equivalent to having one affected first-degree relative, and high familial risk (> 30% full lifetime BOADICEA) is equivalent to having two affected first-degree relatives

Discussion

From studying a prospective and combined (prospective and retrospective) cohort enriched with women having a family history of BC across a wide range of absolute predicted familial BC risk (10-year risk: mean, 5.3%; range, < 0.1–68.5%), we found regular aspirin use to be associated with a 39% and 37% reduction in BC risk in the two cohorts, respectively. Regular use of COX-2 inhibitors was associated with a 61% and 71% reduction in BC risk in the two cohorts, respectively. The strength of these associations did not differ by familial risk or mutation status, and although not nominally significant, negative associations were found for both BRCA1 and BRCA2 mutation carriers. Negative associations were also found for younger women based on attained age models examining risk up to age 45 years.

Our findings are consistent with most, but not all, other studies of NSAIDS and BC risk conducted using samples of average-risk women unselected for family history [24]. As previously noted, the only RCT of aspirin and BC risk as the primary endpoint did not observe an association after 10 years of follow-up (relative risk = 0.98; 95% CI = 0.89–1.08), but this could reflect the fact that participants were randomized to a low dose of aspirin (100 mg every other day) [29]. We found that regular use of aspirin and COX-2 inhibitors was associated with reduced risk of both ER-positive and ER-negative BCs, which again aligns with most, but not all [41, 47, 48], other studies that considered hormone receptor status [49]. This suggests that these NSAIDs might operate through multiple underlying biological pathways to lower BC risk, including a direct effect through ER mediated signaling pathways [13, 41], the COX-2 pathway [13, 50], phosphatidylinositol 3-kinase down-regulation [13, 51], B cell lymphoma 2-mediated apoptosis [13, 52], or epidermal growth factor receptor inhibition and p53 acetylation [13, 53].

We no longer found a statistically significant association between ibuprofen and other NSAIDs and BC risk after we adjusted for regular use of other medications. This could be because ibuprofen does not have an effect on BC risk, as some NSAIDs might inhibit COX-2 more intensely than others [26]. It could also reflect differences in the duration and frequency of ibuprofen and other NSAID use compared with aspirin and COX-2 inhibitors. Women might be more likely to use aspirin regularly because of its anticlotting effect [27, 54, 55], which ibuprofen does not deliver [56]. We also found no evidence of an association between acetaminophen, an analgesic with minimal anti-inflammatory action, and BC risk. This supports the known pharmacological effects of NSAIDs and minimizes concerns that the associations we found are due to confounding from other unmeasured lifestyle factors associated with regular analgesic use [41, 57].

The present study has several strengths. Most notably, we used data from a large cohort of women that is enriched for familial or genetic BC risk. This allowed us to test if associations between NSAIDs and BC risk vary in strength across a wide range of familial BC risk. Another strength is the use of the BOADICEA [34] to estimate a woman’s familial risk profile [35]. We also estimated associations by known BRCA1 and BRCA2 mutation carrier status. Although we had low statistical power, this is the first study to consider carrier-specific associations between NSAIDs and BC risk. One limitation of the present study is the use of binary measures of regular medication use. We also did not have information on dosage, and we had only limited data to examine duration of use. We recognize that these factors need to be considered to fully understand the association of NSAIDs with BC risk. Another limitation is that our exposure measures were retrospective, and thus, recall bias is a concern when interpreting the estimates from the combined cohort analyses. Survival bias is another potential limitation, but the consistency of associations between the prospective and combined cohorts supports that biases that operate differently in prospective and retrospective settings are unlikely to explain these findings. Confounding by indication could also be of concern, given that we did not have information on the reason for medication use. However, attained age models estimated similar associations in young women for whom comorbidities are unlikely, and the sensitivity analysis that further adjusted for diabetes and other cancers produced comparable estimates.

Conclusion

In summary, our findings add to growing evidence for an association between regular use of aspirin and COX-2 inhibitors and reduced BC risk. The potential impact of using these medications for primary BC prevention is underscored by the fact that associations were not modified by familial risk, and suggestive negative associations were found for BRCA1 and BRCA2 mutation carriers. This means that individuals at higher risk of BC could benefit even more in terms of absolute risk reduction from modifying medication use. Additionally, our findings suggest that regular use of aspirin and COX-2 inhibitors are associated with BC risk independent of ER status, which is important because risk-reducing medications are currently only available for ER-positive BC. Although RCTs are ultimately needed to confirm associations of NSAIDs with BC risk, our findings support that the consistent results seen in observational studies of average-risk women may extend to women at the higher range of absolute BC risk.

Abbreviations

ABC:

Aspirin for Breast Cancer

ASPREE:

Aspirin in Reducing Events in the Elderly

BC:

Breast cancer

BCFR:

Breast Cancer Family Registry

BOADICEA:

Breast Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm

CI:

Confidence interval

COX:

Cyclooxygenase

ER:

Estrogen receptor

HR:

Hazard ratio

kConFab:

Kathleen Cuningham Foundation Consortium for Research into Familial Breast Cancer

NSAID:

Non-steroidal anti-inflammatory drug

OR:

Odds ratio

ProF-SC:

Prospective Family Study Cohort

RCT:

Randomized controlled trial

References

  1. 1.

    Collaborative Group on Hormonal Factors in Breast Cancer. Familial breast cancer: collaborative reanalysis of individual data from 52 epidemiological studies including 58,209 women with breast cancer and 101,986 women without the disease. Lancet. 2001;358(9291):1389–99.

  2. 2.

    Kuchenbaecker KB, Hopper JL, Barnes DR, et al. Risks of breast, ovarian, and contralateral breast cancer for BRCA1 and BRCA2 mutation carriers. JAMA. 2017;317(23):2402–16.

  3. 3.

    Hartmann LC, Lindor NM. The role of risk-reducing surgery in hereditary breast and ovarian cancer. N Engl J Med. 2016;374(5):454–68.

  4. 4.

    Cuzick J, Sestak I, Bonanni B, et al. Selective oestrogen receptor modulators in prevention of breast cancer: an updated meta-analysis of individual participant data. Lancet. 2013;381(9880):1827–34.

  5. 5.

    Goss PE, Ingle JN, Alés-Martínez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med. 2011;364(25):2381–91.

  6. 6.

    Cuzick J, Sestak I, Forbes JF, et al. Anastrozole for prevention of breast cancer in high-risk postmenopausal women (IBIS-II): an international, double-blind, randomised placebo-controlled trial. Lancet. 2014;383(9922):1041–8.

  7. 7.

    Meiser B, Wong W, Peate M, Julian-Reynier C, Kirk J, Mitchell G. Motivators and barriers of tamoxifen use as risk-reducing medication amongst women at increased breast cancer risk: a systematic literature review. Hered Cancer Clin Pract. 2017;15(1):14.

  8. 8.

    Smith SG, Sestak I, Forster A, et al. Factors affecting uptake and adherence to breast cancer chemoprevention: a systematic review and meta-analysis. Ann Oncol. 2016;27(4):575–90.

  9. 9.

    Keogh LA, Hopper JL, Rosenthal D, Phillips K-A. Australian clinicians and chemoprevention for women at high familial risk for breast cancer. Hered Cancer Clin Pract. 2009;7(1):9.

  10. 10.

    Noonan S, Pasa A, Fontana V, et al. A survey among breast cancer specialists on the low uptake of therapeutic prevention with tamoxifen or raloxifene. Cancer Prev Res (Phila). 2018;11(1):38–43.

  11. 11.

    Smith SG, Foy R, McGowan JA, et al. Prescribing tamoxifen in primary care for the prevention of breast cancer: a national online survey of GPs’ attitudes. Br J Gen Pract. 2017;67(659):e414–27.

  12. 12.

    Metcalfe KA, Birenbaum-Carmeli D, Lubinski J, et al. International variation in rates of uptake of preventive options in BRCA1 and BRCA2 mutation carriers. Int J Cancer. 2008;122(9):2017–22.

  13. 13.

    Bardia A, Keenan TE, Ebbert JO, et al. Personalizing aspirin use for targeted breast cancer chemoprevention in postmenopausal women. Mayo Clin Proc. 2016;91(1):71–80.

  14. 14.

    Bennett A, McDonald A, Stamford I, Charlier E, Simpson J, Zebro T. Prostaglandins and breast cancer. Lancet. 1977;2(8039):624–6.

  15. 15.

    Su B, Díaz-Cruz ES, Landini S, Brueggemeier RW. Suppression of aromatase in human breast cells by a cyclooxygenase-2 inhibitor and its analog involves multiple mechanisms independent of cyclooxygenase-2 inhibition. Steroids. 2008;73(1):104–11.

  16. 16.

    Sanmuganathan PS, Ghahramani P, Jackson PR, Wallis EJ, Ramsay LE. Aspirin for primary prevention of coronary heart disease: safety and absolute benefit related to coronary risk derived from meta-analysis of randomised trials. Heart. 2001;85(3):265–71.

  17. 17.

    Derry S, Loke YK. Risk of gastrointestinal haemorrhage with long term use of aspirin: meta-analysis. BMJ. 2000;321(7270):1183–7.

  18. 18.

    Bresalier RS, Sandler RS, Quan H, et al. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. N Engl J Med. 2005;352(11):1092–102.

  19. 19.

    Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med. 2000;343(21):1520–8.

  20. 20.

    US Food and Drug Administration. FDA briefing document on nonsteroidal anti-inflammatory drugs and cardiovascular thrombotic risk. Joint Meeting of the Arthritis Advisory Committee and the Drug Safety and Risk Management, Advisory Committee, April, vol. 24-25; 2018. https://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/ArthritisAdvisoryCommittee/UCM605207.pdf. Accessed August 8, 2018

  21. 21.

    McNeil JJ, Nelson MR, Woods RL, et al. Effect of aspirin on all-cause mortality in the healthy elderly. N Engl J Med. 2018. https://doi.org/10.1056/NEJMoa1803955.

  22. 22.

    Rostom A, Dubé C, Lewin G, et al. Nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 inhibitors for primary prevention of colorectal cancer: a systematic review prepared for the US Preventive Services Task Force. Ann Intern Med. 2007;146(5):376–89.

  23. 23.

    Rothwell PM, Wilson M, Elwin C-E, et al. Long-term effect of aspirin on colorectal cancer incidence and mortality: 20-year follow-up of five randomised trials. Lancet. 2010;376(9754):1741–50.

  24. 24.

    Luo T, Yan H-M, He P, Luo Y, Yang Y-F, Zheng H. Aspirin use and breast cancer risk: a meta-analysis. Breast Cancer Res Treat. 2012;131(2):581–7.

  25. 25.

    Lu L, Shi L, Zeng J, Wen Z. Aspirin as a potential modality for the chemoprevention of breast cancer: a dose-response meta-analysis of cohort studies from 857,831 participants. Oncotarget. 2017;8(25):40389–401.

  26. 26.

    de Pedro M, Baeza S, Escudero M-T, et al. Effect of COX-2 inhibitors and other non-steroidal inflammatory drugs on breast cancer risk: a meta-analysis. Breast Cancer Res Treat. 2015;149(2):525–36.

  27. 27.

    Paulose-Ram R, Hirsch R, Dillon C, Gu Q. Frequent monthly use of selected non-prescription and prescription non-narcotic analgesics among US adults. Pharmacoepidemiol Drug Saf. 2005;14(4):257–66.

  28. 28.

    Zhong S, Chen L, Zhang X, Yu D, Tang J, Zhao J. Aspirin use and risk of breast cancer: systematic review and meta-analysis of observational studies. Cancer Epidemiol Biomark Prev. 2015;24(11):1645–55.

  29. 29.

    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(1):47–55.

  30. 30.

    Chen WY, Winer EP, Barry WT, et al. ABC trial (A011502): randomized phase III double blinded placebo controlled trial of aspirin as adjuvant therapy for breast cancer. J Clin Oncol. 2018;15_suppl(36):TPS597.

  31. 31.

    Coyle C, Cafferty FH, Rowley S, et al. ADD-ASPIRIN: a phase III, double-blind, placebo controlled, randomised trial assessing the effects of aspirin on disease recurrence and survival after primary therapy in common non-metastatic solid tumours. Contemp Clin Trials. 2016;51:56–64.

  32. 32.

    Kim S, Shore DL, Wilson LE, et al. Lifetime use of nonsteroidal anti-inflammatory drugs and breast cancer risk: results from a prospective study of women with a sister with breast cancer. BMC Cancer. 2015;15(1):960.

  33. 33.

    Hopper JL. Disease-specific prospective family study cohorts enriched for familial risk. Epidemiol Perspect Innov. 2011;8(2):1–9.

  34. 34.

    Terry MB, Phillips KA, Daly MB, et al. Cohort profile: The Breast Cancer Prospective Family Study Cohort (ProF-SC). Int J Epidemiol. 2015;45(3):683–92.

  35. 35.

    Antoniou A, Pharoah P, Smith P, Easton D. The BOADICEA model of genetic susceptibility to breast and ovarian cancer. Br J Cancer. 2004;91(8):1580–90.

  36. 36.

    Antoniou AC, Cunningham AP, Peto J, et al. The BOADICEA model of genetic susceptibility to breast and ovarian cancers: updates and extensions. Br J Cancer. 2008;98(8):1457–66.

  37. 37.

    MacInnis R, Bickerstaffe A, Apicella C, et al. Prospective validation of the breast cancer risk prediction model BOADICEA and a batch-mode version BOADICEACentre. Br J Cancer. 2013;109(5):1296–301.

  38. 38.

    John EM, Hopper JL, Beck JC, et al. The Breast Cancer Family Registry: an infrastructure for cooperative multinational, interdisciplinary and translational studies of the genetic epidemiology of breast cancer. Breast Cancer Res. 2004;6(4):R375–89.

  39. 39.

    Mann GJ, Thorne H, Balleine RL, et al. Analysis of cancer risk and BRCA1 and BRCA2 mutation prevalence in the kConFab familial breast cancer resource. Breast Cancer Res. 2006;8(1):R12.

  40. 40.

    Neuhausen SL, Ozcelik H, Southey MC, et al. BRCA1 and BRCA2 mutation carriers in the Breast Cancer Family Registry: an open resource for collaborative research. Breast cancer research and treatment. Breast Cancer Res Treat. 2009;116(2):379–86.

  41. 41.

    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(20):2433–40.

  42. 42.

    Surveillance Epidemiology and End Results (SEER) Program. SEER*Stat Database: Incidence - SEER 9 Regs Research Data, Nov 2011 Sub (1973–2009) <Katrina/Rita Population Adjustment> − Linked To County Attributes - Total U.S.; 2011.

  43. 43.

    Surveillance Epidemiology and End Results (SEER) Program. SEER*Stat Database: Incidence - SEER 13 Regs Research Data, Nov 2011 Sub (1992–2009) <Katrina/Rita Population Adjustment> − Linked To County Attributes - Total U.S.; 2011.

  44. 44.

    Surveillance Epidemiology and End Results (SEER) Program. SEER*Stat Database: Incidence - SEER 18 Regs Research Data, Nov 2011 Sub (2000–2009) <Katrina/Rita Population Adjustment> − Linked To County Attributes - Total U.S.; 2011.

  45. 45.

    Antoniou AC, Goldgar DE, Andrieu N, et al. A weighted cohort approach for analysing factors modifying disease risks in carriers of high-risk susceptibility genes. Genet Epidemiol. 2005;29(1):1–11.

  46. 46.

    StataCorp. Stata statistical software: release 15. College Station, TX: StataCorp LLC; 2017.

  47. 47.

    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(11):805–12.

  48. 48.

    Gierach GL, Lacey JV, Schatzkin A, et al. Nonsteroidal anti-inflammatory drugs and breast cancer risk in the National Institutes of Health–AARP Diet and Health Study. Breast Cancer Res. 2008;10(2):R38.

  49. 49.

    Bardia A, Olson JE, Vachon CM, et al. Effect of aspirin and other NSAIDs on postmenopausal breast cancer incidence by hormone receptor status: results from a prospective cohort study. Breast Cancer Res Treat. 2011;126(1):149–55.

  50. 50.

    Gallicchio L, McSorley MA, Newschaffer CJ, et al. Nonsteroidal antiinflammatory drugs, cyclooxygenase polymorphisms, and the risk of developing breast carcinoma among women with benign breast disease. Cancer. 2006;106(7):1443–52.

  51. 51.

    Liao X, Lochhead P, Nishihara R, et al. Aspirin use, tumor PIK3CA mutation, and colorectal-cancer survival. N Engl J Med. 2012;367(17):1596–606.

  52. 52.

    Choi B-H, Chakraborty G, Baek K, Yoon HS. Aspirin-induced Bcl-2 translocation and its phosphorylation in the nucleus trigger apoptosis in breast cancer cells. Exp Mol Med. 2013;45(10):e47.

  53. 53.

    Huang L, Wong CC, Mackenzie GG, et al. Phospho-aspirin (MDC-22) inhibits breast cancer in preclinical animal models: an effect mediated by EGFR inhibition, p53 acetylation and oxidative stress. BMC Cancer. 2014;14(1):141.

  54. 54.

    Bibbins-Domingo K. US Preventive Services Task Force. Aspirin Use for the Primary Prevention of Cardiovascular Disease and Colorectal Cancer: US Preventive Services Task Force Recommendation Statement. Ann Intern Med. 2016;164(12):836–45.

  55. 55.

    Zhou Y, Boudreau DM, Freedman AN. Trends in the use of aspirin and nonsteroidal anti-inflammatory drugs in the general US population. Pharmacoepidemiol Drug Saf. 2014;23(1):43–50.

  56. 56.

    Mazaleuskaya LL, Theken KN, Gong L, et al. PharmGKB summary: ibuprofen pathways. Pharmacogenet Genomics. 2015;25(2):96–106.

  57. 57.

    Hargreave M, Andersen TV, Nielsen A, Munk C, Liaw KL, Kjaer SK. Factors associated with a continuous regular analgesic use—a population-based study of more than 45 000 Danish women and men 18–45 years of age. Pharmacoepidemiol Drug Saf. 2010;19(1):65–74.

Download references

Acknowledgements

We thank the entire team of Breast Cancer Family Registry (BCFR) past and current investigators as well as the kConFab investigators. We also thank Heather Thorne, Eveline Niedermayr, Lucy Stanhope, Sandra Picken, all the BCFR and kConFab research nurses and staff, the heads and staff of the Family Cancer Clinics, and the many families who contribute to the BCFR and kConFab for their contributions to this resource.

Funding

The six sites of the Breast Cancer Family Registry were supported by grant UM1 CA164920 from the USA National Cancer Institute. This work was also supported by grants to kConFab and the kConFab Follow-Up Study from Cancer Australia [grant numbers 809195, 1100868], the Australian National Breast Cancer Foundation [grant number IF 17 kConFab], the National Health and Medical Research Council [grant numbers 454508, 288704, 145684], the National Institute of Health U.S.A. [grant number 1RO1CA159868], the Queensland Cancer Fund, the Cancer Councils of New South Wales, Victoria, Tasmania, and South Australia, and the Cancer Foundation of Western Australia [grant numbers not applicable]. RDK is supported by the National Institutes of Health, National Cancer Institute, Cancer Epidemiology Training Grant [grant number T32-CA009529]. KAP is an Australian National Breast Cancer Foundation Practitioner Fellow.

The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The content of this manuscript does not necessarily reflect the views or policies of the National Cancer Institute or any of the collaborating centers in the BCFR, nor does mention of trade names, commercial products, or organizations imply endorsement by the USA Government or the BCFR.

Availability of data and materials

The datasets analyzed during the current study are available from the corresponding author on reasonable request.

Author information

MBT had full access to all the data in the study and takes responsibility for the integrity and accuracy of the overall content. MBT attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted. RDK and MBT contributed to the study concept and design. All authors contributed to the acquisition, analysis, and interpretation of data. RDK contributed to the statistical analysis. RDK and MBT contributed to drafting the manuscript. All authors contributed to the critical revision of the manuscript for important intellectual content. SN and YL contributed administrative, technical, and material support. MBT and JLH contributed study supervision. All authors read and approved the final manuscript.

Correspondence to Mary Beth Terry.

Ethics declarations

Ethics approval and consent to participate

All participants in the BCFR and kConFab provided written informed consent before participation. Human research ethics committees at the participating institutions granted ethics approval for the six sites of the BCFR and for kConFab:

Northern California—Cancer Prevention Institute of California, Institutional Review Board (2001–033) and Stanford University School of Medicine, Institutional Review Board (45842).

New York—Columbia University Medical Center, Institutional Review Board (AAA7794).

Philadelphia—Fox Chase Cancer Center, Institutional Review Board (95–009).

Utah—Huntsman Cancer Institute, University of Utah, Institutional Review Board (00004965).

Ontario—Mount Sinai Hospital Research Ethics Board (#02–0076-U) and University Health Network Research Ethics Board (#96-U107-CE).

Australia—University of Melbourne, Human Ethics Sub-Committee (1441420.1).

kConFab—Peter MacCallum Cancer Centre, the Peter Mac Ethics Committee (97/27).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Additional files

Additional file 1:

Correlation between regular use of medications in the combined cohort of the Prospective Family Study Cohort (N = 8233). Additional File 1 presents tetrachoric correlations and odds ratios comparing regular use of each of the four medications (aspirin, Cox-2 inhibitors, ibuprofen, and acetaminophen) that were included in the analysis. (DOCX 15 kb)

Additional file 2:

Distribution of BOADICEA 1-year risk scores by medication use in the combined cohort of the Prospective Family Study Cohort (N = 8233). Additional File 2 presents overlapping histograms of the distribution of BOADICEA one-year risk score by medication use (regular users versus non-regular users). (DOCX 45 kb)

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kehm, R.D., Hopper, J.L., John, E.M. et al. Regular use of aspirin and other non-steroidal anti-inflammatory drugs and breast cancer risk for women at familial or genetic risk: a cohort study. Breast Cancer Res 21, 52 (2019) doi:10.1186/s13058-019-1135-y

Download citation

Keywords

  • Breast cancer
  • Non-steroidal anti-inflammatory drugs
  • Family history
  • High-risk population