Prolonged analgesic use failed to reduce postmenopausal breast cancer risk

Zhang X. J Clin Oncol. 2012;doi:10.1200/JCO.2012.42.2006.

  • HemOnc Today, October 25, 2012

No association exists between postmenopausal breast cancer risk and prolonged analgesic use, according to researchers who conducted a prospective study that involved more than 84,000 women.

Some prior studies suggested regular aspirin use may reduce incidence of postmenopausal, hormone receptor-positive breast cancer by up to 25%, while other studies found no association.

The current study was designed to assess the link between aspirin, other NSAIDs and acetaminophen in regard to overall breast cancer incidence in postmenopausal women, as well as by molecular subtype.

The first portion of the study involved the observation of 84,602 postmenopausal women who did not have cancer at the time of study initiation in 1980. They were followed until June 2008.

Xuehong Zhang, MD, ScD, of Brigham and Women’s Hospital and Harvard Medical School, and colleagues used biennial questionnaires to collect data on analgesic use, reproductive history and other lifestyle factors.

The final analysis included 4,734 documented cases of incident invasive breast cancer.

Regular aspirin use was defined as at least two tablets per week for more than 20 years. Compared with nonuse, regular use was linked to a 0.91 RR (95% CI, 0.81-1.01) for overall breast cancer, a 0.9 RR for ER- and PR-positive breast cancer (95% CI, 0.77-1.06), and a 0.91 RR for ER- and PR-negative breast cancer (95% CI, 0.68-1.22).

Researchers observed no variation in results in terms of past or present aspirin use, days per week of use or dosage levels.

The investigators found no significant association between breast cancer risk and use of other NSAIDs or acetaminophen.

Higher doses were defined as at least six tablets per week for more than 10 years. Use of any analgesic at this level was not significantly linked to overall or subtype breast cancer risk.

Similarly, the researchers observed no link for breast cancer subtypes such as luminal A, luminal B, triple negative, basal-like, HER-2–positive, Cox-2–negative and Cox-2–positive.

In an accompanying editorial, Andrew Jess Dannenberg, MD, director of the Weill Cornell Cancer Center at New York Presbyterian Hospital/Weill Cornell Medical College, and co-authors called the report by Zhang and colleagues an important contribution to scientific literature that explores possible associations between anti-inflammatory drugs and breast cancer incidence among postmenopausal women as a whole.

“The next step is to investigate this strategy where it makes the most sense: in the subset of women with identifiable localized inflammation of the breast,” Dannenberg and co-authors wrote. “Many of these women will be obese, some will be aged, but none of them can be currently identified with conventional clinical tests. To truly test the hypotheses, we need to develop reliable biomarkers for the localized inflammatory lesion, and then conduct clinical trials that can demonstrate that the lesion and its consequences can be reversed and that such a reversal truly lowers risk. Given the frequency of the risk factors (aging and obesity) and the disease, this effort has important public health implications.

“At the same time, it should not escape anyone’s attention that postmenopausal hormone receptor–positive breast cancer, while likely linked to inflammation resulting from obesity, is by no means the only consequence of this unfortunate public health problem,” Dannenberg and co-authors wrote. “Reducing inflammation to prevent the consequences of diet and aging could have broad benefits far beyond oncology, but we will need to perform the right studies to find out. This may be financially painful, but we have much to gain.”

Perspective
G. Thomas Budd, MD

G. Thomas Budd

  • The Nurses Health Study is a landmark epidemiologic study that has led to numerous insights in women’s health. This prospective cohort study enrolled and prospectively followed 121,700 registered female nurses aged 30 to 55 years at baseline in 1976. In their manuscript, Zhang and colleagues discuss 28-year follow-up on the effects of aspirin, non-steroidal anti-inflammatory drugs and acetaminophen on the incidence of breast cancer in this group of women. A total of 4,734 documented cases of cancer developed in this group of women, but pre-diagnosis use of aspirin, NSAIDs or acetaminophen was not correlated with the risk of developing breast cancer or any specific molecular sub-type of breast cancer. These findings do not support the use of these agents as chemopreventive agents for breast cancer, and they add to the somewhat conflicting literature related to this subject.

    Although these results differ from those of a meta-analysis published in 2008 [Takkouche B.  J Natl Cancer Inst. 2008;100:1439-1447], they concord with the only randomized trial of aspirin (though at a low dose) as a preventive for breast cancer and other diseases [Cook NR. JAMA. 2005;294:47-55].

    So what is a clinician to do? Simply put, aspirin and NSAIDs may be prescribed for a number of reasons, but they should not be prescribed for the sole purpose of reducing the risk of developing breast cancer. A number of effects of aspirin or NSAIDs have been proposed that might affect breast oncogenesis, including effects of prostaglandin E2 on cell proliferation, angiogenesis and apoptosis, a reduction in aromatase activity, and anti-inflammatory effects. However, additional studies will be needed to delineate a mechanism of action for these agents as cancer preventives, and to identify subsets of patients who might benefit, if they exist.

    • G. Thomas Budd, MD
    • Staff physician
      Department of Solid Tumor Oncology
      Taussig Cancer Institute
      Cleveland Clinic
      Professor of Medicine
      Cleveland Clinic Lerner College of Medicine
      Case Western Reserve University
  • Disclosures: Budd reports no relevant financial disclosures.
Perspective
  • My initial opinion about these data was that, if there is an effect of aspirin, it is minimal or modest, resulting in a 10% decline in overall relative risk of breast cancer. However, there are several issues at hand that may make this a bit more complicated. One issue is that people who normally take that much aspirin over such a long period of time are those who have some chronic condition that calls for it, such as rheumatoid arthritis, which is associated with chronic inflammation. Interestingly, women with rheumatoid arthritis actually have a 16% lower chance of developing breast cancer compared with the general population, as shown in a 2008 meta-analysis (Smitten AL. Arthritis Res Ther. 2008;10:R45). The most likely explanation, as proposed by the authors, is the frequent use of NSAIDs like aspirin and ibuprofen, among this group that is expected to reduce inflammation. Recent evidence has shown that inflammation is related to increased cancer incidence and poorer outcomes, so if you have people taking aspirin for inflammation-related diseases, the reduction in risk observed in this high-risk group might actually be attenuated and appear smaller than it really is. Another point is that inflammation may play a more important role with respect to breast cancer risk in certain groups of women, as well as for specific breast cancer subtypes. Zhang and colleagues tried to address some of these questions, but the number of breast cancer cases was limited for certain molecular subtypes, particularly for ER-negative disease, which is more aggressive than ER-positive disease and appears to be more closely linked with inflammation.  Given that this cohort is comprised of only postmenopausal women, an important group to study in the future will be premenopausal women, since ER-negative cancers are more common among younger women. Hence, reductions in inflammation may be more important for these women. One thing that was interesting was that the data for acetaminophen, which can lower estrogen levels, were at least as strong as the evidence for a modest effect of aspirin. We see this for total risk and for ER-positive and triple-negative disease. At the moment, the evidence for taking aspirin to prevent breast cancer is still pretty weak. The strongest scientific evidence to date points to maintaining a healthy weight, engaging in regular physical activity, and limiting alcohol intake as the best lifestyle strategies to reduce breast cancer risk.

    • Chi-Chen Hong, PhD
    • Assistant professor of oncology
      Department of Cancer Prevention and Control
      Roswell Park Cancer Institute
  • Disclosures: Hong reports no relevant financial disclosures.
Perspective
Harry S. Jacob, MD

Harry S. Jacob, MD, FRCPath(Hon)

  • For now, postmenopausal breast cancer may not join other cancers — especially colon — that seem to manifest reduced incidence in chronic aspirin users.

    • Harry S. Jacob, MD, FRCPath(Hon)
    • HemOnc Today Chief Medical Editor
  • Disclosures: Jacob reports no relevant financial disclosures.