Introduction: Obesity is associated with a worse breast cancer prognosis, with the most prominent effects seen in hormone responsive postmenopausal patients. It has also been linked to elevated levels of inflammation, including greater cyclooxygenase-2 (COX-2) expression and activity in adipose-infiltrating macrophages. The product of this enzyme, the pro-inflammatory eicosanoid prostaglandin E2 (PGE2), stimulates adipose tissue aromatase expression and subsequent estrogen production, which could promote breast cancer progression. Consequently, we hypothesized that non-steroidal anti-inflammatory drug (NSAID) use decreases estrogen receptor (ER) positive breast cancer recurrence in the obese population via inhibition of PGE2-mediated local aromatase expression.

Methods: Four-hundred and forty women treated for invasive, ER positive breast cancer at San Antonio area clinics were retrospectively classified according to NSAID use, body mass index category (Normal Weight (NW): 18.5-24.9 kg/m2; Overweight (OW): 25.0-29.9 kg/m2; Obese (OB): ≥30.0 kg/m2), and disease recurrence. To examine the role of obesity-induced local aromatase expression in the link between NSAID use and disease recurrence, we utilized an in vitro model of obesity in which we exposed macrophages to pooled sera samples from NW or OB postmenopausal breast cancer patients. Adipose stromal cells’ (ASC) aromatase expression was measured following exposure to conditioned media (CM) collected from these sera-exposed macrophages. ER activity and in vitro measures of cancer aggression were assessed in MCF-7 and T47D breast cancer cells cultured in CM from sera-exposed macrophage/ASC co-cultures.

Results: Within our patient population, which had an average BMI in the obese range, NSAID users had significantly lower recurrence rates (p = 0.05) and their time to disease progression was delayed by almost 28 months in comparison to nonusers. Our in vitro studies demonstrated that growth in OB macrophage CM significantly enhances ASC aromatase expression in comparison to NW (p<0.05), while macrophage treatment with celecoxib during the generation of CM neutralizes the difference between OB and NW. This was correlated with significantly greater macrophage PGE2 production following OB versus NW sera exposure (p<0.05). In addition, CM from macrophage/ASC co-cultures exposed to OB patient sera stimulates more breast cancer cell ER activity, proliferation, and S phase activity, and these differences are eliminated by the addition of an aromatase inhibitor during the generation of CM. We also plan to examine how the co-culture CM impacts breast cancer cell expression of a panel of genes related to cancer aggression.

Conclusions: Our results indicate that NSAID use can improve the recurrence rate for hormone-responsive breast cancer patients, particularly those with an elevated BMI. The in vitro model suggests that obesity-related enhancement of PGE2-induced local aromatase expression and estrogen production may be a key mechanism mediating this effect. Further studies designed to examine the clinical benefit of NSAID use in the obese breast cancer patient population are warranted.

Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-06-01.