Overweight and Obese Perimenopausal and Postmenopausal Women Exhibit Increased Abnormal Mammary Epithelial Cytology

It is estimated that roughly 10 million U.S. women are at high risk for breast cancer (1). Excessive body weight, as reflected by a body mass index (BMI ≥ 25 kg/m²), is consistently associated with postmenopausal breast cancer and increases the risk of dying from this disease (2-6). Whereas previous studies have explored associations between body weight and indirect biomarkers associated with disease risk, such as circulating hormonal levels or mammographic density, few have explored associations between BMI and direct measures on target tissue (7, 8).

Random periareolar fine needle aspiration (RPFNA) is a research technique developed to sample mammary cells from the whole breast of asymptomatic women at high risk for development of breast cancer to assess both (a) breast cancer risk and (b) response to chemoprevention (9, 10). RPFNA can be done successfully in a majority of high-risk women (82-89% cell yield; refs. 9-11). RPFNA has been used successfully to predict short-term breast cancer risk; the presence of cellular atypia in a breast RPFNA specimen has been prospectively validated to confer a 5.6-fold increase in breast cancer risk in high-risk women (9).

The purpose of this brief communication is to relay findings obtained from an exploratory study aimed at investigating whether perimenopausal and postmenopausal women who are overweight or obese exhibit increased abnormal Masood Cytology as compared with women of normal weight. The rationale for undertaking this exploration was driven by the hypothesis that these direct measures on the target tissue would indeed be associated with BMI.

This study was conducted among 62 women who underwent RPFNA under an Institutional Review Board–approved protocol at Duke University Medical Center (March 2003-August 2005) and who presented sufficient cells for analysis.

Women were required to have at least one of the following major risk factors for breast cancer: (a) 5-year Gail risk calculation ≥1.7% (12); (b) prior biopsy exhibiting atypical hyperplasia, lobular carcinoma in situ, or ductal carcinoma in situ; (c) known BRCA1/2 mutation carrier; or (d) prior history of contralateral breast cancer. Women were required to be perimenopausal or postmenopausal, defined as <6 menstrual periods/y in the absence of pregnancy, polycystic ovarian syndrome, or thyroid disorder or no menses for >12 months in the absence of pregnancy and/or status postsurgical removal of both ovaries, respectively. Sociodemographic variables (age, race), hormone replacement therapy (HRT) use, and family history of breast cancer were collected. Given the potential for HRT to serve as a confounder of an investigation focused on body weight and breast cancer risk, all women currently using HRT on a routine basis were excluded from the analyses.

Body weights and heights were clinically measured on a platform scale with a fixed stadiometer; BMIs were calculated and the standard cutoff point of 25 kg/m² was used to dichotomize normal weight from overweight women (13).

RPFNA was done as previously published (11). In subjects with prior invasive cancer or ductal carcinoma in situ, only the contralateral breast was aspirated. Slides for cytology were prepared in the laboratory of Dr. Carol Fabian (9, 10, 14). Morphologic assessment, Masood cytology index scores, and cell count were assigned by a single dedicated pathologist who was blinded to BMI (C.M.Z.; refs. 9, 15). Cells were given a score of 1 to 4 for each of six morphologic characteristics including cell arrangement, pleomorphism, number of myoepithelial cells, anisonucleosis, nucleoli, and chromatin clumping; the sum of these points computed the Masood score: ≤10, nonproliferative (normal); 11 to 13, hyperplasia; 14 to 16, atypia; and ≥17, suspicious cytology (9, 15). The number of epithelial cells was quantitated per breast and classified as <10 cells (insufficient quantity for analysis); 10 to 99 cells; 100 to 499 cells; 500 to 999 cells; 1,000 to 4,999 cells; and ≥5,000 cells.

Differences in mean cell counts and Masood cytology index scores between women with BMIs <25 kg/m² versus those with BMIs ≥25 kg/m² were tested using the pooled and Satterthwaite t tests, respectively.

This study was conducted among 62 women who underwent RPFNA from March 2003 to August 2005. Table 1 lists the characteristics of the total study sample, as well as their distribution according to BMI. Most women were Caucasian and postmenopausal and none were currently taking HRT.

The distribution of Masood cytology index score and RPFNA cell counts was reported as a function of BMI. RPFNA aspirates were stratified using the Masood cytology index. Figure 1 depicts the distribution of Masood Cytology and RPFNA cell count as a function of BMI. Masood cytology index scores were significantly higher (P = 0.017) in women with BMIs ≥25 kg/m² (13.9 ± 0.42 kg/m²) than in women with BMIs <25 kg/m² (12.7 ± 0.29 kg/m²). Women with BMIs ≥25 kg/m² also had significantly higher (P = 0.028) RPFNA epithelial cell counts (1,230 ± 272) than women whose BMIs were lower (521 ± 185).

These observations show that high BMI is associated with increased RPFNA cytologic abnormalities as measured by the Masood cytology index. No significant differences were detected between groups defined by BMI including race, prior abnormal biopsy, and age. Importantly, no women were using hormone replacement at the time of undergoing RPFNA. These data show that high BMI is associated with increased RPFNA cytologic abnormalities as measured by the Masood cytology index.

These findings represent the first reported data to show that BMI is associated with direct cytologic abnormalities within the breast. These results are noteworthy given the mean Masood Index score among overweight women was 13.9, which approaches the Masood cutoff point of 14 established for atypia. The presence of atypia in RPFNA from high-risk women has been previously shown to be an independent short-term risk factor for breast cancer (9). Importantly, none of our 62 subjects currently used HRT (Table 1), thus eliminating a potential confounding variable.

The limitations of this study are that (a) we have not tested for the stability of RPFNA cytology in all of our subjects over time and (b) we have only one single dedicated cytopathologist. Currently, there are no published studies documenting the interobserver and intraobserver variability of RPFNA over time. Repeat RPFNA cytology was done in 32 of our 62 subjects as part of ongoing studies: (a) 23 of 32 (72%) subjects had the same RPFNA score on repeat testing; (b) 7 of 32 (22%) had a RPFNA score that differed by one Masood Index point; and (c) 2 of 32 (6%) subjects had RPFNA scores that differed by two points. Multi-institutional studies are under way to test for the interobserver and intraobserver variability and reproducibility of RPFNA cytology in a multi-institutional cohort.

Given the exploratory nature of this study, such findings require further validation. Additional studies are also needed to test whether these RPFNA markers are responsive to changes in body weight. If indeed they are, then these biomarkers may serve as useful intermediate end points that could be used in diet and exercise interventions that are aimed at reducing weight as a means of breast cancer prevention. The discovery of such biomarkers would have tremendous potential import because they would allow prevention studies to be accomplished in far less time, with far fewer participants and using far fewer resources.