Abstract PD11-06: Associations between obesity and metastasis in the Carolina breast cancer study

Abstract

Abstract Background. Metastatic disease accounts for the majority of breast cancer-specific mortality, and identifying women who are most at risk for metastasis remains an important challenge. Genomic risk scores are valuable tools in assessing risk of recurrence but have rarely been applied to assess risk of distant metastasis in large, diverse population-based cohorts. Such cohorts can identify important modifiers of the relationship between tumor genomics and metastasis. In particular, obesity at diagnosis has been implicated as a risk factor for breast cancer metastasis. Waist-to-hip ratio (WHR), as a measure of central adiposity, appears to increase risk of breast cancer mortality. It is important to understand how obesity impacts risk of metastasis and modifies other known prognostic factors. Methods. Data from the Carolina Breast Cancer Study, a prospective cohort of women with incident breast cancer that oversampled young women and black women such that each represented 50% of the study population, were used to study metastasis within five years of follow up (n=239 incident metastases in 2767 eligible participants with stage I-III disease). PAM50-derived risk-of-recurrence (ROR-PT) scores were measured using a research version of the 50-gene assay, and represent a multigene classifier incorporating tumor subtype, size, and proliferation score. Women with de novo metastasis (identified at diagnosis, n=109) were excluded, as incident and de novo metastases have different prognosis. Obesity was defined by WHR (> 0.85). Absolute risk within strata and risk differences between strata were calculated using the Kaplan-Meier estimator; models were adjusted with inverse probability weights conditional on grade, stage, age, and race, where appropriate. To examine the association between obesity and site or multiplicity of metastasis, relative frequency differences were estimated using generalized linear models. Results. The overall risk of metastasis in CBCS differed significantly by race (11.2% in Black women and 6.8% in non-Black women and was associated with higher ROR-PT score (19.8% in high ROR-PT vs. 6.5% in low or medium ROR-PT). 48.2% of women were classified as obese by WHR. Women with high WHR had a significantly higher absolute risk of metastasis (11.2%, 95% CI: 9.4, 13.0) relative to those with lower WHR (6.9, 95% CI: 5.6, 8.2). However, the effects of WHR on risk seemed to be isolated to those with high risk tumors. Non-obese and obese women with low-to-medium ROR-PT scores had similar risk of metastasis [RDLowMed (95% CI) = 1.5% (-1.5, 4.4)], while obese women with high ROR-PT scores had substantially higher risk of metastasis than non-obese women [RDHigh (95% CI) = 13.1% (4.3, 21.9); 26.2% vs. 13.1%]. Considering site of metastasis, more metastases were observed among women with high WHR at every site except brain. More occurrences of multiple metastases were also observed among obese women. Conclusions. Consistent with previous studies, we observed racial disparities in metastasis, with Black women having higher 5-year metastatic risk and with women with high ROR-PT scores also having higher risk. Obesity, herein measured by WHR, is an important predictor of metastatic risk and may modify the relationship between genomic risk scores and risk of metastasis. Citation Format: Linnea T Olsson, Andrea Walens, Alina M Hamilton, Halei C Benefield, Kevin P Williams, Jodie M Fleming, Stephen D Hursting, Melissa A Troester. Associations between obesity and metastasis in the Carolina breast cancer study [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD11-06.