Abstract

Abstract Background: Triple-negative breast cancer (TNBC) accounts for a disproportionate amount of poor outcomes among breast cancers. A subset of TNBCs demonstrates an aggressive course with marked chemoresistance, rapid distant metastatic spread, and poor survival. The clinicopathologic and sociodemographic features associated with rapid relapse among TNBCs remain poorly understood. Primary Objective: To evaluate the relationship between clinicopathologic and sociodemographic features with rapid relapse in TNBC (rrTNBC). Methods: This large multi-institutional study analyzed a cohort of breast cancer patients diagnosed with TNBC who received treatment at one of ten academic centers that previously participated in a National Comprehensive Cancer Network (NCCN) outcomes database between 1998 and 2012. We defined rrTNBC as a distant metastatic recurrence event or death from any cause ≤24 months after diagnosis. We included patients with ≥2 years follow-up or had suffered a survival event within that timeframe. We excluded patients with de novo metastatic disease and those who did not receive chemotherapy. We randomly divided the total dataset into 70% training and 30% validation cohorts, balanced by the number of rrTNBC events. Covariates included study site, age at diagnosis, body mass index (BMI), race/ethnicity, education, median annual household income (2000 census tract), insurance type (Managed Care, Medicare, Medicaid, and Other), Charlson comorbidity index, tumor stage and grade at diagnosis, and adjuvant radiation treatment. Logistic regression was performed among the training dataset univariately for associations with rapid relapse vs. not. Features with a p-value <0.10 were included in a multivariable logistic regression model, using backward elimination performed with a p<0.10 criterion. The final multivariable model was applied to the training dataset and to the independent validation cohort. Results: Among 41,839 patients with invasive breast cancer treated in these ten centers, 5256 had TNBC (12.6%), among whom 3016 had adequate follow-up to be included in the analysis. Bivariable analyses in the training cohort (n=2112) identified tumor stage at diagnosis, insurance type, age at diagnosis, BMI, race, and income to be associated with rrTNBC events (p<0.10). The multivariable model identified tumor stage at diagnosis, insurance type, and age at diagnosis as significant contributors. In the final multivariable model, tumor stage was the most significant factor, with patients who presented at stage III having a >15x increased risk of rapid relapse (adjusted OR [95% CI]: 16.5 [10.3, 26.4]; p<0.0001) compared to stage I. Additionally, patients with Medicare (aOR: 2.25 [1.35, 3.73]; p=0.002) versus Managed Care and patients in the youngest age category (aOR: 1.90 [1.27, 2.86]; p=0.002) versus 60-69 years old were significantly more likely to develop rrTNBC. Patients with Medicaid had a non-significant trend towards an increased risk of rrTNBC (aOR: 1.43 [0.98, 2.09]; p=0.06). In the validation cohort (n=904), tumor stage at diagnosis and insurance remained significantly associated with rrTNBC. Conclusion: Advanced tumor stage at diagnosis was the most influential predictor of rapid relapse among patients who had TNBC, while type of insurance remains an independent predictor in training and validation cohorts. Given the known association of sociodemographic disparities with tumor stage, further study of underlying causes and potential interventions to reduce rapid relapse of TNBC is warranted. Citation Format: Sarah Asad, Carlos H. Barcenas, Richard J. Bleicher, Adam L. Cohen, Sara H. Javid, Ellis G. Levine, Nancy U. Lin, Beverly Moy, Joyce Niland, Antonio C. Wolff, Michael J. Hassett, Daniel G. Stover. Factors associated with rapid relapse in triple negative breast cancer: A multi-institution study [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P5-07-02.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call