Impact of Race and Health Insurance Status on the Outcomes of Neoadjuvant Chemotherapy and Radiation Therapy for Breast Cancer Patients

Abstract

Racial inequalities in breast cancer (BC) mortality rates continue despite general improvement in BC survival.1 We seek to discern racial and socioeconomic disparities in BC outcomes in patients who received neoadjuvant chemotherapy (NAC) and adjuvant radiation therapy (RT). We evaluated how patient characteristics, tumor biology, and radiation treatment plans influence outcomes among BC patients treated with NAC. We identified BC patients treated with NAC from 2006 to 2018 at our institution. Univariable and multivariable binary logistic regression analyses were performed to estimate the effects of race, health insurance status, and other predictors on NAC treatment outcomes. Outcomes of interest included pathologic complete response (pCR), partial response (PR), and cancer recurrence. Predictors that demonstrated significant unadjusted effects were included in the multivariable models. Variables included in the final multivariable models were health insurance status, race, Ki-67, BMI, hormone receptor (HR) status, and clinical group stage. This IRB approved study included 298 BC patients who received NAC. 69 identified as African American (AA), 227 identified as non-AA, and 2 were not reported. 168 had private insurance, 111 had Medicare/Medicaid, 17 had no insurance, and 2 were not reported. There was no significant association between race and any of the treatment outcomes. Insurance status emerged as a significant predictor for any response to NAC (pCR or PR) in both univariable and multivariable analyses (p = 0.01). Odds of pCR or PR attainment were lower for patients with Medicare/Medicaid compared to patients with private insurance (OR 0.36, 95% CI: 0.18 – 0.71). Other variables significant for attainment of either pCR or PR upon multivariable analysis included BMI (p < 0.01), HR status (p = 0.01), and clinical group stage (p < 0.01). Ki-67 (p = 0.02) and HR status (p < 0.01) demonstrated significant effects on pCR alone where odds of pCR were greater for triple negative breast cancer (TNBC) compared to ER+/PR+/HER2- (OR 4.10, 95% CI: 1.50 – 11.23). While AA patients had similar treatment outcomes as non-AA patients in our study, insurance status, Ki-67, BMI, clinical stage, and HR status correlated with outcomes and can potentially serve as useful predictive factors for cancer treatment response. In particular private health insurance and TNBC were correlated with improved outcomes. Further exploration of the impact of health insurance and HR status on treatment outcomes in BC patients treated with NAC may help predict and improve outcomes in vulnerable populations.