The Assessment of Race, the 21-Gene Recurrence Score, and Breast Cancer Outcomes at Kaiser Permanente
Amanda F. Petrik, Charisma L. Jenkins, Ana G. Rosales, Terry Kimes, Ning Smith, Nathalie Johnson, Amy Morris, Suma Vupputuri, A. Blythe Ryerson, Matthew P. Banegas, Robert B. Hufnagel, Lilian G. Perez, David MosenBackground: It is well documented that Black and African American (AA) patients have the highest risk of recurrence of breast cancer. However, less is known about racial differences in gene expression profiling tests such as Oncotype DX that predicts cancer recurrence. This study aimed to assess the predictive differences in Oncotype DX scores for breast cancer outcomes between Black/AA and non-Hispanic White patients. Methods: We identified Oncotype DX testing status in patients ages 18–75 years who were newly diagnosed with breast cancer, assessed race and other factors among patients with different testing statuses, and described the distribution of the Oncotype DX score. Additionally, we assessed treatment, and multifactorial impact on outcomes by Oncotype DX score, including: receipt of chemotherapy, endocrine therapy, stage, and disease-related recurrence and mortality. Results: We identified 2577 eligible patients for the analysis. Black/AA patients had a 49% greater risk/hazard of mortality than non-Hispanic White (NHW) patients (multivariable model p-value 0.0444). Oncotype DX score differences were associated with mortality in the overall sample (multivariable model, p-value = 0.0281). Specifically, in the Oncotype DX-specific models, Black/AA patients with an Oncotype DX score of 0–16 had 73% greater risk of mortality than NHW patients (multivariable model p-value 0.0409). While all other Oncotype DX score categories showed elevated risks of mortality, they were not statistically significant. The Oncotype DX-specific multivariable models showed that, compared to NHW patients, recurrence was lower for Black/AA patients in the lower categories and higher in the upper categories, but results were not statistically significant. Conclusions: This study assessed recurrence and mortality outcomes by the recurrence score and found significantly higher rates of mortality among Black/AA patients in the lowest recurrence score category. Other studies have shown that recurrence scores add prognostic value during diagnosis. However, examining performance of these tools among diverse populations is imperative.