Understanding Racial and Ethnic Inequities in Uptake and Outcomes Following Multigene Prognostic Testing in Early Breast Cancer: The Promise of Real-World Data

Breast cancer remains the most frequently diagnosed cancer and a leading cause of death among women in the United States (1). Although prognosis among patients with breast cancer has greatly improved due to advances in breast cancer screening, diagnosis, and treatment (2), inequities in access to care and survival outcomes still exist across racial and ethnic groups (3, 4). These inequities are evident among women with hormone receptor (HR)-positive, HER2-negative breast cancer, the most common subtype, which tends to have a favorable prognosis in part because of responsiveness to hormone therapy (5). Genomic testing in early-stage HR-positive, HER2-negative breast cancer aids clinical decision making by identifying patients for whom adjuvant chemotherapy would not provide additional benefit over endocrine therapy alone. Forthcoming evidence from real-world data on the uptake and accuracy of genomic testing across diverse populations is increasing our understanding of whether these gains have been equitable.

Multigene assays such as Oncotype DX (ODX) have established prognostic value to determine the likelihood that breast cancer will return and predictive value of the likelihood to benefit from chemotherapy in women with early-stage, HR-positive, HER2-negative invasive breast cancer. Results from the landmark TAILORx trial evaluating noninferiority of endocrine therapy alone versus chemoendocrine therapy for invasive disease-free survival in women with intermediate risk ODX scores have been practice-changing and have been incorporated into National Comprehensive Cancer Network guidelines (6, 7) to inform treatment decisions. Based on the TAILORx trial, 70% of women with HR-positive, HER2-negative, axillary lymph node–negative breast cancer can forgo chemotherapy when guided by ODX Recurrence Scores (RS). Per the American Society of Clinical Oncology endorsement of the Cancer Care Ontario recommendations, patients older than age 50 years and whose tumors have ODX RS < 26, and for patients age 50 years or younger with ODX RS < 16, there is little to no benefit from chemotherapy. Physicians may offer endocrine therapy alone for these patients. Patients with ODX RS > 30 should be considered candidates for chemoendocrine therapy. For premenopausal women under age 50 with scores between 16 and 25, shared decision-making is especially important because a subset of women may benefit from chemotherapy. Previous studies have shown that the uptake in use of ODX RS in clinical practice has increased over time (8). However, increasing use of genomic testing in clinical practice has also raised concerns about whether the testing was offered, conducted, and utilized uniformly among patients from different racial and ethnic groups. In an analysis among 227,259 women with early-stage estrogen receptor–positive (ER⁺) node-negative breast cancer from the National Cancer Database (NCDB), Moore and colleagues (9) reported that among patients who were eligible for ODX, patients of color, especially Black and Hispanic/Latinx women, were less likely to receive ODX tests, compared with White patients. Among those who received testing, Black patients were more likely to have higher RS. However, receipt of chemotherapy was similar across ODX RS groups, emphasizing that the opportunity for genomic testing might be a critical hurdle for consistent treatment choices to be recommended. Moore and colleagues (9) further suggested that factors potentially associated with access-to-care, such as education, income, insurance coverage, and knowledge of genomic testing availability, might be the main contributors to unequal access to genomic testing. These findings provide critical insight into the inequity in access to genomic testing for precision oncology and deepen our understanding of racial inequities among patients with breast cancer.

The NCDB is well positioned for assessing the uptake of ODX testing given the large nationwide sample with adequate representation of diverse racial and ethnic groups, and relatively complete data on ODX use over time. The availability of area-level factors associated with socioeconomic status and access to care also allowed for a closer assessment of the contribution of those factors on genomic testing uptake in diverse populations. Both Moore and colleagues (9) and a previous study (10) showed that economic, access-to-care and provider factors are significant predictors of genomic test use, but neither fully explains the observed racial/ethnic inequities in testing among patients with early-stage breast cancer. This could be due to several reasons. Black patients are more likely to have advanced disease at diagnosis, and the incremental benefit of genomic testing may be minimal because RS results are less likely to change treatment decisions. In a study of breast cancer patients in Georgia, Black patients had higher rates of chemotherapy among those who received ODX test and among those who did not, compared with their White counterparts (11). In addition, it is important to examine uptake of ODX use across racial and ethnic groups and other important potential drivers of testing uptake such as insurance status. For instance, patients of color are more likely to be uninsured or have government insurance, while White patients are more likely to have private insurance. Factors associated with providers may also contribute to differences in rates of genomic testing. It was previously suggested that patients with lymph node–negative disease (the largest group eligible for testing) were more likely to receive ODX testing if they receive care from moderate volume providers and have private insurance (10). A nationwide survey of breast cancer oncologists also suggested that there are potential biases from physicians' knowledge, attitudes and practice in recommending genomic counseling and testing for African American women (12). Socioeconomic variables are believed to be important predictors of ODX use, but area-level socioeconomic status based on zip codes may not be sufficient to provide precise estimates of these variables due to relatively large geographical areas (13, 14). Therefore, the current state of knowledge regarding the factors that drive the use (or nonuse) of genomic testing in early breast cancer and corresponding impact on cancer care disparities remains limited. Comprehensive assessment among real-world patients is needed to assess the differences in patient and provider characteristics between patients with breast cancer who received a genomic test and those who did not.

As noted by Moore and colleagues (9), among those who received ODX testing the receipt of chemotherapy was consistent across ODX RS groups and race/ethnicity. This is consistent with a previous study using the same database between 2010 and 2014 (15), but a study among patients diagnosed between 2004 and 2015 in the Surveillance, Epidemiology, and End Results registry has shown that the uptake of chemotherapy was lower among Black compared with White patients in the high-risk group (16). Whether these disparate findings reflect thoughtful shared decision-making between clinician and patient to forego chemotherapy, incorporation of other clinical factors (e.g., favorable tumor size and/or grade), increased adoption of ovarian suppression in premenopausal patients, and/or the impact of socioeconomic factors are important but unknown.

Previous observational studies have shown that the prognostic accuracy of the ODX RS for breast cancer–specific mortality and overall mortality was lower among patients of color (8, 17), potentially due to the low representation of patients of color involved in the clinical trials that were used to develop the tool. Black patients had higher hazards of breast cancer–specific mortality compared with White patients across all ODX RS risk groups (11, 17). In contrast to racial/ethnic inequities in breast cancer–specific mortality, Moore and colleagues did not find statistically significant interaction between race/ethnicity and ODX RS scores on overall mortality in their study. The variability in the observed inequities in breast cancer–specific mortality and overall mortality across these studies should be further investigated. This could be explained, in part, by the relatively short follow-up period in the present study (approximately 40 months) compared with the natural history of early-stage, HR-positive breast cancer with high relative survival. Therefore, the number of events were relatively small across race and ODX RS groups and interaction analyses were likely under-powered. In addition, it is possible that the interaction between race and ODX RS exists on the additive scale, rather than the multiplicative scale, as previously shown in the analysis of breast cancer–specific mortality among patients in Georgia (11). Black patients were also more likely to be diagnosed at a younger age and with more aggressive disease compared with White patients. It is possible that Black patients were more likely to die from breast cancer than from other competing risks of death, compared with White patients. Moore and colleagues also report that the differences in the association between ODX RS was stronger among White patients aged 50 years or older compared with younger White patients. This could suggest potential effect modification by age or menopausal status in the association between ODX RS and overall survival across different racial and ethnic groups of patients with breast cancer.

The TAILORx trial was designed to determine the noninferiority of endocrine therapy alone for invasive disease-free survival. In the current study, it is worth noting that adjustment for chemotherapy, a mediator in the association between ODX RS and overall mortality, did not change the estimated hazards of deaths materially. This result suggests that there was limited mediation by chemotherapy in addition to endocrine therapy. However, the potential differential association between ODX RS and mortality by treatment with chemotherapy cannot be ruled out. Unmeasured factors that were not adjusted for in the analysis, including adherence to endocrine and chemotherapy, types of chemotherapy, and dose of chemotherapy delivered, may also play a role in the estimated overall survival among patients with breast cancer.

The landscape of genomic testing in cancer care is rapidly changing. Moore and colleagues (9), along with previous clinical trials and observational studies, have consistently shown that Black patients were more likely to have a high risk ODX RS and that utilization of genomic testing in clinical practice remains unequal and inconsistent across racial and ethnic groups. Analyses of contemporary, real-world data from racially/ethnically diverse populations with long-term follow-up needs to keep pace with the expansion of precision breast cancer care to better understand inequities in genomic testing in order to mitigate potentially widening inequities that typically emerge in the wake of medical advances.

X. Wang reports personal fees from Flatiron Health outside the submitted work; in addition, X. Wang reports stock ownership in Roche. C. Cho-Phan reports personal fees from Flatiron Health outside the submitted work; and stock ownership in Roche. K.F. Hoskins reports nonfinancial support from Agendia; and grants from Pfizer outside the submitted work. G.S. Calip reports other support from Flatiron Health, Roche; and grants from Pfizer outside the submitted work.