SAN ANTONIO – Researchers at the San Antonio Breast Cancer Symposium added to a growing body of data suggesting unique tumor biology may be to blame for the worse outcomes seen in Black patients.
Black women have a 4 percent lower incidence of breast cancer but 40 percent higher mortality than white women. In an attempt to explain this disparity, in one analysis of data from the RxPONDER trial, researchers found that among those with Oncotype DX recurrence scores of 25 or lower, Black women with hormone receptor (HR)-positive HER2-negative node-involved breast cancer had worse invasive disease-free survival and five-year distant recurrence-free survival than white women. In another study, researchers looked at tumor microenvironment features that may be causing metastases after neoadjuvant chemotherapy and proposed that the worse outcomes seen in Black women with estrogen receptor (ER)-positive, HER2-negative tumors could be due to higher density of so-called tumor microenvironment of metastasis doorways.
While both analyses yielded data worthy of further follow-up, what the studies could ultimately conclude about tumor biology differences between Black and white patients were limited due to study design constraints, low accrual, and the fact that race was self-reported. Experts at the conference pointed to the need to look at ancestry genomics, consider more carefully social determinants of health, and improve diversity in trials, before drawing any definitive conclusions about associations between race and treatment outcomes and changing treatment paradigms.
"Here we have two abstracts that have observed a worse outcome for Black women," said Lori Pierce, a radiation oncologist at Michigan Medicine and the first African American female to become president of the American Society of Clinical Oncology in 2020. "But we have to consider the … heterogeneity within races, and we also have to think about the arbitrariness of race."
She pointed out how differently the historical system of "blood quantum" was applied when it came to determining whether individuals belonged to the Black and Native American races. For example, if a person had "one drop" of African ancestry in their blood, they were considered Black, Pierce said. On the other hand, Native Americans had to prove they had a specific measurement of "Indian blood" in order to qualify for treaty obligations to tribal members.
"With that backdrop, how should we as researchers analyze outcomes by race? Should we analyze outcomes by race?" Pierce posited, noting that while it is "very important" for researchers to consider race, they must do so "very thoughtfully," so they don't perpetuate some of the concepts they're trying to disprove.
No straight answers on racial disparities
The Phase III RxPONDER trial was designed to explore the utility of the 21-gene expression Oncotype DX breast cancer recurrence test in predicting adjuvant chemotherapy benefit for node-positive, HR-positive HER2-negative breast cancer patients. Researchers had originally enrolled around 5,000 patients between 2011 and 2017 with an Oncotype DX recurrence score of 25 or lower and randomized them to receive either endocrine therapy or endocrine therapy plus chemo. At this same meeting two years ago, researchers reported that adjuvant chemo benefited premenopausal women in this risk score range, but postmenopausal women could forgo chemo without sacrificing outcomes.
In the exploratory RxPONDER analysis presented at this year's meeting, researchers led by Yara Abdou, an early-stage breast cancer specialist at the University of North Carolina at Chapel Hill, considered clinicopathological characteristics, survival outcomes, and race in association with Oncotype DX recurrence scores in the 0 to 25 range. They included around 4,000 women with self-reported race in this subset, of which 70 percent were white, 6.1 percent were non-Hispanic Black, 15.1 percent were Hispanic, 8 percent were Asian, and less than 1 percent were Native American.
Pierce pointed out that representation of Black patients in the study did not reflect the diversity of the US population, since 13.6 percent of the nation is Black. Moreover, the representation of Native Americans was so low that they couldn't be included in this analysis.
Due to the fact that there were only 248 Black women included in this RxPONDER analysis, researchers couldn't draw any conclusions about the extent to which Black women with node-positive HR-positive breast cancer and Oncotype DX recurrence scores of 25 or lower benefited from the addition of chemotherapy to endocrine therapy compared to white women. Of the premenopausal women included in the five-year invasive disease-free survival analysis, for example, there were only 32 Black women who received chemo plus endocrine therapy and 26 Black women who received endocrine therapy compared to more than 400 white women in each treatment arm.
"There was no significant interaction between race and treatment arm in either premenopausal or postmenopausal cohorts," Abdou said, cautioning that due to the limited number of events in the Black cohort, "any apparent differences by race should be considered hypothesis generating."
However, in the overall cohort, unadjusted for recurrence score and other clinical factors, 87.2 percent of Black patients achieved five-year invasive disease-free survival — meaning they were alive and without breast or another invasive cancer for that time — compared to 91.5 percent of White patients. Similarly, the unadjusted five-year distant recurrence free survival rates were 90.1 percent among Black patients and 94.7 percent among white patients.
Abdou pointed out that the recurrence scores, nodal involvement, and tumor size were similar among patients included in the analysis, but Black patients had higher-grade tumors than white patients. "This is just reflecting the more aggressive biology of [Black patients'] cancers," she said. "It's interesting, though, that we see higher-grade, yet there are no differences in recurrence score."
When the five-year invasive disease-free survival analysis was adjusted according to recurrence score and other factors like treatment arm, menopausal status, age, and tumor grade, Black women still had worse outcomes, but the "race effect" was reduced when body mass index was added in.
In reviewing the data, Virginia Kaklamani, professor of medicine in the division of hematology/oncology at University of Texas Health Science Center at San Antonio, observed that other studies have similarly reported racial disparities in breast cancer outcomes. The I-SPY investigators reported last year at SABCS that Black women with ER-positive disease had worse outcomes despite having the same chemo backbone treatment. "Why are we seeing this over and over again?" asked Kaklamani.
In response to such questions, Abdou was clear that the exploratory analysis doesn't offer any straightforward reasons for the observed racial disparities. The researchers looked into social determinants of health as a possible explanation but found that Black women were more likely to accept their treatment assignment in the RxPONDER study and just as likely to remain on treatment as white women at six and 12 months.
"I speculate that there's a compilation of … biological and non-biological factors," Abdou said, adding that researchers should consider differences in endocrine therapy adherence beyond the first year and follow up on prior studies that suggest that women with higher BMI may be more resistant to aromatase inhibitors.
Kaklamani, who wasn't involved in the present RxPONDER analysis but helped recruit patients for the original trial, said it pained her to acknowledge the low number of Black patients enrolled. "This data definitely highlights that we need to do more in regards to diversity and inclusion in clinical trials," Abdou said, "and the only way to overcome this is to build out racial cohorts in our clinical trials prospectively moving forward."
In a second study presented at SABCS, researchers tried to get at some of the biological factors that might be causing more Black women with ER-positive HER2-negative breast cancer to experience distant metastases and have worse survival than white women. They hypothesized that Black women may have worse outcomes because following neoadjuvant chemo they have a more enhanced pro-metastatic response — meaning their cancer cells are more likely to disseminate to distant sites — compared to white women.
To test this hypothesis, researchers led by Maja Oktay, a cytopathologist at Montefiore Einstein Cancer Center, characterized tumor microenvironment metastasis (TMEM) doorways, which are openings in the vasculature caused by the interaction of tumor cells, macrophages, and endothelial cells and allow tumor cells to disseminate. In an analysis involving 96 self-identified Black and 87 self-identified white patients with residual cancer after neoadjuvant chemo, researchers used immunohistochemical staining to visualize the interaction of these cell types in patients' tumor samples and calculated the TMEM doorway density using an automated image scoring method.
They reported that Black patients had more macrophages and higher TMEM doorway scores than white patients in the whole cohort and among those with ER-positive HER2-negative disease but not in the triple-negative subset. And after adjusting for race, age, surgery type, lymph node status, tumor grade, and tumor subtype, a high TMEM score was associated with worse distant recurrence-free survival.
This data, in Oktay's view, may explain the racial disparities in outcomes seen with Black women, and if validated, TMEM doorway density may be used as a prognostic marker of distant recurrence-free survival and a predictive marker of response to TMEM doorway inhibitors. "We are just scratching the surface," she said, "but I think this is a pretty good introduction to where differences may lie in the dissemination of cancer."
Pierce questioned the strength of this conclusion by pointing out a few weaknesses of the analyses. Even though the researchers aimed to investigate whether neoadjuvant chemotherapy was driving a pro-metastatic effect in Black versus white women, they did not have TMEM scores on patients at baseline for comparison. Moreover, if neoadjuvant chemo is driving metastasis, then patients should have inferior outcomes compared to patients on adjuvant treatment, but past studies show comparable outcomes, Pierce noted.
"While these data are very provocative, I certainly wouldn't want Black women, or any women who need neoadjuvant therapy, to be discouraged," she said. "We need these data to be rigorously tested clinically. I look forward to the clinical trials that test this question."
To Pierce, both of these analyses point to the need for researchers to do better when it comes to studying the association between race and outcomes in clinical trials. "Without a doubt, biology, ancestry is so very important, but we also have to dig deeper to understand and address the hidden factors, the social determinants of health, that contribute to disparities attributed to race," she said. "This is our collective responsibility, and I believe the onus is on us to do so."
For example, in RxPONDER, Pierce noted, because so few Black patients were included in the exploratory analysis, the authors couldn't make any conclusions about the benefit of adding chemo to endocrine therapy, which was the goal of this research, and ultimately, the primary point of interest for oncologists. She cited research that showed that, when asked, Black patients participate in clinical trials at similar rates as white patients.
Pierce further underscored the importance of systematically studying the impact of social determinants of health. For example, RxPONDER investigators should consider how genotype-driven benign ethnic neutropenia, a type of neutropenia that is commonly seen among patients of African, West Indian, and Middle Eastern descent, might impact the dose and intensity of chemotherapy given to Black patients. Moreover, Pierce agreed with Abdou that researchers should look at endocrine therapy adherence among Black patients but suggested the need to look deeper and gauge whether Black patients with lower financial means were less likely to be adherent as prior research suggests.
"The underlying causes of the established racial differences in breast cancer risk and outcomes are complex and likely multifactorial, and the effects of socioeconomic factors and systematic racism on breast cancer also need to be further explored," agreed Rick Baehner, chief medical officer of precision oncology at Exact Sciences. He added that the latest RxPONDER analysis does not impact how the company will market Oncotype DX to Black women, since there is sufficient evidence from independently run studies in thousands of patients showing that "all women, regardless of race, can feel confident in their treatment decision made when considering results from the Oncotype DX test."
In terms of next steps, Abdou and Baehner both noted that RxPONDER researchers in collaboration with Exact will study the proliferation of gene scores and gene groups that make up the Oncotype DX recurrence score according to race and if this can explain the racial disparities observed in the RxPONDER trial.
Exact will also try to improve the prognostic ability of the recurrence score to gauge distant recurrence by integrating clinicopathological features; applying data from next-generation sequencing; and conducting real-world data analysis in collaboration with the National Cancer Institute's Surveillance, Epidemiology, and End Results. Lastly, the company has hired a health equity team to improve clinical trial diversity and is strengthening connections with healthcare systems, underserved communities, and other stakeholders.