Tumor Subtype Fails to Explain Racial Disparity in Breast Cancer Survival
Washington, DC—Black women had worse survival across all breast cancer subtypes, indicating that the survival disparity compared with white women cannot be blamed entirely on more frequent diagnosis of less treatable subtypes, based on data from a breast cancer survivor study.
Consistent with previous data, black women more often had difficult-to-treat triple-negative breast cancer and were less likely than white women to have the more treatable luminal A subtype. Mortality was twice as high in blacks as in whites, as documented in most studies.
However, a significant survival disparity in favor of whites persisted across all cancer subtypes, including a 2-fold greater likelihood of dying of luminal A, luminal B, or HER2-positive breast cancers.
“Even though African American women more often had hard-to-treat tumors, [they] appeared to have a poorer prognosis regardless of subtype,” said Candyce Kroenke, MPH, ScD, a scientist at the Kaiser Permanente Division of Research in Oakland, CA, at the 2013 American Association for Cancer Research meeting. “It seems from our data that the black-white breast cancer survival difference cannot be explained entirely by variable breast cancer subtype diagnosis.”
Studies have consistently shown worse outcomes in black versus white patients with breast cancer, including approximately a 40% higher mortality risk. Although the precise reason (or reasons) for the disparity has yet to be identified, explanations include more advanced stage at diagnosis and differences in treatment, comorbidity burden, and disease aggressiveness.
However, survival disparities have been observed in controlled clinical trials wherein stage at diagnosis, uniformity of treatment, and follow-up were comparable for all patients. Previous studies had not examined the race-survival disparity after accounting for the distribution of breast cancer subtype, as determined by the PAM50 Breast Cancer Intrinsic Classifier gene array classifier, said Dr Kroenke.
The Life After Cancer Epidemiology (LACE) and Pathways studies of breast cancer survivors afforded an opportunity to examine the relationship between intrinsic tumor subtype by PAM50 and subsequent survival. A cohort study involved 1643 breast cancer survivors with archived tumor specimens available for analysis. The study population comprised 1213 white women, 132 black women, 144 Latina women, and 154 Asian/Pacific Islander women.
Clinical and demographic information was obtained for each LACE participant by means of a mailed questionnaire and electronic medical records. Investigators analyzed 1-mm punch samples from tumor specimens, determining PAM50 molecular subtype by quantitative polymerase chain reaction assay of tumor RNA.
Luminal A subtype accounted for 76% of specimens from white women compared with 6% from blacks, 7% from Latinas, and 10% from Asian/Pacific Islanders. Compared with white women, black women were 50% as likely to have luminal A or luminal B intrinsic tumor subtypes, including almost a 40% lower likelihood of luminal A tumors, Dr Kroenke reported.
Black women were more than 4 times as likely to have basal-like tumors versus luminal A subtype tumors in comparison with their white counterparts.
During a median follow-up of 6.3 years, 499 LACE participants died, including 268 deaths from breast cancer. The data showed that black women had a 57% risk of recurrence and a 71% increased risk of breast cancer mortality compared with whites. The Latina and Asian/Pacific Islander subgroups did not differ significantly from whites with respect to either outcome.
The hazard ratios for breast cancer death by tumor subtype were consistently higher for blacks compared with whites, but these ratios were not statistically significant:
- Luminal A, 2.30
- Luminal B, 2.59
- Basal-like, 1.27
- HER2, 2.44.
However, a clear trend toward higher mortality risk was observed for all intrinsic tumor subtypes, noted Dr Kroenke.
“We need to examine other potential reasons for the black-white survival disparity. The distribution of different tumor subtypes does not explain the disparity,” she noted.