ORLANDO-A New Mexico study has found that Hispanic women are more likely than white women to be diagnosed with breast cancer at a younger age, are more likely to be estrogen-receptor (ER) negative, and are at greater risk for poor survival. Diane D. Crumley, PhD, research assistant professor, Division of Epidemiology and Preventive Medicine, University of New Mexico, reported the findings at the Era of Hope Department of Defense Breast Cancer Research Program meeting.
ORLANDOA New Mexico study has found that Hispanic women are more likely than white women to be diagnosed with breast cancer at a younger age, are more likely to be estrogen-receptor (ER) negative, and are at greater risk for poor survival. Diane D. Crumley, PhD, research assistant professor, Division of Epidemiology and Preventive Medicine, University of New Mexico, reported the findings at the Era of Hope Department of Defense Breast Cancer Research Program meeting.
A review of breast cancer incidence and mortality in New Mexico over the past 30 years found that Hispanic women’s incidence rates were lower than non-Hispanic white women’s during the period. Both groups saw an increase in incidence. However, the rate of change was much higher in the Hispanic women. Their incidence rate doubled from 36.5 per 100,000 population to 75.4 from 1969 to 1998, compared with an increase from 79.6 per 100,000 to 112.1 among the non-Hispanic white women.
Furthermore, breast cancer mortality rates for the Hispanic women increased more than 58% (from 12 deaths per 100,000 in 1969 to 19 deaths in 1998), compared with a stable mortality rate for non-Hispanic white women (25.7 in 1969 and 24.2 in 1998).
The percentage of women diagnosed with regional disease decreased for both ethnic groups, but Hispanic women’s rates of regional breast cancer were consistently higher throughout a 20-year period (1981 to 1999).
In an attempt to understand the factors responsible for the increase in breast cancer mortality among New Mexico Hispanics, Dr. Crumley examined risk factors and survival data from the Health, Eating, Activity and Lifestyle (HEAL) multicenter study of breast cancer prognosis funded by the National Cancer Institute.
The New Mexico HEAL study recruited 654 women with breast cancer (24% Hispanic, 76% non-Hispanic white) diagnosed between 1996 and 1999 (age range: 29 to 91 years). After 24 months, the researchers interviewed 526 of these women and performed blood tests, measured body composition, and abstracted medical records for all reported breast cancer recurrences. Among the women who completed the 24-month assessment, there were 498 invasive breast cancer cases, 124 Hispanic, 374 non-Hispanic white.
The team found ethnic differences in income, with a larger percentage of Hispanic women reporting a household income of less than $20,000 per year. However, there were no ethnic differences in insurance status or having a regular physician. There were no differences in prior mammography screening between ethnic groups, although slightly more Hispanic women, ages 40 to 49, had not had prior mammography.
There were no significant ethnic differences in tumor stage, but slightly more non-Hispanic white women were diagnosed at local, TNM stage I.
There was a significant ethnic difference in body mass index measurement at baseline. Hispanic women were more likely to be overweight, according to WHO standards. The difference in waist-to-hip ratio was of borderline significance: 42% of the Hispanic women were in the highest third of waste-to-hip ratio vs 31% of the non-Hispanic whites.
"We’re interested in this measure, because we know it is associated oftentimes with insulin levels, which can increase growth of both normal breast cells and breast cancer cells," Dr. Crumley said.
Hispanic women had slightly larger tumors at diagnosis and were more likely to receive surgery plus chemotherapy, 36% vs 24%. Surgery plus radiation therapy was more common in the non-Hispanic white women.
A significant ethnic difference was noted in ER-negative status (15% non-Hispanic white vs 26% Hispanic) and progesterone-receptor (PR)-negative status (24% vs 35%, respectively).
The Hispanic women were more likely to be less than 50 years old at the time of diagnosis (40% vs 21% for whites). "Part of this is due to the fact that Hispanics have a younger population age structure overall," Dr. Crumley said.
At the end of 3 years, 91% of non-Hispanic white women were alive vs 85% of Hispanic women, Dr. Crumley said. During this period, 11% of Hispanic women died of breast cancer vs 4.5% of non-Hispanic white women, a significant difference. The team did not see an ethnic difference in deaths due to other causes.
Of the women who died of breast cancer, 50% of the Hispanic women were under age 50 at the time of death vs 29.5% of the non-Hispanic white women; 53% of the non-Hispanic white women were 65 or older at the time of death vs 21% of the Hispanic women.
Among the various factors examined, only twostage at diagnosis and ER-negative statuswere significantly associated with poorer survival. In both cohorts, women with ER-negative status showed a reduction in short-term survival. The decrease was larger (but not significantly so) in Hispanic women than in non-Hispanic white women.
There were no ethnic differences in survival among women diagnosed with local stage I breast cancer. However, Hispanic women with stage IIA or greater disease had significantly reduced survival, compared with non-Hispanic whites with similar-stage disease, Dr. Crumley said (see Figure).
"Adjusting for all factors, including age, menopausal status, treatment, stage, ER status, and body mass index, the odds of death from breast cancer are twice as high in Hispanic women as in non-Hispanic whites," she said.
In the future, Dr. Crumley plans to investigate factors that may be responsible for development of ER-negative breast cancer. The team also will work to identify factors associated with poorer stage-specific survival among Hispanic women and factors linked to the diagnosis of early- and later-stage breast cancer.