National Action Plan on Breast Cancer Workshop onMulticultural Aspects of Breast Cancer Etiology

Supplement to Cancer

Abstracts of Poster Presentations

Presented at the National Action Plan on BreastCancer Workshop on Multicultural Aspects ofBreast Cancer Etiology, Washington, DC, March17–19, 1999.

Alterations in the Estrogen Receptor mRNA in theBreast Carcinoma Tumors of African-AmericanWomenSailaja Koduri and Indra PoolaOBJECTIVE. This study was conducted to gain insight into the factors that may be

responsible for poorly differentiated aggressive tumors.

BACKGROUND. Several recent reports have shown that the breast carcinoma mor-

tality rate is approximately three times as high in African-American women com-

pared with other populations. In addition, the available data also indicate that the

tumors are very aggressive and poorly differentiated, with a very low frequency of

hormone receptors.

METHODOLOGY. The authors investigated the transcript profiles of the estrogen

receptor (ER), the most important prognostic factor in breast carcinoma, in the

tumors derived from African-American women. They analyzed 15 immunohisto-

chemically ER-positive and 6 ER-negative malignant tumors for estrogen receptor

mRNAs by reverse transcriptase-polymerase chain reaction using a number of

primer pairs. For comparative purposes, five tumor tissues derived from white

women also were included.

RESULTS. Three types of modifications were observed in a majority of tumors: 1)

truncations/alterations in exon 8; 2) the tumors that had alterations/truncations in

exon 8 also had base insertions; and 3) absence of the naturally occurring exon 7

deletion variant ER. Unexpectedly, two of six ER-negative tumors studied had

full-length receptor mRNAs and none of the variant transcripts.

Arabic Women and Breast Carcinoma: Loss ofHeterozygosity and Microsatellite Instability of theBRCA1 LocusR. Ali, G. Wiesner, N. Kaisi, R. Badin, G. Pals, and M. J. WorshamOBJECTIVE. The current study was conducted to investigate the prevalence of loss

of heterozygosity (LOH) and microsatellite instability of the BRCA1 region in Arabic

women with breast carcinoma.

BACKGROUND. LOH of BRCA1, a tumor suppressor gene, is one mechanism of

genetic inactivation in both sporadic and familial forms of breast carcinoma.

Women of northern European descent have shown LOH of 30 –50% in sporadic

tumors. Microsatellite instability (MSI) has served as evidence of the involvement

of DNA repair genes.

METHODOLOGY. Genomic DNA from normal and malignant tissue was analyzed

from 43 Arabic women treated for breast carcinoma at the University of Damascus,

Syria. Using two intragenomic BRCA1 markers (D17S1323 and D17S855), polymer-

ase chain reaction was performed twice on each case, and three independent

reviewers analyzed results. Compared with normal DNA, MSI and LOH were

recognized as a gain and a loss, respectively, of one signal in one allele in the

tumor DNA.

1267

© 2000 American Cancer Society

RESULTS. DNA analysis was possible on tumor and normal

tissue from 18 patients. Five of these 18 cases (28%) showed

MSI of at least 1 marker. In 17 cases, at least 1 marker was

informative for LOH (heterozygous in normal tissue); 11 (65%)

showed LOH in tumor. Thirteen of 18 cases (72%) showed

MSI, LOH, or both. Of these, seven patients had a positive

family history.

CONCLUSIONS. The proportion of aberrant findings of the BRCA1

locus in breast carcinoma appears to be higher in Arabic women

(72%) than in other populations studied to date, many of whom

presented with a family history of breast carcinoma as well as

early onset breast carcinoma. The latter may uncover germline

mutations in this cohort and reveal either a founder effect such

as that observed among Ashkenazi Jews or a different mutational

spectrum.

Breast Carcinoma Screening Practicesamong HispanicsAmelie G. Ramirez, Dr.P.H., Lucina Suarez, Ph.D., andPatricia Chalela, M.P.H.

OBJECTIVE. This study was conducted to learn more about breast

carcinoma risk factors and regional differences in screening

participation among diverse Hispanic groups.

METHODOLOGY. Data collected from Hispanic women age $40

years (n 5 2082) through a random-digit telephone survey were

analyzed using logistic regression.

RESULTS. Significant differences in breast carcinoma screen-

ing behaviors were found among different Hispanic groups.

Recent mammography (within the preceding 2 years) ranged

from ,50% among Mexican-American women living along the

Mexico border in Texas to .70% among Central-American

women living in San Francisco and Cuban women living in

Miami. Although substantial differences were observed in age

structure and health care coverage among the diverse groups

in different regions, variations in mammography participation

rates remained after adjusting for these variables.

CONCLUSIONS. Although traditional demographic characteris-

tics influence mammography behaviors, factors that affect the

use of breast carcinoma screening services among different

ethnoregional populations include other variables that are not

well defined currently. Cultural differences among Hispanic

groups may play a strong role in determining preventive care

behaviors in general and breast carcinoma screening prac-

tices in particular. Clearly, researchers should avoid the temp-

tation to make generalizations regarding the Hispanic popu-

lation. The obvious heterogeneity of Hispanics suggests that

efforts to cluster data under this general classification should

be considered carefully in current and future research. The

development of cancer prevention and health care delivery

services for these audiences and communities should recog-

nize the diversity of Hispanic groups, take into account both

their differences and similarities, and concentrate efforts

where the need is greatest.

Can Racial Differences in BreastCarcinoma Survival Be Explained byAccess to Medical Care?Barbara Wojcik, Ph.D., and Martha Spinks, M.A., M.S.W.

BACKGROUND. This retrospective review of breast carcinoma

cases in the Department of Defense (DoD) Central Tumor Reg-

istry (Phase I Design) evaluated differences in survival patterns

between African-American and white women treated in U.S.

military health care facilities. The authors examined the effects

of age, stage of disease, tumor size, grade, lymph node status,

waiting time between diagnosis and first treatment, marital sta-

tus, military dependent status, alcohol usage, tobacco usage, and

family history of cancer. Results of this review prompted the

follow-up study (Phase II design) of breast carcinoma patients

treated and/or diagnosed in the past 10 years at Brooke Army

Medical Center (BAMC), San Antonio, Texas.

METHODS. Researchers originally reviewed tumor registry

records of 6577 women (5879 whites and 698 African Americans)

diagnosed with breast carcinoma. The subjects, ages 19 –97

years, were diagnosed between 1975 and 1994. Survival models

were used to compare African-American and white patients,

adjusting for various combinations of covariates. Impact of in-

dependent variables on risk of death, prognostic factors signifi-

cantly associated with survival, disease free and overall survival

times, and trends in stage at diagnosis also were examined. In

Phase II, survey questionnaires were sent to .900 women diag-

nosed at BAMC. The survey collected information regarding how

breast carcinoma was diagnosed (self-examination, physician

examination, or mammogram), reasons for delay in seeing a

physician, additional primary tumors, family history of breast

carcinoma, alcohol and tobacco use, dietary patterns, menstrual

periods and menopause, childbirth, breast feeding, etc. Addi-

tional information on these women was retrieved from medical

records by certified tumor registrars: estrogen receptor/proges-

terone receptor status, tumor histology, adjuvant therapy, radi-

ation therapy, type of surgery, and previous primary tumors.

RESULTS. The Phase I study analyzed differences in survival

between African-American and white DoD patients with breast

carcinoma. Approximately 22.8% of African-American women

were diagnosed at age #40 years compared with 9.2% of white

patients, and 12.3% of African-American women were diagnosed

at age $65 years compared with 26.3% of white patients. In

addition 45.9% of African-American women were diagnosed at

Stage 0/I compared with 55.9% of whites. After adjusting for age,

the risk of death was 1.45 times (95% confidence interval [CI],

1.20 –1.76) greater for African-American women than for white

women. Adjusting for stage reduced the risk to 1.41 (95% CI,

1.16 –1.70); further adjustment for demographic variables and

the majority of clinical variables had no effect. Partial analysis of

the Phase II project for 447 women who answered the survey

revealed that the mean age at diagnosis was 47.5 years for

African-American patients, 53.5 years for Hispanic patients, and

58.6 years for white patients. Tumor size ,1 cm was found in

1268 CANCER Supplement March 1, 2000 / Volume 88 / Number 5

11.63% of African-American women, 13.3% of Hispanic women,

and 21.63% of white women. Breast carcinoma was diagnosed by

mammography in 30.23% of African Americans, 46.67% of His-

panics, and 46.77% of whites. An even larger discrepancy was

found for the percentage of in situ cases diagnosed by mammo-

gram: 7.69% for African-Americans, 28.57% for Hispanics, and

25.86% for whites.

CONCLUSIONS. Results of the Phase I study suggest that ready

access to medical facilities and the full complement of treatment

options improves survival rates for African-American women.

However, a significant and unexplained difference still exists

between African-American and white military beneficiaries. The

follow-up Phase II design added Hispanic patients to the study.

Phase II, which currently is in progress, will attempt to explain

differences in survival between ethnic groups. Phase II also will

add more information about lifestyle factors and personal his-

tory that may correlate with breast carcinoma.

Clinical and Mammographic BreastCarcinoma Risk Factors in Four NativeAmerican PopulationsJ. S. Kaur, M. Roubidoux, J. A. Sloan, P. J. Novotny, andM. LobellOBJECTIVE. The current study presents comparative data from

1400 screening mammograms of American Indian and Alaskan

Native women in four geographically disparate populations:

Alaska, Arizona, and urban and rural South Dakota.

METHODOLOGY. Retrospective chart reviews were performed for

women who received screening mammograms in Alaska; Ari-

zona; Rosebud, SD; and Rapid City, SD. Data were collected on

breast carcinoma risk factors and mammographic descriptors.

RESULTS. There were significant differences in the breast density

distributions among the four populations. The percentage of

mammograms indicating moderate or very dense tissue was

45%, 37%, 55%, and 47%, respectively, for the four groups (P 5

0.004). Compared with a University of Michigan normative pop-

ulation, the breast density was lower in the native populations.

The reported incidence rate of a family history of a first-degree

relative with breast carcinoma was similar to that reported in the

general population. Hormonal replacement therapy varied

among the groups and, surprisingly, was not correlated with

breast density. Ages at first birth, menarche, and menopause

were similar among the four native populations.

CONCLUSIONS. Patterns of mammography in Native Americans

appear to be different from a normative white population and

have considerable variability among the four populations stud-

ied. The presence of low breast density in these populations

could make mammographic detection of abnormalities easier.

Biologic risk factors generally are protective. Variation in breast

carcinoma rates among the four groups is not explained com-

pletely by these findings.

THIS WORK WAS SUPPORTED IN PART BY THE NATIONAL CANCERINSTITUTE AND THE NORTH CENTRAL CANCER TREATMENTGROUP.

A Review of Factors that InfluenceSurvival Rates among Minority andMajority PopulationsAaron Banks, Susan Robinson, M.D., M.P.H., and SamuelShacks, Ph.D., M.D.

OBJECTIVE. Minority patients, especially African Americans, are

prone to develop aggressive forms of breast carcinoma and are

more likely to die from the disease compared with majority

groups. The aim of this study was to identify risk factors and

biologic factors that influence breast carcinoma aggression and

survival rates among high-risk populations.

METHODOLOGY. A review of the literature with a focus on iden-

tifying culturally specific risk factors for breast carcinoma was

conducted. The review highlights molecular factors that impact

breast carcinoma pathogenesis and prognosis.

RESULTS. Differences in risk perception, knowledge, and breast

carcinoma screening behavior among minority and majority

populations explain, in part, the poor survival rates among Af-

rican Americans. Body weight and physical activity may contrib-

ute to the disparity of breast carcinoma aggression among mi-

nority and majority populations. Reviews of the literature in

molecular biology suggest that minority patients are predis-

posed to the most aggressive forms of breast carcinoma. The

presence of the aromatase enzyme and cathepsin D protease are

key factors in breast carcinoma growth and metastasis. These

molecules should be of interest in examining breast carcinoma

treatment and survival.

CONCLUSIONS. Additional studies of molecular research may be

beneficial in identifying patients at risk for developing aggressive

forms of breast carcinoma. A better understanding of factors

associated with the aggressiveness of breast carcinoma may lead

to the design of more effective therapies.

THIS PROJECT WAS SUPPORTED BY GRANT DAMD17-94-J-4437FROM THE DEPARTMENT OF DEFENSE.

An Innovative Method for IncreasingParticipation of African-AmericanWomen in Epidemiological StudiesDeborah O. Erwin, Rebecca Morris-Chatta, StephanieLong, and Christine B. AmbrosoneOBJECTIVE. This study was conducted to develop and refine

methodology to effectively encourage participation of African-

American women in research studies.

BACKGROUND. Low participation in epidemiologic studies is a

significant problem. Those who do consent to participate may

not be representative of the general population.

METHODOLOGY. Utilizing successful models for cancer education

and treatment trial recruitment, the authors developed an inno-

vative method of using survivors as advocates for research par-

ticipation. Women who are breast carcinoma survivors are

matched by racial and residential backgrounds to recruit the

potential participants. An introductory postcard with the recruit-

er’s photograph on it is sent, followed by a telephone call. This

Poster Session Abstracts 1269

provides a culturally appropriate process to overcome the bar-

riers to recruitment, especially for African-American women.

Relying on data obtained from focus groups, interviews, and

taped recruitment calls, training procedures for recruiters were

standardized into a training manual.

RESULTS. To date, 22 of the 33 African-American women con-

tacted for the study (67%) have agreed to participate, as have 116

of the 160 white women (73%). An additional 21 recruitment

calls (to 11 African-American women and 10 white women) were

pending.

CONCLUSIONS. As the authors continue to enhance their recruit-

ment methods, rates are much improved compared with rates

using standard methodology in another case– control study in

this difficult-to-reach community (30% for controls and 37% for

cases). Frequent social gatherings with the recruiters are used to

maintain interest and motivation and to share success stories

and difficulties encountered.

Native American Breast CancerSurvivors’ Support NetworkLinda Burhansstipanov, M.S.P.H., Dr.P.H., CHES

OBJECTIVE. This study was conducted to identify patterns of

disease and patterns of care that differ from Native American

breast carcinoma survivors in comparison with non-Native

American breast carcinoma survivors.

BACKGROUND. There are insufficient data on Native American

cancer survivors. The National Native American Breast Cancer

Survivors’ Support Network is a project being conducted by the

National Indian Health Board, the Denver Indian Center, and

the “Gathering of Cancer Support” of Santo Domingo Pueblo.

The project is funded by the Department of Defense (1999), the

Susan G. Komen Breast Cancer Foundation (1997 and 1998), and

the Breast Cancer Fund (1998). The network is designed to

develop a national database to learn more regarding the patterns

of disease (e.g., age at diagnosis) and patterns of care (e.g., ability

to access state-of-the-art care) among Native Americans. Infor-

mation from the database will be used to develop interventions

to improve the survival from breast carcinoma and quality of life

after being diagnosed with breast carcinoma for both the patient

and her loved ones.

METHODOLOGY. A survivors’ intake instrument was developed

throughout 1998. The instrument requires approximately 2 hours

to complete over the phone. The survivors receive support re-

sources (such as prepaid long distance phone cards and print and

video support materials) for taking part in the data collection pro-

cess. Medical records are reviewed to provide validation and cor-

rection of self-reported information from the patient’s intake data.

RESULTS. Preliminary data from the Native American Breast Can-

cer Survivors’ Support Network are presented. The anticipated

outcome is the development of a cancer survivor database com-

prising information on at least 500 Native Americans that may be

used to help learn more about how breast carcinoma affects Native

Americans. A second outcome is the identification of a cohort for

subsequent breast carcinoma survivor research studies.

Roles of Nutrition and Migration inBreast Carcinoma Risk for PolishWomenDorothy Rybaczyk Pathak, Ph.D.

OBJECTIVE. This study was conducted to develop an understand-

ing of nutritional risk factors and their implications in breast

carcinoma etiology. This study brings together research in nu-

trition, epidemiology, and changes in risks due to migration in

strategically selected populations with a threefold difference in

breast carcinoma mortality. The study will contribute to the

understanding of the role that specific foods and phytochemi-

cals play in lowering breast carcinoma risk, with clear implica-

tions for prevention.

BACKGROUND. The incidence of breast carcinoma among Polish

women is approximately 33% that of U.S. women. Yet recent

studies of Polish immigrants to the West showed breast carci-

noma mortality rates for immigrants similar to those of the host

country. No other population of immigrants has shown so rapid

a transition. The short time needed to express this changing risk

for breast carcinoma implicates modifiable environmental fac-

tors as significant determinants of risk.

METHODOLOGY. The authors are conducting parallel, popula-

tion-based, case– control studies of women ages 20 –79 years in 2

populations: 1) Polish-born immigrants in Cook County and the

Detroit metropolitan area, and 2) Polish natives in Warsaw,

Poland. The authors hypothesize that components of current

and past diet are associated with breast carcinoma risk and that,

after adjusting for established risk factors, breast carcinoma risk

will be increased by 1) reduction in the intake of specific foods

such as cruciferous vegetables, root vegetables, whole grain

breads, and other cereals; 2) resulting reduction in the intake of

phytochemicals (e.g., glucobrassicins, carotenoids, flavonoids,

and antioxidants) and dietary fiber; 3) increase in the intake of

meat and dietary fat; and 4) interaction among the dietary con-

stituents in 1–3 above.

Using Guided Focus Groups in BreastCancer ResearchM. E. Ford, D. Hill, J. M. Worsham, C. C. Johnson,and S. WolmanOBJECTIVE. This study was conducted to describe the results of

two age-specific guided focus groups held with African-Ameri-

can women to evaluate a breast carcinoma risk factor survey.

METHODOLOGY. A health system patient database was used to

identify African-American women ages 18 –50 years (Focus

Group 1) and those age $ 50 years (Focus Group 2). From these

listings, 15 women were selected randomly, called, and invited to

each focus group. Eligible and interested women received a

mailed confirmation of their focus group and a reminder call.

Each 2-hour focus group was videotaped.

RESULTS. The women in the younger age group (n 5 12) stated

that the rationale for the item on race/ethnicity was not clear,

the relevance between the parent’s country of origin and breast

1270 CANCER Supplement March 1, 2000 / Volume 88 / Number 5

carcinoma risk was not clear, and it was difficult to remember

the number of menstrual periods they had had in previous

decades. The women in the older age group (n 5 9) stated that

in the past, their physicians did not name their medications. The

meaning of several terms, such as “demographics,” was not

clear, and family medical history often was unknown. Women in

both age groups stated that it was difficult to recall previous

average weight, alcohol consumption, and level of physical ac-

tivity, and that the sports listed were not culturally appropriate.

CONCLUSIONS. The results show that questionnaire items devel-

oped in the general population may not be appropriate for

African-American women.

Health Education via PublicTransportation: The East–West BreastExpressSelma Morris and Bonnie Wheatley, M.P.H.

BACKGROUND. Despite demonstrated benefits of breast carci-

noma screening, breast health remains underutilized by a large

percentage of multicultural women. The East-West Breast Ex-

press successfully linked public transportation with volunteers

to address this epidemic. This nontraditional approach provided

new avenues of reaching multicultural individuals who have

been mislabeled “unreachable.” A plethora of multicultural or-

ganizations/individuals volunteered their time and collabora-

tively exceeded the original goal of reaching 1000 individuals on

each coast by .100%. Funding for this 1997 program was

through the National Office of the Susan G. Komen Breast Can-

cer Foundation. Public transportation allows numerous win-

dows of opportunity for learning. Advertisements of products,

schools, and other organizations are showcased by this medium.

One essential missing link is the humanistic personal touch. The

incorporation of credible, trained, multicultural individuals who

mirror the populations to be reached significantly enhances

programs that address humanistic needs. This program collab-

orated with multiple organizations from inception through com-

pletion. It is important to note that some organizations asked to

participate in this unique program.

OBJECTIVE. This program was a bicoastal partnership between

two inner-city hospitals serving a predominantly diverse multi-

cultural population: Fulton/DeKalb Hospital Authority (Grady

Health System) (Atlanta, Georgia) and the Alameda County Med-

ical Center (Highland General Hospital) (Oakland, California).

The goal was to provide breast health education (clinical/breast

self-examination, mammography) via public transportation at

selected rail stations (16 on each coast of the U.S.) to women and

men age $18 years along with onsite and appointments for free

screening and follow-up opportunities.

METHODOLOGY. A consortium of .60 combined organizations

(women and men) of volunteers from civic, community, health,

county, business, and breast carcinoma survivor organizations

was formed. They met monthly and formed subcommittees (site,

education, evaluation, publicity, and abstracts/manuscripts).

Volunteers participated in the following activities: 1) 4-hour

orientation—prior to implementation; 2) 3-minute baseline sur-

veys—staggered 3-hour shifts (morning, evening, and week-

ends); 3) culturally specific hands-on breast self-examination

demonstrations by participants (teaching models); 4) onsite

screening/appointments at the nearest community-based health

center or hospital; 5) reminder telephone calls prior to appoint-

ments; 6) culturally appropriate breast health take-home pack-

ages; 7) certificates of appreciation; and 8) follow-up telephone

calls at 3– 6 months.

EVALUATION. A total of 2029 women on each coast were reached,

for a total of 4058 nationwide. The process included: coalition

building, ongoing social network activities, identification of is-

sues, monthly statistical reports, and development of consumer

satisfaction tools.

CONCLUSIONS. Multicultural community organizations, clubs,

health care facilities, and breast carcinoma survivors can and

will volunteer together as partners in addressing the humanistic

need to increase breast carcinoma awareness. Collaboration

provides an arena of ongoing program design that has the ca-

pability of utilizing public transportation as a vehicle for other

diseases beyond breast health or nontraditional avenues that

incorporate the “A” Principles: acceptability, accessibility, avail-

ability, affordability, and accountability to the “YET TO BE

REACHED.” The authors will continue to seek funding for the

replication of this volunteer program to the “REACHABLE

STARS” nationwide and, potentially, worldwide.

Poster Session Abstracts 1271