Can this be done 8 pm 4/22

Saltwata2012

AccessIssues.pdf

Home >Social Science homework help >Can this be done 8 pm 4/22

Meta-Analysis of Survival in African American and White American Patients With Breast Cancer: Ethnicity Compared With Socioeconomic Status Lisa A. Newman, Kent A. Griffith, Ismail Jatoi, Michael S. Simon, Joseph P. Crowe, and Graham A. Colditz

A B S T R A C T

Purpose The extent to which socioeconomic disadvantages and inadequate health care access account for the disproportionately elevated mortality hazard observed in African American compared with white American patients with breast cancer is poorly defined.

Methods We identified 20 studies reported between January 1980 and June 2005 that provided survival analyses in patients with breast cancer after adjusting for ethnicity and some measurement of socioeconomic status. These studies also adjusted for age and stage of disease at time of diagnosis.

Results The pooled outcome data yielded estimates for the mortality hazard in 14,013 African American and 76,111 white American patients with breast cancer. Studies varied in their methods for assigning socioeconomic status, with most relying on area-wide measures such as census tract and census block data. The combined analysis (adjusted for age, stage, and socioeconomic status) revealed that African American ethnicity was associated with a statistically significant excess mortality risk in overall survival (mortality hazard, 1.27; 95% CI, 1.18 to 1.38) and in breast cancer–specific survival (mortality hazard, 1.19; 95% CI, 1.10 to 1.29).

Conclusion Our pooled analysis demonstrated that African American ethnicity is a significant and independent predictor of poor outcome from breast cancer, even after accounting for socioeconomic status by conventional measures. These findings support the need for further investigation of the biologic, genetic, and sociocultural factors that may influence survival in African American patients with breast cancer.

J Clin Oncol 24:1342-1349.

INTRODUCTION

The disproportionately high breast cancer mor- tality in African American women, coupled with a paradoxically lower breast cancer incidence rate when compared with white American women is receiving increased attention in the medical liter- ature. Recent studies have attempted to disentan- gle the coexisting influences of primary tumor biology from the socioeconomic, cultural, and behavioral factors that affect outcome. The Afri- can American community is characterized by multiple socioeconomic disadvantages1 that im- pact cancer control. These disadvantages create barriers to breast cancer screening as well as treat- ment, and are widely assumed to explain a sub- stantial proportion of the mortality disparities.2-5

Several patterns related to breast cancer in African American women are not readily ex-

plained by socioeconomic factors. African Amer- ican women are more likely to be diagnosed with early-onset disease; population-based data from the Surveillance, Epidemiology and End Results (SEER) Program confirm poorly understood dif- ferences in breast cancer age-incidence curves. For women living in the United States (ie, Amer- ican women) younger than 45 years, breast cancer incidence is higher among African Americans com- pared with white Americans. A cross over in inci- dence occurs during the fifth decade of life, resulting in the overall decreased lifetime risk.6 Most of the lower lifetime risk is a consequence of diminished incidence rates for estrogen receptor–positive breast cancer.7 African American women have a higher risk of being diagnosed with hormone receptor–nega- tive, aneuploid, and node-positive cancer.7-15 Un- certainty persists regarding the possible existence of hereditary and/or environmental factors that might

From the University of Michigan, Ann Arbor; Karmanos Cancer Institute, Detroit, MI; Uniformed Services Univer- sity, Bethesda, MD; Cleveland Clinic Foundation, Cleveland, OH; Brigham and Women’s Hospital, Boston, MA.

Submitted July 5, 2005; accepted November 15, 2005.

Authors’ disclosures of potential con- flicts of interest and author contribu- tions are found at the end of this article.

Address reprint requests to Lisa A. Newman, MD, MPH, University of Michigan, 1500 E Medical Center Dr, Ann Arbor, MI 48109; e-mail: [email protected].

0732-183X/06/2409-1342/$20.00

DOI: 10.1200/JCO.2005.03.3472

JOURNAL OF CLINICAL ONCOLOGY O R I G I N A L R E P O R T

V O L U M E 2 4 � N U M B E R 9 � M A R C H 2 0 2 0 0 6

1342 Downloaded from jco.ascopubs.org on December 27, 2015. For personal use only. No other uses without permission.

predispose African American women to these biologically more ag- gressive phenotypes.

Single-institution studies of breast cancer survival generally have limited statistical power in addressing socioeconomic status versus ethnicity as prognostic factors because of the small sample sizes of affluent minority-ethnicity patients with breast cancer that are entered into final multivariate survival analyses. In contrast, population-based registries and multicenter cohorts rarely maintain detailed socioeco- nomic status information. Research synthesis (meta-analysis) is a sta- tistical tool that can potentially compensate for these inadequacies by pooling the results of appropriately-designed studies, yielding a more robust evaluation.16,17

Meta-analysis is therefore a valid approach for assessing whether African American ethnic background is an independent predictor of adverse outcome. A pooled analysis of studies published between 1980 and 200118 revealed a 22% excess risk of death (all-cause) for African American patients with breast cancer after accounting for socioeco- nomic background. Many recent studies have investigated the effect of ethnicity on breast cancer survival. We updated the meta-analysis to test the stability of our prior results.

METHODS

We performed a National Library of Medicine MEDLINE literature search to identify articles published from January 1980 to June 2005 that analyzed breast cancer survival in African American and white American patients with breast cancer. The following terms were entered: “breast cancer” and “African Amer- ican”; “breast cancer” and “race”; “breast cancer” and “ethnicity”; “breast cancer” and “black.” This search was augmented by a review of manuscript references and abstracts from conference proceedings. Appropriate studies for inclusion in the meta-analysis utilized a Cox proportional hazards regression model19 for calculation of survival after adjusting for some measure of socioeconomic status. Studies based on outcomes of patients treated in an “equal-access” system (defined as a health care system in which treated patients share the same payor system; eg, military or health maintenance organization–managed-care systems) were also included, and results from these systems were analyzed separately as well. All studies adjusted for stage and age at diagnosis.

Several maneuvers were used to avoid overlapping patient populations for the studies included in the pooled analysis. Geographic locations, sites of treatment, and time frame for breast cancer diagnosis were all recorded. Studies were excluded when these features suggested population overlap with other reports,14,20 in which case the study with longer follow-up or a larger data set was utilized.21,22 Data from the Department of Defense tumor registry were reported in two studies.23,24 Survival results from nonoverlapping years of patient diagnosis were entered into the analysis, with confirmation of the patient population size by the investigator of the more recent study (I. Jatoi, personal communication, September 7, 2004).

All included studies analyzed ethnic background based on patients’ self- reported information, as abstracted from tumor registry or medical record data. Throughout this article, the term “ethnicity” is used as opposed to “race.” This term was selected because of its connotations regarding shared sociocul- tural features, in addition to ancestral nationality.

Two studies conducted separate survival analyses with age stratification. Simon and Severson22 evaluated outcome separately for patients younger than 50 years versus those 50 years and older; Albain et al25 presented outcome stratified by menopausal status. The age-stratified subsets from these studies were entered separately in the meta-analysis.

A total of 20 studies were appropriate for the meta-analysis, and the summary statistic was reported as the mortality hazard (relative risk of death) for African American compared with white American patients with breast cancer. The random-effects model of DerSimonian and Laird26 was applied to

calculate the pooled estimate, as heterogeneity between studies was expected, a priori. To test this assumption explicitly, a Cochran � 2 test statistic (Q statistic) was also calculated and reported.27 All meta-analysis computations were per- formed using STATA 8.1 software (STATA Corp, College Station, TX).

RESULTS

Table 1 summarizes the 20 included studies,2,21-25,28-41 representing a total of 14,013 African American and 76,111 white American women diagnosed with breast cancer from 1961 to 2003. Geographically diverse communities throughout the United States were represented, and most were analyzed in retrospective reviews. A single case-case compara- tive study was included—the National Cancer Institute’s Black White Breast Cancer Survival Study.33 Phase III clinical trials data were reported by Albain et al25; this was the only study included in the analysis that has not yet been published in manuscript form. Selected studies used either the American Joint Commission on Cancer Staging TNM system or the Local-Regional-Distant categories to adjust for extent of disease at presen- tation. Studies varied in method of age classification as a continuous or a categorical variable, as detailed in Table 1.

As Table 1 presents, most studies relied on area-wide measures for assignment of socioeconomic status, based on average income and educational profile for the census block or tract associated with the patient’s address and/or zip code. Four studies23,24,35,36 accounted for socioeconomic status by reporting on breast cancer survival among patients treated in equal-access health care systems. One study33 as- signed socioeconomic status according to self-reported data from individual patients; two other studies2,21 relied on hospital records with regard to income and source of pay for treatment. Regardless of method for assigning socioeconomic status, studies comparing re- sources for the African American and white American patients re- ported significant disparities,2,21,22,28,30,31,33,34,36,38,41 with an excess of poverty-related features among the African American patients.

All studies provided a Cox proportional hazards survival anal- ysis that adjusted for age and stage of disease at diagnosis. In addition, some adjusted for selected primary tumor features (such as hormone receptor status and/or grade)2,22-25,28-30,32-34; comor- bidity 2,33,35,38; and sociodemographic features such as marital sta- tus.22,23,36,39,40 One study29 also accounted for nutritional status, and 11 studies2,21-25,32-34,38,40 considered treatment issues.

Despite methodologic heterogeneity, all studies revealed survival disadvantages for African American patients in crude univariate anal- yses. Adjusting for socioeconomic status resulted in a loss of strength for the association between outcome and ethnic background within individual studies; in several, the correlation was no longer statistically significant.2,21,28-31,33-37,40 The pooled estimate for the hazard of mor- tality (all-cause) after adjustment for socioeconomic status was 1.28 (95% CI, 1.18 to 1.38), with African American ethnicity re-emerging as an independent predictor of worse overall survival (Fig 1). Results were similar for the combined analysis of the eight studies reporting breast cancer–specific mortality (Fig 2), revealing that African Amer- ican patients had a 19% greater risk of death (hazard ratio, 1.19; 95% CI, 1.09 to 1.30). Tests for homogeneity of the hazard ratios across studies for both all-cause mortality (Q11 � 56.6; P � .0001) and breast cancer–specific mortality (Q8 � 11.174; P � .0001) indicate signifi- cant differences between study-specific estimates, and confirmed our use of the random-effects pooled estimates.

Breast Cancer Survival and Ethnicity

www.jco.org 1343 Downloaded from jco.ascopubs.org on December 27, 2015. For personal use only. No other uses without permission.

Subset meta-analyses based on type of socioeconomic status as- sessment are presented in Table 2. Exclusion of the one study32 that used marital status alone as a surrogate for socioeconomic status revealed an even greater survival disadvantage for African American patients with breast cancer (hazard ratio, 1.29; 95% CI, 1.19 to 1.40). Subset meta-analyses of the equal-access systems and the non– equal- access systems consistently revealed that African American ethnicity was significantly and independently associated with worse survival.

DISCUSSION

Nearly 5 years ago, we conducted a meta-analysis18 of studies that reported survival in African American and white American patients with breast cancer. This initial pooled analysis of 14 studies, involving more than 50,000 patients with breast cancer (10,000 of whom were African American) revealed a statistically significant 22% excess in mortality for the African American patients (hazard ratio, 1.22; 95%

CI, 1.13 to 1.30). As shown in Figure 3, literature reviews through MEDLINE using a variety of search terms demonstrate a rapid expan- sion in the volume of studies addressing ethnicity and breast cancer outcome. The number of titles retrieved through a search of the years 1990 to 1994 was notably higher compared with the yield for the entire preceding decade (1980 to 1989); for every subsequent 5 years, the number of titles retrieved nearly doubled. This rapid escalation in available data motivated us to update the meta-analysis.

Six studies2,24,25,39-41 were added to the original series of 14, corresponding to a supplemental 37,062 patients with breast cancer whose outcomes contributed to the summary statistics. This follow-up meta-analysis also includes data derived from prospective, randomized clinical trials. Our earlier finding that self-reported African American ethnic background is an independent adverse prognostic feature was unchanged in the updated patient popula- tion, with an updated mortality hazard of 1.27. These results can be compared with the age-standardized death rates (unadjusted for

Table 1. Characteristics of 20 Selected Studies (including two studies that reported results stratified by age/menopausal categories22,25)

Study (first author), Year Study Type�

Years of Diagnosis

(range) Study Location/Source

No. of AA

Patients

No. of WA

Patients Stage

System Age Stratification

(years) Primary SES Measure SES Data Source

Bassett, 198628 R 1973-1983 Washington state 251 1,255 TNM � 50; � 49 Income Census block Coates, 199029 R 1975-1979 Atlanta, GA 469 1,491 TNM � 50; 50-64; � 64 Sociodemographic Self-report Gordon, 199230 R 1974-1985 Cleveland, OH 253 1,132 TNM � 50; � 49 Income Census tract Ansell, 199331 R 1973-1985 Chicago, IL 887 265 LRD � 50; � 49 Income Census tract Neale, 199432 R 1973-1978 Detroit, MI 1,558 9,230 LRD � 40; 40-54; 55-69;

� 69 Sociodemographic Self-report

Eley, 199433 R, CC 1985-1986 Atlanta, GA; New Orleans, LA; San Francisco, CA

612 518 TNM 20-49; 50-64; 65-79 Income, sociodemographic

Self-report

Perkins, 199621 R 1958-1987 Houston, TX 801 2,581 LRD � 45; � 44 Insurance Hospital designated Simon (� 49), 199622 R 1988-1992 Detroit, MI 1,275 6,705 Simon (� 50), 199622 R 1988-1992 Detroit, MI 605 1,917 LRD Age as a

continuous variable Income Census tract

Franzini, 199734 R 1987-1991 Houston, TX 163 964 LRD � 30; 30-39; 40-49; 50-59; 60-69; � 69

Insurance Hospital designated

Howard, 199835 R 1986-1990 Three states 89 157 TNM � 40; 40-49; 50-64; � 64

Insurance, sociodemographic

Equal-access system

Wojcik, 199823 R 1975-1994 Department of Defense 698 5,879 TNM � 40; 40-49; 50-65; � 65

Insurance Equal-access system

Yood, 199936 R 1986-1996 Detroit, MI 273 613 TNM � 55; � 54 Insurance, income Equal-access system

El Tamer, 199937 R 1982-1995 Brooklyn, NY 1,297 448 TNM Age as a continuous variable

Income Census tract

Roetzheim, 200038 R 1994-1997 Florida state 770 9,318 LRD Age as a continuous variable

Income, insurance Census tract, insurance

Albain (premenopausal), 200425

Clinical trials�

1975-1995 SWOG 250 2,110 TNM Premenopausal v postmenopausal†

Income Census tract, insurance

Albain (postmenopausal), 200425

Clinical trials�

1975-1995 SWOG 414 3,902

Polednak, 200239 R 1988-1995 Connecticut state 868 16,063 LRD � 45; 45-54; 55-64; 65-74; 75�

Income, residence in medically underserved

area

Census tract

Bradley, 200240 R 1996-1997 Michigan state 1,110 4,609 LRD Age as a continuous variable

Income, Medicaid, insurance

Census tract

Jatoi, 200324 R 1995-1999 Department of Defense 726 4,685 TNM 5-year intervals; 20 to � 95

Insurance (Department of Defense system)

Equal-access system

Crowe, 200541 R 1961-2003 Ohio state 313 2,012 TNM Age in 10-year intervals; 20-99

Income Census tract

Du, 20052 R 1994-1997 Detroit, MI 331 257 TNM Age in 10-year intervals; 20-99

Insurance Medical records

Total 14,013 76,111

Abbreviations: R, retrospective; CC, case-case comparative study of African American and white American breast cancer patients from Surveillance, Epidemiology and End Results Program; SWOG, Southwest Oncology Group; SES, socioeconomic status; TNM, American Joint Committee on Cancer TNM staging system; LRD, local, regional, distant disease.

�Pooled analysis of African American and white American breast cancer patients participating in multiple SWOG adjuvant therapy trials. †Survival analysis stratified by menopausal status and adjusted for age, but data not reported regarding whether age was entered as a continuous or a categorical variable.

Newman et al

1344 JOURNAL OF CLINICAL ONCOLOGY

socioeconomic factors and other confounders) reported by the SEER Program for African American as compared with white American patients with breast cancer of 35.9/100,000 and 27.2/ 100,000, respectively (ratio, 1.32).1

Stability of the meta-analysis results strengthens the need for advances in two research areas: (1) methods to measure effects of sociobehavioral issues and poverty on breast cancer risk, and (2) exploration of associations between African American ethnicity and variation in primary breast tumor biology. Current insights regarding these areas are inadequate. Furthermore, while our meta-analysis in-

cludes data on patients diagnosed throughout four decades, we cannot assess temporal trends related to improved breast cancer treatments.

Socioeconomic status is commonly measured as a function of income; however, other factors contribute to this complex descriptor, such as environmental, dietary, cultural, behavioral, education, and access/treatment issues. These elements exert confounding effects on the cancer burden of different ethnic groups. The strategy of classifying an individual’s or a community’s socioeconomic status on the basis of income alone is therefore potentially misleading in the study of cancer control. Developing a successful strategy for

Fig 1. All studies: overall survival mortal- ity hazard (mortality hazard indicates the relative risk for mortality in African Ameri- can compared with white American pa- tients with breast cancer).

Fig 2. Eight studies: disease-specific sur- vival/mortality hazard (mortality hazard indi- cates the relative risk for mortality in African American compared with white American patients with breast cancer).

Breast Cancer Survival and Ethnicity

www.jco.org 1345 Downloaded from jco.ascopubs.org on December 27, 2015. For personal use only. No other uses without permission.

eliminating cancer outcome disparities requires careful scrutiny of all socioeconomic status components for different popula- tion subsets.4,18,42-45

Area-wide measures of socioeconomic status (eg, census block and census tract information) may be less reliable for studies of minority-ethnicity families. These measures involve estimates of the socioeconomic status of residential areas, composed of several hun- dred households, based on average income and educational history. These average socioeconomic estimates are then assigned to all indi- viduals residing in one of these geographic areas by matching address zip codes with the corresponding census block or census tract. Since several neighborhoods will be grouped together within these census regions, socioeconomic heterogeneity within small communities can be underestimated. Clusters of affluent African American families choosing to reside in predominately African American neighbor- hoods will be assigned the socioeconomic status of the larger area. The larger area frequently has a relatively low average income-education level, reflecting the prevalence of socioeconomic disadvantages of the general African American community. While some studies have vali- dated these area-wide measures in capturing socioeconomic hetero- geneity,46 others28 have suggested that the accuracy of estimates for

individual households is inversely related to the size of the census area being averaged.

Alternative measures of socioeconomic status have been pro- posed and require further study as we analyze ethnicity-related cancer disparities. Some candidate models are based on assessment of income inequality; social network; and social deprivation.47-50 Yabroff et al51

reported on the utility of a National Health Interview Survey– based method of measuring community socioeconomic status.

Most studies of ethnicity and cancer outcome rely on patient self-identification of background. This self-reporting is artificial to some extent, because four centuries of intermarriage between the Europeans, Africans, Scandinavians, and Asians who populate the United States’ “melting pot” have resulted in substantial genetic ad- mixture for most contemporary Americans. The Human Genome Project52,53 has ushered in a new era of research that may account for ancestry by gene sequencing, and several investigators have embarked on research of genetic admixture and risk of disease.54-57 These studies have demonstrated the superiority of ancestry-informative genetic markers over skin pigmentation58,59 in the assessment of ethnic heri- tage. The term “ethnicity” is itself imprecise in definition; we chose to use the term ethnicity instead of “race” because the former connotes some cultural commonality in addition to shared ancestry. The extent to which the African American community should be characterized as diverse versus socioculturally homogeneous can be debated. Regard- less, as this meta-analysis demonstrates, self-reported African Ameri- can ethnic background is consistently shown to be associated with worse breast cancer survival.

Recently, Tammemagi et al have shown that comorbidities account for a significant proportion of the mortality risk seen in African American patients with breast cancer.60 Four studies from this meta-analysis adjusted for comorbidity.2,33,35,38 Interestingly, the Tammemagi et al study also demonstrated statistically signifi- cant survival disadvantages for African American patients with breast cancer as reflected by all-cause (1.34; 95% CI, 1.11 to 1.62), breast cancer–specific (1.47; 95% CI, 1.08 to 2.00), or competing-causes– specific (1.27; 95% CI, 1.00 to 1.63) hazard ratios.

Table 2. Summary: Complete and Subset Meta-Analyses

Studies Included in Summary Statistics No. of Studies Reference No. Mortality Hazard 95% CI

All studies with overall survival data 20 13, 20-23, 25-39 1.27 1.18 to 1.38 All studies with disease specific survival data 8 23, 25, 28, 30, 31, 34, 36, 39 1.19 1.10 to 1.29 Studies with overall survival, excluding one study

of SES measured by sociodemographics/marital status only32

19 13, 20-23, 25-35, 37-39 1.29 1.18 to 1.40

Studies with disease specific survival, excluding one study of SES measured by sociodemographics/marital status only32

7 23, 25, 28, 30, 31, 34, 39 1.21 1.09 to 1.35

Studies with overall survival, excluding one study of SES measured by sociodemographics/marital status only,32 and excluding four studies of equal access systems23,24,35,36

15 20, 21, 23, 25, 28-35, 37-39 1.26 1.15 to 1.37

Studies from equal access systems only (overall survival data) and excluding one study of SES measured by sociodemographics/marital status only32

4 13, 22, 26, 27 1.48 1.15 to 1.90

NOTE. Mortality hazard indicates the relative risk for mortality in African American compared with white American patients with breast cancer. Abbreviation: SES, socioeconomic status.

Fig 3. Number of titles retrieved from MEDLINE literature searches with entry of different key expressions, 1989 to 2004.

Newman et al

1346 JOURNAL OF CLINICAL ONCOLOGY

It remains valid to question whether African American ethnicity as an adverse breast cancer prognostic feature is a surrogate marker for socioeconomic disadvantage and inadequate health care. Poverty rates and likelihood of being uninsured are two- to three-times higher among African Americans compared with white Americans, causing delayed diagnostic procedures and treatment. Provider-level inequal- ities in delivery of care have also been documented.61,62 All of these issues ultimately result in advanced– cancer stage distributions and higher mortality rates.1,63 Some investigators have in fact documented comparable responses to therapy in African American and white American patients with breast cancer after controlling for disease stage. For example McCaskill-Stevens et al found similar effectiveness and adverse effect profiles for tamoxifen64; Dignam et al found similar outcomes in an overview of National Surgical Adjuvant Breast Project clinical trials65; and Roach et al reported similar outcomes in patients treated on Cancer and Leukemia Group B protocols.66

Speculation persists regarding ethnicity-related variation in primary tumor biology as an explanation for outcome disparities. Subset analysis from phase III clinical trials of breast cancer should eliminate many confounding effects because random assignment and protocol design should standardize treatment, independent of ethnic background. Unger et al67 presented provocative results from a pooled analysis of clinical trials conducted by the Southwest Oncology Group, and found that survival disparities for African Americans as well as other minority-ethnicity patient populations were eradicated in the context of protocol-standardized manage- ment for the majority of malignancies. Interestingly however, persis- tent survival disadvantages were seen for African American participants in clinical trials for hormonally driven cancers such as breast, prostate, and ovarian. These results suggest that African Amer- ican ethnicity may be associated with genetic or metabolic character- istics that influence breast cancer outcome. Albain et al25 explored this further in a pooled analysis of Southwest Oncology Group adjuvant therapy protocols for breast cancer. They found statistically significant differences in outcome for both pre- and postmenopausal African American participants, despite the expectation that the clinical trial mechanism would control for stage of disease, work-up, and delivery of care. The investigators nonetheless included adjustments for socio- economic status based on census tract estimates for participants, with no substantive changes in their results. These results are included in the present meta-analysis.

Data from the Women’s Health Initiative (WHI)13 provided additional clinical trial– based evidence of a survival disadvantage for African American patients with breast cancer. The WHI was designed to analyze the effect of postmenopausal hormone replacement ther- apy versus placebo on breast cancer incidence and cardiovascular disease. With a median follow-up exceeding 6 years, WHI investiga- tors reported risk factors associated with breast cancer incidence and mortality after stratifying by patient ethnicity. Established risk factors were found to explain differences in breast cancer burden between white Americans and all other ethnic groups, except for African Amer- icans. For African American participants, there remained a significant increase in incidence of estrogen receptor–negative cancer, and an excess breast cancer mortality risk.

The concept that disparities in breast and prostate cancer may have a unique etiology with regard to African Americans is partic- ularly intriguing. There are several parallels observed for the effect of these two hormonally driven malignancies on African American women and men, respectively, such as younger age distribution, more advanced-stage distribution at time of diagnosis, and in- creased prevalence of adverse primary tumor prognostic features. The possibility of hereditary predisposition for aggressive disease related to African ancestry is currently being investigated.68-72

Hershman et al73 reported on the increased prevalence of neu- tropenia among African Americans and the effect of this feature on adjuvant therapy for cancer patients. Baseline WBC counts are known to be relatively lower for African Americans compared with white Americans, and this neutropenia can increase the risk for requiring chemotherapy dose reductions and/or treatment delays. Hershman et al found that this issue did correlate with decreased chemotherapy dose intensity for African American patients with breast cancer treated at Columbia Presbyterian Medical Center.73 The suspected associa- tion between vitamin D levels and breast cancer risk also warrants further study,74-76 as vitamin D levels seem to vary between African Americans and white Americans.77

This meta-analysis of socioeconomic status–adjusted breast can- cer survival demonstrates that self-reported African American ethnic- ity is an independent predictor for worse outcome. While attempts to correct socioeconomic disparities between African Americans and white Americans must be prioritized on the public health care agenda, our study confirms the importance of research regarding primary variation in breast tumor biology and studies of cancer-related factors associated with the environment of poverty.

REFERENCES

1. Ward E, Jemal A, Cokkinides V, et al: Cancer disparities by race/ethnicity and socioeconomic sta- tus. CA Cancer J Clin 54:78-93, 2004

2. Du W, Simon M: Racial disparities in treatment and survival of women with stage I-III breast cancer at a large academic medical center in metropolitan De- troit. Breast Cancer Res Treat 91:243-248, 2005

3. Brawley OW: Disaggregating the effects of race and poverty on breast cancer outcomes. J Natl Cancer Inst 94:471-473, 2002

4. Baquet CR, Commiskey P: Socioeconomic factors and breast carcinoma in multicultural women. Cancer 88:1256-1264, 2000

5. Cross C, Harris J, Recht A: Race, socioeco- nomic status, and breast carcinoma in the U.S. Cancer 95:1988-1999, 2002

6. Ries L, Eisner M, Kosary C, et al: SEER Cancer Statistics Review, 1975-2001. Bethesda, MD, National Cancer Institute, 2004. Available at http://seer.cancer.gov/csr/1975_2001/

7. Anderson WF, Chatterjee N, Ershler WB, et al: Estrogen receptor breast cancer phenotypes in the Surveillance, Epidemiology, and End Results database. Breast Cancer Res Treat 76:27-36, 2002

8. Cunningham JE, Butler WM: Racial disparities in female breast cancer in South Carolina: Clinical evidence for a biological basis. Breast Cancer Res Treat 88:161-176, 2004

9. Porter PL, Lund MJ, Lin MG, et al: Racial differences in the expression of cell cycle-regulatory proteins in breast carcinoma. Cancer 100:2533- 2542, 2004

10. Jones BA, Kasl SV, Howe CL, et al: African- American/White differences in breast carcinoma:

p53 alterations and other tumor characteristics. Can- cer 101:1293-1301, 2004

11. Li CI, Malone KE, Daling JR: Differences in breast cancer hormone receptor status and histology by race and ethnicity among women 50 years of age and older. Cancer Epidemiol Biomarkers Prev 11:601- 607, 2002

12. Shavers VL, Harlan LC, Stevens JL: Racial/ ethnic variation in clinical presentation, treatment, and survival among breast cancer patients under age 35. Cancer 97:134-147, 2003

13. Chlebowski RT, Chen Z, Anderson GL, et al: Ethnicity and breast cancer: Factors influencing dif- ferences in incidence and outcome. J Natl Cancer Inst 97:439-448, 2005

14. Simon MS, Severson RK: Racial differences in breast cancer survival: The interaction of socio- economic status and tumor biology. Am J Obstet Gynecol 176:S233-S239, 1997

Breast Cancer Survival and Ethnicity

www.jco.org 1347 Downloaded from jco.ascopubs.org on December 27, 2015. For personal use only. No other uses without permission.

15. Newman LA: Breast cancer in African- American women. Oncologist 10:1-14, 2005

16. Berkey CS, Hoaglin DC, Antczak-Bouckoms A, et al: Meta-analysis of multiple outcomes by regression with random effects. Stat Med 17:2537-2550, 1998

17. Mosteller F, Colditz GA: Understanding re- search synthesis (meta-analysis). Annu Rev Public Health 17:1-23, 1996

18. Newman LA, Mason J, Cote D, et al: African- American ethnicity, socioeconomic status, and breast cancer survival: A meta-analysis of 14 studies involving over 10,000 African-American and 40,000 White American patients with carcinoma of the breast. Cancer 94:2844-2854, 2002

19. Cox D: Regression models and life-tables. J R Stat Soc B 4:187-220, 1972

20. Vernon SW, Tilley BC, Neale AV, et al: Ethnicity, survival, and delay in seeking treatment for symptoms of breast cancer. Cancer 55:1563-1571, 1985

21. Perkins P, Cooksley CD, Cox JD: Breast can- cer. Is ethnicity an independent prognostic factor for survival? Cancer 78:1241-1247, 1996

22. Simon MS, Severson RK: Racial differences in survival of female breast cancer in the Detroit met- ropolitan area. Cancer 77:308-314, 1996

23. Wojcik BE, Spinks MK, Optenberg SA: Breast carcinoma survival analysis for African American and white women in an equal-access health care sys- tem. Cancer 82:1310-1318, 1998

24. Jatoi I, Becher H, Leake CR: Widening dispar- ity in survival between white and African-American patients with breast carcinoma treated in the U.S. Department of Defense Healthcare system. Cancer 98:894-899, 2003

25. Albain KS, Unger JM, Hutchins LF, et al: Out- come of African Americans on Southwest Oncology Group (SWOG) breast cancer adjuvant therapy trials. 2003 San Antonio Breast Cancer Symposium, San Antonio, TX, 2003 (abstr 21)

26. DerSimonian R, Laird N: Meta-analysis in clin- ical trials. Control Clin Trials 7:177-188, 1986

27. Cochrane WG: The combination of estimates from different experiments. Biometrics 10:101-129, 1954

28. Bassett MT, Krieger N: Social class and black- white differences in breast cancer survival. Am J Public Health 76:1400-1403, 1986

29. Coates RJ, Clark WS, Eley JW, et al: Race, nutritional status, and survival from breast cancer. J Natl Cancer Inst 82:1684-1692, 1990

30. Gordon NH, Crowe JP, Brumberg DJ, et al: Socioeconomic factors and race in breast cancer re- currence and survival. Am J Epidemiol 135:609-618, 1992

31. Ansell D, Whitman S, Lipton R, et al: Race, income, and survival from breast cancer at two public hospitals. Cancer 72:2974-2978, 1993

32. Neale AV: Racial and marital status influences on 10 year survival from breast cancer. J Clin Epidemiol 47:475-483, 1994

33. Eley JW, Hill HA, Chen VW, et al: Racial differences in survival from breast cancer: Results of the National Cancer Institute Black/White Cancer Survival Study. JAMA 272:947-954, 1994

34. Franzini L, Williams AF, Franklin J, et al: Effects of race and socioeconomic status on survival of 1,332 black, Hispanic, and white women with breast cancer. Ann Surg Oncol 4:111-118, 1997

35. Howard DL, Penchansky R, Brown MB: Dis- aggregating the effects of race on breast cancer survival. Fam Med 30:228-235, 1998

36. Yood MU, Johnson CC, Blount A, et al: Race and differences in breast cancer survival in a man-

aged care population. J Natl Cancer Inst 91:1487- 1491, 1999

37. El-Tamer MB, Homel P, Wait RB: Is race a poor prognostic factor in breast cancer? J Am Coll Surg 189:41-45, 1999

38. Roetzheim RG, Gonzalez EC, Ferrante JM, et al: Effects of health insurance and race on breast carci- noma treatments and outcomes. Cancer 89:2202- 2213, 2000

39. Polednak AP: Survival of breast cancer pa- tients in Connecticut in relation to socioeconomic and health care access indicators. J Urban Health 79:211-218, 2002

40. Bradley CJ, Given CW, Roberts C: Race, so- cioeconomic status, and breast cancer treatment and survival. J Natl Cancer Inst 94:490-496, 2002

41. Crowe JP Jr, Patrick RJ, Rybicki LA, et al: Race is a fundamental prognostic indicator for 2325 Northeastern Ohio women with infiltrating breast cancer. Breast J 11:124-128, 2005

42. Muss HB: Factors used to select adjuvant therapy of breast cancer in the United States: An overview of age, race, and socioeconomic status. J Natl Cancer Inst Monogr 30:52-55, 2001

43. Robert SA, Strombom I, Trentham-Dietz A, et al: Socioeconomic risk factors for breast cancer: Distinguishing individual- and community-level ef- fects. Epidemiology 15:442-450, 2004

44. Soler-Vila H, Kasl SV, Jones BA: Prognostic significance of psychosocial factors in African- American and white breast cancer patients: A population-based study. Cancer 98:1299-1308, 2003

45. Burchard EG, Ziv E, Coyle N, et al: The impor- tance of race and ethnic background in biomedical research and clinical practice. N Engl J Med 348:1170- 1175, 2003

46. Krieger N: Overcoming the absence of socio- economic data in medical records: Validation and application of a census-based methodology. Am J Public Health 82:703-710, 1992

47. Subramanian SV, Kawachi I: Income inequal- ity and health: What have we learned so far? Epide- miol Rev 26:78-91, 2004

48. Kawachi I, Kennedy BP, Lochner K, et al: Social capital, income inequality, and mortality. Am J Public Health 87:1491-1498, 1997

49. Taylor A, Cheng KK: Social deprivation and breast cancer. J Public Health Med 25:228-233, 2003

50. Michael YL, Berkman LF, Colditz GA, et al: Social networks and health-related quality of life in breast cancer survivors: A prospective study. J Psy- chosom Res 52:285-293, 2002

51. Yabroff KR, Gordis L: Assessment of a na- tional health interview survey-based method of measuring community socioeconomic status. Ann Epidemiol 13:721-726, 2003

52. Finishing the euchromatic sequence of the human genome. Nature 431:931-945, 2004

53. Lander ES, Linton LM, Birren B, et al: Initial sequencing and analysis of the human genome. Nature 409:860-921, 2001

54. Barnholtz-Sloan JS, de Andrade M, Dyer TD, et al: Admixture effects in the traditional linkage analysis of admixed families. Ethn Dis 12:411-419, 2002

55. Hoggart CJ, Shriver MD, Kittles RA, et al: Design and analysis of admixture mapping studies. Am J Hum Genet 74:965-978, 2004

56. Seldin MF, Morii T, Collins-Schramm HE, et al: Putative ancestral origins of chromosomal segments in individual african americans: Implications for admix- ture mapping. Genome Res 14:1076-1084, 2004

57. Smith MW, Lautenberger JA, Shin HD, et al: Markers for mapping by admixture linkage disequi-

librium in African American and Hispanic popula- tions. Am J Hum Genet 69:1080-1094, 2001

58. Parra EJ, Kittles RA, Shriver MD: Implications of correlations between skin color and genetic an- cestry for biomedical research. Nat Genet 36:S54- S60, 2004

59. Shriver MD, Parra EJ, Dios S, et al: Skin pigmentation, biogeographical ancestry and admix- ture mapping. Hum Genet 112:387-399, 2003

60. Tammemagi CM, Nerenz D, Neslund-Dudas C, et al: Comorbidity and survival disparities among black and white patients with breast cancer. JAMA 294:1765-1772, 2005

61. Breen N, Wesley MN, Merrill RM, et al: The relationship of socio-economic status and access to minimum expected therapy among female breast can- cer patients in the National Cancer Institute Black-White Cancer Survival Study. Ethn Dis 9:111-125, 1999

62. Bach PB, Cramer LD, Warren JL, et al: Racial differences in the treatment of early-stage lung cancer. N Engl J Med 341:1198-1205, 1999

63. Health Insurance Coverage: 2001. Washing- ton, DC, US Census Bureau, 2001

64. McCaskill-Stevens W, Wilson J, Bryant J, et al: Contralateral breast cancer and thromboembolic events in African American women treated with tamoxifen. J Natl Cancer Inst 96:1762-1769, 2004

65. Dignam JJ: Efficacy of systemic adjuvant ther- apy for breast cancer in African-American and Cauca- sian women. J Natl Cancer Inst Monogr 30:36-43, 2001

66. Roach M 3rd, Cirrincione C, Budman D, et al: Race and survival from breast cancer: Based on Cancer and Leukemia Group B trial 8541. Cancer J Sci Am 3:107-112, 1997

67. Unger J, Coltman C, Appelbaum F, et al: African Americans have worse survival in hormone-related cancers: A Southwest Oncology Group (SWOG) study. Proc Am Soc Clin Oncol 22:529, 2003 (abstr 2127)

68. Powell IJ, Carpten J, Dunston G, et al: African- American heredity prostate cancer study: A model for genetic research. J Natl Med Assoc 93:120-123, 2001

69. Royal C, Baffoe-Bonnie A, Kittles R, et al: Recruit- ment experience in the first phase of the African Amer- ican Hereditary Prostate Cancer (AAHPC) study. Ann Epidemiol 10:S68-S77, 2000

70. Olopade OI, Fackenthal JD, Dunston G, et al: Breast cancer genetics in African Americans. Cancer 97:236-245, 2003 (suppl 1)

71. Ikpatt O, Xu J, Kramtsov A, et al: Hormone receptor negative and Basal-like subtypes are overrep- resented in invasive breast carcinoma from women of African ancestry. American Association for Cancer Research, Anaheim, CA, 2005 (abstr 2550)

72. Fregene A, Newman LA: Breast cancer in sub-Saharan Africa: How does it relate to breast cancer in African-American women? Cancer 103: 1540-1550, 2005

73. Hershman D, Weinberg M, Rosner Z, et al: Ethnic neutropenia and treatment delay in African American women undergoing chemotherapy for early-stage breast cancer. J Natl Cancer Inst 95: 1545-1548, 2003

74. Lowe LC, Guy M, Mansi JL, et al: Plasma 25-hydroxy vitamin D concentrations, vitamin D recep- tor genotype and breast cancer risk in a UK Caucasian population. Eur J Cancer 41:1164-1169, 2005

75. Guy M, Lowe LC, Bretherton-Watt D, et al: Vitamin D receptor gene polymorphisms and breast cancer risk. Clin Cancer Res 10:5472-5481, 2004

76. John EM, Schwartz GG, Dreon DM, et al: Vita- min D and breast cancer risk: The NHANES I Epidemi- ologic follow-up study, 1971-1975 to 1992—National Health and Nutrition Examination Survey. Cancer Epi- demiol Biomarkers Prev 8:399-406, 1999

Newman et al

1348 JOURNAL OF CLINICAL ONCOLOGY

77. Nesby-O’Dell S, Scanlon KS, Cogswell ME, et al: Hypovitaminosis D prevalence and determinants

among African American and white women of repro- ductive age: Third National Health and Nutrition Exam-

ination Survey, 1988-1994. Am J Clin Nutr 76:187-192, 2002

■ ■ ■

Authors’ Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest.

Author Contributions

Conception and design: Lisa A. Newman, Michael S. Simon Administrative support: Lisa A. Newman Provision of study materials or patients: Lisa A. Newman, Ismail Jatoi, Joseph P. Crowe Collection and assembly of data: Lisa A. Newman Data analysis and interpretation: Lisa A. Newman, Kent A. Griffith, Graham A. Colditz Manuscript writing: Lisa A. Newman, Kent A. Griffith, Ismail Jatoi, Michael S. Simon, Graham A. Colditz Final approval of manuscript: Lisa A. Newman, Kent A. Griffith, Ismail Jatoi, Michael S. Simon, Joseph P. Crowe, Graham A. Colditz

Breast Cancer Survival and Ethnicity

www.jco.org 1349 Downloaded from jco.ascopubs.org on December 27, 2015. For personal use only. No other uses without permission.