women born and living in Asia.4,6 Breast cancer risk increaseswith the number of years lived in the immigration country,and in subsequent generations risk continues to increase andto approach the risk in native-born populations.6 However,the speed with which breast cancer incidence among immi-grants and their offspring approaches that of their adoptedcountry varies considerably from one ethnic group to another.7 Of additional interest, among Asian American women, immigrants from urban areas arrived in the UnitedStates with a 30% greater risk than immigrants from rural areas.6

International geographic variations in breast cancer inci-dence, in conjunction with the temporal changes in these ratesand results of immigrant studies, have long indicated thatlifestyle and environment play an important role in the devel-opment of breast cancer. However, a recent study that assessedthe risk for breast cancer in women from various racial andethnic groups living in California and Hawaii reported that,after adjusting for seven known lifestyle risk factors (ages at menarche, age at birth of first child, parity, age at and type of menopause (natural or surgical), weight, hormone replace-ment therapy use, and alcohol consumption), the risk forbreast cancer was 65% greater in Native Hawaiian women and11% greater in Japanese American women than in whitewomen.8 These results suggest that genetic factors may alsoplay a role, although other environmental factors, such as diet,could also explain these results.

What is the frequency of breast cancer in the United States?

Breast cancer is the most common cancer among Americanwomen, representing 32% of all new cancers (Fig. 1–2), and isthe second leading cause of cancer deaths in Americanwomen, representing 15% of all cancer deaths, behind lungcancer (25% of all cancer deaths).9 On the basis of currentincidence, it is predicted that one of every eight women in the United States will develop invasive breast cancer in herlifetime.10

In 2005, approximately 211,240 new cases of invasive breastcancer are expected to occur among women in the UnitedStates.11 In addition, it is anticipated that 58,490 new cases ofin situ breast cancer will be diagnosed in the same year.These numbers represent a sharp increase over the past 30years. From 1970 to 2005, the number of invasive breast

What is the frequency of breast cancer in the world?

Breast cancer is the most common cancer (22% of all newcases) as well as the leading cause of cancer deaths (14% of allcancer deaths) in women worldwide.1 The number of newcases in 2000 was estimated to be 1.05 million and the num-ber of deaths 370,000. Breast cancer is the most prevalent can-cer in the world: in 2000, there were an estimated 3.9 millionwomen alive who had had breast cancer diagnosed within theprevious 5 years.1

How does breast cancer incidence vary by country?

Incidence varies more than fourfold internationally, withhigh rates in all the more developed countries except Japan(i.e., North America, Northern and Western Europe,Australia) (Fig. 1–1). High rates are also observed in southernSouth America, especially in Uruguay and Argentina. In con-trast, most African and Asian populations have low rates,whereas rates in Central America and Eastern Europe areintermediate.1 The incidence of breast cancer is increasing inmost countries, with the greatest changes where rates werepreviously low. Whereas the worldwide increase in incidencebetween 1990 and 2000 was about 1.5% per year, many low-risk countries recorded increases greater than this: forinstance, 2% in Japan, and 3% to 5% in some areas of China.1

Although part of these international and temporal variationsmay be spurious, owing to incomplete reporting and variablediagnostic practices, the consistent pattern of higher rates incertain regions suggests true differences in underlying breastcancer risk across nations.2

How does breast cancer incidence change in immigrant populations?

Breast cancer incidence among immigrant populations grad-ually changes from the incidence in the country of origin toapproach the incidence in the immigration country.3–6 Forinstance, Asian Americans born in Asia, where incidence islow, are at lower risk for breast cancer than Asian Americansborn in the United States but are at higher risk than Asian

CHAPTER 1

Epidemiology of Breast CancerAnne Zeleniuch-Jacquotte and Roy E. Shore

3

cancer cases increased by approximately 200% (from 69,000to 211,240).9,12 This increase, however, is due in large part tothe increase in the population of older women, who are athigher risk of developing the disease.13 In the same time pe-riod (1970–2003), the number of deaths from breast canceralso increased, but only by 33% (from 30,000 to 39,800), aresult of early detection by mammography screening andimprovements in treatment.

How has U.S. incidence changed over time?

To assess the temporal changes in breast cancer incidence, it isnecessary to examine age-adjusted incidence that is independ-ent of changes in the age distribution of the U.S. population.Such statistics are available through the National CancerInstitute’s Surveillance, Epidemiology, and End Results(SEER) program, which has collected cancer incidence andsurvival data from population-based cancer registries since1973 (http://www.seer.cancer.gov). Figure 1–3 shows the SEERage-adjusted incidence of invasive breast cancer from 1975 to2000. Three distinct phases are seen: between 1975 and 1980,the incidence was essentially constant; between 1980 and1987, the incidence increased by about 4% per year; between1987 and 2000, the incidence (adjusted for delayed report-ing14) increased by 0.6% per year.15,16 Using the 2000 U.S.standard population, the 2000 age-adjusted incidence was135.1 per 100,000 woman-years.

The increase in invasive breast cancer incidence was notuniform across histologic types and estrogen receptor (ER)status. The incidence of ductal carcinoma appeared to remainessentially constant from 1987 to 1999, whereas the incidenceof lobular carcinoma increased steadily.17 Also, most of the in-crease in breast cancer incidence appeared to be due to anincrease in incidence of estrogen receptor–positive tumors,18–20

which was not accounted for by technical improvements orchanges in tumor size, age, or nodal status.21

SECTION I. MOLECULAR AND EPIDEMIOLOGIC ISSUES4

< 19.3 < 26.1 < 36.0 < 54.1< 91.6

Age-standardized world rateper 100,000

Figure 1–1 Estimated incidence by country, age-standardized by world standard population. (Data from Parkin DM. Global cancer statistics in the year 2000. Lancet Oncol 2001;2:533–543.)

All others35%

Breast32%

Lung andbronchus

12%Colon andrectum11%

Uterine corpus6%

Ovary4%

All others41%

Breast15%

Lung andbronchus

25%

Colon andrectum11%Uterine

corpus3%

Ovary5%

NEW CASES

DEATHS

Figure 1–2 Sites of new cancer cases and deaths—2003 esti-mates. (Data from American Cancer Society. Breast Cancer Facts andFigures 2003–2004 [Online]. Available: http://www.cancer.org/docroot/STT/content/STT_1x_Breast_Cancer_Facts__Figures_2003–2004.asp[accessed January 29, 2004].)

The age-adjusted incidence of in situ breast cancerincreased considerably over the past 20 years, with a 32.3%annual increase from 1982 to 1986 and a 6.1% annual increasefrom 1986 to 2000.16 This dramatic increase is largely attribut-able to an increase in ductal carcinoma in situ (DCIS),22 whichbetween 1996 and 2000 accounted for 88% of all in situ breastcancer cases.11 Between 1975 and 2000, the incidence of DCISincreased five times faster than the incidence of invasive breastcancer. In the same time period, the incidence of lobular car-cinoma in situ (LCIS) also increased, but only twice as fast asthe incidence of invasive breast cancer.11

What are the causes of the increase in breast cancer incidence in the United States?

The rapid increase in invasive breast cancer incidence between1980 and 1987 was largely due to an increase in the incidenceof small tumors (<2.0 cm), which more than doubled, whilerates of larger tumors decreased.23 The increase in incidencewas also limited to the localized stage, whereas the incidenceof breast cancer at regional and distant stages remained stable.These data suggest that the increase may be explained in largepart by greater use of mammography screening, which detectssmall tumors, many of them too small to be palpable. The limited data available on mammography utilization before1987 indicate that only 10% to 20% of women older than 40years had ever had a mammogram (including for diagnosticpurposes) by the early 1980s.24 The rate of mammographyscreening has increased steadily since then. Data from theNational Health Interview Surveys show that the percentageof women who reported having had a screening mammogramin the previous year was 17% in 1987, 33% in 1990, and 55%

in 2000.11,25 Although it may seem surprising that the inci-dence increased at a much slower rate after 1987 than in the1980 to 1987 years whereas use of mammographic screeningwas still increasing sharply, this pattern is not unexpected fora screening procedure that detects early cases that would havebeen diagnosed in any case but at a later date: at the beginningof screening implementation, the incidence increases becausescreen-detected cases are added to symptom-diagnosed cases. However, later on, although the number of screen-detected cases keeps increasing as more women are beingscreened, the number of symptom-detected cases goes down(because some of them were detected earlier through screen-ing), resulting in a leveling off (or even a decrease) in the overall incidence.26

The strong increase in incidence of in situ breast cancer was especially pronounced among women 50 years of age andolder,11,22 which is consistent with the age pattern of mam-mography screening.22 The larger impact of screening onDCIS than LCIS incidence was also anticipated because of thelack of specific radiologic criteria for LCIS, which is generallynot detectable by mammography.27

Are there causes other than mammography that explain the changes in U.S. incidence?

It has been argued that the increase in mammographic screen-ing cannot entirely explain the increase in breast cancer inci-dence,5 and that changes in the profile of risk factors in theU.S. population also may have played a role.23 In particular,changes in patterns of childbearing, with a trend towarddelayed first birth and decreased parity, have been suggested as plausible causes of the rise in breast cancer incidence.21

Indeed, it has been estimated that later age at first birth andnulliparity account for 30% of breast cancer cases in theUnited States.28 Although an increase in fertility rates wasobserved following World War II (baby boom), and thisincrease was shown to be associated with a reduction in breastcancer mortality in women born between 1924 and 1938,29 atrend to lower parity and later age at first birth started in the1960s, which would be expected to result in an increase inincidence. Oral contraceptives and hormone replacementtherapy may also have contributed to the rise in breast cancerincidence because both exposures are associated with smallincreases in breast cancer risk during their use and for sometime after cessation.30,31

Changes in the distribution of risk factors could explain,at least in part, the variations in the temporal trends ofthe incidence according to histologic type and ER status:for instance, nulliparity appears to be more consistentlyassociated with ER-positive than with ER–negative tumors,and increasing use of combined hormone replacement therapy32 appears to increase preferentially the risk for lobular carcinoma.33–36

What is the risk for breast cancerin American men?

Breast cancer occurs infrequently in men, with about 1500new cases diagnosed each year in the United States.37 The life-time risk for being diagnosed with breast cancer is 0.11% for

5Chapter 1. Epidemiology of Breast Cancer

Year of diagnosis

Rat

e pe

r 10

0,00

0

1975 1980 1985 1990 1995 2000

160

140

120

100

80

60

40

20

0

Figure 1–3 Incidence, age-adjusted to the 2000 standard U.S.population, SEER 1975–2000. (Data from Ries LAG, Eisner MP,Kosary CL, et al. SEER Cancer Statistics Review, 1975–2000 [Online].Available: http://seer.cancer.gov/csr/1975_2000 [accessed January 28,2004].)

American men, in contrast to 13.5% for women,10 a 120-foldratio.

What is the impact of age on breast cancer incidence?

The incidence for invasive breast cancer strongly increaseswith age. Overall, the increase in incidence with age is muchsteeper before than after 50 years of age, suggesting thatmenopause has a protective effect.38 Furthermore, two recentstudies, one in Denmark and one in the United States, haveshown that only the incidence of ER-positive breast cancerincreases after 50 years of age, whereas the incidence of ER-negative breast cancer stops increasing after 50 years of age39,40

(Fig. 1–4). These observations, suggesting that ER-negativetumors are more dependent on the premenopausal sex hormone environment than ER-positive tumors, need to be investigated further. An extensive review of the role ofreproductive factors in the development of breast cancer ispresented in Chapter 4.

What is the impact of race and ethnicity on breast cancer incidence?

Breast cancer is the most common cancer among women inevery major ethnic group in the United States.41 However,the age-adjusted incidence varies greatly according to race or ethnic group. It is highest among white women (140.8 per 100,000 in years 1996–2000), lowest among AmericanIndian or Alaska Native women (58 per 100,000), and

intermediate among African American (121.7 per 100,000),Asian or Pacific Islander (97.2 per 100,000), and Hispanicwomen (89.8 per 100,000).15 During 1992 to 2000, incidenceincreased overall in Asians and Pacific Islanders (2.1% peryear), Hispanics (1.3% per year), and whites (0.9% per year)whereas rates decreased in American Indians and AlaskaNatives (-3.7%) and remained stable in African Americans(Fig. 1–5).

White non-Hispanic women are more likely to present with localized stage breast cancer than African Americans(Fig. 1–6) and other racial or ethnic groups.15,42 Differences inmammography screening rates may explain this result, at leastin part: in 2000, the percentage of women 40 years and olderreporting having had a mammogram in the previous 2 yearswas 72% for white women, 68% for African Americanwomen, 63% for Hispanic women, 57% for Asian and PacificIslander women, and 52% for American Indian and AlaskaNative women.43 However, the difference in mammographyuse between African American and white women decreasedsubstantially since 1987.44

Other differences in tumor characteristics in relation torace or ethnicity have been reported. SEER data show that,compared with non-Hispanic whites, tumors in AfricanAmericans are more likely to be hormone receptor negativeand of medullary histology, two characteristics associatedwith poor prognosis, and are less likely to have a lobular his-tology, which is associated with lower mortality.45–48 Althoughdata are limited, the risk for hormone receptor–negativebreast cancer also appears higher in Native Americans, Asiansor Pacific Islanders, and Hispanic whites 50 years of age orolder than in non-Hispanic whites in the same age group,although not as high as in African Americans.46

SECTION I. MOLECULAR AND EPIDEMIOLOGIC ISSUES6

Age

-spe

cific

rat

es p

er 1

00,0

00

20 30 40 50 60 70 80 90

100

10

1

0.1

ER+ ER–

20 30 40 50 60 70 80 90

100

10

1

0.1

Age (yr) Age (yr)

Figure 1–4 Five-year age-specific breast cancer incidence for white women (open circles) and African American women (closed circles)from 25 through 84 years of age by estrogen receptor (ER) status of breast cancer. (Data from Tarone RE, Chu KC. The greater impact ofmenopause on ER- than ER+ breast cancer incidence: A possible explanation [United States]. Cancer Caus Cont 2002;13:7–14.)

Comparisons of the age-specific incidence in AfricanAmerican and white women show an unusual racial crossoverin risk: up to age 40 years, African American women have a higher incidence than white women, whereas after age 40 years, rates for white women exceed those for AfricanAmerican women, and the disparity increases with age15 (Fig.

1–7). The incidence difference observed among older womenappears to reflect racial differences in socioeconomic status49

and related reproductive patterns50: white women have a laterage at menarche, fewer births, and a later age at first full-termpregnancy than African American women, all factors that areknown to increase the long-term risk for breast cancer.

Differences in reproductive patterns in young AfricanAmerican and white women could also explain, at least inpart, the excess risk observed in young African Americanwomen.51,52 It has been shown that pregnancy has a dual effecton breast cancer risk, with a transient risk increase for severalyears after childbirth, followed by a risk reduction in lateryears.53–55 African American women tend to have higher parity than white women,56 which could contribute to theirhigher breast cancer incidence at ages younger than 40 yearsand lower incidence at older ages. Other factors possibly con-tributing to the higher premenopausal breast cancer rates inAfrican Americans compared with whites are less breast feed-ing,57 younger age at menarche,51,58,59 use of oral contracep-tives at a younger age,51 and lower physical activity.51 On theother hand, African Americans have a younger age at first full-term pregnancy,50 are more likely to be overweight or obese,60

and are less likely to consume alcohol than whites,61 all factorsthat should lower their risk for premenopausal breast cancer,as compared with white women. Additional research to assesswhether differences in known risk factor profiles are sufficientto explain the racial age-related crossover in breast cancerincidence is warranted.50

What is the impact of place of residence in the United States on breast cancer incidence?

For several decades, mortality from breast cancer has beenhighest in the Northeast of the United States and lowest in theSouth.62,63 Rates are also somewhat higher in the Midwest andWest than in the South.62 This pattern is more pronouncedamong postmenopausal than among premenopausal women, for whom little geographic variation was observed.A cluster analysis further suggested that the New YorkCity–Philadelphia metropolitan area had a 7.4% higher mor-tality rate than the rest of the Northeast.64 The elevated breast

7Chapter 1. Epidemiology of Breast Cancer

160

140

120

100

80

60

40

20

01975 1980 1985

Year

White African AmericanAsian/Pacific IslanderHispanicAmerican Indian/Alaska Native

Rat

e pe

r 10

0,00

0

1990 1995 2000

Figure 1–5 Incidence, age-adjusted to the2000 standard U.S. population, by race andethnicity, SEER 1975–2000. (Data fromAmerican Cancer Society. Breast Cancer Facts andFigures 2003–2004 [Online]. Available: http://www.cancer.org/docroot/STT/content/STT_1x_Breast_Cancer_Facts__Figures_2003–2004.asp [accessedJanuary 29, 2004].)

Regional28%

Localized65%

Distant5%

Unstaged2%

Distant9%

Unstaged4%

Regional34%

Localized53%

WHITE

AFRICAN AMERICAN

Figure 1–6 Stage distribution for African American and whitewomen, based on patients diagnosed between 1992 and 1999.(Data from American Cancer Society. Breast Cancer Facts and Figures2003–2004 [Online]. Available: http://www.cancer.org/docroot/STT/content/STT_1x_Breast_Cancer_Facts__Figures_2003–2004.asp [accessedJanuary 29, 2004].)

cancer mortality observed in the Northeast has led to the sus-picion that unknown environmental hazards, more prevalentin this region than in others, increase the risk for breast can-cer of Northeastern residents. Two studies, though, showedthat geographic variations in mortality or incidence could beaccounted for, at least in part, by differing regional patterns ofknown risk and prognostic factors for breast cancer.62,65 Oneof the major contributors to the high rates of breast cancer inthe Northeast was the tendency of Northeastern women tohave first births at later ages.66

What environmental exposures may affect breast cancer risk?

A number of studies have demonstrated that ionizing radia-tion causes female breast cancer. The studies range from the Japanese atomic bomb study,67 to diagnostic x-rays,68,69 tostudies of radiation treatment for benign conditions70,71 orcancer.72

The studies show agreement that the dose-response rela-tionship for breast cancer is approximately linear73 and thatradiation exposure in the first two decades of life confers sev-eral times more breast cancer risk than exposure in the adultyears, whereas exposures beyond age 50 years confer littlerisk.67 For irradiation at age 30 years, the excess relative risk for breast cancer increases about 1% per 1 cGy (1 rad).73 Ofspecial interest is the finding of a substantial excess of breastcancer in patients who had received on the order of 100 fluo-roscopic examinations in the course of pneumothorax treat-ment for tuberculosis68,74; these studies show that small dosefractions have a cumulative breast cancer induction effectsimilar to the risk one would expect if the total dose had beengiven as one exposure. The bottom line, however, is that mostmedical diagnostic radiation exposures with modern equip-ment and techniques are sufficiently low that any breast can-cer risk from them would be undetectable.

It has been proposed that exposure to 60-Hz electromag-netic fields (EMFs) may confer risk for breast cancer by

inhibiting the synthesis of melatonin by the pineal gland,which, in turn, may result in higher levels of circulating estro-gens, or by disrupting calcium homeostasis so as to increaseoxidative stress and tumor-promoting protein kinases.75 Mostof the studies of residential EMF exposure and breast cancerrisk in females have shown no association, including thosestudies that carefully included both the measurements and theelectrical transmission wire assessments of home EMF expo-sure levels.76–78 Because electric blankets or heating pads areclose to the body for extended periods of time, several studieshave been conducted to evaluate whether EMF exposure fromthis source poses a carcinogenic risk to the breast. Nearly allstudies of breast cancer and electric blanket use have beennegative.79–84

In recent years, there has been concern over whether envi-ronmental chemicals that mimic or antagonize endogenoussex hormones (so-called endocrine disruptors) may haveadverse effects on health. Human breast cancer risk fromendocrine-disrupting chemicals has been studied largely forpersistent organochlorine compounds, most notably PCBs(which were widely used in electrical transformers) and DDE(a metabolite of the insecticide DDT). Studies subsequent tothe initial ones have generally not been positive,85,86 and meta-analyses of the several studies have shown no indication of anassociation of breast cancer risk with PCBs or with DDE.87,88 Astudy of occupational exposure to some 29 different estrogen-mimicking chemicals did not find any statistically significantassociations with breast cancer, but it had limited statisticalpower.89

More than 50 studies have shown that alcohol consumptionis a moderate risk factor for breast cancer. Recent pooledanalyses have found that the relative risk for breast cancerincreases on the order of 7% to 10% for each additional drinkper day on average, and the association is reasonably linear.90,91

The mechanism by which alcohol influences breast cancer riskappears to be primarily by increasing endogenous estrogenlevels92,93; hence, one might expect alcohol to influence hormonally responsive tumors. Although the data are notentirely consistent, alcohol consumption appears to be more

SECTION I. MOLECULAR AND EPIDEMIOLOGIC ISSUES8

600

550

500

450

400

350

300

250

200

150

100

50

0 15–19 25–29 30–34 35–3920–24 40–44 45–49

Age (yr)

Incidence: WhiteIncidence: African American

Rat

e pe

r 10

0,00

0

50–54 55–59 60–64 65–69 70–74 75–79 80–84 85+

Figure 1–7 Age-specific incidence for African American and white women, SEER 1996–2000. (Data from American Cancer Society. BreastCancer Facts and Figures 2003–2004 [Online]. Available: http://www.cancer.org/docroot/STT/content/STT_1x_Breast_Cancer_Facts__Figures_2003–2004.asp [accessed January 29, 2004].)

strongly associated with ER-positive and progesterone receptor–positive breast cancers than with others, as expect-ed.94 Similarly, alcohol use is more strongly associated withlobular breast cancer, probably because lobular breast canceris more often ER postive and progesterone negative than is ductal breast cancer. There are suggestions that alcohol may confer greater risk for breast cancer among women whohave genetic polymorphisms in the CYP19 gene (which playsa key role in estrogen biosynthesis)95 or in the alcohol-metabolizing genes, alcohol dehydrogenase III (ADH3)96 orCYP2E1,97 but these results require confirmation by additionalstudies.

The weight of evidence from more than 35 studies,including the large Women’s Health Initiative,98 indicates that increased physical activity is protective against post-menopausal breast cancer.99 At least 16 studies have found aninverse dose-response association, and the average decrease inrisk among those who exercise regularly is 30% to 40%.99

However, the seeming protective effect of exercise on breastcancer risk may be confounded with, or secondary to, associ-ated variations in body mass index, because obesity is a knownrisk factor for breast cancer. Exercise may also be a surrogatefor a composite of lifestyle factors that collectively diminishrisk. The modest number of studies available on pre-menopausal breast cancer are more mixed in their results andprobably indicate a smaller protective effect of physical exer-cise at premenopausal ages.98,100–102

Although most studies suggest that regular use of non-steroidal anti-inflammatory drugs (NSAIDs), such as aspirinand ibuprofen, affords a modest protection against breast cancer risk, the case is not fully convincing because of a number of null results and conflicting evidence aboutwhether the protective effect is limited to aspirin or includesall NSAIDs.103–105 More importantly, a randomized controlledtrial is needed to ensure that the apparent beneficial effect ofNSAIDs on breast cancer risk is not an artifact of lifestyle dif-ferences between users and nonusers.

What is the impact of family history on breast cancer risk?

One long-known feature of breast cancer occurrence has beenthe familial aggregation of the disease. The recent, strikingadvances in our understanding of the genetic basis for thisfamilial aggregation are discussed in Chapters 2 and 20; theepidemiologic features of familial aggregation of breast canceris briefly outlined here. A first question is, at the populationlevel, what percentage of breast cancer is of heritable origin?One large study of monozygotic and dizygotic twin pairs esti-mated the heritability as 27%,106 but statistically more appro-priate analyses of those twin data and of large family-cancerdatabases in Sweden and among Utah Mormons suggest thatheritability is on the order of 10% to 15%.107–109 If one consid-ers only early breast cancer (<50 years of age), heritability isup to two times as high.108

Several major studies have conducted more detailed analy-ses and agree that about 6% of breast cancers before age 55years are linked to a family history of breast cancer in first-degree relatives.110 If one also includes any first-degree relativewith a history of ovarian cancer and second-degree relativeswith breast cancer, then about 13% to 15% of breast cancer

cases before age 55 years have a positive family history. On theother hand, for breast cancers occurring at ages 60 to 80 years,only 3% to 5% have a familial component.111 It is notable thatabout one third of the familial aggregation of breast cancer isnot accounted for by BRCA1 or BRCA2 mutations; thus, thereis clearly more to be learned about genetics and shared envi-ronmental factors in breast cancer.

Table 1–1 presents a summary of familial breast cancer risks in younger women from a large population-based studyconducted by the U.S. Centers for Disease Control andPrevention.110 As expected for autosomal dominant traits, theelevation in relative risk is appreciably greater when a first-degree relative is affected than when only a second-degree rel-ative is affected. Having more than one relative affected greatlyincreases risk. It is of interest that another major study, whichshowed a twofold breast cancer risk before age 40 years whena first-degree relative had breast cancer, also showed a 1.5-foldrisk for breast cancer even after the age of 70.112 Studies seemto disagree on whether having a relative with bilateral breastcancer confers more risk than having a relative with unilateralbreast cancer.110 However, a woman with a family history ofbreast cancer is more likely to develop metachronous bilateralbreast cancer than one with a negative family history.

The risk for breast cancer from a family history of severalother cancers has been studied. Ovarian cancer in a first-degree relative appears to confer a breast cancer relative risk ofabout 1.5 to 2 overall, although the subset for whom the ovar-ian cancer is associated with BRCA1 and BRCA2 mutationsmay have much higher risks. There is some indication,although it is not firmly established, that a family history ofprostate cancer or thyroid cancer may confer breast cancerrisks of about 1.5 and 1.7, respectively.107,110 In contrast, a fam-ily history of endometrial cancer does not appear to conferbreast cancer risk.

What is the impact of endogenous hormone levels on breast cancer risk?

Studies of reproductive risk factors and exogenous hormonesand theories of the roles of various hormones in breast cancerare reviewed in Chapter 4. This section briefly summarizesresults of studies on breast cancer risk in relation to endoge-nous levels of estrogens and androgens. These studies have

9Chapter 1. Epidemiology of Breast Cancer

Table 1–1 Odds Ratios (Relative Risks) for Reported FamilyHistory of Breast Cancer among 2080 Breast Cancer Patients cf.2058 Control Subjects, Ages 20–44 Years

Data from Thompson WD. Genetic epidemiology of breast cancer. Cancer1994;74:279–287.

Odds Ratio (95% Breast Cancer in Female Relatives Confidence Interval)

First-degree relativeMother affected 3.1 (2.3%–4.2%)Sister affected 3.1 (1.8%–5.3%)Two or more affected 6.8 (2.5%–20%)

Second-degree relativeMaternal grandmother/aunt affected 1.4 (1.2%–1.8%)Paternal grandmother/aunt affected 1.6 (1.2%–2.0%)Two or more affected 2.2 (1.4%–3.4%)

African American White Women (%) Women (%)

Current Age (yr) +10 yr +20 yr +30 yr +10 yr +20 yr +30 yr

20 0.04 0.43 1.90 0.07 0.51 1.86

30 0.39 1.86 4.64 0.44 1.80 4.06

40 1.48 4.29 7.98 1.39 3.70 6.34

50 2.90 6.70 10.4 2.43 5.21 7.66

60 4.06 8.05 10.4 3.09 5.80 7.26

been conducted mostly in postmenopausal women becausecyclic variations in estrogen levels among premenopausalwomen have made meaningful assessments difficult in largepopulations. Nine prospective studies have now reported onthe association between circulating estrogen and androgenlevels in postmenopausal women and subsequent breast can-cer risk, and a pooled analysis of the original data of thesestudies (663 cases and 1765 controls) concluded that bothestrogen and androgen hormones were strongly associatedwith risk.113 The breast cancer risk for women whose estradiollevels were in the top quintile was twice the risk of womenwhose estradiol levels were in the bottom quintile.

Similar results were observed for the other hormones studied, that is, estrone, estrone sulfate, androstenedione,testosterone, dehydroepiandrosterone, and dehydroepian-drosterone sulfate. As expected, the strongest associationswere observed with free estradiol and estradiol not bound tosex hormone–binding globulin (SHBG), which comprisesboth free and albumin-bound estradiol, because only estra-diol not bound to SHBG can readily enter cells. The hormoneand breast cancer risk associations remained after subjectswhose date of diagnosis was within 2 years of blood donationwere excluded from the analysis. A more recent study showedthat levels of sex hormones measured 5 to 13 years beforediagnosis were elevated in breast cancer cases as comparedwith controls.114 The association of circulating estrogens withbreast cancer risk appeared to be present regardless of whetherthe breast cancer was ER positive or ER negative.115

The New York University Women’s Heath Study also sug-gested that the contribution of androgens to breast cancer riskis largely through their role as substrates for estrogen produc-tion,114 whereas in the pooled analysis of prospective studies,androgens appeared to be associated with risk independentlyof estrogens.113 However, sorting out the role of estrogens andandrogens is complicated by the high correlations betweenhormones and the fact that usually only a single hormonalmeasurement is available for each participant in epidemio-logic studies, whereas the hormonal levels of interest are thelong-term average levels.113,114

How can the individual risk for developing breast cancer be estimated?

The lifetime probability of developing breast cancer (1 in 8 inwomen in the United States) is a striking statistic that conveysthe public health importance of the disease. However, it is a valid estimate for a girl born today only if incidence andmortality remain unchanged over her lifetime,116 an unlikelyassumption. Furthermore, it does not convey the fact that thelonger a woman lives without breast cancer the lower her riskis over the remainder of her lifetime.117 Estimates of risk at dif-ferent ages and for a more tangible time horizon, for example,the next 10 or 20 years, are more informative to women andtheir physicians, as well as more reliable.118 Table 1–2 providessuch estimates, calculated with the method of Feuer and col-leagues,10,114 for both African American and white women.

The estimates provided in Table 1–2 reflect the average risksfor developing breast cancer in women from the general pop-ulation. However, the risk is not homogeneous across womenof the same race and age. Models to predict individual riskhave been developed and can be useful decision-making tools

for women considering preventive use of tamoxifen or pro-phylactic mastectomy.117 The most frequently used is themodel developed by Gail and associates, for which a softwareprogram is available from the National Cancer Institute athttp://bcra.nci.nih.gov/brc/q1.htm.119,120 Graphs derived fromthis model to estimate individual risks and correspondingconfidence intervals have also been published.121 The Gailmodel estimates the chance that a woman with given age and risk factors who is screened annually will develop breastcancer over the next 5 years and for her lifetime. The risk fac-tors taken into account are age, ages at menarche and first livebirth, number of previous breast biopsies, presence of atypicalhyperplasia on biopsy, and number of first-degree relativeswith breast cancer. The model was shown to perform well forwhite women who receive annual mammograms.120,122–124

However, it has several limitations: it overestimates risk forwomen with infrequent mammography screening; it fails totake into account certain risk factors, such as the personal his-tory of in situ breast cancer125 or major genetic risks such asBRCA1, BRCA2, Cowden syndrome, or Li-Fraumeni syn-drome125,126; and it has been validated only among whites.127

This simple model, though based on data easily obtained fromthe medical history, is quite useful and is currently being usedto assess eligibility for the NSABP P-2 chemoprevention trial,a randomized study of tamoxifen versus raloxifene in womenat high risk for breast cancer.

Several other models have been developed to predict breast cancer risk,128–130 but they focus on the family history of breast cancer in high-risk families and are not broadlyapplicable.117,131

What is the survival from breast cancer in the United States?

Overall, survival from invasive breast cancer has improved inthe United States: the 5-year relative survival rate was 86.6%for women diagnosed in 1995 versus 74.9% for women diagnosed between 1975 and 1979.15 The increase in survivallikely results from both the “stage shift” toward increasingdiagnosis at early stage, owing to greater use of mammo-graphic screening, and therapeutic improvements.132 Factorsinfluencing survival include age, stage at diagnosis, and race

SECTION I. MOLECULAR AND EPIDEMIOLOGIC ISSUES10

Table 1–2 Risk for Being Diagnosed with Breast Cancer in 10,20, and 30 Years, Given that a Woman Is Cancer Free at CurrentAge, for White and African American Women

Data from Feuer E, Wun LM. DEVCAN: Probability of developing or dyingof cancer software, version 5.0 [Online]. Available: http://www.seer.cancer.gov/faststats/html/dev_breast.html (accessed January 29, 2004).

African American White Women (%) Women (%)

Stage <50 yr ≥50 yr <50 yr ≥50 yr

Localized 95.0 98.4 85.7 91.9

Regional 80.8 80.2 66.3 65.7

Distant 32.6 22.2 19.5 13.0

(Table 1–3). For patients diagnosed between 1992 and 1999,the 5-year relative survival rate was slightly lower for womendiagnosed before age 50 years (84.5%) than for older women(87.5%), which may be due to younger women’s havingtumors that are more aggressive and less responsive to hor-monal therapy. Race and stage at diagnosis also affect survivalrates (see Table 1–3). The lower stage-specific survival amongAfrican American women may be due to more aggressivetumors: tumors appear to be more often of medullary histol-ogy,46 hormone receptor negative,46 and of higher grade133 inAfrican American than in white women, all factors associatedwith poor prognosis. Less access to health care and differencesin treatment may also be contributing factors.134

What is the breast cancer mortality in the United States?

In the years 1996 to 2000, the female breast cancer age-adjustedmortality rates (per 100,000) were 27.2 in white women,35.9 in African American women, 17.8 in Hispanic women,14.9 in American Indian or Alaska Native women, and 12.5 inAsian or Pacific Islander women135 (Fig. 1–8). Trends overtime indicate that between 1970 and 1990, the mortality rateremained fairly constant among white women but increasedamong African American women, with an annual change of+0.9%. Between 1990 and 2000, the mortality rate decreased

in both white and African American women, but the decreasewas steeper among white (annual percent change, -2.5%)than among African American (-0.9%) women. In these same years, annual percent changes were -1.1% in Hispanicwomen, -1.5% in Asian or Pacific Islander women, and +0.2%in American Indian or Alaska Native women. The overalldecline in breast cancer mortality since 1990 has been attrib-uted to both improvements in treatment and early detection.Several factors may contribute to the 30% greater mortality inAfrican American than in white women. Although in recentyears rates of mammography screening in African Americanwomen have been approaching the rates in white women,African American women tend to present with higher-stagedisease than white women, raising the question of whetherthey receive screening at appropriate time intervals. In addi-tion to higher stage, tumors in African American women present more often with poor prognosis characteristics thantumors in white women. Finally, differences in access to healthcare and treatment may also contribute to the excess in mor-tality observed among African American women.136

CONCLUSION

The steady increase in breast cancer incidence observed in theUnited States in the 1980s appears to have subsided in the1990s, whereas mortality has decreased since 1990.16 Thesetrends are attributable, at least in part, to large-scale use ofmammographic screening. It is important to emphasize, how-ever, that early detection of breast cancer requires periodicscreening, not just one-time mammograms.137,138 Adjuvantmultiagent chemotherapy and use of antiestrogen therapy,such as tamoxifen, have also contributed to the decrease inmortality.139 Further research is needed to assess differences in incidence and mortality in African American and whitewomen as well as epidemiologic differences according totumor characteristics.

Breast cancer remains the most common cancer and thesecond most common cause of cancer death in women in theUnited States. Incidence of breast cancer has been increasingin most countries, and the increases have been particularlyrapid in several Asian countries where a “westernization” of

11Chapter 1. Epidemiology of Breast Cancer

*Relative survival rates are adjusted for expected mortality from causesother than breast cancer.

Data from http://surveillance.cancer.gov/statistics/types/survival.html.

Table 1–3 Five-Year Relative* Survival Rates by Race, Stage, andAge at Diagnosis (1992–1999 SEER Data)

45

40

35

30

25

20

15

10

5

01975 1980 1985

Year

White African AmericanAsian/Pacific IslanderHispanicAmerican Indian/Alaska Native

Rat

e pe

r 10

0,00

0

1990 1995 2000

Figure 1–8 Mortality rate, age-adjusted tothe 2000 standard U.S. population, by raceand ethnicity, SEER 1975–2000. (Data fromAmerican Cancer Society. Breast Cancer Facts andFigures 2003–2004 [Online]. Available: http://www.cancer.org/docroot/STT/content/STT_1x_Breast_Cancer_Facts__Figures_2003–2004.asp [accessedJanuary 29, 2004].)

the lifestyle has been observed. These rapid changes, togetherwith studies in migrants, underscore the importance of envi-ronmental factors in the development of the disease. However,to date, well-established “lifestyle” risk factors (later age at firstbirth, nulliparity, and family history of breast cancer) accountfor only about 40% of breast cancer in the United States. It isnot clear that these risk factors may be altered, which limitstheir relevance for prevention.28 The two most importantmodifiable risk factors known currently are to decrease alco-hol consumption and increase exercise and physical activity.Additional research to identify modifiable risk factors forbreast cancer, as well as to elucidate underlying biologicmechanisms, may help devise preventive strategies.

REFERENCES

1. Parkin DM. Global cancer statistics in the year 2000. Lancet Oncol2001;2:533–543.

2. Lacey JV Jr, Devesa SS, Brinton LA. Recent trends in breast cancer inci-dence and mortality. Environ Molec Mutagen 2002;39:82–88.

3. Geddes M, Parkin DM, Khlat M, et al. Cancer in Italian MigrantPopulations. IARC Scientific Publication No. 123. Lyon: IARC Press;1993.

4. Stanford JL, Herrinton LJ, Schwartz SM, et al. Breast cancer incidencein Asian migrants to the United States and their descendants.Epidemiology 1995;6:181–183.

5. Tyczynski J, Tarkowski W, Parkin DM, et al. Cancer mortality amongPolish migrants to Australia. Eur J Cancer 1994;30A:478–484.

6. Ziegler RG, Hoover RN, Pike MC, et al. Migration patterns and breastcancer risk in Asian-American women. J Natl Cancer Inst 1993;85:1819–1827.

7. Kelsey JL, Horn-Ross PL. Breast cancer: Magnitude of the problemand descriptive epidemiology. Epidemiol Rev 1993;15:7–16.

8. Pike MC, Kolonel LN, Henderson BE, et al. Breast cancer in a multi-ethnic cohort in Hawaii and Los Angeles: Risk factor-adjusted inci-dence in Japanese equals and in Hawaiians exceeds that in whites.Cancer Epidemiol Biomarkers Prev 2002;11:795–800.

9. American Cancer Society. Cancer Facts & Figures 2005 [Online].Available: http://www.cancer.org/docroot/STT/STT_0.asp (accessedDecember 8, 2003).

10. Feuer E, Wun LM. DEVCAN: Probability of developing or dying ofcancer software, version 5.0 [Online]. Available: http://www.seer.cancer.gov/faststats/html/dev_breast.html (accessed January 29, 2004).

11. American Cancer Society. Breast Cancer Facts and Figures 2005[Online]. Available: http://www.cancer.org/docroot/STT/content/STT_1x_Breast_Cancer_Facts__Figures_2005.asp (accessed January 2005).

12. Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer JClin 2005;55:10–30.

13. Alberg AJ, Singh S. Epidemiology of breast cancer in older women:Implications for future heath care. Drugs Aging 2001;18:761–772.

14. Clegg LX, Feuer EJ, Midthune DN, et al. Impact of reporting delay andreporting error on cancer incidence rates and trends. J Natl CancerInst 2002;94:1537–1545.

15. Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review,1975–2000 [Online]. Available: http://seer.cancer.gov/csr/1975_2000(accessed January 28, 2004).

16. Weir HK, Thun MJ, Hankey BF, et al. Annual report to the nation onthe status of cancer, 1975–2000, featuring the use of surveillance datafor cancer prevention and control. J Natl Cancer Inst 2003;95:1276–1299.

17. Li CI, Anderson BO, Daling JR, et al. Trends in incidence rates ofinvasive lobular and ductal breast carcinoma. JAMA 2003;289:1421–1424.

18. Glass AG, Hoover RN. Rising incidence of breast cancer: Relationshipto stage and receptor status. J Natl Cancer Inst 1990;82:693–696.

19. Li CI, Daling JR, Malone KE. Incidence of invasive breast cancer byhormone receptor status from 1992 to 1998. J Clin Oncol 2003;21:28–34.

20. Chu KC, Anderson WF. Rates for breast cancer characteristics by estro-gen and progesterone receptor status in the major racial/ethnicgroups. Breast Cancer Res Treat 2002;74:199–211.

21. Pujol P, Hilsenbeck SG, Chamness GC, et al. Rising levels of estrogenreceptor in breast cancer over 2 decades. Cancer 1994;74:1601–1606.

22. Zheng T, Holford TR, Chen Y, et al. Time trend of female breast carci-noma in situ by race and histology in Connecticut, U.S.A. Eur J Cancer1997;33:96–100.

23. Garfinkel L, Boring CC, Heath CW. Changing trends: An overview ofbreast cancer incidence and mortality. Cancer 1994;74:222–227.

24. Howard J. Using mammography for cancer control: An unrealizedpotential. CA Cancer J Clin 1987;37:33–48.

25. Breen N, Kessler L. Changes in the use of screening mammography:Evidence from the 1987 and 1990 National Health Interview Surveys.Am J Public Health 1994;84:62–67.

26. Wun L-M, Feuer EJ, Miller BA. Are increases in mammographicscreening still a valid explanation for trends in breast cancer incidencein the United States? Cancer Caus Cont 1995;6:135–144.

27. Dershaw D, Abraham A, Kinne DW. Ductal carcinoma in situ:Mammographic findings and clinical implications. Radiology 1989;170:411–415.

28. Madigan MP, Ziegler RG, Benichou J, et al. Proportion of breast can-cer cases in the United States explained by well-established risk factors.J Natl Cancer Inst 1995;87:1681–1685.

29. Tarone RE, Chu KC, Gaudette LA. Birth cohort and calendar periodtrends in breast cancer mortality in the United States and Canada. JNatl Cancer Inst 1997;89:251–256.

30. Collaborative Group on Hormonal Factors in Breast Cancer. Breastcancer and hormonal contraceptives: Collaborative reanalysis of indi-vidual data on 53,297 women with breast cancer and 100,239 womenwithout breast cancer from 54 epidemiologic studies. Lancet 1996;347:1713–1727.

31. Collaborative Group on Hormonal Factors in Breast Cancer. Breastcancer and hormone replacement therapy: Collaborative reanalysis ofdata from 51 epidemiological studies of 52,705 women with breastcancer and 108,411 women without breast cancer. Lancet 1997;350:1047–1059.

32. Brett KM, Madans JH. Use of postmenopausal hormone replacementtherapy: Estimates from a nationally representative cohort study. Am JEpidemiol 1997;145:536–545.

33. Li CI, Weiss NS, Stanford JL, et al. Hormone replacement therapy inrelation to risk of lobular and ductal breast carcinoma in middle-agedwomen. Cancer 2000;88:2570–2577.

34. Chen CL, Weiss NS, Newcomb P, et al. Hormone replacement therapyin relation to breast cancer. JAMA 2002;287:734–741.

35. Newcomb PA, Titus-Ernstoff L, Egan KM, et al. Postmenopausal estro-gen and progestin use in relation to breast cancer risk. CancerEpidemiol Biomarkers Prev 2002;11:593–600.

36. Daling JR, Malone KE, Doody DR, et al. Relation of regimens of com-bined hormone replacement therapy to lobular, ductal, and other histologic types of breast carcinoma. Cancer 2002;95:2455–2464.

37. American Cancer Society. Breast Cancer Facts & Figures 2001–2002 [Online]. Available: http://www.cancer.org/downloads/STT/BrCaFF2001.pdf (accessed December 8, 2003).

38. Key TJA, Pike MC. The role of oestrogens and progestagens in the epidemiology and prevention of breast cancer. Eur J Cancer 1988;24:29–43.

39. Tarone RE, Chu KC. The greater impact of menopause on ER- thanER+ breast cancer incidence: A possible explanation (United States).Cancer Caus Cont 2002;13:7–14.

40. Yasui Y, Potter JD. The shape of age-incidence curves of female breastcancer by hormone-receptor status. Cancer Caus Cont 1999;10:431–437.

41. U.S. Cancer Statistics Working Group. United States Cancer Statistics:2000 Incidence [Online]. Available: http://www.cdc.gov/cancer/npcr/uscs/2000/working_group.htm (accessed January 24, 2004).

42. Hsu JL, Glaser SL, West DW. Racial/ethnic differences in breast cancersurvival among San Francisco Bay Area women. J Natl Cancer Inst1997;89:1311–1312.

43. Swan J, Breen N, Coates RJ, et al. Progress in cancer screening practicesin the United States: Results from the 2000 National Health InterviewSurvey. Cancer 2003;2003:1528–1540.

44. Breen N, Wagener DK, Brown ML, et al. Progress in cancer screeningover a decade: Results of cancer screening from the 1987, 1992, and1998 National Health Interview Surveys. J Natl Cancer Inst 2001;93:1704–1713.

45. Swanson GM, Haslam SZ, Azzouz F. Breast cancer among youngAfrican-American women: A summary of data and literature and ofissues discussed during the “Summit meeting on breast cancer among

SECTION I. MOLECULAR AND EPIDEMIOLOGIC ISSUES12

African American women,” Washington, DC, September 8–10, 2000.Cancer 2003;97(1 Suppl):273–279.

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

47. Joslyn SA. Hormone receptors in breast cancer: Racial differences indistribution and survival. Breast Cancer Res Treat 2002;73:45–59.

48. Li CI, Moe RE, Daling JR. Risk of mortality by histologic type of breastcancer among women aged 50 to 79 years. Arch Intern Med 2003;163:2149–2153.

49. Krieger N. Social class and the black/white crossover in the age-specific incidence of breast cancer: A study linking census-deriveddata to population-based registry records. Am J Epidemiol 1990;131:804–814.

50. Brinton LA, Benichou J, Gammon MD, et al. Ethnicity and variationin breast cancer incidence. Int J Cancer 1997;73:349–355.

51. Bernstein L, Teal CR, Joslyn S, et al. Ethnicity-related variation inbreast cancer risk factors. Cancer 2003;97 (1 Suppl):222–229.

52. Pathak DR, Osuch JR, He J. Breast carcinoma etiology: Currentknowledge and new insights into the effects of reproductive and hor-monal risk factors in black and white populations. Cancer 2000;88:1230–1238.

53. Bruzzi P, Negri E, La Vecchia C, et al. Short term increase in risk ofbreast cancer after full term pregnancy. BMJ 1994;297:1096–1098.

54. Lambe M, Hsieh C, Trichopoulos D, et al. Transient increase in the riskof breast cancer after giving birth. N Engl J Med 1994;331:5–9.

55. Rosner B, Colditz GA, Willett WC. Reproductive risk factors in aprospective study of breast cancer: The Nurses’ Health Study. Am JEpidemiol 1994;139:819–835.

56. Ventura SJ, Mosher WD, Curtin SC, et al. Highlights of trends in preg-nancies and pregnancy rates by outcome: Estimates from the UnitedStates, 1976–1996. Hyattsville, MD: National Center for HealthStatistics; 1999.

57. Abma J, Chandra A, Mosher W, et al. Fertility, family planning,and women’s health: New data from the 1995 National Survey ofFamily Growth. Hyattsville, MD: National Center for Health Statistics,1997.

58. Burke GL, Savage PJ, Manolio TA, et al. Correlates of obesity in youngblack and white women: The CARDIA Study. Am J Public Health1992;82:1621–1625.

59. Wattigney WA, Srinivasan SR, Chen W, et al. Secular trend of earlieronset of menarche with increasing obesity in black and white girls:The Bogalusa Heart Study. Ethnicity Dis 1999;9:181–189.

60. National Center for Health Statistics. Prevalence of BMI, overweightand obesity: United States, 1960–1994 [Online]. Available: www.cdc.gov/nchs/about/major/nhanes/overweight.pdf (accessed January 14,2004).

61. Schoenborn CA, Adams PF. Alcohol use among adults: United States,1997–98. Advanced data from vital and health statistics, No. 324.Hyattsville, MD: National Center for Health Statistics; 2001.

62. Sturgeon SR, Schairer C, Gail MH, et al. Geographic variation in mor-tality from breast cancer among white women in the United States. JNatl Cancer Inst 1995;87:1846–1853.

63. Blot WJ, Fraumeni JF, Stone BJ. Geographic patterns of breast cancerin the United States. J Natl Cancer Inst 1977;59:1407–1411.

64. Kulldorff M, Feuer EJ, Miller BA, et al. Breast cancer clusters in theNortheast United States: A geographic analysis. Am J Epidemiol1997;146:161–170.

65. Laden F, Spiegelman D, Neas LM, et al. Geographic variation in breastcancer incidence rates in a cohort of U.S. women. J Natl Cancer Inst1997;89:1373–1378.

66. Blot WM, McLaughlin JK. Geographic patterns of breast canceramong American women (editorial). J Natl Cancer Inst 1995;87:1819–1820.

67. Land CE, Tokunaga M, Koyama K, et al. Incidence of female breastcancer among atomic bomb survivors, Hiroshima and Nagasaki,1950–1990. Radiat Res 2003;160:707–717.

68. Howe G, McLaughlin J. Breast cancer mortality between 1950 and1987 after exposure to fractionated moderate-dose-rate ionizing radi-ation in the Canadian fluoroscopy cohort study and a comparisonwith breast cancer mortality in the atomic bomb survivors study.Radiat Res 1996;145:694–707.

69. Doody MM, Lonstein JE, Stovall M, et al. Breast cancer mortality afterdiagnostic radiography: Findings from the U.S. Scoliosis CohortStudy. Spine 2000;25:2052–2063.

70. Hildreth N, Shore R, Dvoretsky P. Risk of breast cancer after irradia-tion of the thymus in infancy. N Engl J Med 1989;321:1281–1284.

71. Shore RE, Hildreth N, Woodard E, et al. Breast cancer among womengiven x-ray therapy for acute postpartum mastitis. J Natl Cancer Inst1986;77:689–696.

72. Travis LB, Hill DA, Dores GM, et al. Breast cancer following radio-therapy and chemotherapy among young women with Hodgkin dis-ease. JAMA 2003;290:465–475.

73. Preston DL, Mattsson A, Holmberg E, et al. Radiation effects on breastcancer risk: A pooled analysis of eight cohorts. Radiat Res 2002;158:220–235.

74. Boice JD, Preston D, Davis F, et al. Frequent chest x-ray fluoroscopyand breast cancer incidence among tuberculosis patients inMassachusetts. Radiat Res 1991;125:214–222.

75. Stevens RG. Biologically based epidemiological studies of electricpower and cancer. Environ Health Perspect 1993;101(Suppl 4): 93–100.

76. Davis S, Mirick DK, Stevens RG. Residential magnetic fields and therisk of breast cancer. Am J Epidemiol 2002;155:446–454.

77. London SJ, Pogoda JM, Hwang KL, et al. Residential magnetic fieldexposure and breast cancer risk: A nested case-control study from amultiethnic cohort in Los Angeles County, California. Am J Epidemiol2003;158:969–980.

78. Schoenfeld ER, O’Leary ES, Henderson K, et al. Electromagnetic fieldsand breast cancer on Long Island: A case-control study. Am JEpidemiol 2003;158:47–58.

79. Gammon MD, Schoenberg JB, Britton JA, et al. Electric blanket useand breast cancer risk among younger women. Am J Epidemiol1998;148:556–563.

80. Zheng T, Holford TR, Mayne ST, et al. Exposure to electromagneticfields from use of electric blankets and other in-home electrical appli-ances and breast cancer risk. Am J Epidemiol 2000;151:1103–1111.

81. McElroy JA, Newcomb PA, Remington PL, et al. Electric blanket ormattress cover use and breast cancer incidence in women 50–79 yearsof age. Epidemiology 2001;12:613–617.

82. Laden F, Neas LM, Tolbert PE, et al. Electric blanket use and breastcancer in the Nurses’ Health Study. Am J Epidemiol 2000;152:1–9.

83. Kabat GC, O’Leary ES, Schoenfeld ER, et al. Electric blanket use andbreast cancer on Long Island. Epidemiology 2003;14:514–520.

84. Zhu K, Hunter S, Payne-Wilks K, et al. Use of electric bedding devicesand risk of breast cancer in African-American women. Am JEpidemiol 2003;158:798–806.

85. Lopez-Carrillo L, Blair A, Lopez-Cervantes M, et al. Dichlorodi-phenyltrichloroethane serum levels and breast cancer risk: A case-control study from Mexico. Cancer Res 1997;57:3728–3732.

86. Wolff MS, Zeleniuch-Jacquotte A, Dubin N, et al. Risk of breast cancerand organochlorine exposure. Cancer Epidemiol Biomarkers Prev2000;9:271–277.

87. Adami H-O, Lipworth L, Titus-Ernstoff L, et al. Organochlorine com-pounds and estrogen-related cancers in women. Cancer Caus Cont1995;6:551–566.

88. Laden F, Collman G, Iwamoto K, et al. 1,1-Dichloro-2,2-bis(p-chlorophenyl)ethylene and polychlorinated biphenyls and breast cancer: Combined analysis of five U.S. studies. J Natl Cancer Inst2001;93:768–776.

89. Aschengrau A, Coogan PF, Quinn MM, et al. Occupational exposureto estrogenic chemicals and the occurrence of breast cancer: Anexploratory analysis. Am J Indust Med 1998;34:6–14.

90. Collaborative Group on Hormonal Factors in Breast Cancer. Alcohol,tobacco and breast cancer: Collaborative reanalysis of individual datafrom 53 epidemiological studies, including 58,515 women with breastcancer and 95,067 women without the disease. Br J Cancer 2002;87:1234–1245.

91. Ellison RC, Zhang Y, McLennan CE, et al. Exploring the relation ofalcohol consumption to risk of breast cancer. Am J Epidemiol 2001;154:740–747.

92. Dorgan JF, Baer DJ, Albert PS, et al. Serum hormones and the alcohol-breast cancer association in postmenopausal women. J Natl CancerInst 2001;93:710–715.

93. Singletary KW, Gapstur SM. Alcohol and breast cancer: Review of epi-demiologic and experimental evidence and potential mechanisms.JAMA 2001;286:2143–2151.

94. Li CI, Malone KE, Porter PL, et al. The relationship between alcoholuse and the risk of breast cancer by histology and hormone receptorstatus among women 65–79 years of age. Cancer EpidemiolBiomarkers Prev 2003;12:1061–1066.

13Chapter 1. Epidemiology of Breast Cancer

95. Lee KM, Abel J, Ko Y, et al. Genetic polymorphisms of cytochromeP450 19 and 1B1, alcohol use, and breast cancer risk in Koreanwomen. Br J Cancer 2003;88:675–678.

96. Freudenheim JL, Ambrosone CB, Moysich KB, et al. Alcohol dehydro-genase 3 genotype modification of the association of alcohol consumption with breast cancer risk. Cancer Caus Cont 1999;10:369–377.

97. Choi JY, Abel J, Neuhaus T, et al. Role of alcohol and genetic polymor-phisms of CYP2E1 and ALDH2 in breast cancer development.Pharmacogenetics 2003;13:67–72.

98. McTiernan A, Kooperberg C, White E, et al. Recreational physicalactivity and the risk of breast cancer in postmenopausal women. TheWomen’s Health Initiative Cohort Study. JAMA 2003;290:1331–1336.

99. Friedenreich CM. Physical activity and cancer prevention: Fromobservational to intervention research. Cancer Epidemiol BiomarkersPrev 2001;10:287–301.

100. Colditz GA, Feskanich D, Chen WY, et al. Physical activity and risk of breast cancer in premenopausal women. Br J Cancer 2003;89:847–851.

101. John EM, Horn-Ross PL, Koo J. Lifetime physical activity and breastcancer risk in a multiethnic population: The San Francisco Bay areabreast cancer study. Cancer Epidemiol Biomarkers Prev 2003;12:1143–1152.

102. Steindorf K, Schmidt M, Kropp S, et al. Case-control study of physicalactivity and breast cancer risk among premenopausal women inGermany. Am J Epidemiol 2003;157:121–130.

103. Harris RE, Chlebowski RT, Jackson RD, et al. Breast cancer and non-steroidal anti-inflammatory drugs: Prospective results from theWomen’s Health Initiative. Cancer Res 2003;63:6069–6101.

104. Johnson TW, Anderson KE, Lazovich D, et al. Association of aspirinand nonsteroidal anti-inflammatory drug use with breast cancer.Cancer Epidemiol Biomarkers Prev 2002;11:1586–1591.

105. Khuder SA, Mutgi AB. Breast cancer and NSAID use: A meta-analysis.Br J Cancer 2001;84:1188–1192.

106. Lichtenstein P, Holm NV, Verkasalo PK, et al. Environmental and heritable factors in the causation of cancer. N Engl J Med 2000;343:78–85.

107. Goldgar DE, Easton DF, Cannon-Albright LA, et al. Systematic population-based assessment of cancer risk in first-degree relatives ofcancer probands. J Natl Cancer Inst 1994;86:1600–1608.

108. Risch N. The genetic epidemiology of cancer: Interpreting family andtwin studies and their implications for molecular genetic approaches.Cancer Epidemiol Biomarkers Prev 2001;10:733–741.

109. Hemminki K, Czene K. Attributable risks of familial cancer from thefamily-cancer database. Cancer Epidemiol Biomarkers Prev 2002;11:1638–1644.

110. Thompson WD. Genetic epidemiology of breast cancer. Cancer1994;74:279–287.

111. Claus EB, Schildkraut JM, Thompson WD, et al. The genetic attribut-able risk of breast and ovarian cancer. Cancer 1996;77:2318–2324.

112. Colditz G, Willett W, Hunter D, et al. Family history, age, and risk ofbreast cancer: Prospective data from the Nurses’ Health Study. JAMA1993;270:338–343.

113. Endogenous Hormones and Breast Cancer Collaborative Group.Endogenous sex hormones and breast cancer in postmenopausalwomen: Reanalysis of nine prospective studies. J Natl Cancer Inst2002;94:606–616.

114. Zeleniuch-Jacquotte A, Shore RE, Koenig KL, et al. Postmenopausallevels of estrogen, androgen, and SHBG and breast cancer risk: Long-term results of a prospective study. Br J Cancer 2004;90:153–159.

115. Zeleniuch-Jacquotte A, Toniolo P, Levitz M, et al. Endogenous estro-gens and risk of breast cancer by estrogen receptor status: A prospec-tive study in postmenopausal women. Cancer Epidemiol BiomarkersPrev 1995;4:857–860.

116. Feuer EJ, Wun LM, Boring CC. The lifetime risk of developing breastcancer. J Natl Cancer Inst 1993;85:892–897.

117. Armstrong K, Eisen A, Weber B. Primary care: Assessing the risk ofbreast cancer. N Engl J Med 2000;342:564–571.

118. Swanson GM. Breast cancer risk estimation: A translational statisticfor communication to the public [editorial]. J Natl Cancer Inst1993;85:848–849.

119. Gail MH, Brinton LA, Byar DP, et al. Projecting individualized proba-bilities of developing breast cancer for white females who are beingexamined annually. J Natl Cancer Inst 1989;81:1879–1886.

120. Costantino JP, Gail MH, Pee D, et al. Validation studies for modelsprojecting the risk of invasive and total breast cancer incidence. J NatlCancer Inst 1999;91:1541–1548.

121. Benichou J, Gail MH, Mulvihill JJ. Graphs to estimate an individual-ized risk of breast cancer. J Clin Oncol 1996;14:103–110.

122. Bondy ML, Lustbader ED, Halabi S, et al. Validation of a breast cancerrisk assessment model in women with a positive family history. J NatlCancer Inst 1994;86:620–625.

123. Spiegelman D, Colditz GA, Hunter D, et al. Validation of the Gail et almodel for predicting individual breast cancer risk. J Natl Cancer Inst1994;86:600–607.

124. Rockhill B, Spiegelman D, Byrne C, et al. Validation of the Gail et al.model of breast cancer risk prediction and implications for chemo-prevention. J Natl Cancer Inst 2001;93:358–366.

125. Gail MH, Costantino JP. Validating and improving models for pro-jecting the absolute risk of breast cancer [editorial]. J Natl Cancer Inst2001;93:334–335.

126. Gail MH, Benichou J. Validation studies on a model for breast cancerrisk [editorial]. J Natl Cancer Inst 1994;86:573–575.

127. Bondy ML, Newman LA. Breast cancer risk assessment models:Applicability to African American women. Cancer 2003;97(1Suppl):230–235.

128. Claus EB, Risch NJ, Thompson D. Age at onset as an indicator offamilial risk of breast cancer. Am J Epidemiol 1990;131:961–972.

129. Anderson DE. Risk of familial breast cancer. Cancer 1985;56:383–387.130. Ottman R, Pike MC, King MC, et al. Practical guide for estimating risk

for familial breast cancer. Lancet 1983;2:556–558.131. McGuigan KA, Ganz PA, Breant C. Agreement between breast cancer

risk estimation methods. J Natl Cancer Inst 1996;88:1315–1317.132. Chu KC, Tarone RE, Kessler LG, et al. Recent trends in U.S. breast

cancer incidence, survival, and mortality rates. J Natl Cancer Inst1996;88:1571–1579.

133. Henson DE, Chu KC, Levine PH. Histologic grade, stage, and survivalin breast carcinoma: Comparison of African American and Caucasianwomen. Cancer 2003;98:908–917.

134. Chu KC, Lamar CA, Freeman HP. Racial disparities in breast carcino-ma survival rates: Separating factors that affect diagnosis from factorsaffecting treatment. Cancer 2003;97:2853–2860.

135. Surveillance Epidemiology and End Results (SEER) Program. SEER*Stat Database: Mortality. All COD, public use with state, total U.S.for expanded races/Hispanics (1990–2000) [Online]. Available:www.seer.cancer.gov (accessed January 15, 2004).

136. Chu KC, Tarone RE, Brawley OW. Breast cancer trends of blackwomen compared with white women. Arch Fam Med 1999;8:521–528.

137. U.S. Preventive Services Task Force (USPSTF). Recommendations and rationale for screening for breast cancer [Online]. Available:http://www.ahcpr.gov/clinic/3rduspstf/breastcancer/brcanrr.htm(accessed December 15, 2003).

138. Smith RA, Saslow D, Sawyer KA, et al. American Cancer Society guide-lines for breast cancer screening: Update 2003. Ca Cancer J Clin2003;53:141–169.

139. Mariotto A, Feuer EJ, Harlan LC, et al. Trends in use of adjuvant multi-agent chemotherapy and tamoxifen for breast cancer in the UnitedStates: 1975–1999. J Natl Cancer Inst 2002;94:1626–1634.

SECTION I. MOLECULAR AND EPIDEMIOLOGIC ISSUES14