Cancer Statistics, 2006

Cancer Statistics, 2006

Ahmedin Jemal, DVM, PhD; Rebecca Siegel, MPH; Elizabeth Ward, PhD; Taylor Murray;
Jiaquan Xu; Carol Smigal, MPH; Michael J. Thun, MD, MS

Dr. Jemal is Program Director, Cancer

Occurrence, Department of Epidemiol- ABSTRACT Each year, the American Cancer Society estimates the number of new cancer cases
ogy and Surveillance Research, Ameri- and deaths expected in the United States in the current year and compiles the most recent data on
can Cancer Society, Atlanta, GA.
cancer incidence, mortality, and survival based on incidence data from the National Cancer Institute
Ms. Siegel is Manager, Surveil-
lance Information Services, De- and mortality data from the National Center for Health Statistics. Incidence and death rates are
partment of Epidemiology and age-standardized to the 2000 US standard million population. A total of 1,399,790 new cancer
Surveillance Research, American
Cancer Society, Atlanta, GA. cases and 564,830 deaths from cancer are expected in the United States in 2006. When deaths are
Dr. Ward is Director, Surveillance Re- aggregated by age, cancer has surpassed heart disease as the leading cause of death for those
search, Department of Epidemiology younger than age 85 since 1999. Delay-adjusted cancer incidence rates stabilized in men from 1995
and Surveillance Research, American
Cancer Society, Atlanta, GA. through 2002, but continued to increase by 0.3% per year from 1987 through 2002 in women.
Mr. Murray is Manager, Surveil- Between 2002 and 2003, the actual number of recorded cancer deaths decreased by 778 in men,
lance Data Systems, Department of but increased by 409 in women, resulting in a net decrease of 369, the first decrease in the total
Epidemiology and Surveillance Re-
search, American Cancer Society, number of cancer deaths since national mortality record keeping was instituted in 1930. The death
Atlanta, GA. rate from all cancers combined has decreased by 1.5% per year since 1993 among men and by
Mr. Xu is Epidemiologist, Mortality 0.8% per year since 1992 among women. The mortality rate has also continued to decrease for the
Statistics Branch, Division of Vital
Statistics, Centers for Disease Con- three most common cancer sites in men (lung and bronchus, colon and rectum, and prostate) and
trol and Prevention, Hyattsville, MD. for breast and colon and rectum cancers in women. Lung cancer mortality among women contin-
Ms. Smigal is Epidemiologist, ues to increase slightly. In analyses by race and ethnicity, African American men and women have
Department of Epidemiology and
Surveillance Research, American 40% and 18% higher death rates from all cancers combined than White men and women, respec-
Cancer Society, Atlanta, GA. tively. Cancer incidence and death rates are lower in other racial and ethnic groups than in Whites
Dr. Thun is Vice-President, Depart- and African Americans for all sites combined and for the four major cancer sites. However, these
ment of Epidemiology and Surveil-
lance Research, American Cancer groups generally have higher rates for stomach, liver, and cervical cancers than Whites. Further-
Society, Atlanta, GA. more, minority populations are more likely to be diagnosed with advanced stage disease than are
This article is available online at Whites. Progress in reducing the burden of suffering and death from cancer can be accelerated by
http://CAonline.AmCancerSoc.org
applying existing cancer control knowledge across all segments of the population. (CA Cancer J Clin
2006;56:106–130.) © American Cancer Society, Inc., 2006.

INTRODUCTION

Cancer is a major public health problem in the United States and other developed countries. Currently, one in
four deaths in the United States is due to cancer. In this article, we provide an overview of cancer statistics, including
updated incidence, mortality, and survival rates and expected number of new cancer cases and deaths in 2006.

MATERIALS AND METHODS

Data Sources

Mortality data from 1930 to 2003 in the United States were obtained from the National Center for Health
Statistics (NCHS).1 Incidence data (1975 to 2002), 5-year relative survival rates, and data on lifetime probability of

106 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

TABLE 1 Estimated New Cancer Cases and Deaths by Sex, United States, 2006*

*Rounded to the nearest 10; estimated new cases exclude basal and squamous cell skin cancers and in situ carcinomas except urinary bladder.
About 61,980 carcinoma in situ of the breast and 49,710 melanoma in situ will be newly diagnosed in 2006.
†Estimated deaths for colon and rectum cancers are combined.
‡More deaths than cases suggests lack of specificity in recording underlying causes of death on death certificates.
Source: Estimates of new cases are based on incidence rates from 1979 to 2002, National Cancer Institute’s Surveillance, Epidemiology and End
Results program, nine oldest registries. Estimates of deaths are based on data from US Mortality Public Use Data Tapes, 1969 to 2003, National
Center for Health Statistics, Centers for Disease Control and Prevention, 2006.

Volume 56 Y Number 2 Y March/April 2006 107

 Cancer Statistics, 2006

TABLE 2 Age-standardized Incidence Rates for All Cancers Combined, 1998-2002, and Estimated New Cases* for
Selected Cancers by State, United States, 2006

*Rounded to the nearest 10; excludes basal and squamous cell skin cancers and in situ carcinomas except urinary bladder.
†Average annual rates for 1998–2002, age-adjusted to the 2000 US standard population; source: Cancer in North America; 1998–2002, Vol. One: Incidence,
NAACCR, based on data collected by cancer registries participating in NCI’s SEER Program and CDC’s National Program of Cancer Registries.
‡Estimate is fewer than 50 cases. Note: These estimates are offered as a rough guide and should be interpreted with caution. State estimates are
calculated according to the distribution of estimated cancer deaths in 2006 by state. State estimates may not add to US total due to rounding and
exclusion of state estimates fewer than 50 cases.
§Combined incidence rate is not available.
¶Incidence rate is for the Metropolitan Atlanta area.

108 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

FIGURE 1 Ten Leading Cancer Types for the Estimated New Cancer Cases and Deaths, by Sex, US, 2006.
*Excludes basal and squamous cell skin cancers and in situ carcinoma except urinary bladder. Estimates are rounded to the nearest 10.
Note: Percentage may not total 100% due to rounding.

developing cancer were obtained from the Sur- coded and classified according to the International
veillance, Epidemiology, and End Results Classification of Diseases (ICD-8, ICD-9, and
(SEER) program of the National Cancer Insti- ICD-10).7–9 Cancer cases were classified accord-
tute, covering about 14% of the US popula- ing to the International Classification of Diseases
tion.2–5 State-specific incidence rates were for Oncology.10
abstracted from Cancer in North America (1998-
2002) Volume One, based on data collected by Estimated New Cancer Cases
cancer registries participating in the SEER pro-
gram and Centers for Disease Control and Pre- The precise number of cancer cases diagnosed
vention (CDC)’s National Program of Cancer each year in the nation is unknown because com-
Registries. Population data were obtained from plete cancer registration has not yet been achieved in
the US Census Bureau.6 Causes of death were many states. Consequently, for the national estimate

Volume 56 Y Number 2 Y March/April 2006 109

 Cancer Statistics, 2006

TABLE 3 Age-standardized Death Rates for All Cancers Combined, 1998 to 2002, and Estimated Deaths* from All
Cancers Combined and Selected Sites by State, United States, 2006

*Rounded to the nearest 10.

†Average annual rates for 1998 to 2002 are age-adjusted to the 2000 US standard population.
‡Estimate is fewer than 50 deaths.
Note: State estimates may not add to US total due to rounding and exclusion of state estimates fewer than 50 deaths.
Source: US Mortality Public Use Data Tapes, 1969 to 2003, National Center for Health Statistics, Centers for Disease Control and Prevention, 2006.

110 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

TABLE 4 Cancer Incidence Rates* by Site and State, US, 1998 to 2002

*Per 100,000, age-adjusted to the 2000 US standard population. Not all states submitted data for all years.
†This state’s registry has submitted five years of data and passed rigorous criteria for each single year’s data including: completeness of reporting,
non-duplication of records, percent unknown in critical data fields, percent of cases registered with information from death certificates only, and internal
consistency among data items.
‡This state’s registry did not submit incidence data to the North American Association of Central Cancer Registries (NAACCR) for 1998 to 2002.
Source: Cancer in North America: 1998 to 2002, Volume One: Incidence, North American Association of Central Cancer Registries.

Volume 56 Y Number 2 Y March/April 2006 111

Cancer Statistics, 2006

FIGURE 2 Annual Age-adjusted Cancer Incidence and Death Rates* for All Sites, by Sex, US, 1975 to 2002.
*Rates are age-adjusted to the 2000 US standard population. Incidence rates are delay-adjusted. Source: Incidence data from
Surveillance, Epidemiology, and End Results (SEER) program, nine oldest registries, 1975 to 2002, Division of Cancer Control
and Population Sciences, National Cancer Institute, 2005. Mortality data from US Mortality Public Use Data Tapes, 1960 to
2002, National Center for Health Statistics, Centers for Disease Control and Prevention, 2005.

we first estimated the number of new cancer cases using an autoregressive quadratic time-trend model
occurring annually in the United States from 1979 fitted to the annual cancer case estimates.11 For
through 2002, using age-specific cancer incidence estimates of new cancer cases in individual states, we
rates collected by the SEER program2 and popula- projected the number of deaths from cancer in each
tion data reported by the US Census Bureau.6 We state in 2006 and assumed that the ratio of estimated
then forecast the number of cancer cases expected to cancer deaths to cases in each state equaled that in
be diagnosed in the United States in the year 2006 the United States.

112 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

FIGURE 3 Annual Age-adjusted Cancer Incidence Rates* Among Males and Females for Selected Cancers, US, 1975 to 2002.
*Rates are age-adjusted to the 2000 US standard population and adjusted for delays in reporting with the exception of melanoma. Source:
Surveillance, Epidemiology, and End Results (SEER) program, nine oldest registries, 1975 to 2002, Division of Cancer Control and Popula-
tion Sciences, National Cancer Institute, 2005.

Estimated Cancer Deaths 2003 were coded and classified according to

We used the state-space prediction method 12 ICD-10.7 This report also provides updated
to estimate the number of cancer deaths expected statistics on trends in cancer incidence and
to occur in the United States and in each state in mortality rates, the probability of developing
the year 2006. Projections arebased on underly- cancer, and 5-year relative survival rates for
ing cause-of-death from death certificates as re- selected cancer sites based on data from 1974
ported to the NCHS.1 This model projects the through 2002.3 All age-adjusted incidence
number of cancer deaths expected to occur in and death rates are standardized to the 2000
2006 based on the number that occurred each US standard population and expressed per
year from 1969 to 2003 in the United States and 100,000 population.
in each state separately. The long-term incidence rates and trends
(1975 to 2002) are adjusted for delays in re-
Other Statistics porting where possible. Delayed reporting af-
fects the most recent 1 to 3 years of incidence
We provide mortality statistics for the data (in this case, 2000 to 2002), especially for
leading causes of death as well as deaths from cancers such as melanoma and prostate that are
cancer in the year 2003. Causes of death for frequently diagnosed in outpatient settings.

Volume 56 Y Number 2 Y March/April 2006 113

Cancer Statistics, 2006

FIGURE 4 Annual Age-adjusted Cancer Death Rates* Among Males for Selected Cancers, US, 1930 to 2002.
*Rates are age-adjusted to the 2000 US standard population. Note: Due to changes in ICD coding, numerator information has changed over
time. Rates for cancers of the lung and bronchus, colon and rectum, and liver are affected by these changes. Source: US Mortality Public Use
Data Tapes, 1960 to 2002, US Mortality Volumes, 1930 to 1959, National Center for Health Statistics, Centers for Disease Control and Pre-
vention, 2005.

The NCI has developed a method to account invasive cancer does not include carcinoma
for expected reporting delays in SEER registries in situ of any site except urinary bladder, nor
for all cancer sites combined and several specific does it include basal cell and squamous cell
cancer sites when long-term incidence trends are cancers of the skin. Over 1 million cases of
analyzed.13 Delay-adjusted incidence provides a basal cell and squamous cell skin cancer, about
more accurate assessment of trends in the most 61,980 cases of breast carcinoma in situ, and
recent years for which data are available. 49,710 cases of in situ melanoma are expected
to be newly diagnosed in 2006. The estimated
SELECTED FINDINGS numbers of new cancer cases by state for se-
lected cancer sites are shown in Table 2.
Expected Numbers of New Cancer Cases
Figure 1 indicates the most common cancers
Table 1 presents estimated numbers of expected to occur in men and women in 2006.
new cancer cases expected among men and Among men, cancers of the prostate, lung and
women in the United States in 2006. The bronchus, and colon and rectum account for
estimate of about 1.4 million new cases of over 56% of all newly diagnosed cancer. Pros-

114 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

FIGURE 5 Annual Age-adjusted Cancer Death Rates* Among Females for Selected Cancers, US, 1930 to 2002.
*Rates are age-adjusted to the 2000 US standard population.
Note: Due to changes in ICD coding, numerator information has changed over time. Rates for cancers of the uterus, ovary, lung and bronchus, and
colon and rectum are affected by these changes.
†Uterus includes uterine cervix and uterine corpus.
Source: US Mortality Public Use Data Tapes, 1960 to 2002, US Mortality Volumes 1930 to 1959, National Center for Health Statistics, Centers for
Disease Control and Prevention, 2005.

tate cancer alone accounts for about 33% Expected Number of New Cancer Deaths
(234,460) of incident cases in men. Based on
cases diagnosed between 1995 and 2001, an Table 1 also shows the expected number of
estimated 91% of these new cases of prostate cancer deaths in 2006 for men, women, and both
cancer are expected to be diagnosed at local or sexes combined. It is estimated that about
regional stages, for which 5-year relative sur- 564,830 Americans will die from cancer, corre-
vival approaches 100%. sponding to over 1,500 deaths per day. Cancers of
The three most commonly diagnosed the lung and bronchus, colon and rectum, and
cancers among women in 2006 will be cancers prostate in men, and cancers of the lung and
of the breast, lung and bronchus, and colon and bronchus, breast, and colon and rectum in
rectum, accounting for about 54% of estimated women continue to be the most common fatal
cancer cases in women. Breast cancer alone is cancers. These four cancers account for half of the
expected to account for 31% (212,920) of all total cancer deaths among men and women (Fig-
new cancer cases among women. ure 1). Lung cancer surpassed breast cancer as the

Volume 56 Y Number 2 Y March/April 2006 115

 Cancer Statistics, 2006

TABLE 5 Trends in Cancer Incidence and Death Rates for Selected Cancers by Sex, US, 1975 to 2002

Line Segment 1 Line Segment 2 Line Segment 3 Line Segment 4

Year APC* Year APC* Year APC* Year APC*

All sites
Incidence
Male and female 1975–1983 0.9† 1983–1992 1.8† 1992–1995 -1.7 1995–2002 0.3
Male 1975–1989 1.3† 1989–1992 5.2† 1992–1995 -4.7† 1995–2002 0.2
Female 1975–1979 -0.2 1979–1987 1.5† 1987–2002 0.3†
Death
Male and female 1975–1990 0.5† 1990–1993 -0.3 1993–2002 -1.1†
Male 1975–1979 1.0† 1979–1990 0.3† 1990–1993 -0.4 1993–2002 -1.5†
Female 1975–1992 0.5† 1992–2002 -0.8†
Lung & bronchus
Incidence
Male and female 1975–1982 2.5† 1982–1991 1.0† 1991–2002 -0.7†
Male 1975–1982 1.5† 1982–1991 -0.4 1991–2002 -1.8†
Female 1975–1982 5.5† 1982–1990 3.5† 1990–1998 1.0† 1998–2002 -0.5
Death
Male and female 1975–1980 3.0† 1980–1990 1.8† 1990–1994 -0.1 1994–2002 -0.9†
Male 1975–1978 2.4† 1978–1984 1.2† 1984–1991 0.3† 1991–2002 -1.9†
Female 1975–1982 6.0† 1982–1990 4.2† 1990–1995 1.7† 1995–2002 0.3†
Colon & rectum
Incidence
Male and female 1975–1985 0.8† 1985–1995 -1.8† 1995–1998 1.1 1998–2002 -1.8†
Male 1975–1986 1.1† 1986–1995 -2.1† 1995–1998 1.0 1998–2002 -2.5†
Female 1975–1985 0.3† 1985–1995 -1.8† 1995–1998 1.5 1998–2002 -1.5†
Death
Male and female 1975–1984 -0.5† 1984–2002 -1.8†
Male 1975–1978 0.8 1978–1984 -0.4 1984–1990 -1.3† 1990–2002 -2.0†
Female 1975–1984 -1.0† 1984–2002 -1.8†
Female breast
Incidence 1975–1980 -0.4 1980–1987 3.7† 1987–2002 0.4†
Death 1975–1990 0.4† 1990–2002 -2.3†
Prostate
Incidence 1975–1988 2.6† 1988–1992 16.5† 1992–1995 -11.2† 1995–2002 1.7†
Death 1975–1987 0.9† 1987–1991 3.1† 1991–1994 -0.6 1994–2002 -4.0†

*APC, annual percent change based on rates age-adjusted to the 2000 standard population.
†The APC is significantly different from zero.
Note: Trends were analyzed by Joinpoint Regression Program, version 3.0, with a maximum of three joinpoints (ie, four line segments).
Trends in incidence are based on rates adjusted for delay in reporting.
Source: Ries LAG, Eisner MP, Kosary CL, et al.3

leading cause of cancer death in women in 1987. 21.5 in women in Utah to 138.2 in men and
Lung cancer is expected to account for 26% of all 72.3 in women in Kentucky. In contrast, the
cancer deaths among females in 2006. Table 3 variation in female breast cancer incidence
provides the estimated number of cancer deaths rates was small, ranging from 116.6 cases per
in 2006 by state for selected cancer sites. 100,000 populations in New Mexico to
149.5 cases in Washington. Factors that con-
Regional Variations in Cancer Rates tribute to the state variations in the incidence
rates include differences in the prevalence of
Table 4 depicts cancer incidence for select risk factors, access to and utilization of early
cancers by state. Rates vary widely across states. detection services, and completeness of re-
For example, among the cancers listed in porting. For example, the state variation in
Table 4, the largest variation in the incidence lung cancer incidence rates reflects differ-
rates (in proportionate terms) occurred in ences in smoking prevalence; Utah ranks
lung cancer in which rates (cases per 100,000 lowest in adult smoking prevalence and Ken-
population) ranged from 42.3 in men and tucky highest.

116 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

TABLE 6 Fifteen Leading Causes of Death, United States, 2003

†Rates are per 100,000 population and age-adjusted to the 2000 US standard population.
Note: Percentages may not total 100 due to rounding. Symptoms, signs, and abnormalities, events of undetermined
intent, and pneumonitis due to solids and liquids were excluded from the cause of death ranking order.
Source: US Mortality Public Use Data Tape, 2003, National Center for Health Statistics, Centers for Disease Control and
Prevention, 2006.

Trends in Cancer Incidence and Mortality and women; cigarette smoking in women peaked
about 20 years later than in men. Colorectal
Figures 2 to 5 depict long-term trends in cancer incidence rates have decreased from 1998
cancer incidence and death rates for all cancers through 2002 in both males and in females. Pros-
combined and for selected cancer sites by sex. tate and female breast cancer incidence rates have
Table 5 shows incidence and mortality patterns continued to increase, although at a slower rate
for all cancer sites and for the four most com- than in previous years. The continuing increase
mon cancer sites based on joinpoint analysis. may be attributable to increased screening
Trends in incidence were adjusted for delayed through prostate-specific antigen (PSA) testing
reporting. Delay-adjusted cancer incidence for prostate cancer and mammography for breast
rates stabilized in men from 1995 to 2002 and cancer. Use of postmenopausal hormone therapy
increased in women by 0.3% per year from and increased prevalence of obesity may also be
1987 to 2002. Death rates for all cancer sites factors influencing the increase in female breast
combined decreased by 1.5% per year from cancer incidence.14
1993 to 2002 in males and by 0.8% per year in
females from 1992 to 2002. Changes in the Recorded Number of Deaths from
Mortality rates have continued to decrease Cancer from 2002 to 2003
across all four major cancer sites in men and in
women, except for female lung cancer in which A total of 556,902 cancer deaths were re-
rates continued to increase by 0.3% per year from corded in the United States in 2003, the most
1995 to 2002 (Table 5). The incidence trends are recent year for which actual dates are available.
mixed, however. Lung cancer incidence rates are About 369 fewer deaths were recorded in 2003
declining in men and have leveled off after in- than in 2002, the first decrease since national
creasing for many decades in women. The lag in mortality record keeping was instituted in
the temporal trend of lung cancer incidence rates 1930. Cancer accounted for about 23% of all
in women compared to men reflects historical deaths, ranking second only to heart disease
differences in cigarette smoking between men (Table 6). When cause of death is ranked

Volume 56 Y Number 2 Y March/April 2006 117

Cancer Statistics, 2006

TABLE 7 Ten Leading Causes of Death by Age and Sex, United States, 2003

Note: Symptoms, signs, and abnormalities, events of undetermined intent, certain perinatal conditions, and pneumonitis due to solids and liquids were
excluded from the cause of death ranking order. All ages excludes deaths with unknown age.
Source: US Mortality Public Use Data Tapes, 2003, National Center for Health Statistics, Centers for Disease Control and Prevention, 2006.

within each age group, categorized in 20-year to 79 and among men ages 60 to 79. When
age intervals, cancer is one of the five leading age-adjusted death rates are considered (Figure
causes of death in each age group among both 6), cancer is the leading cause of death among
males and females (Table 7). Cancer is the men and women under age 85. A total of
leading cause of death among women ages 40 476,844 people under age 85 died from cancer

118 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

FIGURE 6 Death Rates* From Cancer and Heart Disease for Ages Younger than 85 and 85 and Older.
*Rates are age-adjusted to the 2000 US standard population.
Source: US Mortality Public Use Data Tapes, 1960 to 2002, National Center for Health Statistics, Centers for Disease Control and Prevention, 2005.

in the US in 2003, compared with 436,258 From 2002 to 2003, the number of recorded
deaths from heart disease. cancer deaths decreased by 778 in men, but
Table 8 presents the number of deaths from increased by 409 in women (Table 9). The
all cancers combined and the five most com- largest change in the total number of deaths
mon cancer sites for males and females at var- from the major cancers was for prostate cancer
ious ages. Among males under age 40, in men (decreased by 892) and for lung cancer
leukemia is the most common cause of cancer in women (increased by 575).
death, whereas cancer of the lung and bronchus
predominates in men age 40 years and older.
Colon and rectum and prostate cancer are the CANCER OCCURRENCE BY RACE/ETHNICITY
second most common causes of cancer death
among men 40 to 79 years old and age 80 years Cancer incidence and death rates vary con-
and older, respectively. Among females, leuke- siderably among racial and ethnic groups (Ta-
mia is the leading cause of cancer death before ble 10). For all cancer sites combined, African
age 20, breast cancer ranks first at ages 20 to 59 American men have a 23% higher incidence rate
years, and lung cancer ranks first at age 60 years and 40% higher death rate than White men.
and older. African American women have a 7% lower inci-

Volume 56 Y Number 2 Y March/April 2006 119

Cancer Statistics, 2006

TABLE 8 Reported Deaths for the Five Leading Cancer Sites by Age and Sex, United States, 2003

*ONS ⫽ Other nervous system.

Note: Others and Unspecified Primary excluded from cause of death ranking order. All ages excludes deaths with unknown age.
Source: US Mortality Public Use Data Tapes, 2003, National Center for Health Statistics, Centers for Disease Control and
Prevention, 2006.

dence rate but an 18% higher death rate than ular screening (breast, cervical, and colorectal
White women for all cancer sites combined. For cancers), and timely treatment (for many
the specific cancer sites listed in Table 10, inci- cancers). The higher breast cancer incidence
dence and death rates are consistently higher in rate among Whites is thought to reflect a
African Americans than in Whites, except for combination of factors that affect diagnosis,
breast cancer (incidence) and lung cancer (mor- such as more frequent mammography in
tality) among women. Death rates from prostate, White women, and factors that affect disease
stomach, and cervical cancers among African risk, such as later age at first birth and greater
Americans are more than twice those in use of hormone replacement therapy among
Whites. Factors known to contribute to ra- White than African American women.14
cial disparities in mortality include differ- Among other racial and ethnic groups, cancer
ences in exposure (eg, Helicobacter pylori for incidence and death rates are lower for all cancer
stomach cancer), access to high-quality reg- sites combined and for the four most common

120 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

TABLE 9 Trends in the Recorded Number of Deaths for Selected Cancers by Sex, United States, 1989
to 2003

Note: Effective with the mortality data for 1999, causes of death are classified by ICD-10, replacing ICD-9 used for 1979 to 1998
data.
Source: US Mortality Public Use Data Tapes, 1989 to 2003, National Center for Health Statistics, Centers for Disease Control
and Prevention, 2006.

cancer sites than are rates in Whites and African Lifetime Probability of Developing Cancer
Americans. However, incidence and death rates
for cancers of the uterine cervix, stomach, and The lifetime probability of developing cancer
liver are generally higher in minority population is higher for men (46%) than for women (38%)
than in Whites. Stomach and liver cancer inci- (Table 11). However, because of the relatively
dence and death rates are more than twice as high early age of onset of breast cancer, women have a
slightly higher probability of developing cancer
in Asian/Pacific Islanders as in Whites, reflecting
before the age of 60 years. It is noteworthy that
increased exposure to infectious agents such as H.
these estimates are based on the average experi-
pylori and Hepatitis B virus.15
ence of the general population and may over or
Trends in cancer incidence can only be adjusted
under estimate individual risk because of differ-
for delayed reporting in Whites and African Amer- ences in exposure and/or genetic susceptibility.
icans, and not in other racial and ethnic subgroups
because the long-term incidence data required for
Cancer Survival by Race
delay adjustment are available only for Whites and
for African Americans. From 1992 to 2002, inci-
Compared with Whites, African American
dence rates for all cancer sites combined, not ad-
men and women have poorer survival once a
justed for delayed reporting, decreased by 2.7% per cancer diagnosis is made. As shown in Figure 7,
year among American Indians/Alaskan Natives, by African Americans are less likely than Whites to
1.0% per year in African Americans, by 0.6% among be diagnosed with cancer at a localized stage,
Asian/Pacific Islanders, and by 0.4% among when the disease may be more easily and suc-
Hispanic-Latinos and Whites. Similarly, the death cessfully treated, and are more likely to be
rate for all cancers combined decreased from 1992 diagnosed with cancer at a regional or distant
through 2002 by 1.7% per year in Asian/Pacific stage of disease. Five-year relative survival is
Islanders, by 1.5% among African Americans, by lower in African Americans than Whites within
0.9% among Whites, and by 0.6% among Hispanic- each stratum of stage of diagnosis for nearly
Latinos. The death rate from all cancers combined every cancer site (Figure 8). These disparities
stabilized during this time period among American may result from inequalities in access to and
Indians/Alaskan Natives.3 receipt of quality health care and/or from dif-

Volume 56 Y Number 2 Y March/April 2006 121

Cancer Statistics, 2006

TABLE 10 Age-standardized Incidence and Death Rates* for Selected Cancers by Race and Ethnicity,
US, 1998 to 2002

Asian American
American/ Indian/
All African Pacific Alaskan Hispanic-
Races White American Islander Native Latino†

Incidence Rates

All sites
Male 553.3 556.4 682.6 383.5 255.4 420.7
Female 413.5 429.3 398.5 303.6 220.5 310.9
Breast (Female) 134.4 141.1 119.4 96.6 54.8 89.9
Colon & rectum
Male 62.1 61.7 72.5 56.0 36.7 48.3
Female 46.0 45.3 56.0 39.7 32.2 32.3
Lung & bronchus
Male 77.8 76.7 113.9 59.4 42.6 44.6
Female 48.9 51.1 55.2 28.3 23.6 23.3
Prostate 173.8 169.0 272.0 101.4 50.3 141.9
Stomach
Male 12.3 10.7 17.7 21.0 15.9 17.2
Female 6.1 5.0 9.6 12.0 9.1 10.1
Liver & bile duct
Male 9.3 7.4 12.1 21.4 8.7 14.1
Female 3.6 2.9 3.7 7.9 5.2 6.1
Uterine cervix 8.9 8.7 11.1 8.9 4.9 15.8

Death Rates

All sites
Male 247.5 242.5 339.4 148.0 159.7 171.4
Female 165.5 164.5 194.3 99.4 113.8 111.0
Breast (Female) 26.4 25.9 34.7 12.7 13.8 16.7
Colon & rectum
Male 24.8 24.3 34.0 15.8 16.2 17.7
Female 17.4 16.8 24.1 10.6 11.8 11.6
Lung & bronchus
Male 76.3 75.2 101.3 39.4 47.0 38.7
Female 40.9 41.8 39.9 18.8 27.1 14.8
Prostate 30.3 27.7 68.1 12.1 18.3 23.0
Stomach
Male 6.3 5.6 12.8 11.2 7.3 9.5
Female 3.2 2.8 6.3 6.8 4.1 5.3
Liver & bile duct
Male 6.8 6.2 9.5 15.4 7.9 10.7
Female 3.0 2.7 3.8 6.5 4.3 5.1
Uterine cervix 2.8 2.5 5.3 2.7 2.6 3.5

*Rates are per 100,000 and age-adjusted to the 2000 US standard population.
†Hispanics-Latinos are not mutually exclusive from Whites, African Americans, Asian Americans/Pacific Islanders, and
American Indians/Alaskan Natives.
Source: Ries LAG, Eisner MP, Kosary CL, et al.3

122 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

TABLE 11 Probability of Developing Invasive Cancers Within Selected Age Intervals, by Sex, US, 2000
to 2002*

70 and Birth to
Birth to 39 40 to 59 60 to 69 Older Death
(%) (%) (%) (%) (%)

All sites† Male 1.43 (1 in 70) 8.57 (1 in 12) 16.46 (1 in 6) 39.61 (1 in 3) 45.67 (1 in 2)
Female 1.99 (1 in 50) 9.06 (1 in 11) 10.54 (1 in 9) 26.72 (1 in 4) 38.09 (1 in 3)
Urinary bladder Male .02 (1 in 4375) .40 (1 in 250) .93 (1 in 108) 3.35 (1 in 30) 3.58 (1 in 28)
Female .01 (1 in 9513) .12 (1 in 816) .25 (1 in 402) .96 (1 in 104) 1.14 (1 in 88)
Breast Female .48 (1 in 209) 4.11 (1 in 24) 3.82 (1 in 26) 7.13 (1 in 14) 13.22 (1 in 8)
Colon & rectum Male .07 (1 in 1399) .90 (1 in 111) 1.66 (1 in 60) 4.94 (1 in 20) 5.84 (1 in 17)
Female .06 (1 in 1567) .70 (1 in 143) 1.16 (1 in 86) 4.61 (1 in 22) 5.51 (1 in 18)
Leukemia Male .15 (1 in 650) .22 (1 in 459) .35 (1 in 284) 1.17 (1 in 85) 1.50 (1 in 67)
Female .13 (1 in 788) .14 (1 in 721) .19 (1 in 513) .78 (1 in 129) 1.07 (1 in 93)
Lung & bronchus Male .03 (1 in 3244) 1.00 (1 in 100) 2.45 (1 in 41) 6.33 (1 in 16) 7.58 (1 in 13)
Female .03 (1 in 3103) .80 (1 in 125) 1.68 (1 in 60) 4.17 (1 in 24) 5.72 (1 in 17)
Melanoma of the skin Male .13 (1 in 800) .51 (1 in 195) .51 (1 in 195) 1.25 (1 in 80) 1.94 (1 in 52)
Female .21 (1 in 470) .40 (1 in 248) .26 (1 in 381) .56 (1 in 178) 1.30 (1 in 77)
Non-Hodgkin lymphoma Male .14 (1 in 722) .47 (1 in 215) .56 (1 in 178) 1.57 (1 in 64) 2.18 (1 in 46)
Female .09 (1 in 1158) .31 (1 in 320) .42 (1 in 237) 1.29 (1 in 77) 1.82 (1 in 55)
Prostate Male .01 (1 in 10149) 2.66 (1 in 38) 7.19 (1 in 14) 14.51 (1 in 7) 17.93 (1 in 6)
Uterine cervix Female .15 (1 in 657) .28 (1 in 353) .15 (1 in 671) .22 (1 in 464) .74 (1 in 135)
Uterine corpus Female .06 (1 in 1641) .72 (1 in 139) .83 (1 in 120) 1.36 (1 in 74) 2.61 (1 in 38)

*For those free of cancer at beginning of age interval. Based on cancer cases diagnosed during 2000 to 2002. The “1 in”
statistic and the inverse of the percentage may not be equivalent due to rounding.
†All sites excludes basal and squamous cell skin cancers and in situ cancers except urinary bladder.
Source: DevCan Software, Probability of Developing or Dying of Cancer Software, Version 6.0. Statistical Research and
Applications Branch, National Cancer Institute, 2005. http://srab.cancer.gov/devcan.

ferences in comorbidities. The extent to which ferences in age at diagnosis.18,19 For the four
these factors, individually or collectively, con- major cancer sites (prostate, female breast, lung
tribute to the overall differential survival is un- and bronchus, and colon and rectum), minority
clear.16 However, recent findings suggest that populations are more likely to be diagnosed at
African Americans who receive similar cancer distant stage, compared with non-Hispanic
treatment and medical care as Whites experi- Whites.19
ence similar outcomes.17
There have been notable improvements
over time in relative five-year survival rates
CANCER IN CHILDREN
for many cancer sites and for all cancers
combined (Table 12). This is true for both
Whites and African Americans. However, Cancer is the second leading cause of death
5-year relative survival is still very poor (less among children between the ages of 1 and 14
than 25%) for many cancers, including pan- years in the United States; accidents are the most
creas, liver, esophagus, lung, and stomach. frequent cause of death in this age group (Table
Relative survival rates cannot be calculated 13). The most common cancers in children (0 to
for other racial and ethnic populations because 14 years) are leukemia (particularly acute lym-
accurate life expectancies are not available. phocytic leukemia), cancer of the brain and other
However, based on cause-specific survival rates nervous system, soft tissue sarcomas, non-
of cancer patients diagnosed from 1992 to 2000 Hodgkin Lymphoma, and renal (Wilms) tu-
in SEER areas of the United States, all minority mors.3 Over the past 25 years, there have been
populations, except Asian/Pacific Islander significant improvements in the 5-year relative
women, have a greater probability of dying survival rate for many childhood cancers (Table
from cancer within 5 years of diagnosis than 14). The 5-year relative survival rate among chil-
non-Hispanic Whites after accounting for dif- dren for all cancer sites combined improved from

Volume 56 Y Number 2 Y March/April 2006 123

Cancer Statistics, 2006

FIGURE 7 Distribution of Selected Cancers by Race and Stage at Diagnosis, US, 1995 to 2001.
*The distribution for localized stage represents localized and regional stages combined.
Note: Staging according to Surveillance, Epidemiology, and End Results (SEER) historic stage categories rather than the American
Joint Committee on Cancer (AJCC) staging system. For each type and race, stage categories do not total 100% because sufficient in-
formation is not available to assign a stage to all cancer cases.
Source: Ries LAG, Eisner MP, Kosary Cl. et al.3

124 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

FIGURE 8 Five-year Relative Survival Rates Among Patients Diagnosed with Selected Cancers, by Race and Stage at Diagnosis, US,
1995 to 2001.
*Data for distant stage melanoma of the skin for African American is not shown.
†The rate for localized stage represents localized and regional stages combined. Note: Staging according to Surveillance, Epidemiology,
and End Results (SEER) historic stage categories rather than the American Joint Committee on Cancer (AJCC) staging system.
Source: Ries LAG, Eisner MP, Kosary Cl, et al.3

Volume 56 Y Number 2 Y March/April 2006 125

 Cancer Statistics, 2006

TABLE 12 Trends in Five-year Relative Survival Rates* (%) for Selected Cancers by Race and Year of Diagnosis, US,
1974 to 2001.

Relative Five-year Survival Rate (%)

White African American All Races
1974 1983 1995 1974 1983 1995 1974 1983 1995
to to to to to to to to to
Site 1976 1985 2001 1976 1985 2001 1976 1985 2001

All sites 51 54 66† 39 40 56† 50 53 65†

Brain 22 26 33† 26 32 38† 22 27 33†
Breast (female) 75 79 90† 63 64 76† 75 78 88†
Colon 51 58 65† 46 49 55† 50 58 64†
Esophagus 5 9 16† 4 6 10† 5 8 15†
Hodgkin lymphoma 72 79 86† 69 78 80† 71 79 85†
Kidney 52 56 65† 49 55 64† 52 56 65†
Larynx 66 68 68 60 55 51 66 67 66
Leukemia 35 42 49† 31 34 38 34 41 48†
Liver & bile duct 4 6 9† 1 4 5† 4 6 9†
Lung & bronchus 13 14 16† 11 11 13† 12 14 15†
Melanoma of the skin 81 85 92† 67‡ 74§ 76‡ 80 85 92†
Multiple myeloma 24 27 32† 28 31 33 25 28 32†
Non-Hodgkin lymphoma 48 54 61† 48 45 52 47 54 60†
Oral cavity 55 56 62† 36 35 40 54 54 59†
Ovary¶ 37 40 44† 41 42 38 37 41 45†
Pancreas 3 3 4† 3 5 4† 3 3 5†
Prostate 68 76 100† 58 64 97† 67 75 100†
Rectum 49 56 65† 42 44 56† 49 55 65†
Stomach 15 16 21† 16 19 23† 15 17 23†
Testis 79 91 96† 76‡ 88‡ 88 79 91 96†
Thyroid 92 93 97† 88 91 95 92 93 97†
Urinary bladder 74 78 83† 48 60 64† 73 78 82†
Uterine cervix 70 71 75† 64 61 66 69 69 73†
Uterine corpus 89 85 86† 62 55 62 88 83 84†

*Survival rates are adjusted for normal life expectancy and are based on cases diagnosed from 1974 to 1976, 1983 to 1985, and 1995 to 2001,
and followed through 2002.
†The difference in rates between 1974 to 1976 and 1995 to 2001 is statistically significant (P⬍0.05).
‡The standard error of the survival rate is between 5 and 10 percentage points.
§The standard error of the survival rate is greater than 10 percentage points.
¶Recent changes in classification of ovarian cancer, namely excluding borderline tumors, have affected 1995 to 2001 survival rates.
Source: Ries LAG, Eisner MP, Kosary CL, et al.3

56% for patients diagnosed in 1974 to 1976 to reflecting the high prevalence of Hepatitis B
79% for those diagnosed in 1995 to 2001.3 virus in that country. The death rate for
cervical cancer in Zimbabwe (43.1 per
100,000) is about 20 times that in the United
CANCER AROUND THE WORLD States (2.3 per 100,000) and more than 25
times the rate in Australia (1.7 per 100,000).
Table 15 provides cancer death rates for 50
selected countries around the world for all sites
combined and for 9 major sites, by sex. The LIMITATIONS AND FUTURE CHALLENGES

highest lung cancer death rates are found in

Hungary for men and in Denmark for Estimates of the expected numbers of new
women. China has the highest mortality rate cancer cases and cancer deaths should be inter-
for liver cancer in both men and women, preted cautiously. These estimates may vary

126 CA A Cancer Journal for Clinicians

 CA Cancer J Clin 2006;56:106–130

TABLE 13 Fifteen Leading Causes of Death Among Children Ages 1 to 14, United States, 2003

* Rates are per 100,000 population and age-adjusted to the 2000 US standard population.
Note: Percentages may not total 100 due to rounding. Symptoms, signs, and abnormalities, events of undetermined intent,
certain perinatal conditions, and pneumonitis due to solids and liquids were excluded from ranking order.
Source: US Mortality Public Use Data Tape, 2003, National Center for Health Statistics, Centers for Disease Control and
Prevention, 2006.

TABLE 14 Trends in Five-year Relative Cancer Survival Rates* (%) for Children Under Age 15, US, 1974
to 2001

Five-year Relative Survival Rates (%)

Year of Diagnosis
1974 1977 1980 1983 1986 1989 1995
to to to to to to to
Site 1976 1979 1982 1985 1988 1991 2001

All sites 56 62 65 68 71 74 79†

Acute lymphocytic leukemia 53 68 71 69 78 80 86†
Acute myeloid leukemia 14 27‡ 25‡ 29‡ 33‡ 37 52†
Bones & joints 55‡ 53‡ 54‡ 57‡ 63‡ 62 71†
Brain & other nervous system 55 56 56 62 63 63 73†
Hodgkin disease 78 83 91 90 90 94 95†
Neuroblastoma 53 53 53 55 60 68 66†
Non-Hodgkin lymphoma 45 50 61 71 70 75 86†
Soft tissue 61 68 65 76 67 78 73†
Wilms’ tumor 74 78 86 87 91 93 92†
*Survival rates are adjusted for normal life expectancy and are based on follow-up of patients through 2002.
†The difference in rates between 1974 to 1976 and 1995 to 2001 is statistically significant (P ⬍ 0.05).
‡The standard error of the survival rate is between 5 and 10 percentage points.
Note: “All sites” excludes basal and squamous cell skin cancers and in situ carcinomas except urinary bladder.
Source: Ries LAG, Eisner MP, Kosary CL, et al.3

Volume 56 Y Number 2 Y March/April 2006 127

 TABLE 15 Cancer Around the World, 2002, Death Rates* Per 100,000 Population for 50 Countries

Cancer Statistics, 2006

128 CA
A Cancer Journal for Clinicians
 Note: Figures in parentheses are rank order within site and gender group.
*Rates are age-adjusted to the World Health Organization world standard population.
Source: Ferlay J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2002: Cancer Incidence, Mortality, and Prevalence Worldwide IARC CancerBase No. 5, version 2.0. IARC Press, Lyon, 2004.
Volume 56 Y Number 2 Y March/April 2006

CA Cancer J Clin 2006;56:106–130

129
 Cancer Statistics, 2006

considerably from year to year, particularly for tional Center for Health Statistics and cancer in-
less common cancers and in states with smaller cidence rates from SEER are the preferred data
populations. Unanticipated changes may occur sources for tracking cancer trends, even though
that are not captured by our modeling efforts. these data are 3 and 4 years old, respectively, by
The estimates of new cancer cases are based on the time that they become available. Despite their
incidence rates for the geographic locations that limitations, the American Cancer Society esti-
participate in the SEER program and, therefore, mates of the number of new cancer cases and
may not be representative of the entire United deaths in the current year provide reasonably
States. For these reasons, we discourage the use of accurate estimates of the burden of new cancer
these estimates to track year-to-year changes in cases and deaths in the United States. Such esti-
cancer occurrence and death. Age-standardized mates will assist in continuing efforts to reduce
or age-specific cancer death rates from the Na- the public health burden of cancer.

REFERENCES SEER*Stat Database: Incidence—SEER 11 mortality counts for the current calendar year. CA
Regs Public-Use, Nov 2004 Sub for Hispanics Cancer J Clin 2004;54:30–40.
1. National Center for Health Statistics, Division (1992-2002), National Cancer Institute, DC- 13. Clegg LX, Feuer EJ, Midthune DN, et al.
of Vital Statistics, Centers for Disease Con- CPS, Surveillance Research Program, Cancer Impact of reporting delay and reporting error on
trol. Available at: http://www.cdc.gov/nchs/ Statistics Branch, released April 2005, based on cancer incidence rates and trends. J Natl Cancer
nvss.htm. Accessed January 2006. the November 2004 submission. Inst 2002;94:1537.
2. Surveillance, Epidemiology, and End Re- 6. US Census Bureau. Available at: http://www. 14. Ghafoor A, Jemal A, Ward E, et al. Trends in
sults (SEER) Program (www.seer.cancer.gov) census.gov. Accessed September 2005. breast cancer by race and ethnicity. CA Cancer
SEER*Stat Database: Incidence—SEER 9 J Clin 2003;53:342–355.
7. World Health Organization. International Statis-
Regs Public-Use, Nov 2004 Sub (1973-2002),
National Cancer Institute, DCCPS, Surveil- tical Classification of Diseases, Injuries, and Causes 15. Ward E, Jemal A, Cokkinides V, et al.
lance Research Program, Cancer Statistics of Death. Vol. 1, 10th Rev. Geneva: WHO, 1992. Cancer Disparities by race/ethnicity and socio-
Branch, released April 2005, based on the No- 8. World Health Organization. International Statis- economic status. CA Cancer J Clin 54:78–93.
vember 2004 submission. tical Classification of Diseases, Injuries, and Causes 16. Ghafoor A, Jemal A, Cokkinides V, et al.
3. Ries LAG, Eisner MP, Kosary CL, et al. of Death. Vol. 1, 9th Rev. Geneva: WHO, 1975. Cancer statistics for African Americans. CA
(eds). SEER Cancer Statistics Review, 1975- 9. World Health Organization. International Statis- Cancer J Clin 2002;52:326–341.
2002. Bethesda, MD: National Cancer Insti- tical Classification of Diseases, Injuries, and Causes
tute. Available at: http://seer.cancer. 17. Bach PB, Schrag D, Brawley OW, et al.
of Death. Vol. 1, 8th Rev. Geneva: WHO, 1967. Survival of Blacks and Whites After a Cancer
gov/csr/1975_2002/.
10. Fritz A, Percy C, Jack A, et al (eds.). Inter- Diagnosis. JAMA 2002;287:2106–2112.
4. Surveillance, Epidemiology, and End Results national Classification of Diseases for Oncology,
(SEER) Program (www.seer.cancer.gov) 18. Jemal A, Clegg LX, Ward E, et al. Annual
3rd Ed. Geneva: World Health Organization, Report to the Nation on the status of cancer,
SEER*Stat Database: Incidence—SEER 12 Regs
2000. 1975-2001, with a special feature regarding sur-
Public-Use, Nov 2004 Sub for Expanded Races
(1992-2002), National Cancer Institute, DCCPS, 11. Wingo PA, Landis S, Parker S, et al. Using cancer vival. Cancer 2004;101:3–27.
Surveillance Research Program, Cancer Statistics registry and vital statistics data to estimate the number 19. Clegg LX, Li FP, Hankey BF, et al. Cancer
Branch, released April 2005, based on the Novem- of new cancer cases and deaths in the US for the survival among US whites and minorities: a
ber 2004 submission. upcoming year. J Reg Management 1998;25:43–51. SEER (Surveillance, Epidemiology, and End Re-
5. Surveillance, Epidemiology, and End Re- 12. Tiwari RC, Ghosh K, Jemal A, et al. A new sults) Program population-based study. Arch In-
sults (SEER) Program (www.seer.cancer.gov) method of predicting US and state-level cancer tern Med 2002;162:1985–1993.

130 CA A Cancer Journal for Clinicians