Cancer statistics, 2025

Data sources

Population‐based cancer incidence data in the United States have been collected by the National Cancer Institute's (NCI's) Surveillance, Epidemiology, and End Results (SEER) program since 1973 and by the Centers for Disease Control and Prevention's (CDC's) National Program of Cancer Registries (NPCR) since 1995. The SEER program is the only source for historic population‐based incidence data from the eight oldest SEER areas (Connecticut, Hawaii, Iowa, New Mexico, Utah, and the metropolitan areas of Atlanta, San Francisco‐Oakland, and Seattle‐Puget Sound), representing approximately 8% of the US population.7 Historical survival data (1975–1977 and 1995–1997) are based on the SEER 8 areas plus the Detroit metropolitan area, as published previously.8 Contemporary survival statistics are based on data from the SEER 8 registries plus the Alaska Native Tumor Registry and the registries in California, Georgia, Idaho, Kentucky, Louisiana, New Jersey, New York, and Texas,9 representing 42% of the US population, with vital status follow‐up through 2021. All 22 SEER registries (with the addition of Massachusetts and Illinois), covering 48% of the United States, were the source for the probability of developing cancer.

The North American Association of Central Cancer Registries (NAACCR) compiles and reports incidence data from 1995 forward for registries that participate in the SEER program and/or the NPCR. These data approach 100% coverage of the US population for the most recent years and were the source for the projected new cancer cases in 2025, contemporary incidence trends, cross‐sectional incidence rates, and stage distribution.10^, 11 The incidence data presented herein differ slightly from those published online in the NAACCR's Cancer in North America (CiNA) Explorer (apps.naaccr.org/explorer/, September 1, 2024) because rates are adjusted for delays in case reporting using national delay factors. Incidence data for Puerto Rico were abstracted from the CiNA Explorer.

Mortality data from 1930 to 2022 were provided by the National Center for Health Statistics (NCHS).12^, 13 Forty‐seven states and the District of Columbia met data quality requirements for reporting to the national vital statistics system in 1930, and Texas, Alaska, and Hawaii began reporting in 1933, 1959, and 1960, respectively. The methods for abstraction and age‐adjustment of mortality data before 1969 are described elsewhere.13^, 14 Contemporary 5‐year mortality rates for Puerto Rico were obtained from State Cancer Profiles (statecancerprofiles.cancer.gov, August 1, 2024), a collaboration between the CDC and the NCI.

All cancer cases were classified according to the International Classification of Diseases for Oncology, third edition, except childhood and adolescent cancers, which were classified according to the International Classification of Childhood Cancer.15^, 16^, 17 Causes of death were classified according to the International Classification of Diseases.18

Statistical analysis

Expected number of new cancer cases and deaths

Table 1 presents the estimated numbers of new invasive cancer cases in the United States in 2025 by sex and cancer type. In total, there will be approximately 2,041,910 new cancer diagnoses, the equivalent of about 5600 cases each day. In addition, there will be about 59,080 new cases of ductal carcinoma in situ in women and 107,240 new cases of melanoma in situ of the skin in 2025. The estimated numbers of new cases for selected cancers by state are shown in Table 2. The lifetime probability of being diagnosed with invasive cancer is only slightly higher for men (39.9%) than for women (39.0%; Table 3).

Open multimedia modal

Open multimedia modal

Open multimedia modal

Although cancer risk increases exponentially with advancing age, the proportion of new diagnoses in adults aged 65 years and older decreased from 61% in 1995 to 59% in 2021 despite the growth of this age group in the general population from 13% to 17%. In contrast, the proportion of adults aged 50–64 years increased in both the cancer patient population, from 25% to 29%, and the general population, from 13% to 19%. The shift toward more middle‐aged patients partly reflects steep declines in incidence among older men for smoking‐related cancers and prostate cancer from the early 1990s through the late 2000s. Also contributing more recently is increased cancer risk in people born since the 1950s because of changing patterns in known exposures (e.g., higher obesity rates), as well as other factors yet to be elucidated.26 Although people younger than 50 years were the only age group to experience an increase in cancer incidence from 1995 through 2021, the steep drop in population size (from 74% to 64%) drove a decrease in the proportion of cancer diagnoses in this age group from 15% to 12%.

Figure 1 depicts the most common cancers diagnosed in men and women in 2025. Prostate cancer, lung and bronchus (hereinafter lung) cancer, and CRC account for almost one half (48%) of all incident cases in men, with prostate cancer alone accounting for 30% of diagnoses. For women, breast cancer, lung cancer, and CRC account for 51% of all new diagnoses, with breast cancer alone accounting for 32% of cases.

Open multimedia modal

An estimated 618,120 people in the United States will die from cancer in 2025, corresponding to approximately 1700 deaths per day (Table 1). Despite steep declines in cigarette smoking prevalence, from 42% in 1965 to 12% in 2022,27 smoking continues to be the leading cause of preventable death in the United States, accounting for almost 500 cancer deaths each day in 2025, most from lung cancer.28 Approximately 85% (106,150) of lung cancer deaths in 2025 (124,730) will be caused by cigarette smoking directly, with an additional 3500 caused by second‐hand smoke28 and 15,100 caused by a combination of other combustible tobacco products (e.g., cigar or pipe smoking),29^, 30 radon,31 occupational exposures,32 air pollution, and other environmental exposures.33 There is growing concern that e‐cigarettes and vaping may contribute to this burden in the future given their carcinogenic potential and wide popularity.34 Lung cancer causes nearly 2.5 times more deaths than second‐ranking CRC and third‐ranking pancreatic cancer. Table 4 provides the estimated number of deaths for these and other common cancers by state.

Open multimedia modal

Trends in cancer incidence

Figure 2 illustrates long‐term trends in overall cancer incidence rates from 1975 through 2021 by sex. Observed rates in 2020 are shown separate from the trend line because of the large deficit in cancer diagnoses that resulted from pandemic‐related health care closures. The 2020 rate for cancer overall was 9% lower than the 2019 rate, with the largest reductions for asymptomatic (in situ and localized stage) disease because of both less screening and less incidental detection through provider visits and imaging.19^, 35 Analyses of 2021 incidence rates indicate a continuation of prepandemic trends without an obvious recovery of deficient cases from 2020 except for breast cancer.1^, 36

Open multimedia modal

Overall cancer incidence in men spiked during the early 1990s because of a surge in the detection of asymptomatic prostate cancer as a result of rapid, widespread uptake of prostate‐specific antigen (PSA) testing.37 The steep decline from 2007 through 2013 was similarly driven by prostate cancer, although rates leveled off thereafter and have remained stable through 2021 (Table 5). In contrast, cancer incidence in women has been much less erratic and gradually rose from a nadir of 361.2 per 100,000 in 1978 to 443.2 in 2021, an overall increase of 23%. Generally declining trends in men alongside increasing incidence in women has narrowed the male‐to‐female rate ratio from a peak of 1.6 in 1992 to 1.1 in 2021. (Even preceding the prostate/PSA peak in men, the rate ratio was 1.4.) Higher risk in men for most cancer types reflects greater exposure to carcinogenic environmental and lifestyle factors, such as smoking, as well as other factors, including height,38^, 39 endogenous hormone exposure, and immune function and response.40^, 41

Open multimedia modal

However, sex differences in cancer risk vary markedly by age because of variation in both cancer distribution and age‐specific temporal trends. For example, incidence rates in middle‐aged women (50–64 years) have already surpassed those of men and were statistically equivalent in 2021 (832.5 vs. 830.6 per 100,000, respectively) after being 21% lower than in men in 2007 (780.1 vs. 992.6 per 100,000; Figure 3). Among people younger than 50 years, for whom the burden in women is already greater largely because of breast cancer, the chasm is widening. The incidence rate in women younger than 50 years was 82% higher than in their male counterparts in 2021 (141.1 vs. 77.4 per 100,000), up from 51% higher in 2002. During this period, incidence declined slightly in young men, but rose in young women by nearly 20%, largely because of increasing trends for breast and thyroid cancers, which make up almost one half (46%) of all cancers in this age group. Although rates in men younger than 50 years are increasing for the four leading cancers (colorectal, testicular, kidney, and leukemia), these trends were offset by declines for other common cancers (e.g., melanoma, non‐Hodgkin lymphoma, and prostate). Notably, among those younger than 65 years, lung cancer incidence was higher in women than in men in 2021 for the first time (15.7 versus 15.4 per 100,000; p = .03). These patterns harken back to the pretobacco epidemic era, when the cancer burden was higher in women than in men,13 and may be a bellwether for the future cancer landscape.

Open multimedia modal

Figure 4 shows incidence rates by sex for all ages combined for the most common cancers. The incidence rate for prostate cancer dropped by almost 40% from 2007 to 2014 because of reductions in PSA testing and the diagnosis of localized tumors that coincided with recommendations against screening by the US Preventive Services Task Force (USPSTF) for men aged 75 years and older in 2008 and for all men (temporarily) in 2012.42^, 43 Since 2014, however, the prostate cancer incidence rate has risen by 3% per year, ranging from an increase of 2.4% per year for localized disease to more than 4.5% per year for advanced (regional or distant‐stage) disease diagnoses that began as early as 2011.10 A recent study estimated that more than one half of men in the United States living with metastatic prostate cancer were initially diagnosed with localized or regional‐stage disease.44

Open multimedia modal

Untethered PSA screening of the 1990s and 2000s was plagued by diagnosis and treatment of clinically insignificant prostate cancer while simultaneously credited with steep reductions in prostate cancer mortality.45^, 46 Fortunately, refined early detection strategies, genetic testing for disease aggressiveness, and more conservative treatment have emerged rapidly. A Swedish population‐based trial reported that the diagnosis of clinically insignificant tumors was reduced by more than one half when biopsy was limited to men with positive magnetic resonance imaging (MRI) results.47 Similarly, a recent meta‐analysis found that MRI‐integrated screening reduced the odds of biopsy by 72% and the diagnosis of clinically insignificant cancers by 66%.48 The inclusion of MRI and genomic testing is associated with increased active surveillance management, although uptake of these strategies is low and geographically uneven.49^, 50 In the United States, approximately 60% of men with low‐risk disease are managed with active surveillance51^, 52 versus 9% with intermediate‐risk disease.53 PSA testing increased slightly after the USPSTF upgraded their recommendation to informed decision making in men aged 55–69 years in 2018,54^, 55^, 56 but it remains underused at only 37% in men aged 50 years and older overall and 34% among Black men,57 who tend to have more aggressive disease and thus greater benefit.46^, 58^, 59^, 60 A recent review supports screening Black men from ages 45–75 years at potentially more frequent intervals than other men depending on baseline PSA,61 consistent with long‐standing American Cancer Society recommendations.62

Female breast cancer incidence rates have been slowly increasing since the mid‐2000s (Table 5), largely driven by diagnoses of localized‐stage and hormone receptor‐positive disease.63 In the past decade (2012–2021), the rate increased by 1% per year overall, with the steepest slope in women younger than 50 years (1.4% per year) and Hispanic (1.6% per year) or Asian American and Pacific Islander (AAPI, 2.6% per year) women. Although rising incidence is largely attributed to a decreasing fertility rate and increasing obesity prevalence,64 excess body weight is associated with a decreased risk of breast cancer in premenopausal women.65 Physical inactivity and alcohol consumption account for 7% and 16% of breast cancers, respectively,28 and may also contribute. Binge and heavy drinking is increasing in women younger than 50 years,66 especially among those with high education and income67 and without children.68 Many of these same risk factors, including increased adiposity and sedentary lifestyle and less and/or later childbearing, also likely contribute to the increase in uterine corpus cancer incidence of about 1% annually also since the mid‐2000s.

After decades of increase, thyroid cancer incidence rates have declined since 2014 by 2% per year (Table 5) because of changes in clinical practice designed to mitigate overdiagnosis, including recommendations against thyroid cancer screening by the USPSTF and for more restrictive criteria by professional societies for performing and interpreting biopsies.69^, 70 Data from autopsy studies indicate that the occurrence of clinically relevant tumors has remained stable since 1970 and is generally similar in males and females,71^, 72 despite three‐fold higher incidence rates in females. Even among adolescents (aged 15–19 years), incidence rates are five times higher in girls than in boys (5 vs. 1 per 100,000), increased steeply through 2018, but show signs of leveling in more recent years.

Lung cancer incidence declined over the past decade (2012–2021) by 3% per year in men and 1.4% per year in women (Table 5). The downturn began later and remains slower in women than in men because women took up cigarette smoking in large numbers later and were also slower to quit, including upticks in smoking prevalence in some generations born after 1965.73^, 74 In contrast, CRC incidence patterns have long been similar by sex, with rates declining since 2011/2012 by 1.2% per year in both men and women. However, these declines are increasingly driven by adults aged 65 years and older because CRC is rising in people born after the 1950s, some of whom are now middle‐aged. From 2012 through 2021, incidence increased by 0.4% per year in those aged 50–64 years and by 2.4% per year in younger adults. Rising incidence in the United States and several other high‐income countries since the mid‐1990s75 remains unexplained but likely reflects changes in lifestyle exposures that began with generations born during the last one half of the 20th century.76 In response, the American Cancer Society recommended beginning CRC screening at age 45 years instead of 50 years in 2018,77 and the USPSTF followed suit in a draft statement in 2020 that was finalized in 2021.78 Although only 20% of people aged 45–49 years reported being up to date for CRC screening in 2021,79 we may already be seeing signs of prevalent disease detection in this newly eligible cohort. The CRC incidence rate jumped 17% in people aged 45–49 years from 2018 to 2021 (vs. 1.2% in the previous 3 years). Claims data on 10 million private health insurance beneficiaries aged 45–49 years showed a three‐fold increase in CRC screening from January 2021 to December 2022, when it reached the same level as that in people aged 50–75 years.80

After a long history of increase, non‐Hodgkin lymphoma incidence decreased by about 1% per year in both men and women from 2017 through 2021, and liver cancer and melanoma have stabilized in men, although both cancers continue to increase in women by about 2% per year (Table 5). In adults younger than 50 years, melanoma has stabilized in women, but liver cancer continues to increase by about 2% per year; whereas rates for both cancers have decreased in men by about 1% and 2.5% per year, respectively. The decline in urinary bladder cancer since the mid‐2000s accelerated from <1% per year to 1.4% per year during 2015 through 2021 overall. Incidence rates appear to have finally stabilized for kidney cancer after increasing since the 1970s, but they continue to increase by about 1% per year for cancers of the pancreas and oral cavity and pharynx. Increasing trends for oral cavity are mostly confined to cancers of the tongue, tonsil, and oropharynx (increasing by 1.9% per year), which are usually associated with human papillomavirus (HPV).

Cervical cancer incidence has decreased by more than one half since the mid‐1970s because of the widespread uptake of vaccination, screening, and treatment of precursor lesions. Although incidence overall has stabilized in the past decade, trends vary widely by age. For example, rates in women aged 30–44 years have increased by 11%, from 12.7 in 2013 to 14.1 in 2021 after decades of decline, consistent with persistent reports of underscreening among young women in recent studies.81^, 82 In contrast, rates in women aged 20–24 years who were first to be exposed to the HPV vaccine83^, 84 have plummeted by 69% during this time. A recent study based on data from the New Mexico HPV Pap Registry found that the incidence of cervical intraepithelial neoplasia grades 2 and 3 had dropped from 153 and 41 cases, respectively, in 2007 to zero in 2019 among adolescents 15‐19 years.85 Vaccine efficacy against other HPV‐related cancers is also emerging, including a 70% reduction in high‐grade anal squamous intraepithelial lesions and cancer among women who were vaccinated before age 17 years.86 Surprisingly large herd immunity87 and single‐dose efficacy88^, 89 spurred the UK Health Security Agency to move to single‐dose vaccination in 2023.90 In the United States, where the annual number of HPV‐associated cancers is estimated at greater than 37,000,91 77% of adolescents had received at least one vaccine dose in 2023, and 61% were up to date (2 doses if initiated before 15th birthday, 3 doses otherwise), down from 63% in 2022.92

Cancer survival

The 5‐year relative survival rate for all cancers combined has increased from 49% for diagnoses during the mid‐1970s to 69% during 2014–2020 (Table 6),8^, 9 with the highest contemporary survival for cancers of the thyroid (98%), prostate (97%), testis (95%), and melanoma (94%). Earlier diagnosis through screening and incidental detection through imaging has contributed to gains in survival for some cancers (e.g., breast, prostate, thyroid, and kidney) both by increasing the likelihood of successful treatment and through lead‐time bias and the detection of indolent cancers (e.g., overdiagnosis).62^, 93^, 94 Liver cancer has the largest increase in survival (in relative terms), from 3% to 22%, but remains among the least favorable prognosis, along with cancers of the lung (27%), esophagus (22%), and pancreas (13%).

Open multimedia modal

Gains in survival have been especially rapid for hematopoietic and lymphoid malignancies because of improvements in treatment protocols, including the development of targeted therapies and immunotherapies, likely in part because of a history of high research funding.95 The 5‐year relative survival rate for chronic myeloid leukemia has more than tripled, from 22% in the mid‐1970s to 70% for those diagnosed during 2014–2020, with tyrosine‐kinase inhibitors providing most patients with near‐normal life expectancy.96 Although three generations of tyrosine‐kinase inhibitors have been approved, drug resistance and risk of progression to acute disease occurs in 5% to 10% of patients with chronic myeloid leukemia and is an active area of research.97 Metabolic intervention is an emerging therapeutic strategy for resistant chronic myeloid leukemia98 as well as acute myeloid leukemia,99 for which 5‐year relative survival remains just 32%.

A cascade of new regimens of immunotherapy and targeted therapy has also revolutionized the management of cancer, particularly melanoma, over the past 2 decades, especially in the treatment of metastatic disease.100 Consequently, 5‐year relative survival for distant‐stage melanoma has doubled, from 18% for patients diagnosed in 2009 to 35% in 2014–2020 (Figure 5). Immunotherapy has also shown efficacy for nonsmall and small cell lung cancer.101 Advances in treatment for lung cancer have been almost exclusively confined to nonsmall cell disease, for which 3‐year relative survival has increased from 26% in 2004 to 43% in 2018, compared with an increase from 9% to 12% for small cell lung cancer.9 Progress not only reflects improved disease management102^, 103^, 104 but also advances in staging105 and earlier detection.106^, 107

Open multimedia modal

Open multimedia modal

The only cancer for which survival has decreased over the past 4 decades is uterine corpus cancer (Table 6), which is the fourth most common cancer in women and the fifth most common cause of cancer death. Uterine corpus cancer is one of only a handful of cancers with increasing mortality (Table 7) and also has among the largest Black–White mortality disparity (Table 8), with a 5‐year relative survival rate of just 63% among Black women versus 84% among White women (Figure 5). Aside from modest advances in treatment108 and slightly improved survival for women with regional stage disease,109 progress has lagged behind other common cancers, at least in part reflecting persistent underfunding.95 The NCI allocated almost six times more research dollars in 2019 for cervical cancer ($86 million) than for uterine corpus cancer ($14.9 million), which killed three times more women in 2022 (4051 vs. 12,763, respectively).

Open multimedia modal

Open multimedia modal

Cervical cancer survival rates have also stagnated, although this partly reflects an increased proportion of adenocarcinoma, which has less favorable survival, as a result of high cytology screening, which disproportionately detects and removes squamous cell carcinoma.110^, 111 Cervical cancer has the highest screening prevalence of any screen‐detectable cancer at 75% among women aged 25–65 years in 2021.60

Trends in cancer mortality

Mortality rates are a better indicator of progress against cancer than incidence or survival because they are less affected by detection biases, such as those that can occur for screen‐detected cancers.112 Cancer mortality rose during most of the 20th century (Figure 7), largely because of a rapid increase in lung cancer among men as a consequence of the tobacco epidemic. However, reductions in smoking as well as improvements in disease management and the uptake of screening have resulted in a 34% overall drop in the cancer death rate from 1991 through 2022, translating to nearly 4.5 million fewer cancer deaths (3,021,200 in men and 1,437,500 in women) than if mortality had remained at its peak (Figure 8). The number of averted deaths is twice as large for men than for women because the death rate in men peaked higher and declined faster (Figure 7).

Open multimedia modal

Open multimedia modal

Despite decades of decline, lung cancer continues to dwarf other cancers in the number of deaths, causing more deaths in 2022 than colorectal, breast, and prostate cancers combined (Figure 7). The lung cancer death rate has dropped by 61% from the peak in 1990 among men and by 38% from the peak in 2002 among women, consistent with the later downturn in incidence. Since 2013/2014, the pace of decline accelerated from 3% to almost 5% per year in men and from 2% to 4% per year in women (Table 7) because of earlier detection and treatment advances that have extended survival.107 A recent study found that mortality declines accelerated for both smoking‐related and smoking‐unrelated lung cancer but are slower for the latter because they are not influenced by reductions in incidence.113 Low‐dose computed tomography screening reduces the odds of lung cancer mortality by 16%–24% among high‐risk individuals114^, 115 and is associated with a substantial survival advantage.116 The potential for reducing the mortality burden is further increased by updated American Cancer Society guidelines that expanded screening eligibility to an additional 5 million people by eliminating cessation time among former smokers because of protracted excess risk.117 Nevertheless, lung cancer screening prevalence remains low, ranging from one of 10 eligible adults in most Western states to three of 10 in the Northeast,118 and most cases (43%) are diagnosed at a distant stage (Figure 6).

Sustained reductions in mortality for CRC, the second‐most common cause of cancer death in men and women combined, are the result of changing patterns in risk factors, such as declines in smoking, screening uptake, and improved treatment. The CRC death rate has dropped by 55% among males since 1980 and by 60% among females since 1969. (The rate in women began declining before 1969, but those data are not exclusive of cancer in the small intestine.) Contemporary trends in CRC are remarkably similar by sex, with rates decreasing during the most recent decade (2013–2022) by 1.7% per year in both men and women (Table 7).

In contrast, death rates for pancreatic cancer, the third leading cause of cancer death, have gradually increased since cancer mortality reporting began, from about five per 100,000 in both men and women in the 1930s to 13 per 100,000 men and 10 per 100,000 women today (Figure 7). This trend reflects improved classification on death certificates as well as increased incidence since the mid‐1990s, in part because of the obesity epidemic. In contrast to most cancers, however, therapeutic advances are also lacking despite substantial effort that includes national legislation to focus attention on pancreatic cancer (House Resolution 733: Recalcitrant Cancer Research Act of 2012). Since 2000, there have been 481 phase 1 clinical trials and 85 phase 3 trials conducted on metastatic pancreatic cancer, along with five new drug approvals; yet median survival for the disease remains less than 1 year.119 The appearance of a three‐fold increase in the 5‐year relative survival rate, from 4% in the mid‐1990s to 13% in 2014–2020, is partly an artifact of a burgeoning number of incidentally detected, well differentiated neuroendocrine tumors.120^, 121 For the nine in 10 patients diagnosed with pancreatic adenocarcinoma, 5‐year survival remains just 8%.

Female breast cancer mortality peaked in 1989 and has since decreased by 44% through 2022, translating to the avoidance of more than 517,900 deaths. A recent study attributed three fourths of this progress to treatment advances and the remainder to earlier diagnosis through screening.122 Declines in breast cancer mortality have slowed from 2% per year during the 2000s to 1% per year since 2010 (Table 7), reflecting relatively stable mammography prevalence over the past 2 decades and perhaps increased incidence. Reductions in prostate cancer mortality have similarly decelerated from 3.6% per year during the late 1990s and 2000s to 0.5% per year from 2012 through 2022 (Table 7, Figure 7). The prostate cancer death rate in 2022 was 52% lower than the peak in 1993 because of earlier detection through PSA screening and advances in treatment.45^, 123 Steep declines in melanoma mortality during 2013–2017 (6.1% per year) because of breakthrough treatments noted earlier have slowed to 1.4% during 2017–2022 (Table 7). Declining mortality trends for melanoma, leukemia, and cancers of the kidney, prostate, and breast, despite increasing or stable (leukemia) incidence, underscore the impact of advances in treatment, but perhaps also signal some overdiagnosis.

The decades‐long increase in liver cancer mortality has finally reversed in men, as rates decreased by 1.2% per year from 2018–2022, but continued in women with an increase of 0.7% per year during this time period. Mortality rates also continue to rise for uterine corpus cancer by 1.5% per year from 2013 through 2022 and for HPV‐associated oral cancers (tongue, tonsil, and oropharynx) by 2% per year (Table 7).

Recorded number of deaths in 2022

In 2022, a total of 3,279,857 deaths were recorded in the United States: 184,374 fewer than in 2021 (Table 9). The large drop is because the number of COVID‐19 deaths fell by more than one half, from 12% of total deaths in 2021 (416,893) to 6% in 2022 (186,552). This large shift altered the proportion of other deaths as well. For example, the percentage of deaths from cancer increased from 17% in 2021 to 19% in 2022; however, in the absence of COVID‐19 deaths, cancer accounted for 20% of deaths in both years. The age‐adjusted cancer death rate decreased by 1.5% from 2021 to 2022 despite 3158 more cancer deaths in 2022 because of the aging and growth of the population.

Open multimedia modal

Although cancer remained the second leading cause of death after heart diseases in 2022 overall, it is the leading cause of death among men aged 60–79 years and women aged 40–79 years (Table 10). Notably, suicide was the second leading cause of death in young adult men (aged 20–39 years). Table 11 presents the number of deaths in 2022 for the five leading cancer types by age and sex. Brain and other nervous system (hereafter brain) tumors are the leading cause of cancer death among children and adolescents younger than 20 years (480 deaths), followed closely by leukemia (450 deaths). Brain is also the leading cause of cancer death in men aged 20–39 years, causing twice as many deaths as in women in this age group (624 vs. 305). Breast cancer is the leading cause of cancer death among women younger than 50 years, and lung cancer leads among both men and women aged 50 years and older.

Open multimedia modal

Open multimedia modal

Cancer disparities by race and ethnicity

This section highlights the most striking cancer disparities among the five broadly defined racial and ethnic groups shown in Table 8. Although race and ethnicity are social constructs that aggregate heterogeneous population groups, they are useful for examining the influence of discrimination and inequality in health disparities. Cancer incidence and mortality are highest among AIAN people overall and in women, whereas Black men have the highest sex‐specific rates (Table 8). The incidence rate in Black men during 2017–2021 was about 80% higher than in AAPI men (535 vs. 298 per 100,000), who have the lowest rate of any sex‐race group. AAPI people are the only group for which the incidence in men is lower than in women.

AIAN people have the highest cancer mortality of any group for cancers of the colorectum, kidney, liver, lung, stomach, and cervix. Black men have the highest prostate cancer mortality, from two‐fold to four‐fold that in any other group, whereas Black women have the highest breast and uterine corpus cancer mortality, with the latter from two‐fold to three‐fold that of all other women. Although Hispanic and AAPI people have relatively low risk for most common cancers, their rates of liver cancer mortality are approximately 40%–50% higher than in White people, and their stomach cancer mortality is two‐fold higher. Aggregated data for broad racial and ethnic groups mask substantial heterogeneity. For example, combined AAPI data mask even larger disparities among Pacific Islander people, who have three‐fold higher stomach cancer mortality and from two and one half–fold to three‐fold higher rates of uterine corpus and cervical cancer mortality compared with White women.124

Disparities in cancer occurrence and outcomes are often the result of socioeconomic deprivation as a consequence of structural racism, which limits opportunities for education and other mechanisms of upward mobility.125 For example, the historical practice of mortgage lending discrimination known as redlining is associated with later stage cancer diagnosis, lower likelihood of receiving recommended treatment, and higher cancer mortality.126^, 127^, 128^, 129^, 130 Inequalities in wealth lead to differences in the prevalence of risk factor exposures as well as access to high‐quality cancer prevention, early detection, and treatment.131^, 132 Even beyond higher prevalence of common risk factors like smoking and obesity,60 exposure to carcinogenic air emissions is up to 50% higher among people experiencing poverty, regardless of race or ethnicity.133 In 2022, 25% of AIAN people lived below the federal poverty level ($27,750 for a family of four), as well as 17% of Black and Hispanic people, compared with 9% of White and Asian people.134 Persistent poverty ranks among the leading causes of death alongside smoking135 and is consistently associated with higher cancer incidence, later stage diagnosis, and worse outcomes.136^, 137^, 138 The effects of poverty are both compounded by and independent of racial status. A recent study found that living in a disadvantaged neighborhood increased the likelihood of aggressive prostate cancer by 30% among Black men but had no impact on risk in White men, suggesting the contribution of race‐specific factors, such as minority stress.139

Cancer survival after diagnosis is lower among Black people than among White people for almost every cancer type shown in Figure 5, even after controlling for stage at diagnosis (Figure 6) and socioeconomic status.138 A higher likelihood of more aggressive disease (e.g., triple‐negative breast cancer, nonendometrioid uterine corpus cancer) explains a small portion of this gap, but the largest contributor is less access to high‐quality care across the cancer continuum from prevention to diagnosis and treatment. Black women are one half as likely as White women to receive guideline‐concordant diagnostic procedures for the workup of uterine corpus cancer,140 less likely to receive a provider referral for mammography,141 and less likely to receive timely follow‐up after an abnormal mammogram.142 These inequalities directly translate to later stage disease, as only 56% of uterine corpus cancers and 58% of breast cancers are diagnosed at a localized stage in Black women versus 71% and 68%, respectively, in White women (Figure 6). A recent study estimated that CRC mortality rates in Black people would be reduced by 19%, eliminating two thirds of the Black–White disparity, merely by ensuring the same quality screening as White people.143 Asian, Black, and Hispanic people are less likely to receive both a physician recommendation for cancer screening144 and recommended germline genetic testing necessary for receipt of game‐changing targeted therapies.145 The economic burden of racial and ethnic health inequalities was estimated at $421–$451 billion in 2018146 and will surely rise in the wake of the COVID‐19 pandemic because of a slower rebound in screening and other routine care among people of color.6

Geographic variation in cancer occurrence

Tables 12 and 13 show the 5‐year average annual incidence and mortality rates for selected cancers by state. Geographic variation reflects population demographic characteristics and differences in the prevalence of cancer risk factors, early detection practices, and access to care, including state and local public health policies. For example, states have a large influence on the health of their residents by controlling accessibility and affordability of health insurance through the Marketplace and Medicaid.147^, 148 Cancer death rates range from <150 per 100,000 people in Utah, Hawaii, and New York to >210 per 100,000 people in West Virginia, Kentucky, and Mississippi (Table 13). This gap is largely driven by lung cancer and mirrors patterns in smoking prevalence, which have remained consistent geographically despite temporal declines. In 2022, the highest smoking prevalence was in West Virginia (22%); Arkansas (20%); Tennessee (19%); and Mississippi, Missouri, Ohio, and Kentucky (18%) compared with 7% in Utah and 10% in California.27

Open multimedia modal

Open multimedia modal

Although it is mostly preventable, cervical cancer incidence varies two‐fold by state, ranging from five per 100,000 women in Massachusetts, New Hampshire, and Connecticut to ≥10 per 100,000 women in Texas, Kentucky, West Virginia, Oklahoma, and Puerto Rico (Table 12). Advances in cancer control create or exacerbate disparities because of the unequal dissemination of interventions. Although HPV vaccination can virtually eliminate cervical cancer149 and prevent against numerous other cancers, large state differences in coverage will likely widen existing disparities. In 2022, up‐to‐date HPV vaccination among boys and girls aged 13 to 17 years per 100,000 people ranged from 38% in Mississippi and 46% in Oklahoma to 78% in the District of Columbia and 85% in Rhode Island.27

Cancer in children and adolescents

Cancer is the second most common cause of death among children aged 1–14 years (after accidents) and the fourth most common cause of death among adolescents (aged 15–19 years). In 2025, an estimated 9550 children (aged birth to 14 years) and 5140 adolescents (aged 15–19 years) will be diagnosed with cancer, and 1050 and 600, respectively, will die from the disease. An estimated one in 264 children and adolescents will be diagnosed with cancer before age 20 years.

Leukemia is the most common childhood cancer, accounting for 28% of cases, followed closely by central nervous system tumors (27%), one third of which are benign or borderline malignant (Table 14). Cancer types and their distribution differ in adolescents, among whom the most common cancer is central nervous system tumors (22%), more than one half of which are benign or borderline malignant, followed by lymphoma (19%) and leukemia (13%). In addition, Hodgkin lymphoma is much more common than non‐Hodgkin lymphoma among adolescents, whereas the reverse is true among children. Thyroid carcinoma accounts for 12% of cancers in adolescents but only 2% in children.

Open multimedia modal

After increasing since at least 1975, the overall invasive cancer incidence rate in children declined slightly from 2015 through 2021 by 0.8% per year driven by a recent rapid decline in malignant brain tumors (from 37.3 per million in 2017 to 31.9 per million in 2021) and stabilized rates of lymphoid leukemia.11 In contrast, overall incidence continued a slow increase in adolescents (by 0.7% per year) because of climbing rates for both lymphoid leukemia and non‐Hodgkin lymphoma. Malignant brain tumors decreased rapidly, consistent with the pattern in children, but represent only 9% of all malignancies versus 20% in children. The steep (4% per year) rise in thyroid cancer incidence rates since at least 1998 may be stabilizing in recent years.

In contrast, cancer mortality has declined steadily since 1970, from 6.3 to 1.9 per 100,000 in 2020–2022 in children and from 7.2 to 2.7 in adolescents, for overall reductions of 70% and 63%, respectively. Much of this progress reflects the dramatic declines in mortality for leukemia of 83% in children and 73% in adolescents (based on the average rate during 2018–2022 compared to 1970). Remission rates of 90%–100% have been achieved for childhood acute lymphocytic leukemia over the past 4 decades, primarily through the optimization of established chemotherapeutic regimens as opposed to the development of new therapies.150 However, progress among adolescents has lagged behind that in children, partly because of differences in tumor biology, clinical trial enrollment, treatment protocols, and tolerance and adherence to treatment.151

The 5‐year relative survival rate for all cancers combined improved from 58% for diagnoses during the mid‐1970s to 85% during 2014 through 2020 in children and from 68% to 87% in adolescents, but it varies substantially by cancer type and age at diagnosis. Survival is higher in children than in adolescents for lymphoid leukemia (92% vs. 76%), Ewing sarcoma (81% vs. 68%), hepatic tumors (79% vs. 53%), and rhabdomyosarcoma (66% vs. 53%) but is higher in adolescents (although not statistically significantly) for neuroblastoma (82% vs. 86%) and malignant brain tumors (75% vs. 78%; Table 14). Survival rates are lowest in both children and adolescents for hepatic tumors, osteosarcoma, and rhabdomyosarcoma. Long‐term survivors of childhood cancer are at substantially elevated risk of treatment‐related adverse health effects as they age. For example, one longitudinal study found that 18% of childhood cancer survivors had experienced a major cardiovascular event by the age of 50 years compared with 0.9% of community controls.152 Thus survivorship care plans are particularly critical for young survivors to help facilitate informed prevention and early detection interventions in addition to surveillance for subsequent cancers.

Limitations

The estimated numbers of new cancer cases and deaths in 2025 provide a reasonably accurate portrayal of the contemporary cancer burden, but these model‐based 3‐year to 4‐year projections should not be used to track temporal trends for several reasons. First, the methodology changes, most recently in 2021,23^, 24 to take advantage of improved modeling techniques and cancer surveillance coverage. Second, although the models are robust, they can only account for trends through the most recent data year (currently, 2021 for incidence and 2022 for mortality) and cannot accommodate abrupt fluctuations caused by changes in detection practice, such as those that occurred for prostate cancer because of changes in PSA testing. Third, the model can be oversensitive to sudden or steep changes in observed data. The most informative metrics for tracking cancer trends are age‐standardized or age‐specific cancer incidence rates from SEER, NPCR, and/or NAACCR and cancer death rates from the NCHS.

Errors in reporting race and ethnicity in medical records and on death certificates result in underestimated cancer incidence and mortality in persons who are not White, particularly Native American populations. Racial misclassification for cancer incidence is reduced by confining cases to those diagnosed in Purchased/Referred Care Delivery Area counties, which are associated with the Indian Health Service, but thus excludes one third of the Native American population who reside in urban areas. In addition, mortality data for the Native American population are adjusted for racial misclassification using factors for all cancers combined and may overestimate or underestimate rates for individual cancer types.20 Cancer data for other non‐White groups in the United States are mostly limited to the broadly defined racial and ethnic categories developed by the Office of Management and Budget and mask important differences in the cancer burden within these heterogenous populations. For example, breast cancer mortality is approximately 40% lower in AAPI women than in White women overall (Table 8), but it is 30% higher in Native Hawaiian and other Pacific Islander women, who are classified within this category.124 Finally, the lack of sexual orientation and gender identity data collection precludes an analysis of cancer occurrence in the LGBTQ+ (lesbian, gay, bisexual, transgender, and queer or questioning) population, which would undoubtedly inform targeted cancer control efforts given the high prevalence of smoking in this group.153

ACKNOWLEDGMENTS

The authors gratefully acknowledge all cancer registries and their staff for their hard work and diligence in collecting cancer information, without which this research could not have been accomplished.