Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries

Introduction

Noncommunicable diseases (NCDs) are now responsible for the majority of global deaths,1and cancer is expected to rank as the leading cause of death and the single most important barrier to increasing life expectancy in every country of the world in the 21st century. According to estimates from the World Health Organization (WHO) in 2015, cancer is the first or second leading cause of death before age 70 years in 91 of 172 countries, and it ranks third or fourth in an additional 22 countries (Fig. 1).

Cancer incidence and mortality are rapidly growing worldwide. The reasons are complex but reflect both aging and growth of the population, as well as changes in the prevalence and distribution of the main risk factors for cancer, several of which are associated with socioeconomic development.2, 3 With rapid population growth and aging worldwide, the rising prominence of cancer as a leading cause of death partly reflects marked declines in mortality rates of stroke and coronary heart disease, relative to cancer, in many countries. The extent to which cancer’s position as a cause of premature death reflects national levels of social and economic development can be seen by comparing the maps in Figures 1 and 2A (the latter map depicts the 4‐tier Human Development Index [HDI]).

Cancer transitions are most striking in emerging economies, where an increasing magnitude of the disease is paralleled by a changing profile of common cancer types. A recurring observation is the ongoing displacement of infection‐related and poverty‐related cancers by those cancers that already are highly frequent in the most developed countries (eg, in Europe, North America, and high‐income countries in Asia and Oceania). These cancers are often ascribed to a so‐called westernization of lifestyle,3-5 yet the differing cancer profiles in individual countries and between regions signify that marked geographic diversity still exists, with a persistence of local risk factors in populations at quite different phases of social and economic transition. This is illustrated by the prominent differences in rates of infection‐associated cancers, including cervix, stomach, and liver, observed in countries at opposite ends of the human development spectrum.4

Against this backdrop, the current article provides a status report on the cancer burden worldwide in 2018, based on the GLOBOCAN 2018 estimates of cancer incidence and mortality produced by the International Agency for Research on Cancer (IARC).6 As in previous reports for 2002,7 2008,8 and 2012,9 the primary focus is on a description of cancer incidence and mortality at the global level and an assessment of the geographic variability observed across 20 predefined world regions. We describe the magnitude and distribution of the disease overall and for the major cancer types, commenting briefly on the associated risk factors and prospects for prevention of the major cancers observed worldwide. We conclude by stating the limitations of the exercise and the need for population‐based national and subnational cancer surveillance data to improve the accuracy of the GLOBOCAN estimates and inform on‐the‐ground initiatives in cancer control.

Data Sources and Methods

The sources and methods used in compiling the estimates in GLOBOCAN 2018 are described in detail elsewhere6 and also are available online at the Global Cancer Observatory (gco.iarc.fr). The Global Cancer Observatory website includes facilities for the tabulation and graphical visualization of the GLOBOCAN database for 185 countries and 36 cancers (as well as all cancers combined) by age and sex.

The profile of cancer, globally and by world region, is built up in GLOBOCAN using the best available sources of cancer incidence and mortality data within a given country; therefore, validity of the national estimates depends on the degree of representativeness and the quality of source information. The methods used to compile the 2018 estimates are largely based on those developed previously, with an emphasis on the use of short‐term predictions and modeling of incidence‐to‐mortality ratios, where applicable. The list of cancer sites, however, has been extended to 36 cancer types in GLOBOCAN 2018, with one of the major additions being estimates of the incidence of, and mortality from, nonmelanoma skin cancer (NMSC) (excluding basal‐cell carcinomas).6 Together with all cancers combined, cancer‐specific estimates are provided for 185 countries or territories worldwide by sex and for 18 age groups (ages birth‐4, 5‐9, …, 80‐84, and >85 years).

The numbers of new cancer cases and cancer deaths were extracted from the GLOBOCAN 2018 database for all cancers combined (International Statistical Classification of Diseases and Related Health Problems‐10th Revision codes C00‐C97) and for 36 cancer types: lip, oral cavity (C00‐C06); salivary glands (C07‐C08); oropharynx (C09‐C10); nasopharynx (C11); hypopharynx (C12‐C13); esophagus (C15); stomach (C16); colon (C18); rectum (C19‐C20); anus (C21); liver (C22, including intrahepatic bile ducts); gallbladder (C23‐C24, including extrahepatic ducts); pancreas (C25); larynx (C32); lung (C33‐C34, including trachea and bronchus); melanoma of the skin (C43); NMSC (C44, excluding basal cell carcinoma for incidence); mesothelioma (C45); Kaposi sarcoma (C46); female breast (C50); vulva (C51); vagina (C52); cervix uteri (C53); corpus uteri (C54); ovary (C56); penis (C60); prostate (C61); testis (C62); kidney (C64‐C65, including renal pelvis); bladder (C67); brain, central nervous system (C70‐C72); thyroid (C73); Hodgkin lymphoma (C81); non‐Hodgkin lymphoma (C82‐C86 and C96); multiple myeloma (C88 and C90, including immunoproliferative diseases); and leukemia (C91‐C95). For the purposes of consistency with previous exercises,10 we combine cancers of the colon, rectum, and anus as colorectal cancer (codes C18‐C21) and exclude NMSC (C44) when making global comparisons of the magnitude of different cancers; however, NMSC is included in the overall estimation of the total cancer burden, unless otherwise stated.

Incidence is the number of new cases occurring in a specified period and geographic area, conveyed either as an absolute number of cases per annum or as a rate per 100,000 persons per year. Rates are used to approximate the average risk of developing a cancer in the year 2018 and allow comparisons between countries and world regions. Primary prevention strategies aim to reduce this measure, although increasing incidence rates do not necessarily reflect failure in this domain, given that early detection (tests or programs) can result in a transient rise in incidence rates as subclinical cancer cases are discovered. This increase, however, will be maintained if some of the detected cases represent overdiagnosis (those cancers that would not otherwise have been diagnosed in an individual’s lifetime). Recent illustrations of this phenomenon include prostate cancer in the era of prostate‐specific antigen (PSA) testing and thyroid cancer after the introduction of new diagnostic techniques,11 including ultrasonography.12

Incidence data are produced by population‐based cancer registries (PBCRs). Although PBCRs may cover national populations, more often they cover subnational areas, such as selected urban areas, particularly in countries undergoing economic development. According to the most recent data compiled in Volume XI of the IARC’s Cancer Incidence in Five Continents,13 approximately 15% of the world population was covered by high‐quality cancer registries around 2010, with lower registration in South America (7.5% of the total population), Asia (6.5%), and Africa (1%). The coverage in Africa rises to 13% when we consider additional PBCR data from Sub‐Saharan African registries in the African Cancer Registry Network (https://afcrn.org/) that did not meet the criteria for inclusion in Cancer Incidence in Five Continents.13 However, such recorded data from lower resource countries are the only relatively unbiased source of information on the distribution of common cancer types in a defined population and are vital for planning local cancer prevention and control as well as developing national estimates.

Similarly, mortality is the number of deaths occurring in a specified region and period, and the mortality rate is the number of deaths per 100,000 persons per year. Mortality is the product of the incidence and the fatality rate (the proportion of patients who die); thus, the mortality rates given in this report measure the average risk of death in the population from a specific cancer in 2018. Mortality data are available in many countries through the WHO, although the degree of detail and quality of the data (both the accuracy of the recorded cause of death and the completeness of registration) vary considerably; currently, only about 1 in 5 countries can report high‐quality death registrations. Mortality rates are often used (instead of incidence) as a proxy measure of the risk of acquiring the disease across populations, but this assumes that survival is constant between the populations being compared. This still may be the case for cancers associated with poor prognoses, but it is much less likely for cancers that can be detected early and treated successfully in view of significant variations in the availability of, and access to, cancer care services among populations.

A novel feature of GLOBOCAN 2018 is the inclusion of 95% uncertainty intervals (95% UIs) for the estimated sex‐specific and site‐specific, all‐ages number of new cancer cases and cancer deaths.6 The 95% UIs take into account uncertainties linked to the extent of geographic coverage of the cancer registry and death registration (because the data used in the estimation may be subregional rather than national), the timeliness of data reporting (because the data come from years before 2018), and the quality of data (because the completeness and accuracy of the data are variable, depending on the source).

Age‐standardized rates (ASRs) per 100,000 person‐years are calculated using the direct method and the world standard population.14 The cumulative risk of developing or dying from cancer before age 75 years, assuming the absence of competing causes of death, also was calculated using the age‐specific rates and is expressed as a percentage. Both of these indicators allow comparisons between populations that are not influenced by differences in their age structures.

We present the incidence and mortality rates globally and for 20 aggregated regions, as defined by the United Nations Population Division (Fig. 2B). We also characterize the burden according to the HDI, which was created by the United Nations Development Program to highlight the importance of national policy decisions beyond economic growth in assessing development outcomes.10 As noted, the HDI can help identify cancer transitions, and we use the 4‐tier HDI (Fig. 2A) to further assess the cancer burden according to a binary proxy of development (low and medium HDI vs high and very‐high HDI).

We summarize the estimated numbers of new cases and deaths by cancer type and point to the variations in the incidence and mortality rates observed in the world regions and individual countries. We use the terms developing, transitioning, emerging, and lower HDI countries/economies as synonyms for countries classified as low or medium HDI, and we use developed, transitioned, or higher HDI countries/economies for countries classified as high or very‐high HDI.