Global Burden of 5 Major Types Of Gastrointestinal Cancer

1. Esophageal cancer

With about 572,000 new cases and 508,000 deaths in 2018, esophageal cancer is the 7^th most common cause of cancer morbidity and the 6^th most common cause of cancer-related death worldwide. Both incidence and mortality vary greatly across countries and world regions (Figure 3), and rates are 2–3 fold higher in men relative to women. Esophageal cancer is most common in Eastern Asia (ASR 12.2 per 100,000 person-years) and Eastern Africa (8.3), followed by Southern Africa (7.4); incidence rates are below 2 per 100,000 in the remaining regions of Africa as well as in Central America (Annex Table 1, Annex Figure 1). The vast majority of cases occur in Asia, with an estimated 307,000 cases (54% of the global burden) in China alone (Figure 2). Patterns of mortality closely follow those of incidence, reflecting the poor average prognosis associated with these tumors. Globally, the highest mortality rates were in the Eastern Asian and Eastern African regions, where mortality from esophageal cancer ranks among the top five causes of cancer-related death. Even larger variations in incidence rates between and within world regions are evident when examining the cancer registry data worldwide (Annex Figure 2), with the highest incidence rates of 129 per 100,000 observed in Cixian, China, and a greater than 100-fold difference in rates across registries. However, patterns have changed substantially over time and decreasing incidence trends in most Asian populations can be observed that contrast with the increasing rates seen in historically low-risk countries including the United Kingdom, the Netherlands and United States Whites (Figure 4A). These trends are most evident in males; incidence rates in females have remained at a much lower level and have changed little over time.

Open in a separate window

Figure 3.

Esophageal cancer: Estimated age-standardized incidence rates per 100,000 person-years in 2018. Source: GLOBOCAN 2018

Open in a separate window

Figure 4.

Esophageal cancer: Trends in age-standardized incidence and mortality rates by country. Source: CI5plus and WHO Mortality Database.

The geographical differences and trends over time are striking and can partly be explained by varying distributions of the two main histological subtypes of esophageal cancer, adenocarcinoma (AC) and squamous cell carcinoma (SCC), which represent about 90–95% of all esophageal cancers.¹⁸ The latter histological subtype is linked to tobacco smoking and heavy alcohol consumption, but also opium intake, air pollution and diet. SCC is the most common form of esophageal cancer globally, particularly in high-risk regions, where it represents more than 90% of all esophageal cancer cases.^18–20 Other contributors to the high incidence observed in some countries, including the so-called central Asian esophageal cancer belt²¹ and Eastern Africa’s esophageal cancer corridor²², remain to be elucidated. AC has been associated with obesity and gastroesophageal reflux disease (GERD) and today represents up to two-thirds of esophageal cancer in a relatively small number of high-income countries including the United Kingdom and the United States.^18,23,24 While SCC incidence rates are declining in most countries, rates of esophageal AC appear to be rising in an expanding number of countries. These trends are likely attributable to shifts in the prevalence of key risk factors, including increases in the prevalence of obesity paralleled by decreasing levels of both smoking and chronic infection with Helicobacter pylori (H. pylori), which has been inversely related with esophageal AC incidence.²⁵ These trends are predicted to continue in the near future, with AC surpassing SCC in many mainly high-income countries and obesity becoming an increasingly important contributor to the future burden of esophageal cancer.²⁵

With no specific symptoms of early esophageal cancers, most tumors are diagnosed at a late and rapidly progressing to an advanced stage when treatment options are limited and cure is not possible. In high-income settings, only about a quarter of all patients diagnosed with esophageal cancer are still alive five after diagnosis.¹¹ Efforts to screen for esophageal cancer in high-incidence areas of China using endoscopy showed some evidence of a significant reduction in mortality²⁶ and survival has been increasing in China over the past decades.²⁷ Yet, to-date, population-based endoscopic screening has not been recommended given the associated relatively high cost and risk of complication, and non-endoscopic cell-sampling methods have been found to lack the necessary specificity.²⁸ Clinical trials of other, less invasive screening methods, (e.g. inflatable balloons and sponges) are currently underway, and may be a more favorable approach to screen for esophageal cancer in the future, given improvements in specificity and risk stratification.^28,29. For now, primary prevention measures, including obesity control in high-income settings and continuing interventions that seek to reduce the smoking and alcohol prevalence in the community, remain key to reducing the incidence of disease.

2. Gastric cancer

Even though uniform declines in incidence have been observed in many parts of the world for decades, gastric cancer remains an important cause of cancer-related incidence and mortality, with over one million new cases and close to 800,000 deaths in 2018. It is the fifth most frequently diagnosed cancer and the third-leading cause of cancer death. Incidence rates are generally two-fold higher in men relative to women and range from 22 per 100,000 person-years in Eastern Asia to less than 5 per 100,000 in in Africa, as well as Northern America and Northern Europe (Annex Table 2, Annex Figure 3). Countries with the highest incidence rates of gastric cancer include the Republic of Korea, Mongolia and Japan, while variations in recorded incidence rates remains high globally, most markedly in Asia (Figure 5, Annex Figure 4). As gastric cancer is rarely detected early, mortality from this cancer is often high. Geographic and temporal variations in mortality are therefore often closely aligned with those of incidence, yet with a proportionally lower share of deaths from the disease occurring in very high HDI countries compared to medium and low HDI countries, where gastric cancer often ranks among the top causes of cancer-related death. Since the beginning of cancer registration several decades ago, steady declines in the incidence and mortality from gastric cancer have been observed consistently across world regions in both men and women (Figure 6).

Open in a separate window

Figure 5.

Gastric cancer: Estimated age-standardized incidence rates per 100,000 person-years in 2018. Source: GLOBOCAN 2018

Open in a separate window

Figure 6.

Gastric cancer: Trends in age-standardized incidence and mortality rates by country. Source: CI5plus and WHO Mortality Database.

The observed declines in the incidence and mortality from gastric cancer were once termed “an unplanned triumph” of cancer prevention, as this took place in the absence of active primary prevention programs and, outside of Japan and Korea, population-based screening.³⁰ Part of these decreases has been attributed to better food preservation practices, linked to refrigeration during the transportation and storage of food. At the same time, economic development has led to improved living conditions, including reduced overcrowding and better hygiene, leading to a reduced prevalence of H. pylori, an established cause of gastric cancer.³¹ Other risk factors include tobacco smoking, low consumption of fruits and vegetables, as well as high intake of salt-preserved foods and possibly alcohol.^6,32,33 Although often reported as a single entity, gastric cancers can be broadly classified into two major topographical subsites, the cardia (CGC) and non-cardia (NCGC). CGC arises in the area of the stomach adjoining the esophageal-gastric junction, while NCGC develops from more distal regions of the stomach, and the two entities differ in terms of etiology, risk factors and geographical patterns. While CGC is associated with obesity and GERD in Western populations (resembling etiological factors attributed to esophageal AC), the vast majority of NCGC cases are associated with H. pylori infection.^34–36 The contributions of NCGC and CGC to the overall burden from gastric cancer varies greatly across countries; most are NCGC in global terms, but this is particularly the case in low HDI settings. The proportion of CGCs tends to be highest in high HDI regions and small in most high-risk populations such as China or Japan, although exceptions include a high proportion of CGC in the high-incidence regions of Iran.^37–39 Risks for each subsite also differ across ethnic groups, with, for example, CGC more common in non-Hispanic Whites in the United States than in other ethnic groups.⁴⁰ Opposing the overall decreases in gastric cancer incidence and mortality rates, are the recent increases in the incidence of both CGC and cancer of the gastric corpus (a non-cardia subsite), as seen among younger generations, e.g. those aged 25–39 years in the United States.^41,42 Such trends may lead to a deceleration or a reversal of the overall declining gastric cancer rates.

A recent global study indicated that mortality rates have decreased more rapidly than incidence, ascribed to socioeconomic development and better access to diagnostic and treatment facilities.⁴³ As early-stage gastric cancer rarely causes symptoms, early detection continues to pose many challenges and five-year survival ranges between 20–40% in most countries.^10,11 Japan and the Republic of Korea, two high-risk countries, have implemented national screening programs using endoscopic and/or radiographic methods (biennial gastro-fibre scope (endoscopy) for adults aged 50 and over in addition to the existing upper GI series (barium meal followed by radiography). These efforts have led to an increasing number of cases diagnosed at an early, curable stage and relatively high survival proportions of over 60% in both countries.^44,45 As the introduction of screening also led to an increase in incidence, it remains to be determined as to how much of these survival increases are attributable to lead time and detection bias. Although routine screening in low-risk populations has not proven to be cost-effective, the implementation of population-based screening in high-risk regions has shown some promise, although whether this will translate into a true mortality reduction is yet to be confirmed.^46,47 In parallel, the cost-effectiveness of H. pylori eradication is being explored in several ongoing randomized controlled trials indicating that treatment lowers gastric cancer risk⁴⁸ and may aid in gastric cancer prevention in some settings.⁴⁹

3. Colorectal cancer

In 2018, colorectal cancer remains the most commonly diagnosed GI cancer, representing 1.8 million cases and 881 000 deaths globally, and constituting one in ten cancer cases and deaths. Incidence rates vary by up to 10-fold across world regions, with Australia/New Zealand having the highest and South-Central Asia the lowest rates (Annex Table 3, Annex Figure 5). At a country level, elevated incidence is found in a diverse set of countries, with rates in Hungary, the Republic of Korea, Slovakia and Norway above 40 per 100,000 (Figure 7). Incidence rates are about 3-fold higher in high and very high HDI when compared with low and medium HDI countries. Recorded incidence rates from cancer registries reveal high risks in two Canadian populations (Northwest Territories, Newfoundland and Labrador) and substantial variation in rates within world regions (Annex Figure 6). Variations in colorectal cancer mortality rates tend to be less pronounced due to the higher case fatality in lower HDI settings. Colorectal cancer incidence rates are still rising rapidly in many less-developed countries, particularly in Eastern Europe, Asia and South America, while trends are stabilizing or decreasing in highly developed countries, such as Canada and in Northern Europe, where rates remain among the highest in the world. (Figure 8)

Open in a separate window

Figure 7.

Colorectal cancer: Estimated age-standardized incidence rates per 100,000 person-years in 2018. Source: GLOBOCAN 2018

Open in a separate window

Figure 8.

Colorectal cancer: Trends in age-standardized incidence and mortality rates by country. Source: CI5plus and WHO Mortality Database.

Colorectal cancer is one of the clearest markers of the cancer transition, with infection-related cancers being gradually displaced with those linked to rapid societal and economic change.⁵⁰ Colorectal cancers can be broadly subdivided into colon, rectal and anal cancers; while colon and rectal cancer share a number of risk factors, a large proportion of anal cancers have been attributed to infection with human papilloma virus (HPV), most notably HPV-16.⁵¹ That combined incidence rates are steadily increasing in formerly low incidence regions can primarily be attributed to changes in lifestyle and diet, shifts towards an increased intake of fat, sugar and animal-source foods, paralleled by increases in sedentary behavior and obesity. Established risk factors for colorectal cancer include alcohol consumption⁵², high consumption of red and processed meat^53,54, obesity⁵⁵ and physical inactivity⁴. Additional risk factors include and smoking⁵⁶. In contrast, diets rich in wholegrains, dairy products and foods containing dietary fiber have been found to protect from colorectal cancer.⁴ A recent global study concluded that the three most important contributors to age-standardized disability-adjusted life years (DALYs) of colorectal cancer globally were diets low in calcium (20.5%) and alcohol use (15.2%).⁵⁷ Decreasing or stabilizing incidence rates in some high HDI and high-risk countries have been mainly attributed to population-level changes towards healthier lifestyle choices, coupled with the introduction and implementation of population-based screening. Precursor lesions of colorectal cancer can be detected and removed during screening, typically carried out using either stool-based tests (e.g. guaiac-based fecal occult blood test (gFOBT) or fecal immunochemical test (FIT)) and/or endoscopic methods such as sigmoidoscopy and colonoscopy, all of which have been associated with reductions in colorectal cancer mortality.⁵⁸ The introduction of population-based screening in a growing number of countries over recent years and decades may thus have contributed to decreasing incidence and mortality rates in some regions.⁵⁹ Yet, despite overall decreasing trends in some high-income countries, recent increases in incidence rates have been observed in younger generations, specifically in adults younger than 50 years.^60–62 While these trends point towards changes in early-life exposures, further etiologic research is needed to elucidate the causal factors driving these trends.

While the prospects of cure from colorectal cancer are generally good, marked survival differences have long existed, even across high-income countries.¹⁰ These disparities can partly be explained by stage at diagnosis, an important determinant for prognosis affected by early detection and screening programs as well as diagnostic pathways and clinical procedures. Yet, as the implementation of screening programs is yet to be fully implemented in most countries⁵⁹, the impact of colorectal cancer screening on stage distributions and mortality trends is likely to be limited to a handful of countries where activities started more than a decade ago. Treatment is also an important determinant of survival and considerable advances have been made, including improved surgical techniques, radiotherapy, chemotherapy, targeted therapy and palliative care. The adoption of best practices in treatment and management have led to declining mortality rates in most parts of the world. Colorectal cancer remains more common in older people and due to ongoing population aging and growth alone, the number of older adults with cancer is predicted to double in all world regions by 2035, with colorectal cancer representing one of the main contributors.⁶³

4. Liver cancer

Liver cancer was the sixth most commonly diagnosed cancer (841,000 cases) and the fourth leading cause of cancer death (782,000 deaths) globally. Similar to other GI cancers, both incidence and mortality rates are up to 2–3 fold higher in men compared to women. The overall burden from liver cancer is most pronounced in transitioning countries; rates are highest in Eastern Asia (17.7), Micronesia (15.2), and Northern Africa (14.1), with Mongolia the country with by far the highest incidence and mortality rates worldwide (93.7 and 75.4, respectively) (Figure 9, Annex Table 4, Annex Figure 7). Recorded data from individual cancer registries emphasize the variation in incidence rates between and within world regions, particularly in Asia (Annex Figure 8). Due to the poor prognosis associated with liver cancer, incidence and mortality patterns are typically closely aligned and show similar patterns across world regions. Yet country-level trends in incidence and mortality are quite heterogeneous; while rising incidence rates have been observed in many parts of the world over recent decades, including the United States, Australia and most parts of Europe, decreasing rates have been observed in some Asian countries such as the Republic of Korea, Japan and China (Figure 10).

Open in a separate window

Figure 9.

Liver cancer: Estimated age-standardized incidence rates per 100,000 person-years in 2018. Source: GLOBOCAN 2018

Open in a separate window

Figure 10.

Liver cancer: Trends in age-standardized incidence and mortality rates by country. Source: CI5plus and WHO Mortality Database.

Different sets of risk factors are likely responsible for these patterns and trends. Primary liver cancer, which can be broadly subdivided into hepatocellular carcinoma (HCC) (typically representing 75–85% of all liver cancer cases) and intrahepatic cholangiocarcinoma (ICC) (representing about 10–15% of cases), and other, more rare types. While ICC has been linked with primary sclerosing cholangitis, fibropolycystic liver disease, parasitic infection and intrahepatic biliary stones⁶⁴, HCC has been mainly associated with chronic infection with hepatitis B or C virus (HBV or HCV) as well as exposure to aflatoxin, excess alcohol consumption, obesity, type 2 diabetes and smoking.

In high-risk areas such as Eastern Asia, infection with HBV and aflatoxin are considered the main underlying risk factors for HCC, whereas HCV is the predominant cause in Japan and Egypt. Decreases in the prevalence of both infections and in exposure to aflatoxin have likely contributed to declines seen in the incidence of liver cancer in these regions.⁶⁵ Increasing incidence trends in low-risk areas have been postulated to be related to the rising prevalence of obesity and metabolic disorders like diabetes, offsetting the gains made through reductions in the prevalence of HBV and HCV.^66,67 Thus, while obesity is becoming more prevalent and its importance on the burden of liver cancer is growing, HBV vaccination remains key for primary prevention of HCC globally, as HBV is responsible for more cases than any other factor. Where mass vaccination has been introduced during the early-1980s, as in Taiwan, there has been a marked reduction in the prevalence of HBV infection as well as HCC incidence in childhood⁶⁸ and early adulthood⁶⁹. While HCC trends largely reflect those of primary liver cancer, a notable exception is Thailand, where ICC is the dominant type of liver cancer⁷⁰ and rates of HCC have been in decline⁷¹. Liver fluke infections are a major risk factor for ICC, particularly in the Northeast of the country where Opisthorchis viverrini is endemic.⁷²

Early detection of liver cancer remains challenging as the disease is commonly diagnosed at an advanced stage. In high-income settings, screening can be carried out for patients at high risk of developing liver cancer (e.g. with long-standing cirrhosis), using alpha-fetoprotein (AFP) blood tests and ultrasound exams every 6 months. Prognosis remains poor with the maximum 5-year survival estimated at 18% in high-income countries.¹⁰

5. Pancreatic cancer

Pancreatic cancer is only the 12th most common cancer, but is the 7th most common cause of cancer death. In 2018 the number of deaths from pancreatic cancer was almost the same as the number of cases (432,000 and 459,000 respectively). Incidence and mortality rates are about 3–4 fold higher in high compared with low and medium HDI countries and are slightly higher in men than in women. Incidence and mortality rates are highest in Europe, North America and Australia/New Zealand and differ only marginally in their magnitude due to the poor prognosis of the disease (Annex Table 5, Annex Figure 9). Hungary and Uruguay are the countries with the globally highest incidence and mortality rates, while marked differences in incidence have been recorded across populations living in the same world region (Figure 11, Annex Figure 10). Historically, both incidence and mortality rates have been either stable or slightly increasing in most recent decades (Figure 12).

Open in a separate window

Figure 11.

Pancreatic cancer: Estimated age-standardized incidence rates per 100,000 person-years in 2018. Source: GLOBOCAN 2018

Open in a separate window

Figure 12.

Pancreatic cancer: Trends in age-standardized incidence and mortality rates by country. Source: CI5plus and WHO Mortality Database.

To-date, risk factors of pancreatic cancer remain poorly understood, although evidence suggests that smoking, diabetes, body fatness, and some dietary factors including the consumption of red and processed meat as well as excessive alcohol intake are likely all associated with an increased disease risk.⁵ According to recent estimates, about 21% of all deaths from pancreatic cancer are due to smoking, followed by high fasting plasma glucose levels (9%) and excess body fatness (6%).⁷³ Variations in the prevalence of these underlying factors might therefore explain parts of the geographic differences in the burden of the disease. Trends in incidence could furthermore be affected by differences in (access to) diagnostic practices. With patients rarely exhibiting specific symptoms, the diagnosis of pancreatic cancer is reliant on imaging, even though this can be challenging for early-stage tumors.⁷⁴ Increasing incidence rates in some countries could therefore also reflect improvements in the registration of pancreatic cancer.

Even in the highest income countries worldwide most pancreatic cancer patients are diagnosed with metastatic or locally advanced disease precluding curative surgery, and only about 10% are still alive five years after diagnosis.¹¹ Prognosis improves in cases detected at an operable stage with 5-year survival rates after surgical resection ranging up to 25%.⁷⁵ Yet, in the absence of sensitive and specific tumor markers that would allow detecting the disease early, screening for pancreatic cancer remains extremely challenging. Current screening activities (in high-income settings) are limited to people with family histories and to those with inherited mutations known to increase the risk of developing the disease, often carried out using MRI or endoscopic ultrasound.⁷⁴ Despite rather constant incidence and mortality rates in many high-risk countries, the number of new cases will continue to rise in the future, largely due to population ageing and growth. Due to this development, coupled with ongoing advances and improvements in the prognosis of other cancer sites, pancreatic cancer has become or is set out to become one of the leading causes of cancer-related death in many countries.⁷³ As an example, in the European Union (EU), it has been predicted that deaths from pancreatic cancer surpassed those of breast cancer in 2017, with the disease becoming the third most important cause of cancer death in the EU, after lung and colorectal cancer.⁷⁶

The authors gratefully acknowledge all cancer registries and their staff who have contributed in sharing their data needed for this study.

Where authors are identified as personnel of the International Agency for Research on Cancer / World Health Organization, the authors alone are responsible for the views expressed in this article and they do not necessarily represent the decisions, policy or views of the International Agency for Research on Cancer / World Health Organization.