Gastric cancer is one of the major malignancies in the world. This article summarizes the current understanding of the worldwide burden of this disease, its geographic variation, and temporal trends. An overview is presented of known risk factors, including genetic, dietary, and behavioral, but focuses on Helicobacter pylori infection as the most important factor in noncardia gastric cancer. When the data and the literature allow, we distinguish between cardia and noncardia sub-sites, as it is now clear that these two anatomic locations present distinct and sometimes opposite epidemiological characteristics.
Key points
- •
There is a 10-fold international variation in the occurrence of gastric cancer, with rates in men double those of women.
- •
Globally, rates are declining by approximately 2.5% per annum but this is compensated by growth of and aging in the population.
- •
Helicobacter pylori infection is the major risk factor for noncardia gastric cancer and is now regarded by some as a necessary cause of such tumors.
- •
The best route to control noncardia gastric cancer would be through eradication of H pylori , although further research is required to evaluate such a strategy.
- •
Cardia gastric cancer is not associated with H pylori but is associated with cigarette smoking, increased body mass index, and with high socioeconomic status.
Introduction
Gastric cancer has long been one of the world’s major cancers and remains one of the major causes of malignant disease morbidity and mortality. This article provides an overview of current understanding regarding the epidemiology of gastric cancer. A global picture is presented of incidence and mortality patterns and this is then followed by a consideration of the major identified risk factors. It is now apparent that the most important of these risk factors is infection with the bacteria Helicobacter pylori and this is, therefore, discussed in some detail with other factors. It is now clear that there is heterogeneity in the epidemiology of gastric cancer according to its location within the stomach with a distinction between cancers arising in the proximal cardia region and those arising more distally (noncardia). As a consequence, and where the data and literature allow, the epidemiology is reviewed in relation to each of these 2 groupings.
Introduction
Gastric cancer has long been one of the world’s major cancers and remains one of the major causes of malignant disease morbidity and mortality. This article provides an overview of current understanding regarding the epidemiology of gastric cancer. A global picture is presented of incidence and mortality patterns and this is then followed by a consideration of the major identified risk factors. It is now apparent that the most important of these risk factors is infection with the bacteria Helicobacter pylori and this is, therefore, discussed in some detail with other factors. It is now clear that there is heterogeneity in the epidemiology of gastric cancer according to its location within the stomach with a distinction between cancers arising in the proximal cardia region and those arising more distally (noncardia). As a consequence, and where the data and literature allow, the epidemiology is reviewed in relation to each of these 2 groupings.
Descriptive epidemiology
Incidence and Mortality: Worldwide Estimates
GLOBOCAN 2008 provides the most recent figures available for the worldwide cancer burden. Almost 1 million new cases of gastric cancer (988,000 cases; 7.8% of all cancers) were estimated to have occurred globally in 2008, making it the fourth most common malignancy in the world, after cancers of the lung (1.68 million cases; 12.7% of all cancers), breast (1.31; 10.9%), and colorectum (1.24; 9.8%). More than 73% (728,000 cases) of gastric cancer cases occur in Asia, and almost half the world’s total (47%) gastric cancer occurs in China ( Fig. 1 ). Europe contributes nearly 15% of the global burden (146,000 cases), whereas Central and South America contribute a further 7% (65,000 cases) (see Fig. 1 ).
A global map of male age-standardized incidence rates using GLOBOCAN 2008 data ( Fig. 2 ) shows that the highest rates occur in Eastern and South Eastern Asia, Eastern Europe, and parts of Central and South America. High rates are also evident in some countries in Africa and Southern Europe. The equivalent global map of female incidence rates (data not shown) is almost identical except that, in any given country, female rates are approximately half those seen in male rates.
Gastric cancer is the second leading cause of cancer death worldwide (738,000 deaths, 9.7% of all cancer deaths) after lung cancer (1.38 million deaths; 18.2% of all cancer deaths). Because prognosis following a diagnosis of gastric cancer is usually poor, the pattern of mortality is very similar to that for incidence and the proportional breakdowns of the 2 indicators by continent do not seem substantially different (see Fig. 1 ). For this reason, global maps of incidence and mortality are also very similar (data not shown). The above estimates represent the figures for cases of adenocarcinoma, the most common histologic type of gastric cancer. The authors recently estimated a worldwide total of lymphomas of gastric origin in 2008 to be 18,000 (ie, less than 2% of the number of adenocarcinomas). Other gastric malignant histologies would be less common than this.
The summarized GLOBOCAN 2008 results present the best available estimates of the worldwide burden of gastric cancer. Because many assumptions are made in deriving these estimates and, in some cases, the observed data on which they are based are of suboptimal quality, it is important to consider also more reliable incidence information available from high-quality population-based cancer registries. Such information is provided in Fig. 3 , which shows recent age-standardized incidence rates, by sex, from almost 50 cancer registries around the world, all of which have met the quality standards for publication in the definitive publication: Cancer Incidence in Five Continents. The geographic distribution of these cancer registries reflects population coverage by high-quality cancer registration and there is a clear and notable deficit of such information from countries in Africa, Asia, and Central and South America.
Fig. 3 shows several important features of the descriptive epidemiology of gastric cancer. There is an approximate 10-fold variation in incidence rates ranging from, in men, around 60 per 100,000 in Japan and Korea to less than 6 per 100,000 in US whites, Jordan, and Saudi Arabia. Some part of the extremely high rates in Japan and Korea may be due to intensive radiographic or endoscopic surveillance programs and the detection of very small lesions that may rarely be diagnosed in other populations but, even excluding these countries, one can still observe more than a fivefold variation in male incidence with rates over 30 per 100,000 in Belarus, the Russian Federation, and Costa Rica. Restricting attention to Europe, there is considerable variation between the highest rates, generally in Eastern European countries, and the lowest, in Scandinavian countries, Switzerland, and France. There is also wide variation within Asia and other interesting heterogeneities between, for example, countries within South America and between white and black US populations. Fig. 3 also shows the worldwide consistency in the approximate 2:1 male/female ratio in incidence rates. The rank order of countries within each continent is more or less the same for both sexes. Although etiologic explanations can be proposed for the geographic distribution between populations, the sex ratio difference in risk is more complex to understand.
Incidence and Mortality: Trends Over Time
Cancer registry data from specific populations can also be used to analyze trends over time. Fig. 4 provides such data from a group of registries selected to be representative of different world regions and different underlying incidence rates. Time trends in corresponding national rates of mortality are shown alongside those for incidence. As with the previous figures, the trends for women in each population (see Fig. 4 B) are substantively the same as those for men (see Fig. 4 A) but with reduced rates. The time span considered is usually from around 1980 to around 2010, depending on the availability of data. Over this 30-year period, and after making allowance for some sporadic year-to-year fluctuations, the tendency is toward a marked decline in age-standardized incidence and mortality apparent in nearly all populations. This marked decline in age-standardized incidence and mortality is irrespective of whether the population is at the high (eg, Japanese men) or low (eg, white US women) end of the risk spectrum. The one exception would seem to be incidence rates among the Indian population, which have remained relatively constant at around 15 per 100,000 in men and 6 per 100,000 in women. For those populations with long-running series, the extent of decline seems slightly greater in magnitude for mortality than for incidence, indicating that the former may be a result of a combined effect of improved survival with reduction in underlying risk. However, the extent of and consistency in risk reduction between apparently heterogeneous populations are quite remarkable and perhaps indicates a reduction in exposure to a globally ubiquitous risk factor such as H pylori . The downward trajectory in gastric cancer incidence worldwide is responsible for the substantive change in its ranking against other major cancers since the first global estimates were produced for 1975. At this time, gastric cancer represented the most common neoplasm worldwide.
Analysis of these and other cancer registry data over time has provided an average Estimated Annual Percentage Change in gastric cancer incidence of −2.5% per annum. One can apply this to GLOBOCAN 2008 estimates of the likely burden of gastric cancer in 2030, taking account of known changes in population demographics (ie, the growth in the population and the changes in its age structure). On this basis, the current global annual burden of gastric cancer, 988,000 new cases in 2008, is estimated to increase by a small extent to 991,000 new cases by 2030. Thus, the substantial decline in gastric cancer incidence rate is, more or less, balanced by the demographic changes in the world’s population and, unless the Estimated Annual Percentage Change can be decreased still further, the absolute burden of this neoplasm will remain static.
Distribution by Subtype: Cardia Versus Noncardia
There is increasing interest in the distribution of gastric cancer by subsite of the stomach, in particular, the distinction between cancers originating in the most proximal cardia region and those originating more distally (noncardia). This interest is driven in part by evidence suggesting that these 2 categories may have different etiologies. There have also been reports indicating that gastric cardia cancers have been increasing in incidence in recent decades, thus providing a contrasting trend from that observed for gastric cancer overall. It has recently been estimated that, for 2008, approximately 88% of gastric cancers globally would likely originate from regions other than the cardia, thus implying a current annual burden of 118,000 cardia and 880,000 noncardia cancers. Fig. 5 shows cardia cancer incidence rates, derived from Cancer Incidence in Five Continents data, for the same populations as shown in Fig. 4 . Some populations, such as those from Japan and China, which have high rates of gastric cancer overall, also have relatively high rates of cardia cancer. These high rates of cardia cancer were, however, also the case for certain populations, such as from Denmark and among USA whites that have low rates of gastric cancer overall. As a proportion of the total gastric cancer incidence, the contribution from the cardia is, therefore, much higher in such low-risk populations.
Because gastric cardia cancers are not readily distinguished from lower esophageal adenocarcinomas, there has undoubtedly been some variation in pathologic practice in the organ to which these cancers have been assigned for classification purposes. Such variation may be present both between countries and over different time periods, thereby complicating interpretation of geographic and secular trends. Comparisons within the same populations over the same time period may, therefore, be more informative. Thus within the SEER cancer registries in the United States, white men had higher rates of cardia cancer than black men in contrast to the inverse pattern for gastric cancer overall. Comparisons by sex (see Fig. 5 ) also show that male cardia cancer rates were 3 to 4 times greater than those in women, indicating a somewhat larger excess than the 2:1 ratio for gastric cancer overall. In looking at these rates of cardia cancer, it should be noted that most cancer registry data sets include many gastric cancer notifications where the subsite is unspecified. Thus, for the registries shown in Fig. 5 , more than 20% of gastric cancers (and often more than 50%) will not be specified regarding a cardia/noncardia location.
Migrant Studies
Migration studies show that first-generation migrants coming from countries with high incidence to countries of low incidence share the risk of their country of origin but that the incidence rate decreases in the following generation. This decrease is particularly important for women. The global picture suggests that environmental factors acting early in life have a crucial role in gastric carcinogenesis.
Risk factors for gastric carcinoma
Helicobacter pylori
H pylori is a spiral Gram-negative bacterium that colonizes the stomach. Although most infections are asymptomatic, H pylori is associated with chronic gastritis, peptic ulcer disease, gastric B-cell mucosa-associated lymphoid tissue lymphoma, and gastric adenocarcinoma. It is thought that H pylori was once ubiquitous in the human population, but in many populations its prevalence is declining in successive birth cohorts and it is now rare among children in Western Europe, North America, Oceania, and Japan. The risk of H pylori infection is associated with low socioeconomic status, particularly overcrowding and poor sanitation, so the gradual disappearance of H pylori in these populations may be a byproduct of economic development. The widespread use of antibiotics may also have played a role. It is noteworthy that the reduction in H pylori prevalence generally matches the decline in gastric cancer incidence and mortality.
In 1994, an expert working group convened by the International Agency for Research on Cancer (IARC) classified infection with H pylori as carcinogenic to humans, based on its association with gastric adenocarcinoma and mucosa-associated lymphoid tissue lymphoma. This conclusion was confirmed in 2009 by a second IARC working group, with the added precision that H pylori causes noncardia gastric carcinoma, because the risk of H pylori infection seems to be limited to the distal part of the stomach. The authors recently estimated that 75% of noncardia gastric cancers worldwide are attributable to H pylori . However, it now seems that the strength of the association between gastric cancer and H pylori may have been underestimated, because of inaccurate assessment of H pylori infection status. Indeed, it has been hypothesized that H pylori is a necessary cause of gastric cancer.
Almost all of the epidemiologic evidence on the relationship between H pylori and gastric cancer comes from serologic assessment of anti- H pylori immunoglobulin G (IgG) antibodies. It is now widely accepted that retrospective serologic assessment of H pylori infection in gastric cancer cases has poor sensitivity, so that case-control studies systematically underestimate the strength of the association. This problem is caused by atrophic gastritis, a precancerous lesion, which leads to a reduction in the burden of H pylori infection and a subsequent reduction in IgG antibody titers to the extent that an H pylori infection may become serologically undetectable. For this reason, most of the evidence base for the carcinogenicity of H pylori comes from prospective studies.
The most comprehensive relative risk estimates for H pylori and gastric cancer come from a pooled analysis of 12 prospective studies, which included 762 cases of noncardia gastric cancer and 2250 controls. The pooled odds ratio (OR) was 2.97 (95% CI 2.34–3.77) for H pylori infection. The same study included 274 cases of cardia gastric cancer and 827 controls with an OR of 0.99 (95% CI 0.40–1.77) for H pylori infection. When the pooled analysis was restricted to cases occurring at least 10 years after the blood draw used for H pylori diagnosis, the OR increased to 5.93 (95% CI 3.41–10.3) for noncardia cancer but reduced to 0.46 (95% CI 0.23–0.90) for cardia cancer. This subgroup analysis underscores both the difference between cardia and noncardia cancer and the need to account for the effect of gastric carcinogenesis on H pylori measurement, even in prospective studies. Further follow-up of the individual studies contributing to this pooled analysis is reviewed in the IARC Monographs, volume 100, part B, and did not change the substantive conclusions.
The cag pathogenicity island
H pylori is genomically highly diverse and this diversity may contribute to the clinical outcome of the infection. Several genetic factors associated with H pylori colonization ( babA, sabA, alphAB, hopZ ) and virulence ( cagA , vacA ) have been identified. The genetic marker that has attracted the most attention in epidemiologic studies is the presence of the cag pathogenicity island, a DNA sequence of 40 kbp that is present in 70% of H pylori strains in Europe and North America, but is ubiquitous in Asia and most of Africa. The CagA protein encoded by the cagA gene within the cag pathogenicity island is highly immunogenic, which allows the serologic detection of an infection with a cag -positive H pylori strain by the presence of anti-CagA antibodies. CagA-positive strains are associated with higher risk of gastric cancer than Cag-negative strains. A meta-analysis of 16 cohort and case-control studies including 778 cases of noncardia gastric cancer and 1409 matched controls found an elevated risk of CagA-positive H pylori infections, with an OR of 2.01 (95% CI 1.21–3.32) for CagA-positivity among all H pylori -infected individuals.
The cag pathogenicity island is also associated with precancerous gastric lesions. Plummer and colleagues analyzed a cross-sectional endoscopic survey of 2145 individuals from Venezuela, in which both the presence of H pylori DNA and the presence of the cagA gene were determined by polymer chain reaction on gastric biopsies. Infection with cagA -positive H pylori strains but not cagA -negative strains was associated with severity of precancerous lesions. Using individuals with normal gastric mucosa or superficial gastritis as controls, the OR for dysplasia was 15.5 (95% CI 6.4–37.2) for cagA -positive H pylori compared with 0.90 (95% CI 0.37–2.17) for cagA -negative H pylori . Gonzalez and colleagues analyzed a follow-up study of 312 individuals from Spain with an average of 12.8 years of follow-up between 2 endoscopies, also using polymer chain reaction detection and genotyping of H pylori . The relative risk for progression of precancerous lesions was 2.28 (95% CI 1.13–4.58) for cagA -positive strains compared with cagA -negative strains.
Enzyme-linked immunobsorbent assay versus immunoblot
The standard laboratory method for assessing H pylori antibodies in epidemiologic studies is enzyme-linked immunobsorbent assay (ELISA). The few studies that have compared ELISA with immunoblot have consistently found stronger OR for noncardia gastric cancer with immunoblot assays. These studies are summarized in Table 1 .
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
