Epidemiology of ESCC

Abstract

Esophageal cancer is the eighth most common cancer worldwide and the sixth most common cause of death from cancer. More than 80 % of esophageal cancer cases and deaths occur in developing countries, and approximately 90 % are squamous cell carcinomas in the high-incidence regions. The incidence rates of esophageal cancer show wide variation internationally. It has been shown to be two to four times more common among men than women in general; however in Japan it is approximately seven times more common among males. Both incidence and mortality are on the rise since 1960 due to the aging Japanese population, while age-adjusted rates are consistently decreasing with the exception of the increasing male incidence rate. Established risk factors for esophageal squamous carcinoma include tobacco smoking and alcohol consumption, while fruit and vegetable intake show high probability in preventing esophageal cancer. Likewise, intake of high-temperature beverages and foods show high probability of increasing risk through heat damage in the esophagus. Approximately 88 % of male esophageal cancer (52 % for females) in Japan is thought to have been avoidable by lifestyle improvement such as refraining from smoking of tobacco and alcohol use, while maintaining sufficient fruit and vegetable intake.

1.1 Esophageal Cancer in the World and Japan

1.1.1 Esophageal Cancer in the World: Burden, Geographical Difference, and Trends

1.1.1.1 Global Burden and Geographical Difference [1]

Esophageal cancer is the eighth most common cancer worldwide, with 481,000 new cases (3.8 % of the total) estimated in 2008, and the sixth most common cause of death from cancer with 406,000 deaths (5.4 % of the total). These figures encompass both adenocarcinoma and squamous cell carcinoma types. More than 80 % of esophageal cancer cases and deaths occur in developing countries.

The incidence rates of esophageal cancer vary internationally more than 15-fold in men (age-standardized incidence rate to the world population (ASR) 22.3 per 100,000 in Southern Africa compared to 1.4 in Western Africa) and almost 20-fold in women (ASR 11.7 per 100,000 in Southern Africa compared to 0.6 in Micronesia/Polynesia) (Fig. 1.1). The incidence rate in China is one of the highest (22.9 in men and 10.5 in women), while it is also relatively high in Japan (10.6 in men and 1.5 in women).

Fig. 1.1

Age-standardized incidence rate (World) of esophageal cancer in the world (2008)

Esophageal cancer is two to four times more common among men than women in general; however it is approximately seven times more common among men in Japan and 17 times more common among men in Korea. These differences in sex ratio may suggest different etiologies by region. In Japan and Korea, tobacco smoking and alcohol drinking are assumed to be major causes of esophageal cancer and the predominant incidence rate among males is associated with much higher prevalence of smoking of tobacco and alcohol use among men versus women. In China and Southern Africa, an important risk factor, in addition to tobacco smoking and alcohol drinking, is thought to be nutrient deficiency such as vitamins and micronutrients, which occurs equally in both men and women. However, the apparent reason of geographic variations is unspecified.

1.1.1.2 Histological Type [2]

In those high-incidence regions that provide information on histological type, approximately 90 % are squamous cell carcinomas (Fig. 1.2). This is in contrast to some lower-risk populations, such as Caucasian Americans and Europeans, where adenocarcinomas are predominant. For example, in the USA, SEER (non-Hispanic White) indicated ASR 5.3 in men where 67 % of cases are coded as adenocarcinoma as opposed to 25 % squamous cell carcinoma. In contrast, Japan, Nagasaki indicated ASR 9.1 in men where only 2 % of cases are coded as adenocarcinoma as opposed to 93 % squamous cell carcinoma.

Fig. 1.2

Esophageal cancer—histological distribution % (1998–2002)

1.1.2 Esophageal Cancer in Japan

In 2011, 10,141 men and 1,829 women died from esophageal cancer, representing 4.8 and 1.3 % of the total cancer death in men and women, respectively. Mortality rates increased with age rapidly after 40 years (Fig. 1.3). The mortality rate of esophageal cancer is estimated to be 0.67 % in men and 0.09 % in women up to 75 years, increasing to 1.17 % in men and 0.20 % in women over lifetime. Regarding incidence, 17,004 men and 2.990 women were estimated to be diagnosed with esophageal cancer in 2007 and the probability of esophageal cancer diagnosis was 1.35 % in men and 0.19 % in women up to 75 years, increasing to 2.16 % in men and 0.43 % in women for lifetime. Five-year survival rates were 32.3 % in men and 41.3 % in women who were diagnosed with esophageal cancer in 2003–2005 based on the population-based cancer registry.

Fig. 1.3

Age-specific mortality rate of esophageal cancer in Japan (2011)

Both incidence and mortality rates are observed to have increased in number since 1960 due to the aging of Japanese population, while age-standardized rates are observed to have been consistently decreasing with the exception of the increasing male incidence rate (Fig. 1.4). Histological distribution trends were analyzed using eight population-based cancer registries with high level of reliability from 1993 to 2001 (Table 1.1) [3]. Squamous cell carcinoma remains the predominant type of esophageal cancer in Japan, and a remarkable increase in adenocarcinoma has not been observed before 2001. Disparity in the classification of esophageal and gastric cardia adenocarcinoma may have led to underestimation of esophageal adenocarcinoma incidence.

Fig. 1.4

Time trends in the incidence and mortality of esophageal cancer in Japan

Table 1.1 Trends in incidence of esophageal cancer by histological subtype in Japan^a

An increased trend of adenocarcinoma of the esophagogastric junction was observed among patients who had underwent surgery for advanced gastric adenocarcinoma in the National Cancer Center Hospital in Tokyo, from 2.3 % in 1962–1965 to 10.0 % in 2001–2005; however the proportion of Siewert type I (defined as adenocarcinoma of the distal esophagus) had remained very rare (approximately 1 % among adenocarcinoma of the esophagogastric junction) [4]. Since this finding was confined to operative cases with advanced gastric adenocarcinoma, the proportion of Siewert's type I tumors may have been underestimated.

1.2 Risk Factors

Established risk and protective factors for esophageal cancer are listed according to the level of certainty (Table 1.2). Tobacco smoking and alcohol consumption are convincing risk factors for esophageal cancer, especially squamous cell carcinoma. Mate, a traditional herbal beverage consumed in parts of South America, has been identified as a probable cause of esophageal cancer. Non-starchy vegetables, fruits, and foods containing beta-carotene and/or vitamin C probably prevent esophageal cancer.

Table 1.2 Established risk and protective factors for esophageal cancer

1.2.1 Tobacco Smoking and Alcohol Consumption

The main risk factors for esophageal squamous cell carcinoma (ESCC) are tobacco smoking and alcohol consumption, which in individual studies have been found to account for 75–90 % of cases [6]. The risk of esophageal cancer increases rapidly with the amount of both tobacco smoking and alcohol consumption, with no evidence of any threshold effect for either.

In Japan, four cohort studies and eleven case–control studies tested the association between tobacco smoking and esophageal cancer risk [7]. With the exception of three case–control studies, all cohort studies and eight case–control studies showed strong positive associations and dose–response relationships. Meta-analysis of 12 studies indicated that the summary estimate for current and former smokers relative to lifetime nonsmokers was 3.73 (95 % CI, 2.16–6.43) and 2.21 (95 % CI, 1.60–3.06), respectively. Similarly four cohort studies and nine case–control studies tested the association between alcohol consumption and esophageal cancer [8]. With the exception of three case–control studies, all cohort studies and case–control studies showed strong positive associations and dose–response relationship. Meta-analysis of 12 studies indicated that the summary estimate for ever drinkers relative to never drinkers was 3.30 (95 % CI, 2.30–4.74) and 3.36 (95 % CI, 1.66–6.78) across the four studies adjusted for smoking.

We examined the effect of tobacco smoking and alcohol consumption on ESCC in a large-scale population-based cohort study [9] (Fig. 1.5). A total of 44,970 middle-aged and older Japanese men were followed up for up to 14 years, and a total of 215 cases of ESCC were newly diagnosed among participants during this time. Regular alcohol consumers of 150–299 and >300 g ethanol per week had a 2.59- (95 % CI, 1.57–4.29) and 4.64-fold (95 % CI, 2.88–7.48) higher risk of ESCC than nondrinkers, respectively (p for trend = 0.001). Past smokers as well as current smokers had a higher risk than never smokers. Among current smokers, pack-year and cigarettes per day were also associated with the incidence of ESCC, with risk increasing in a dose-dependent manner (p for trend = 0.001). With regard to the interaction of tobacco smoking (pack-years, <40 vs. >40) and alcohol consumption (ethanol g/weeks, <300 vs. >300), no statistically significant results were identified (p for interaction = 0.70).

Fig. 1.5

Smoking of tobacco, alcohol consumption, and subsequent risk of esophageal squamous cell carcinoma in men—JPHC study

1.2.2 Genetic Susceptibility to Tobacco Smoking and Alcohol Drinking

Regarding genetic susceptibility, esophageal cancer does not exhibit any strong familial aggregation, and genetic studies of esophageal cancer have instead focused on genes such as cytochrome P-450 (CYP), glutathione S-transferase (GST), alcohol dehydrogenase (ADH), and acetaldehyde dehydrogenase (ALDH), which metabolize suspected tobacco- and alcohol-derived carcinogens. No consistent findings have emerged for tobacco-derived pathways, although the majority of studies have been limited in sample size.

Conversely, strongly significant effect modifications have been observed with ADH1B and ALDH2 genotype. Among those with ADH1B who have the His allele, approximately 95 % of Japanese and 10–20 % of Caucasians show a rapid increase of blood acetaldehyde due to the high alcohol metabolizing activity of the ADH1B enzyme, compared with those who have the Arg allele. Among those with ALDH2 Lys allele, approximately 50 % of Japanese and <10 % of Caucasians show a higher concentration of blood acetaldehyde after alcohol consumption compared to those who have the ALDH Glu allele, due to the low catalytic activity of ALDH2 enzyme.

A meta-analysis of 19 case–control studies was conducted to evaluate the effect of alcohol consumption modification by ADH1B and ALDH2 polymorphism to the risk of esophageal cancer [10]. The majority of the studies focused on ESCC and was conducted in Asian populations. A meta-analysis of 13 case–control studies on ADH1B showed that ADH1B*1/*1 (Arg/Arg) increased the risk of esophageal cancer among never/rare [odds ratio = 1.56 (95 % CI, 0.93–2.61)], moderate [2.71 (95 % CI, 1.37–5.35)], and heavy alcohol consumers [3.22 (95 % CI, 2.27–4.57)], compared with ADH1B*2/*2 (His/His). Similarly a meta-analysis of 18 case–control studies on ALDH1 showed that ALDH2*1/*2 (Glu/Lys) increased the risk among never/rare [1.28 (95 % CI, 0.91–1.80)], moderate [3.12 (95 % CI, 1.95–5.01)], and heavy [7.12 (95 % CI, 4.67–10.86)] alcohol consumers, compared with ALDH*1*1 (Glu/Glu). The analysis of combined effects of ADH1B and ALDH2 genotypes showed that ADH1B*1/*1 plus ALDH2*1/*2 was associated with the highest risk of esophageal cancer among heavy drinkers [12.45 (2.9–53.46)] (Fig. 1.6), but no significant increase in risk was seen among never/rare drinkers. Recent large-scale genome-wide gene–alcohol consumption interaction analysis of ESCC in China also showed that drinkers with both of the ADH1B and ALDH2 risk alleles experienced a fourfold increase in risk compared to drinkers without the aforementioned risk alleles, while no increased risk was observed among nondrinkers [11].

Fig. 1.6

Risk of esophageal cancer associated with combinations of alcohol dehydrogenase (ADH)1B and aldehyde dehydrogenase (ALDH)2 genotypes

1.2.3 Fruit and Vegetable Intake

Although tobacco smoking and alcohol consumption are the primary lifestyle risk factors for esophageal cancer, dietary factors are also likely to be important [5]. Intake of non-starchy vegetables and fruits appears to have a protective effect. Although the relationship for particular types of fruits and vegetables is unclear, citrus fruits and green leafy vegetables appear to possess greater effects than other families of fruits and vegetables.

The Research Group for the Development and Evaluation of Cancer Prevention Strategies in Japan evaluated that fruit and vegetable intake probably prevent esophageal cancer based on systematic review of epidemiologic evidence among the Japanese population (unpublished data, available at http://epi.ncc.go.jp/can_prev/). Seven studies, two cohort and five case–control studies, tested the association of esophageal cancer prevention with fruit intake and all studies showed a significant protective effect. Eight studies, three cohort and five case–control studies, tested the association with vegetable intake as a whole and green-yellow or cruciferous vegetables. The majority of studies showed significant relationship between the intake of such vegetables and esophageal cancer prevention. However, residual confounding by tobacco smoking and alcohol consumption cannot be ruled out even after adjusting for and stratified by these variables. Both of the variables are strong risk factors for esophageal cancer as well as correlate with the amount of fruit and vegetable intake. The casual association between such lifestyle behaviors and esophageal cancer should be investigated further.

We examined the effect of fruit and vegetable intake on ESCC in a large-scale population-based cohort study [12] (Fig. 1.7). An increase in consumption of total fruits and vegetables by 100 g per day (g/day) was associated with an 11 % decrease in the incidence of ESCC (95 % CI, 1–21 %). In particular, a higher intake of cruciferous vegetables was associated with a significant decrease in risk (HR per 100 g/day, 0.44; 95 % CI, 0.23–0.82). Stratified analyses revealed that the beneficial effect of fruits and vegetables was observed regardless of smoking of tobacco and alcohol use; however it did not completely offset the harmful effects of smoking of tobacco and alcohol consumption.

Fig. 1.7

Fruit and vegetable intake and subsequent risk of esophageal squamous cell carcinoma—JPHC study

1.2.4 Mate and Hot Beverages

Regarding the intake of hot beverages, consumption of hot mate, a traditional herbal beverage consumed in parts of Southern Brazil, Argentina, and Uruguay, there appears to be a strong association with consumption of the beverage and development of esophageal cancer. Meta-analysis of five case–control studies, all adjusted for smoking, showed a summary estimate of 1.16 (95 % CI, 1.07–1.25) per cup/day. Mate is typically consumed very hot through a metal straw. This can cause burns in the esophagus and repeated damage of this nature can lead to cancer, although some have proposed that this may also be a result of chemical carcinogenesis from the composition of mate.

Although there are several studies that show high-temperature drinks and foods are associated with the increased risk of esophageal cancer in Western populations, this evidence is not consistent and most studies have not adequately adjusted for tobacco smoking and alcohol consumption. In a UK population-based case–control study on ESCC comprising 159 female case–control pairs, tea consumption was identified as a risk factor, demonstrating a significant positive correlation with the temperature at which the tea was consumed (p = 0.03) [13].

The Research Group for the Development and Evaluation of Cancer Prevention Strategies in Japan evaluated that intake of hot tea and food is likely to have increased the risk of esophageal cancer based on systematic review of epidemiologic evidence (two cohort and three case–control studies) among the Japanese population (unpublished data, http://epi.ncc.go.jp/can_prev/). A cohort study showed an increased risk of 1.6-fold (95 % CI, 1.2–2.0) for consumption of hot tea (drinking green tea at high temperatures) in comparison with not-hot tea (drinking green tea at moderate temperatures) [14], while another cohort showed that green tea consumption was significantly associated with an increased risk of esophageal cancer [15].

1.2.5 Causes of Esophageal Cancer in Japan

We estimated the population attributable fractions (PAFs) of esophageal cancer attributable to known risk factors from relative risks derived primarily from Japanese pooled analyses (e.g., tobacco smoking), the JPHC study (e.g., alcohol consumption, fruits, and vegetables), and the prevalence of exposure in the period around 1990 [16]. PAFs of tobacco smoking, alcohol consumption, and insufficient intake of vegetables and fruits were estimated to be 58.9, 53.8, 10.4, and 10.9 % in men and 14.7, 28.9, 10.4, and 10.9 % in women. Thus, 88 % of esophageal cancer in men was estimated to be avoidable by lifestyle improvement such as quitting smoking, refraining from too much alcohol consumption, and sufficient intake of fruits and vegetables, after considering combined effect of risk factors. The corresponding statistic for women was estimated at 52 %. Therefore esophageal cancer can be regarded as a lifestyle-related disease.