Gastric cancer in young patients: Clues on a possible separate entity requiring a watchful approach


  • Annemarie C de Vries,

    Corresponding author
    1. Department of Gastroenterology and Hepatology, Erasmus MC University Medical Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Ernst J Kuipers

    1. Department of Gastroenterology and Hepatology, Erasmus MC University Medical Center, Rotterdam, the Netherlands
    2. Department of Internal Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands
    Search for more papers by this author

Dr Annemarie C de Vries, Department of Gastroenterology and Hepatology, Erasmus MC, P.O. Box 2040, 3000 CA Rotterdam, the Netherlands. Email:


See article in J. Gastroenterol. Hepatol. 2011; 26: 1626–1629.

Gastric cancer is a common disease and the fourth most common cancer worldwide, but it is an uncommon condition at younger age. Although the overall incidence of gastric cancer declines worldwide, recent reports from the USA have remarkably suggested that the incidence in Caucasian patients aged younger than 40 years has increased.1 This in particularly pertaining to gastric cancer localized in the corpus. Helicobacter pylori is considered the most important risk factor for the development of gastric cancer, and its prevalence has declined over the past decades following a birth cohort phenomenon.2 This implicates that the prevalence of H. pylori tends to decrease in subsequent birth cohorts. A further decline in gastric cancer incidence in young patients would consequently be expected, but this is in evident contrast to these recent epidemiological observations.

Possible explanations for these findings include the widespread use of proton pump inhibitors (PPI) resulting in increased risk of atrophic changes of the gastric mucosa in a proportion of H. pylori-positives. As these atrophic changes in H. pylori-positive PPI users were, in the past, in particular observed in the corpus mucosa,3 this would explain the specific increase in cancers of the gastric corpus, instead of the usual antral predilection. It may be hypothesized that a putative effect of a PPI-H. pylori interaction may have a relatively lesser influence on gastric cancer incidence in the elderly in whom the overall decrease of H. pylori colonization outweighs other effects on gastric cancer incidence. Other explanations for the observed increase in corpus cancers in younger subjects include changes in the gastric microbial environment after loss of H. pylori colonization, and environmental risk factors, such as changes in diet.

This issue clearly requires further elucidation. Furthermore, evaluation of incidence trends of gastric cancer in young patients in other geographical areas is required. For instance, in the Netherlands, overall gastric cancer incidence in patients aged between 20 and 40 years decreased from 0.88/100.000 persons in the 5-year period from 1989 to 1993 to 0.74/100.000 persons in the period from 2004 to 2008, but data on intra-gastric location of these carcinomas are lacking.4

Although the abovementioned epidemiological trends may implicate the presence of other pathogenetic factors, H. pylori still plays a pivotal role in the pathogenesis of gastric carcinomas in a large number of young patients with gastric cancer. Several studies have shown a high prevalence of H. pylori infection among young gastric cancer patients, for instance, 81% of young Korean gastric cancer patients tested H. pylori positive, as compared to 53% of controls.5 In addition, the prevalence of atrophic gastritis and intestinal metaplasia at the lesser curvature of the corpus was significantly higher in the group of young gastric cancer patients as compared to a control group, while no significant difference was observed for pre-malignant changes at the greater curvature of the corpus, or for atrophic gastritis in the antrum.5 These findings suggest that the progression along the carcinogenic cascade of pre-malignant conditions may be less important in young gastric cancer patients. As the progression along the cascade from H. pylori, to pre-malignant conditions and finally invasive gastric adenocarcinomas of the intestinal type typically takes several decades, it seems conceivable that many patients with a diagnosis of gastric cancer at young age did not progress through all these stages. Indeed, diffuse-type carcinomas, which are commonly not associated with extensive pre-malignant conditions, are the predominant histology in young gastric cancer patients; they account for up to 73% of gastric cancer cases in European patients aged below 45 years.6

Several studies have also observed a female predominance in young gastric cancer patients, while in older patients a male predominance is common.7 Yet, an explanation for this finding is lacking.

Gastric cancer at young age may also occur in the setting of familial gastric cancer. Overall, about 10% of all gastric cancer cases show familial clustering, whereas in young gastric cancer patients a positive family history of gastric cancer is present in up to 19% of patients.7 In the majority of patients with familial clustering of gastric cancer, a specific gene involvement cannot be demonstrated. In these patients, familial clustering probably results from a combination of genetic susceptibility, environmental risk factors, such as H. pylori infection, and lifestyle risk factors, such as smoking and diet.

It has been suggested that the role of genetic factors is larger in young patients, whereas environmental factors seem less important.8 For instance, E-cadherin gene (CDH1) mutations are relatively common in gastric cancer patients younger than 35 years.9 Siblings of young gastric cancer patients share their risk factors and have a higher prevalence of H. pylori infection and pre-malignant conditions than age-matched controls.10 It seems reasonable to offer these subjects endoscopic screening and surveillance, although this approach may be less effective in cases of diffuse carcinomas. Unfortunately, there are, as yet, no data that show that such a strategy has an effect on survival of gastric cancer, nor are there any data that allow determination of surveillance intervals.

In a minority of patients, approximately 1–3% percent of gastric cancer patients, familial clustering occurs in the setting of an inherited familial syndrome, such as hereditary diffuse gastric cancer, Lynch syndrome, Peutz-Jeghers syndrome, or familial adenomatous polyposis (FAP). These syndromes require specific prophylactic measurements. Hereditary diffuse gastric cancer is caused by a germline mutation of the CDH1 gene, and mutation carriers run a more than 70% lifetime gastric cancer risk.11 A prophylactic gastrectomy is commonly recommended at the age of 20 years. The lifetime risk of gastric carcinomas in Lynch syndrome patients is approximately 5% and 9%, in MLH1 and MSH2 mutation carriers, respectively.12 In these patients, surveillance gastroscopy needs to be considered. The concept that this should only be done in subjects with a positive family history for gastric cancer needs to be abandoned, as most gastric cancers in Lynch families occur as a single case.11 In Peutz Jeghers syndrome, the life-time risk of gastric cancer is approximately 14%.13,14 Surveillance gastroscopy is recommended, and has been proposed at intervals between 2 and 5 years.14 In familial FAP, cancer risk in the upper gastrointestinal tract is principally caused by duodenal adenomas and its associated cancer risk; however, gastric cancer can develop after malignant degeneration of fundic gland polyps.15 Recommendations on surveillance strategies should be based on duodenal cancer risk.

In this issue of JGH, Bai and Li describe an interesting series of 210 Chinese gastric cancer patients aged below 35 years,16 and compared the endoscopic, clinicopathologic features, and survival of these patients with a control group of 210 patients with gastric cancer diagnosed at older age. In this patient series, several of the specific abovementioned features characterizing gastric cancer diagnosed at young age were confirmed; female predominance, high proportion of carcinomas of the diffuse type, high percentage of patients with a family history of gastric cancer, and a majority of cancers localized in the gastric corpus.

An important clinical aspect is that early diagnosis of gastric cancer in young patients is hindered by the absence of alarm symptoms in a large proportion of patients. The study of Bai and Li shows that about two thirds of their young gastric cancer patients lacked alarm symptoms, as compared to less than half of patients at older age. Unfortunately, young patients with gastric cancer also tend to show a poor prognosis as a result of advanced disease at diagnosis so that only limited curative options are available. Although published reports data on this point are conflicting. Bai and Li show that in cases where the stage of gastric cancer allows surgical resection, survival rates of young and older patients are similar.

In conclusion, the study of Bai and Li provides a fine overview on the specific patient characteristics of young gastric cancer patients. Individual cases should always be assessed for the presence of an underlying hereditary condition, which is of outmost importance for follow-up, and for the management of family members. Additional investigation of a potential increase of the incidence of gastric cancer diagnoses at young age is required, as this may identify important risk factors for cancer development in this selection of patients.