Gastric Cardia Cancer: How Much Is It from Fat?

Abstract

The overall age-adjusted incidence of gastric cancer hasdecreased in the United States from 11.7 to 7.1 persons per100,000 from 1975 to 2008 according to the surveillanceepidemiology and end results (SEER-9) database [1]. Simi-lar trends have been observed worldwide. Although theincidence of gastric non-cardia adenocarcinoma (GNCA)has decreased worldwide the incidence of gastric cardiaadenocarcinoma (GCA) has steadily increased over the lastthree to four decades along with a rise in esophageal ade-nocarcinoma (EAC) incidence [2, 3].It is possible that the incidence of GCA has increaseddue to better understanding and appropriate classificationof distal esophageal and GCA rather than a true increase incancer incidence rate [4]. Current studies, however, are notcompletely devoid of the error of misclassification andhence concluding that the increase reflects a prior falseunderestimation is not completely justified. There is roomfor error even in studies that have selected patients basedon standardized classifications. Various studies have usedthe International Statistical Classification of Diseases andRelated Health Problems (ICD-9) and International Clas-sification of Diseases for Oncology (ICD-O) codes [5–7].Misclassification in the ICD-9 coding system could lead toincorrect estimation of cancer incidence (Table 1). First,EAC and cardio-esophageal junction cancers were classi-fied under one group (ICD-9-151.0). Second, GCAcould be misclassified under ‘‘unspecified gastric cancers’’(ICD-9-151.9) or ‘‘overlap sites of stomach’’ (ICD-9-151.8) [8]. In the ICD-O 2/3 system GCA tumors areclassified under C16.0 but they could be incorrectly labeledas ‘‘overlapping lesions of esophageal cancer’’ (C15.8) [9](Table 1). Nyre´n et al. in 1999 studied the possibility ofmisclassification in the Swedish cancer registry from 1989and 1994 and reported that the incidence of GCA could be45 % higher or 15 % lower than that reported [10].The Siewert classification sorts tumors as esophagealtumor (type I), true cardia tumor (type II) and subcardialtumor (types III). This is based on the location of thecancer with respect to the cardia, which is defined as oralend of the typical longitudinal gastric mucosa folds. Thismight be a better way of differentiating junctional cancers[11] (Table 2; Fig. 1).Obesity as a Risk Factor for Cardia CancerAmong the risks factors for EAC and GCA, obesity standsout as an important modifiable factor (Table 3). Theprevalence of obesity has increased from 13 to 32 %between the 1960s and 2004. Currently, 68 % of US adultsaged 20 and over are overweight or obese and 34 % areobese [12]. With the rise in the epidemic of obesity in theWestern world there has been an increase in the incidenceof GCA. One study calculated that the attributable riskpercentage of GCA due to obesity has increased steadily inthe United States to 21 and 10 % in men and women,respectively [13].Several case control studies and meta-analysis haveshown a positive association between higher BMI, ECAand GCA. The association appears to be much stronger forEAC. A UK-based case-control study in 2005 showed apositive association between BMI[25 kg/m