Diet and lifestyle factors and risk of subtypes of esophageal and gastric cancers: classification tree analysis

Abstract

Purpose: Although risk factors for squamous cell carcinoma of the esophagus and adenocarcinomas of the esophagus (EA), gastric cardia (GC), and other (noncardia) gastric (OG) sites have been identified, little is known about interactions among risk factors. We sought to examine interactions of diet, other lifestyle, and medical factors with risks of subtypes of esophageal and gastric cancers.

Methods: We used classification tree analysis to analyze data from a population-based case-control study (1095 cases, 687 controls) conducted in Connecticut, New Jersey, and western Washington State.

Results: Frequency of reported gastroesophageal reflux disease symptoms was the most important risk stratification factor for EA, GC, and OG, with dietary factors (EA, OG), smoking (EA, GC), wine intake (GC, OG), age (OG), and income (OG) appearing to modify the risk of these cancer sites. For esophageal squamous cell carcinoma, smoking was the most important risk stratification factor, with gastroesophageal reflux disease, income, race, noncitrus fruit, and energy intakes further modifying risk.

Conclusion: Various combinations of risk factors appear to interact to affect risk of each cancer subtype. Replication of these data mining analyses are required before suggesting causal pathways; however, the classification tree results are useful in partitioning risk and mapping multilevel interactions among risk variables.

Keywords: CART; Classification tree; Diet; Esophageal adenocarcinoma; Gastric cardia adenocarcinoma; Gastroesophageal reflux disease.

Figure 1

Classification Tree: Lifestyle factors and risk of esophageal adenocarcinoma, from United States multicenter, population-based study (1993–1995). Each group contains the number of controls (top number of uppermost box) and the number of cases (bottom number of uppermost box). Terminal subsets are represented by square boxes and are identified by letter in the lower left corner. The proportion in the bottom right corner of each terminal subset gives the probability of being a case in that group. Because the prevalence of cases in the total sample was 29%, terminal subsets comprised of more than 29% cases are considered higher risk groups for classification purposes and are highlighted in bold italics.

Figure 2

Classification Tree: Lifestyle factors and risk of gastric cardia adenocarcinoma, from United States multicenter, population-based study (1993–1995). Each group contains the number of controls (top number of uppermost box) and the number of cases (bottom number of uppermost box). Terminal subsets are represented by square boxes and are identified by letter in the lower left corner. The proportion in the bottom right corner of each terminal subset gives the probability of being a case in that group. Because the prevalence of cases in the total sample was 27%, terminal subsets comprised of more than 27% cases are considered higher risk groups for classification purposes and are highlighted in bold italics.

Figure 3

Classification Tree: Lifestyle factors and risk of esophageal squamous cell carcinoma, from United States multicenter, population-based study (1993–1995). Each group contains the number of controls (top number of uppermost box) and the number of cases (bottom number of uppermost box). Terminal subsets are represented by square boxes and are identified by letter in the lower left corner. The proportion in the bottom right corner of each terminal subset gives the probability of being a case in that group. Because the prevalence of cases in the total sample was 23%, terminal subsets comprised of more than 23% cases are considered higher risk groups for classification purposes and are highlighted in bold italics.

Figure 4

Classification Tree: Lifestyle factors and risk of non-cardia gastric adenocarcinoma, from United States multicenter, population-based study (1993–1995). Each group contains the number of controls (top number of uppermost box) and the number of cases (bottom number of uppermost box). Terminal subsets are represented by square boxes and are identified by letter in the lower left corner. The proportion in the bottom right corner of each terminal subset gives the probability of being a case in that group. Because the prevalence of cases in the total sample was 34%, terminal subsets comprised of more than 34% cases are considered higher risk groups for classification purposes and are highlighted in bold italics.