Tl. Vaughan et al., OBESITY, ALCOHOL, AND TOBACCO AS RISK-FACTORS FOR CANCERS OF THE ESOPHAGUS AND GASTRIC CARDIA - ADENOCARCINOMA VERSUS SQUAMOUS-CELL CARCINOMA, Cancer epidemiology, biomarkers & prevention, 4(2), 1995, pp. 85-92
Adenocarcinomas of the esophagus and gastric cardia were once rare. Ho
wever, for unknown reasons, their incidence has been increasing rapidl
y over the past 15 years in the United States and parts of Western Eur
ope. In contrast, the incidence of esophageal squamous cell carcinomas
has remained relatively constant. To investigate possible reasons for
these diverging incidence rates we analyzed data from two population-
based case-control studies of cancers of the esophagus and gastric car
dia that were conducted among male and female residents of western Was
hington between 1983 and 1990. Information on body mass index, cigaret
te use, alcohol intake, and other possible risk factors was collected
via personal interviews with 404 cases or their next of kin (including
298 adenocarcinomas and 106 squamous cell carcinomas) and 724 control
s identified by random digit dialing. Use of alcohol and cigarettes we
re significant risk factors for both histological types. The increase
in risk for current smokers of 80 or more pack-years compared to nonsm
okers was substantially higher for squamous cell cancer [odds ratio (O
R) = 16.9; 95% confidence interval (CI) = 4.1-69.1] than for adenocarc
inoma (OR = 3.4; 95% Cl = 1.4-8.0), as was the increase for persons wh
o typically drank 21 or more drinks/week compared to those who drank <
7/week (OR = 9.5; 95% CI = 4.1-22.3 versus OR = 1.8; 95% CI = 1.1-3.1)
For squamous cell carcinoma,,body mass index was inversely associated
with risk, whereas for adenocarcinoma, the highest risk was observed
among persons who were in the highest decile of body mass index (OR =
1.9; 95% CI = 1.1-3.2). The risks associated with alcohol and tobacco
use appeared to be similarly elevated for adenocarcinomas arising in t
he gastric cardia and esophagus; however, there was some evidence that
obesity was more strongly associated with esophageal adenocarcinoma (
OR = 2.5; 95% CI = 1.2-5.0 for persons in the highest decile). Togethe
r, obesity, cigarette smoking, and alcohol use accounted for approxima
tely 50% of the adenocarcinoma cases diagnosed in the Seattle area ove
r the period 1983-1990. In comparison, cigarette smoking and alcohol i
ntake alone accounted for 87% of the squamous cell cases. Additional s
tudies should focus on the role of obesity to determine whether adenoc
arcinoma risk varies by the pattern of fat deposition and by patterns
of weight over a lifetime. Intervention trials aimed at reducing weigh
t among persons at high risk of esophageal adenocarcinoma, such as per
sons with Barrett's metaplasia, are also indicated to establish the ca
usal role of obesity and to determine the usefulness of dietary interv
ention in reducing rates of neoplastic progression to cancer.