SEATTLE-Diabetics may face
three times the risk of liver cancer,
according to data from a new study. A
number of epidemiological studies
have "suggested a relationship between
some cancer types and diabetes, and
the most consistent results have been
for cancers of the liver and the pancreas,"
said Marie-Claude Rousseau, PhD,
a postdoctoral fellow, in the Department
of Social and Preventive Medicine,
at the University of Montreal
Faculty of Medicine.
"Some biological mechanisms have
been proposed to suggest how diabetes
affects the risk of cancer, but to this
day, they remain only hypothetical. So
clearly, the question of whether cancer
risk is associated with diabetes is
still open," Dr. Rousseau said. She
presented the findings at the American
Association for Cancer Research
Third Annual International Conference
on Frontiers in Cancer Prevention
Research (abstract 195).
The investigators used data collected
from a study conducted in Mon
treal during the 1980s designed to address
workplace factors and the risk of
cancer among men. The large population-
based multisite case-control
study included information about prior
diagnosis for diabetes, age at diagnosis,
and medication use for diabetes.
The study sample was comprised
of 3,288 male patients who had been
recently diagnosed with cancer and
509 healthy controls.
Patients with diabetes also provided
information about their usual
height and weight and lifestyle factors.
"These are important to consider when
studying diabetes and cancer risk,"
Dr. Rousseau said.
The investigative team analyzed the
association between diabetes and 12
different types of cancer. "Out of the
20 different cancer types that were
documented in this study, our analyses
focused on 12 types for which the
number of patients was large enough
to provide sufficient statistical power
for the analyses," she explained.
After adjusting for body mass index
and other covariates, the risk of
liver cancer was found to be elevated
among men who reported a prior diagnosis
of diabetes. The adjusted odds
ratio (OR) was 3.1. The proportion of
diabetics was 24% among liver cancer
patients, while only 8% of controls
reported having diabetes.
"These estimates were slightly higher
when we used a more stringent definition
of diabetes," Dr. Rousseau said.
This definition included only those
subjects who were taking medication.
The adjusted odds ratio for liver cancer
for this subgroup was 3.9.
An association was observed with
pancreatic cancer, mostly due to recently
diagnosed diabetes. "This may
suggest that diabetes was a result of the
pancreatic cancer rather than the opposite,"
Dr. Rousseau said.
The researchers did not observe any
association with other cancers studied,
including melanoma, non-
Hodgkin's lymphoma, and cancers of
the esophagus, stomach, colon,
rectum, lung, prostate, bladder, and
"While this was not the first study
to examine the link between cancer
and diabetes, key features of our studies
were that we could assess the risk of
several different cancer types in the
same population and that the patients
were newly diagnosed," Dr. Rousseau
said. "Also, we had information that
allowed us to adjust for important factors,
such as body mass index, smoking,
and alcohol consumption. In conclusion,
our study is strengthening the
evidence of a link between diabetes
and liver cancer."
Dr. Rousseau's colleagues in the
study were Jack Siemiatycki, PhD, University
of Montreal, and Marie-Elise
Parent, PhD, INRS-Institut Armand-
Frappier, Laval, Quebec.