Reuters Health Information (2004-09-30): Antioxidant supplements do not prevent GI cancers Epidemiology
Antioxidant supplements do not prevent GI cancers
Last Updated: 2004-09-30 18:30:03 -0400 (Reuters Health)
NEW YORK (Reuters Health) - Findings from a
meta-analysis of trial data indicate that antioxidant supplements offer
no protection against gastrointestinal (GI) cancers. In fact, use of
such supplements was actually tied to an increase in overall mortality.
The one possible exception was selenium, but further research with
better methodology is needed to verify the anti-cancer effect seen, the
investigators note.
Several reports have implicated oxidative stress in the pathogenesis
of cancer. However, it has been unclear if regular use of antioxidant
supplements provides protection against GI malignancies.
To investigate, Dr. Goran Bjelakovic, from the University of Nis in
Serbia and Montenegro, and colleagues conducted a pooled analysis of
data from 14 randomized trials that compared antioxidant supplements
with placebo in their ability to prevent GI cancers. A total of 170,525
patients were included in the analysis.
The researchers' findings appear in the October 2nd issue of The Lancet.
There was no evidence that use of supplements containing
beta-carotene, vitamins A, C, E, or selenium, alone or in combination,
protected against esophageal, stomach, colorectal, pancreatic, or liver
cancer. Selenium use may have had a beneficial effect on cancer risk,
but this was only observed in four trials, three of which had unclear
or inadequate methodology.
In seven trials, use of antioxidant supplements was tied to an
increase in mortality. In particular, beta-carotene combined with
vitamin A or E significantly increased mortality.
In a related editorial, Dr. David Forman, from the University of
Leeds in the UK, and Dr. Douglas Altman, from the Centre for Statistic
in Medicine in Oxford, UK, urge caution in interpreting the mortality
results, noting that the researchers' analysis is a "work in progress,
and does not offer convincing proof of hazard."
Lancet 2004;364:1193-1194,1219-1228.
|