Chester J. Zelasko, Ph.D. |
July 8, 2008
A reader recently
sent me a link to a lecture by Larry Norton, MD, Deputy Physician-in-Chief
for Breast Cancer Programs at Sloan-Kettering Cancer Center (1). The reader
was concerned because he’s a leading expert on cancer treatment and said he
wouldn’t recommend taking vitamins because research has shown an increased
mortality rate. Needless to say, that’s a provocative statement so I decided
to listen to the discussion about supplements and provide my analysis of his
comments. Before I do, let me say upfront that Dr. Norton is a brilliant
physician; I wouldn’t hesitate to recommend him to anyone seeking cancer
treatment. But it’s always the rest of the story where things get muddy.
Let’s take a look.
Are Vitamins Dangerous?
Dr. Norton made a statement that people who take vitamins die sooner. He
based that on a study published in Lancet in 2003 (2). The study was not a
double-blind, placebo-controlled trial (DBPC). Rather, it was a
meta-analysis of seven DBPC trials on vitamin E and eight DBPC trials on
beta-carotene supplementation that included a combined total of 200,000
subjects. The objective of the individual studies was to see whether
supplementation with these antioxidants reduced morbidity and mortality.
Vitamin E had no positive or negative effect; beta-carotene resulted in a
slight increased risk of overall mortality and an increased risk of death
from heart disease.
The problem is that this statistical study combined studies that used
varying amounts of the supplements, used mostly synthetic sources of the
vitamins, and lasted varying lengths of time from 1.4 to 12 years. Even with
no background in statistics, you can see the problem--too many different
types of studies are being combined. In a review of the problems with
meta-analysis, Dr. Craig Coleman who did a similar type of meta-analysis on
the efficacy of echinacea said, “It’s not that we are comparing apples and
oranges. Rather, we’re comparing different types of apples (3).” But that
doesn’t lend itself to an accurate analysis.
There’s one more issue and that’s related to the number of subjects.
Whenever the number of subjects gets so high, it increases the probability
that differences will be found; an analysis of the data confirms that risk.
While there was a statistically significant increase in mortality for those
who used beta-carotene, the numbers work out to an additional four deaths
per 1,000 subjects over 12 years. People are not dropping in the street from
taking beta-carotene as Dr. Norton implied. He acknowledged that some people
who are already sick might be attracted to taking vitamins. With so few
deaths, that’s a more reasonable explanation than saying that vitamins are
bad in general. Vitamins are not a substitute for a healthy lifestyle and
quality health care--they complement it.
Why Would You Say That?
Dr. Norton was asked about the use of substances such as curcumin and
coenzyme Q10 during cancer treatment. He replied with a question: “Why would
you say that?” What he was after was the research: are there randomized DBPC
trials done on those supplements in human trials during cancer treatment?
Not test-tube or rodent studies and not theories based on logic--human
trials. Have those data been accepted for publication in a peer-reviewed
journal? If it passes those criteria, then he’d consider it.
But Dr. Norton violated his own criteria. He made the statement that cancer
cells are more metabolically active and therefore, will need the vitamins
more than the normal cells; the vitamins could therefore increase the
tumor’s ability to defend itself and grow. So let’s ask him the same
question: “Why would you say that?”
Are there published randomized DBPC trials on cancer patients undergoing
treatment who were given antioxidants and had an increase in mortality or
tumor growth? The answer is no. There are test-tube studies on
antioxidants and isolated tumor cells, but that doesn’t fit his criteria.
Neither does his logic when he states that cancer cells are dividing fast
and thus must need more antioxidants. There are only test-tube studies
showing that may be possible but nothing in human trials. In fact, there are
a small number of DBPC trials on human subjects undergoing cancer treatment
who were given antioxidants with beneficial effects (4).
This was not an attempt to catch Dr. Norton making an error. Rather, it’s to
demonstrate that we don’t know everything about antioxidants and cancer
treatment. If a well-respected doctor makes a statement, he must back it up
with research, not logic. So if you’re undergoing cancer treatment and your
doctor says, “I don’t want you to take any vitamins during treatment,” I
would start by asking one question: “Why would you say that?”
- Larry Norton, MD. The Changing Face of Breast Cancer: Where We Are,
Where We Are Going, and How.
- Vivekananthan DP, et al. Use of Antioxidant Vitamins for the
Prevention of Cardiovascular Disease: Meta-Analysis of Randomized
Trials. Lancet. 2003; 361:2017-23.
- Carey, John. When Medical Studies Collide: Contradictory reports?
Meta-analysis may make things more confusing. BusinessWeek. August 6,
2007. Page 38.
- Better Life Newsletter.
Vitamin C and Cancer Treatment. March 16, 2007.