As the obesity epidemic in the United States and around the world
continues to get worse, the links between excess weight and cancer incidence
and mortality have come under increased investigation. With close to one-third
of all U.S. adults classified as obese, according to the CDC, determining just
what the risk between obesity and cancer is and what can be done about it is
more important than ever.
At the beginning of this decade, evidence of the most solid links
between specific types of cancer and obesity began emerging. Malignancies of
the colon, endometrium, kidney and esophagus, as well as postmenopausal breast
cancer, were found to be linked to excess weight. Now, the list has expanded,
and experts are starting to learn more about how many cancer cases can be
attributed to obesity. Furthermore, issues have arisen as to obesity’s
effect on screening and diagnosis, as well as treatment, and there is strong
evidence that it can increase mortality among cancer patients as well.
Although most research in recent years has looked for links between
specific cancer types and obesity, some large cohort studies have shown an
overall correlation between excess weight and cancer incidence. The Million
Women Study in the United Kingdom enrolled more than 1.2 million women and
followed them for 5.4 years for cancer incidence and for seven years to measure
cancer mortality. For all cancers combined, there was a RR increase per 10
units of BMI of 1.12 (95% CI, 1.09-1.14). The study found that about 5% of all
cancers among women in the United Kingdom, or about 6,000 total cases, are
attributable to overweight or obesity.
James Abbruzzese, MD, of M.D. Anderson Cancer
Center, researches links between obesity and pancreatic cancer.
Photo by: John Everett
In the United States, numbers appear to be similarly high. A study
published in Cancer Detection and Prevention showed that, in 2002,
about 41,000 new cases of all types of cancer were due to obesity; this was
about 3.2% of all new cancers. Some speculate that these numbers may increase
in the coming years, as the lag time for malignancies to develop will follow
the recent upward spikes in obesity prevalence.
“I think this is a big worry in overall trends of cancer
[incidence],” said James Abbruzzese, MD, of The University of Texas
M.D. Anderson Cancer Center, who has worked on links between pancreatic cancer
and obesity. “Between the overall aging of the population and the obesity
epidemic, we may be poised to lose some of the gains that we’ve made
[toward decreasing cancer mortality]. I think it is a huge problem.”
Although there is an association between obesity and overall cancer
statistics, drilling down into specific malignancies may be a more important
first step to understanding the mechanisms of the relationship between obesity
and cancer, and combating the problem, according to the experts who spoke with
Some research being conducted today is beginning to explore these
relationships. The Million Women Study found that with the excess 10
kg/m2 of BMI, the RR for endometrial cancer was 2.89; other
significant associations were seen with adenocarcinoma of the esophagus
(RR=2.38), kidney cancer (RR=1.53), pancreatic cancer (RR=1.24) and others.
A 2008 meta-analysis — published in The Lancet —
of 221 distinct data sets also found highly significant associations between
obesity and a number of specific malignancies. In this case, a 5
kg/m2 increase in BMI in men was associated with a RR of 1.52 for
esophageal carcinoma and of 1.33 for thyroid cancer. Other associations were
seen between cancers of the colon, rectum and kidney, malignant melanoma,
multiple myeloma, as well as leukemia and non-Hodgkin’s lymphoma.
Similarly, in women, strong correlations between obesity and malignancies of
the gallbladder, pancreas and postmenopausal breast also reached statistical
With so many cancers being linked to excess weight, determining the
mechanisms underlying such connections is exceptionally difficult. As Roberts
and colleagues wrote in a recent review in Annual Review of
Medicine, “it is unlikely that there is a ‘one system fits
One of the most often cited mechanisms is the insulin and insulin-like
growth factor I axis. Although the details are not yet fully understood, it
seems likely that insulin can strongly promote tumor growth in a number of
locations, and the abnormal insulin production and tolerance that characterizes
obesity could lead to malignancies, according to Roberts’ review.
Another mechanism that is particularly associated with postmenopausal
breast and endometrial cancer — two of the stronger associations between
obesity and cancer — is estrogen level. As Cleary and colleagues wrote
recently in Endocrinology, among postmenopausal women, estrogen
production is primarily performed at peripheral sites (rather than the
ovaries), and among obese women, the primary source is adipose tissue.
Some studies have shown that women with postmenopausal breast cancer
have higher levels of estrogen than control patients, and the higher rates of
estrogen production in obese women can at least partially explain the increased
risk of breast tumors. Aromatase, an enzyme that mediates estrogen synthesis,
is found in breast adipose tissue as well as tumor tissue, leading to
substantially higher estrogen levels in breast tumors compared with circulating
levels. This results in an increased risk for breast tumors as weight and
estrogen levels rise, Cleary wrote.
Other potential mechanisms that mediate the obesity-cancer link are
under investigation as well. One involves adipocytokines, including leptin and
adiponectin. Adiponectin is produced in adipose tissue, and its concentration
is inversely proportional to weight: As a person becomes more obese,
adiponectin concentrations diminish.
The protein acts as a strong angiogenesis inhibitor and tumor cell
inhibitor, so as concentrations decrease, the ability to ward off tumor growth
may decrease. Research into adiponectin and prostate cancer — the
incidence of which obesity has a very small effect on, but appears to be
associated with more aggressive tumors — has indicated that the lower
levels of the protein could contribute to obesity’s negative effects.
Stephen J. Freedland
“If I’m given $10,000 and I can bet on which pathway is the
most important, [the one] I’m putting most of it on is either estrogen or
insulin,” said Stephen J. Freedland, MD, an associate professor of
urology and pathology in the Duke Prostate Center at Duke University, who has
done work on the adipocytokine connections to prostate cancer. “The
adiponectin story is interesting, but it’s a side bet. It may turn out to
be really important, but it may turn out to be not so important.”
Finally, a paper in the International Journal of Cancer in
2002 postulated that the abdominal pressure brought on by increasing obesity
could cause increased gastroesophageal reflux, which in turn can contribute to
malignancies of the esophagus as well as the gastric cardia. This mechanical
explanation most likely does not extend to most other forms of cancer, but
could explain the increasing rates around the world for those two types of
Although Abbruzzese said that we still lack a good understanding of
mechanisms underlying any specific links, “there is an increasing
understanding that fat tissue is not inert, it is a metabolically active organ.
It changes hormonal balance in women, and I think there is only going to be
more information in the next few years toward really understanding what is
behind all this.”
Obesity also plays a role in screening for cancer. There are only a few
recommended screening tests for women in the United States, and a 2008 review
in Cancer found that screening rates for cancers of the cervix and
breast tend to drop as BMI rises (no association was seen for colorectal
Freedland said there is some indication that with prostate cancer
screening in men, obesity might actually be linked to higher rates of
“It’s the only cancer we screen by just checking a box on the
blood collection form, and thus obese men with their greater comorbidities are
seeing their doctors more often and getting blood drawn more often. Also for
prostate cancer, we have the specific problem that the excess blood volume
among obese men dilutes out the PSA making PSA levels lower in obese men and
more cancers can go undetected leading to a potential delay in diagnosis,”
“With other types of screening, due to comorbidities, you may be
less worried about particular cancers. Also, certainly a very large breast can
make mammography more challenging, and doing a good rectal exam can be more
difficult as well.”
Nisa Maruthur, MD, an assistant professor of medicine at Johns
Hopkins University in Baltimore, and colleagues published two papers in early
2009 that further showed the low rates of screening among obese women; the
association was strongest with Pap testing for cervical cancer, with the most
obese women (those 40 kg/m2 or more) showing an OR for screening of
0.62 compared with normal weight women. “It is probably a multifactorial
issue, with both patient- and physician-related barriers,” Maruthur said.
“There is literature that shows that obese women delay medical care, and
then certainly there is a perceived lack of respect from health care
Additional issues with treatment and outcomes in obese patients also
occur after a malignancy is found. There is now very strong evidence that
mortality increases in many cancers among obese individuals, and as Freedland
said, treatments are often made more difficult among people with substantial
“There are a lot of technical difficulties in screening, detection
and, ultimately, treatment. It is hard to operate on an obese man. It is
physically and technically more challenging,” he said. “With prostate
cancer, we have the luxury of saying, ‘Look, go lose 20 lb and come back
in three months and we’ll do your surgery.’” Other more
aggressive cancers do not provide such an advantage and can lead to difficult
treatment decisions with obese patients.
With regard to mortality, many malignancies have higher rates among
obese individuals. In a large 2003 analysis published in The New England
Journal of Medicine, the RR of death for men with a BMI of at least 40
was 1.52; for women the RR was even higher at 1.62. Those risks were for all
cancers combined, and increased risks of mortality for the heaviest members of
the cohort were also found for cancers of the prostate, esophagus, colon,
rectum, liver, gallbladder, pancreas and kidney, as well as for NHL and
multiple myeloma. Even those cancers that did not have a significant
correlation with increased mortality showed a trend in that direction; these
included cancers of the stomach in men and of the breast, uterus, cervix and
ovary in women.
Meir J. Stampfer, MD, DrPH, a professor of medicine at Harvard
Medical School, said that prostate cancer stands out in this realm for the way
obesity contributes to the disease. “[Obesity] also contributes
substantially — perhaps a quarter or more — to prostate cancer
mortality, though not to incidence.”
The Million Woman Study cohort in the United Kingdom found increased
mortality risk as BMI increased for a number of cancers. The strongest
correlations in that study were found in malignancies of the esophagus and the
Obese individuals often have far more comorbidities than normal weight
patients, but the effects of excess weight on mortality are hard to ignore,
Freedland said. And although persuading people to lose weight as a method of
potentially avoiding some cancers is difficult, his group’s efforts to
convince already-diagnosed prostate cancer patients that taking off some of
their excess weight will help with treatment has been extremely successful, at
least at getting the men to lose weight. Whether that slows tumor growth
remains to be seen.
With the amount of data on the connections between obesity and cancer
mounting, the question of how to translate those data into practice may be
lost. There are a limited number of studies that have actually looked at weight
loss as a means of cancer prevention; in other words, are the mechanisms that
may be responsible for increasing an obese individual’s risk reversible?
This is one of the many questions that researchers have as the field moves
forward, Abbruzzese said.
For now, though, adding the apparent cancer risks to the already robust
list of reasons to maintain a healthy weight could play a role in public health
“I think most people associate overweight/obesity with risk of
diabetes and heart disease and not so much with cancer,” Stampfer said.
“Cancer is understandably highly feared, in part because there are fewer
obvious avenues for prevention — compared with heart disease, for example.
If people understood better the strong link of overweight with some aspects of
cancer, it could provide additional motivation to maintain good body
weight.” – by Dave Levitan
cancer be prevented through weight loss or the reversal of the mechanisms
identified as increasing an obese individual’s cancer risk?
For more information:
- Calle EE. NEJM. 2003;348:1625-1638.
- Cleary MP. Endocrinology. 2009;150:2537-2542.
- Cohen SS. Cancer. 2008;112:1892-1904.
- La Vecchia C. Int J Cancer. 2002;102:269-270.
- Maruthur NM. Obesity. 2009;17:375-381.
- Maruthur NM. J Gen Intern Med. 2009;24:665.-677.
- Polednak AP. Cancer Detection and Prevention.
- Reeves GK. BMJ. 2007;335:1134.
- Renehan AG. Lancet. 2008;371:569-578.
- Roberts DL. Annu Rev Med.