Obesity's Risks Include Cancer, Too

Reprinted with permission, J Natl Cancer Inst, 94:1704-11, 2002 OF MICE AND DUCTS: Ductal branching (arrows), part of normal mammary-gland development, is absent in tissues from leptin-deficient (panel B) and leptin-receptor-deficient (D) mice, but is present in tissues from wild-type controls (A and C). By fostering mammary development, the leptin pathway might contribute to tumorigenesis. An estimated 30.5% of American adults--nearly 59 million people--were obese in 2000, after their ranks had swelled 33% since the mid-1990s.1 But growing obesity rates are not confined to the United States, and the attendant health risks do not just include cardiovascular disease, diabetes, asthma, and arthritis. Cancers of the colon, breast, endometrium, kidney, and esophagus also are solidly associated with excess adiposity.2 (A link to prostate cancer is controversial.) In addition, obese patients with cancer have poorer prognoses. Caloric restriction offers some protection against the disease, but eating trends are clearly heading in the opposite direction. Despite the situation's apparent urgency, only a handful of research-ers are examining the obesity- cancer link through experiments. Instead, the evidence is overwhelmingly epidemiological. The human data and some of the underlying mechanisms are "pretty clear," says Walter C. Willett, professor of epidemiology and nutrition at the Harvard School of Public Health. "It's sort of a little bit going backwards to then go look at [the link] in a mouse or rat." Sharon E. Fleming, a professor of nutritional sciences and toxicology at the University of California, Berkeley, suggests another reason for the paucity of obesity-related cancer experiments: their subject is the whole animal in all its complexity. In contrast, she observes, "Scientists are now focusing very strongly on molecular and cellular-level work, and that's what's fundable." Fleming, who compares colon-cancer rates in obese and lean mice, nevertheless argues that animal studies could help set priorities for cancer prevention. Laboratory projects could indicate, for example, whether targeting obesity would be more preemptive than tackling insulin resistance. OBESITY TRUMPS DIET Most research on the obesity-cancer link deals with the large intestine. The colon's surface is easily scanned for tumors or abnormal areas, known as aberrant crypt foci, that correlate with tumors and might be their precursors. The obese rodents used in these studies typically lack the receptor for the hormone leptin. As a result, leptin cannot signal the animal to stop eating. Several years ago, surgeon John G. Kral, at State University of New York Health Science Center at Brooklyn, and his colleagues set out to determine whether obesity trumps diet as a risk factor for colon cancer.3 His team fed obese rats low-fat Purina rat chow--expected to discourage tumor growth--while lean rodents ate a presumptively cancer-promoting diet loaded with corn oil and lard. The tumor initiator was subcutaneously administered azoxymethane. Obesity won out over diet. Eight of the nine fat rats, but no thin ones, developed colon tumors, despite their low-fat fare. Kral concludes, "Obesity, via adipose tissue and a multitude of growth factors in adipose tissue, can facilitate the growth of a colon cancer." Kral says he has since found increased levels of vascular endothelial growth factor in obese rats and is now applying his model to breast, endometrial, and prostate cancers. Kral recounts stiff resistance to his findings from investigators who maintain that improper diet boosts colon cancer. Meanwhile, Harvard Medical School's Edward Giovannucci and other researchers have proposed an alternative, though not mutually exclusive, theory. They posit that obesity creates insulin resistance and the high insulin levels known as hyperinsulinemia; these, in turn, encourage colon carcinogenesis. Some supporting evidence: Insulin injections promote colon tumors in nonobese rats,4 and colon-cancer risk is higher in patients with type 2 diabetes. Many questions remain. Could insulin resistance without obesity advance cancer? "Right now, we really don't know," replies W. Robert Bruce, professor emeritus of nutritional sciences at the University of Toronto. Some thin people, he notes, are insulin-resistant, but attempts to create animal models would likely introduce confounding factors. Another issue is whether the tumor booster is insulin resistance or hyperinsulinemia. "We have sat and argued about that for many hours," recalls Bruce. Fleming hopes to decide whether hyperinsulinemia or excess adiposity promotes aberrant crypt foci and colon tumors in mice. Using azoxymethane or dimethylhydrazine as the cancer initiator, she is now testing a particular murine strain for its susceptibility to colon tumors. Later she plans to compare lean mice who have normal insulin levels to two groups of hyperinsulinemic mice. One group will be obese after being fed a coconut oil-rich diet; the other will be of normal weight after receiving a high-fructose diet. Finally, Fleming intends to use the chemical metformin to reduce plasma insulin levels in obese mice. "In my view, there are no good data now that address body fatness in the absence of hyperinsulinemia," she says. She expects to detect some linkage between colon-cancer risk and hyperinsulinemia but is not sure if adiposity alone will foster malignancies. "We used to think of adipose tissue as being pretty much of an inert material," she notes, "and we know now that it isn't." If body fat alone turns out to promote tumorigenesis, Fleming envisions greater efforts at obesity prevention, given the odds against permanently shedding megapounds. Because of "relatively few people with sustained, substantial weight loss," Willett says that no study nails down a statistical link between such weight loss and reduced cancer risk. MENOPAUSE, THE DIVIDING LINE Breast cancer is associated with obesity in postmenopausal women, and the main mechanism appears to be excessive estrogen levels. Obese women, Willett notes, have about three times the circulating estrogen levels as lean women. He adds that obesity lowers breast-cancer risk in premenopausal women, probably by inhibiting ovulation. This mechanism might have been at work several years ago when Michael C. Archer, a nutritional sciences professor at the University of Toronto, and his colleagues examined the impact of a mammary-tumor initiator on lean and obese rats.5 Half the lean rodents sprouted mammary malignancies, compared to only 10% of the fat ones. Archer remarks that "something strange" was going on in the obese animals--epidermal cysts, for instance, formed around the mammary glands, and some rats even developed colon cancer. In his paper he speculates that the incidence of lower mammary tumors was related to abnormal estrous cyclicity, even though estradiol concentrations did not differ between the lean and fat animals. Courtesy of Margot P. Cleary Margot P. Cleary Margot P. Cleary, an associate professor at the University of Minnesota's Hormel Institute, has implicated leptin as a growth factor promoting breast cancer. (It might play a role in colon cancer as well.6) Last year, she and the Mayo Clinic's Nita J. Maihle reported the presence of leptin receptors in normal and malignant human breast cells. They also found that leptin turned on signal transduction in both cell types, but it increased anchorage-independent growth on a soft agar medium only in the malignant cells.7 Bypassing the need for a carcinogenic initiator, the group generated mice that harbored a transforming growth factor-a gene linked to a mouse mammary-tumor virus promoter. Of the mice with at least one normal copy of the leptin gene, one-half to two-thirds showed mammary tumors, while not a single leptin-deficient obese mouse did.8 The Cleary-Maihle team discovered that tumorigenesis also requires the leptin receptor. In a study slated for presentation at the American Association for Cancer Research's 2003 meeting, no mammary tumors appeared in fat mice deficient for the receptor. Cancer, however, developed in 69% and 82% of the lean wild-type and heterozygous strains, respectively. Leptin generally increases in proportion to the amount of body fat, Cleary says. Given the presence of leptin receptors in mammary tissue and maybe in precursor tumors, she hypothesizes that "if women gain weight and increase their leptin levels, and these receptors are there, then [leptin] would turn on proliferation of these alleged precursors." She stresses, however, that her theory does not preclude an estrogenic role in breast cancer and might not apply to all obese women, whose leptin levels can vary considerably. ANY EFFECT SO FAR? As human obesity becomes more common, the effect on cancer incidence is not yet clear. "Postmenopausal breast-cancer rates have been going up, and obesity is a fairly strong risk factor," observes Willett. "So that's probably part of the reason." But he notes other possible contributory factors, including changes in the components of hormone replacement therapy and increased alcohol consumption. The colon cancer situation is even murkier. Rates have dropped slightly, a development that Willett suggests might stem from lower red-meat consumption and higher folic-acid intake. Bruce points out another mystery: If obesity facilitates sporadic human cancers, what initiates them? Douglas Steinberg (dougste@attglobal.net) is a freelance writer in New York City. References 1. K.M. Flegal et al., "Prevalence and trends in obesity among US adults, 1999-2000," J Am Med Assoc, 288:1723-7, Oct. 9, 2002. 2. F. Bianchini et al., "Overweight, obesity, and cancer risk," Lancet Oncol, 3:565-74, September 2002. 3. R.V. Weber et al., "Obesity potentiates AOM-induced colon cancer," Dig Dis Sci, 45:890-5, 2000. 4. T.T. Tran et al., "Insulin promotion of colon tumors in rats," Cancer Epidemiol Biomarkers Prev, 5:1013-5, 1996. 5. W.M. Lee et al., "Susceptibility of lean and obese Zucker rats to tumorigenesis induced by N-methyl-N-nitrosourea," Cancer Lett, 162:155-60, 2001. 6. Z. Liu et al., "High fat diet enhances colonic cell proliferation and carcinogenesis in rats by elevating serum leptin," Int J Oncol, 19:1009-14, 2001. 7. X. Hu et al., "Leptin--a growth factor in normal and malignant breast cells and for normal mammary gland development," J Natl Cancer Inst, 94:1704-11, 2002. 8. M.P. Cleary et al, "Genetically obese MMTV-TGF-a/ LepobLepob female mice do not develop mammary tumors," Breast Cancer Res Treat, 77:205-15, February 2003.

Obesity's Risks Include Cancer, Too

Interested in reading more?

Become a Member of

Reprinted with permission, J Natl Cancer Inst, 94:1704-11, 2002
OF MICE AND DUCTS: Ductal branching (arrows), part of normal mammary-gland development, is absent in tissues from leptin-deficient (panel B) and leptin-receptor-deficient (D) mice, but is present in tissues from wild-type controls (A and C). By fostering mammary development, the leptin pathway might contribute to tumorigenesis.