Research findings link weight gain and diabetes to a variety of cancers

Dec 10 2007

During this holiday season with its tempting bounty of edible delights, new research calls attention to the role of the expanding American waistline in health and medicine.

Researchers at the American Association for Cancer Research's Sixth Annual International Conference on Frontiers in Cancer Prevention Research, being held from December 5 to 8 in Philadelphia, Pennsylvania, present some of the latest research linking obesity, diabetes and metabolism to cancer risk. Their findings link weight gain and diabetes to a variety of cancers affecting both men and women, including breast, prostate and colorectal cancer

Women with diabetes are 1.5 times more likely to develop colorectal cancer than those who do not have the metabolic disorder, according to researchers at the University of Minnesota. The findings, they say, add to the complex body of evidence linking diet and colorectal cancer and also provide new evidence that furthers our understanding of the role of insulin in cancer promotion.

“Colorectal cancer and type II diabetes share a number of common factors, including obesity, so it is interesting to see the direct line between these two conditions,” said Andrew Flood, Ph.D., assistant professor in the Division of Epidemiology and Community Health at the University of Minnesota School of Public Health and the University of Minnesota Cancer Center ”In general, the idea is that if elevated insulin levels create a biochemical environment conducive to cancer growth, it provides one mechanism by which diet and lifestyle can really influence cancer risk.”

With funding from the National Cancer Institute, Flood and his colleagues examined data from a massive screening study called the Breast Cancer Detection Demonstration Project, initiated at 29 centers throughout the United States in the 1970s. Flood's team subsequently followed more than 45,000 study participants with no history of colorectal cancer or self-reported diabetes for eight years, (from 1987-1989 and from 1995-1998), to identify which of them subsequently developed colorectal cancer. According to their findings, women with diabetes had a greatly increased risk of developing colorectal cancer. “These results remained statistically significant even after controlling for all known and suspected confounding variables,” Flood said.

According to Flood, it is not exactly clear what aspect of diabetes is the underlying cause for this increased risk, but one hypothesis centers on the elevated concentration of insulin typically seen in people with type II diabetes. “In the early stages of the disease process, people become insulin resistant, meaning they must produce more and more insulin to regulate their blood sugar,” Flood said.

“Even after frank diabetes begins, insulin levels remain chronically elevated for extended periods before the pancreas can no longer supply the level of insulin the body demands,” Flood said. “If the elevated insulin is the problem, then pre-diabetics, who are also hyper-insulinemic, should also be at increased risk (for developing colorectal cancer).”

To test that idea, Flood and his colleagues re-analyzed the data, this time including women who were likely pre-diabetic at the beginning of the follow-up period. The idea, Flood says, is that these women were likely hyper-insulinemic at that stage. Surprisingly, the elevated risk, while still significant, had dropped slightly in comparison with that of known diabetics, Flood says.

According to Flood, this suggests that either the pre-diabetic women had not had elevated insulin long enough or intensely enough to increase risk as they observed in the diabetic women, or alternatively, something other than or in addition to hyper-insulinemia could explain the significant, increased risk for colorectal cancer they observed in people with diabetes.

Fasting C-peptide levels and breast cancer death in women with breast cancer: The Health, Eating, Activity and Lifestyle (HEAL) Study. Abstract no. B99:

Women with invasive breast cancer and high blood levels of C-peptide (a marker of insulin secretion) face a risk of death nearly three times higher than women with lower blood levels of C-peptide, according to findings from the Health, Eating, Activity and Lifestyle (HEAL) Study, a long-term observational study of breast cancer patients. The effect was most notable, researchers say, among women in their 40s.

While previous research has demonstrated that insulin stimulates the growth of breast cancer cells in the laboratory, few studies have examined the link between fasting insulin or C-peptide levels and breast cancer prognosis. Women with invasive breast cancer – meaning the cancer had spread throughout the breast tissue or to surrounding tissues – faced the greatest risk from high C-peptide levels, the researchers say, but the association was detected in nearly all women studied, regardless of whether or not their cancer had spread.

“When looking at risk of diabetes and hypertension, breast cancer survivors really should talk to their oncologist about how to lower their insulin levels,” said Melinda L. Irwin, Ph.D., M.P.H., assistant professor at Yale University's School of Public Health. “The simple message is that breast cancer patients should take proven steps to lower their blood insulin levels, including exercise and eating a diet rich in fruits and vegetables and low in fat.”

The HEAL Study is a National Cancer Institute initiative designed to examine the links between diet, physical activity, body fat, and breast cancer prognosis. Patients enrolled in the HEAL Study -- including those participating in the study reported here – were diagnosed or treated at the Fred Hutchinson Cancer Research Center, the University of New Mexico or the University of Southern California.

Between 1995 and 1998, the researchers followed 689 women enrolled in the HEAL program who were diagnosed with breast cancer, but who did not have type 2 diabetes. They monitored their health at periodic intervals beginning six months after diagnosis until September 2004 or the patient's death. From each patient, they collected a fasting blood sample – a common technique for measuring a baseline of insulin or C-peptide levels – and information on prognostic, demographic, and lifestyle factors, including weight and height.

In order to determine the relationship between C-peptide levels and prognosis, Irwin and her colleagues statistically adjusted the data they collected for confounding variables such as body mass index, age, race, disease stage and therapy used in treatment. They found that, when arranged into three groups based on C-peptide levels, women in the top third of the group (highest levels) had twice the risk of death compared to women in the bottom third. When looking at just women with invasive breast cancer, the risk of death among women with high C-peptide levels was three times higher than among women with low C-peptide levels. “Our findings clearly show that C-peptide and most likely insulin, in and of itself, is a marker for breast cancer prognosis,” Irwin said.

According to Irwin, the association was also common in women in their 40s with early stage breast cancer, and less pronounced in women in their 50s or 60s. “The higher death rate among younger women suggests that these women may have had more aggressive tumors, possibly related to tumor genetics or family history,” said Irwin.

Association of C-peptide concentration with prostate cancer incidence in a prospective cohort. Abstract no. B89:

While studies have consistently shown that men with diabetes are at a decreased risk for prostate cancer, the reasons have been unclear. By evaluating prostate cancer data from a large, long-term cohort study, researchers at Johns Hopkins University have shown that those with high concentrations of C-peptide – a marker of high insulin secretion that is a hallmark of diabetes – had a measurable decrease in prostate cancer risk.

“Metabolic perturbations influence cancer risk, that much is becoming clear to us, and we are learning more about the fundamental issues in biology that guide prostate cancer development,” said Gabriel Lai, a doctoral student in the Department of Epidemiology at Johns Hopkins Bloomberg School of Public Health. “One interesting possibility is that, over time, diabetics generally have less testosterone in their bloodstream than non-diabetics, which might lower risk of prostate cancer.”

Lai and his colleagues used data from a large-scale study known as CLUE II, which had enrolled almost 23,000 adults in Washington County, Maryland in 1989. With funding from the National Cancer Institute, they examined the history of 264 men with confirmed prostate cancer and matched them with a group of 264 men without prostate cancer with a similar distribution of age and race.

For each participant, the researchers measured the amount of C-peptide in the blood they donated when they enrolled in the study. Researchers consider C-peptide to be a surrogate marker for insulin secretion because both molecules derive from the same precursor molecule, with insulin degrading faster than C-peptide. They found that patients that had elevated levels of C-peptide in their bloodstream when they started the study were about one-third less likely to develop prostate cancer later. This was true even among men without diabetes.

The researchers also report a markedly lower risk of non-metastasized prostate cancer. Men with higher C-peptide levels in their blood were half as likely to develop organ-confined prostate cancer, Lai says.

“Even though diabetes and obesity are often linked to different types of cancer, our findings illustrate the idea that the link between cancer and metabolic diseases is not the same for every variety of cancer,” Lai said. “Obviously, having high levels of insulin does not promote health but perhaps such disorders can provide insight into the mechanisms of prostate cancer to help us learn how to eventually prevent prostate cancer.”

Post-diagnosis weight change, body mass index, and breast cancer survival. Abstract no. B95:

Gaining weight following a diagnosis of invasive breast cancer could increase a woman's risk of death from the disease by more than half, according to researchers leading the Collaborative Women's Longevity Study. In fact, the researchers associated weight gain with a measurable increase in risk of death due to all causes, not just breast cancer.

“Our findings provide additional support for the benefits of maintaining a healthy weight and exercising,” said Hazel B. Nichols, a doctoral student in the Department of Epidemiology at Johns Hopkins Bloomberg School of Public Health. “According to our results, there is a 14 percent increase in risk for every five kilograms -- about 11 pounds -- of weight gained.”

To analyze the effect of weight gain on breast cancer survival, Nichols and her colleagues contacted women who had taken part in one of three previous studies begun in 1988 at sites in Wisconsin, Massachusetts and New Hampshire. Between 1998 and 2001, Nichols' team surveyed the women about post-diagnosis weight, weight gain, physical activity, diet and related items.

Of the original 4,021 breast cancer patients, the researchers identified 121 breast cancer-related deaths and 428 total deaths. For women classified as obese by body mass index – a measure of weight and height – the risk of dying from breast cancer was nearly 2.4 times that of women classified with a normal body weight. “Obesity was associated with risk of death even after accounting for age, menopausal status or smoking,” Nichols said.

Nichols' study was funded by the Susan G. Komen for the Cure Breast Cancer Foundation.