Home » Publications » Public Health » Spring 2009 » Feature story

Detecting colon cancer

Postdoctoral fellows Joy Owen and Veronika Fedirko examine samples in Robin Bostick's lab at the Emory Winship Cancer Institute.

By Sylvia Wrobel

Robin Bostick aims to develop a simple test to identify risk of disease long before it occurs.

Sometimes an idea grabs your life and won't let go. As a family medicine physician in South Carolina, Robin Bostick became interested in how, as humans evolved and moved from one part of the world to another, dietary changes might affect disease. He was particularly interested in the relationship between colorectal cancer and calcium, which impacts multiple points along the complex molecular pathways that turn normal cells into cancerous ones. Could higher levels of calcium (and vitamin D, then regarded as a helpful sidekick to calcium) actually lower risk for colon cancer, which kills more American men and women than any other cancer except lung?

The idea seemed quixotic at the time but became so consuming that the one-time evolution hobbyist left his medical practice, earned an MPH, and took up the research lance. Bostick, a Georgia Cancer Coalition Distinguished Cancer Scholar in the RSPH, is now one of the country’s leading researchers on identifying biomarkers for colorectal cancer risk and how these biomarkers may soon lead to a simple blood or urine test to detect risk. While such a test would not replace the more invasive colonoscopy, it could be used by physicians to screen many more patients at risk of disease. As head of a national study and co-investigator on several others, Bostick continues riding forth to determine how calcium and vitamin D might lower colorectal cancer risk before the disease develops.

A biomarker is any biologic measurement that correlates with disease risk. For example, cholesterol levels have become a standard biomarker for development of cardiovascular disease. Biomarkers also measure treatment impact. Increasingly, they point to treatments that can lower the likelihood of a disease developing or worsening. In fact, clinicians credit treatments and behavior modifications aimed at changing biomarkers such as cholesterol, blood pressure, blood sugar, and body size/shape with dramatically reducing death rates from heart disease.

No such biomarkers existed for colorectal cancer until Bostick discovered a number of differences in mucosal tissue of the colon among people who later developed the disease compared with those who remained cancer-free. These differences include early alterations in the genes involved in the normal structure and function of the colon; subtle aberrations in the normal growth, repair, and death cycles of cells; inflammation; and an increase in hormones that stimulate proliferation and differentiation of cells.  

To be effective and widely used for screening, biomarker measures must be fast, inexpensive, simple to use, and easy on the patient, unlike the sigmoidoscopy and colonoscopy dreaded by patients. 

Bostick is getting help with the fast and inexpensive part from nanotechnology researchers at Emory and the Georgia Institute of Technology. This collaboration has allowed him to create software that automatically scans slides to quantify the presence and quantity of individual biomarkers in tissue, doing in 15 minutes what used to take a researcher most of the day. The tissue studies are expected to continue as an important way to discover new biomarkers.

The simple, more patient-friendly part of developing screening is evolving as Bostick discovers in blood and urine some of the same biomarkers found earlier in mucosal tissue.

Cox-2, an enzyme involved in inflammation, has proved to be a strong biomarker of colon cancer risk when detected in tissue. It cannot be measured in blood. What can be measured there are other inflammation markers, such as IL-6 and TNF-alpha. Under Bostick’s direction, MPH student Myffy Hopkins and postdoctoral fellow Joy Owen are making sure these measures in blood correlate strongly with similar measures in tissue. Their work is funded by the Wilson and Anne Franklin Foundation, created by Steve and Richard Franklin in honor of their father, Wilson, who died of colon cancer. The funding serves a dual purpose by helping Bostick move his biomarker research from tissue to blood and urine and train young researchers.

The leap from biomarkers found in tissue to similar ones in urine is related to oxidative stress, a measure of damage done to DNA by exposure to free radicals, by-products of oxygen use that damage cells. Tobacco use, for example can cause oxidative stress. Bostick’s research team already has determined that oxidative stress can be reduced in colon tissue by calcium and vitamin D and in blood by antioxidant micronutrients.

Lowering risk

In phase one of this research, Bostick found biomarkers to be a powerful predictor of risk for developing pre-cancerous colon polyps. Phase two of the research, conducted among Emory Clinic patient/volunteers with pre-cancerous colon polyps, demonstrated that treating patients with calcium and vitamin D reduces biomarkers in ways thought to reduce colon cancer risk. The big question in phase three, says Bostick, is whether—and how—changing biomarkers actually translates into improved clinical outcomes. Early evidence from several studies indicates that it does—and through mechanisms and by a source (vitamin D) that researchers are only beginning to understand. 

With funding from the National Cancer Institute, Bostick has piggybacked these questions onto a multi-center study based at Dartmouth to determine whether calcium and vitamin D supplementation reduces polyp recurrence in 2,457 people who have regular colonoscopies. Bostick is the principal investigator of the South Carolina component, while RSPH colleague Michael Goodman, a pediatrician and expert in genetic and epidemiologic factors of cancer, leads the Georgia component. Their work strengthened minority enrollment in the study, allowing multi-center scientists to look at differences in race and ethnicity as well as those of age and gender. Results from this 12-year study will be available in about five years.

Vitamin D once was considered crucial to maintaining calcium levels in the body. Today, like calcium before it, the vitamin is undergoing a renaissance of interest that appears to confirm Bostick’s initial sense that evolution and migration patterns of ancestors affect risk.

Various lines of evidence suggest that more than half of Americans and Europeans maintain vitamin D blood levels below the optimum. The recommended daily allowance (about 200 units for adults, 400 for children) was set at the beginning of the 20th century to reduce the risk of rickets. But raising blood levels to optimal levels would require four or five times that amount.

Given these low levels and several studies that found a correlation between lack of sun exposure and colorectal cancer, should people stop using sunscreen and start popping more vitamins? Bostick warns no, recalling that large supplements of beta-carotene turned out to increase rather than lower cancer risk. “Studying biomarkers will allow us to proceed cautiously and look for what dosage of vitamin D would change the biomarkers related to colon cancer without causing negative changes in the pattern of other biomarkers.”

To examine this question, Bostick again used thousands of stored samples from earlier studies funded by the Emory Winship Cancer Institute, the Georgia Cancer Coalition, and the Franklin Foundation. Postdoctoral fellow Veronika Fedirko is correlating vitamin D levels in these blood samples
with risk for colon polyps—and with variation of genes involved with vitamin D pathways.  

Bostick believes these genetic variations make a big difference in vitamin D’s impact on colon cancer and other health problems. “Early ancestors who migrated from Africa to less sunny lands developed lighter skin because of selective evolutionary pressures to get more vitamin D,” says Bostick. “It makes sense that, especially during the Ice Age, evolution in already fair-skinned Northern Europeans would have selected for genes that more efficiently used what vitamin D they did get.”  

That’s why Bostick is planning a clinical trial to compare the effect on biomarkers of 1,000, 2000, and 4000 IU/day of vitamin D. “It may turn out that not only do all of us need more vitamin D but that one dosage does not fit all,” he says. “Vitamin D needs will vary depending on our genetic makeup.”

Why screening matters

Margaret Shaw and Elaine Scales are interested in the promise that Bostick’s research offers. They are part of a national five-year study to help determine if calcium taken alone or with vitamin D can lower risk of polyps like those discovered through their regular colonoscopies. Shaw’s father and Scales’ mother died of colon cancer.

Both women are participating in two related studies. For one, they undergo a series of rectal biopsies, providing tissue to help develop biomarkers and future screening mechanisms. For the other, researchers have access to their mammography results to determine if different calcium/vitamin D levels affect breast tissue density, a factor in early breast cancer detection.

“We know how important such studies are,” says Scales. “If we don’t do it, who will?"