In the battle against cancer, research discoveries often seize the headlines. The world celebrates the latest breakthrough in the hunt for a cure. But behind the curtain, other scientists are hard at work. Instead of in vitro, they toil in silico, harnessing computing power to sift through mountains of data for answers about the origins of cancer.
It’s a determined group of public health researchers at the University of Florida Health Cancer Center. It includes an epidemiologist focused on the molecules that give rise to breast cancer, a computer expert developing a tool to get more people screened for colorectal cancer and a epidemiologist using big data to guide prostate cancer treatment decisions.
Their backgrounds are diverse, but they’re united by one mission: Catch the killer before it strikes.
Travis Gerke, Sc.D., didn’t set out to become an epidemiologist. It wasn’t until he was studying biostatistics at Harvard University that cancer epidemiology caught Gerke’s eye. That led to a doctor of science degree in epidemiology and, in 2014, to an assistant professorship at his alma mater: UF. Ask Gerke about the lure of cancer epidemiology, and he’ll talk about a chance to do good things for the world.
Press him further, and he’ll describe the thrill of the hunt — determining how gene expression signatures can help identify non-aggressive types of prostate cancer. The implications of such detailed findings are significant: Prostate cancer isn’t always lethal or likely to spread, and knowing that can reduce the number of procedures.
“There’s a saying that most men will die with prostate cancer, but not from the disease itself,” he said. “It’s very much true.”
That’s where the power of epidemiology comes in. Some 2.9 million men in the United States have prostate cancer, so fewer procedures can lower medical costs on a massive scale.
Gerke’s work in bioinformatics is also driving different ways of thinking about inactive prostate cancer. He was part of a research group that found a way to better identify slow-growing indolent tumors, which can be left alone and monitored. That lets patients avoid unnecessary procedures and the difficult aftereffects that come with them, such as urinary, bowel and sexual dysfunction.“You avoid all the costs and all the side effects of things like a radical prostatectomy,” he said.
Lusine Yaghjyan, M.D., Ph.D., has been on both sides of the cancer fight. As a physician in Armenia, Yaghjyan saw the suffering of cancer patients. Now an assistant professor of epidemiology, she focuses on preventing cancer by studying breast density.
Dense tissue increases the chance of getting breast cancer, so Yaghjyan is studying the molecular processes that drive breast tissue density changes. The goal is to keep breast cancer at bay by thwarting its opportunity to flourish.
Yaghjyan also recently established the North Florida Breast Cancer Cohort through the department of epidemiology. Women who volunteer for the program provide information about breast cancer risk factors and permission to access their mammograms and medical records. Yaghjyan will use these data to study the associations between breast tissue density and breast health outcomes.
The mammogram images that are compiled will also be used to improve computerized techniques for breast density estimation in research studies. Treating cancer as a physician and preventing it as an epidemiologist is also deeply personal for Yaghjyan. Cancer struck twice in her family, which led Yaghjyan to her life’s work.
“That’s my biggest motivator — being able to produce something that prevents disease from occurring in the first place,” she said.
Colorectal cancer screening saves lives, and François Modave, Ph.D., wants to make it easier and more commonplace. Before long, there will be an app for that. Modave, an associate professor of biomedical informatics in the department of health outcomes and policy, is building a bilingual computer application to be used in physicians’ offices to help people decide which colorectal screening method is best for them. The app also provides education about the disease, its diagnosis and screening methods, which helps patients understand why they need to get screened.
For example, less than 60 percent of patients who are at risk for colorectal cancer follow screening guidelines. Among Hispanics, the rate is even lower. Screening rates in the El Paso, Texas area, where Modave started the project while working at Texas Tech University’s Health Sciences Center, are as low as 35 percent.
Knowledge of colorectal cancer, its screening and diagnosis are issues among all patients, but Hispanics are known to have a more fatalistic outlook, Modave said. Nimodo, the saying goes in El Paso, Texas: Oh well, never mind.
That’s where Modave’s software comes in. He plans to recruit 140 Hispanic patients and 20 medical providers to collect data about colorectal cancer screening rates and physician-patient interaction. The goal: Make it easier for patients to decide to get screened and facilitate that discussion with their doctor. Modave expects to have the data collected by early next year.
“The reality is that colorectal cancer evolves slowly,” he said. “If you get tested early and it’s positive, that improves the patient’s chances of survival.”
For Elizabeth Shenkman, Ph.D., big data has big potential for preventing cancer — which is why Shenkman, director of the Institute for Child Health Policy and chair of the College of Medicine’s department of health outcomes and policy, is extremely optimistic about the OneFlorida Cancer Control Alliance and data trust.
Essentially, the OneFlorida data trust is a repository of anonymous medical data collected through OneFlorida’s vast statewide network that comprises 22 hospitals, including UF Health facilities; 914 medical practices; and more than 4,000 physicians providing care for 10 million people. It is led by UF’s Clinical and Translational Science Institute in partnership with the University of Miami, Florida State University and the statewide clinical partners. For researchers, the trust is a way to implement cancer prevention and population science studies while including diverse patient populations, Shenkman said.
Having reliable statewide medical data will accelerate the pace and efficiency of research, Shenkman said. For example, researchers who want to study pregnant women who smoke can find out how many subjects are available in a general area.
“The efficiency is enormous,” Shenkman said. “It will decrease the startup time for researchers and allow them to design better studies.”
The data trust should be fully functional in mid-2016.
As the UF Health Cancer Center works toward one of its goals, becoming a National Cancer Institute-designated cancer center, maintaining a strong cancer control and population sciences program is crucial. As one of the program leaders of the cancer human services program and the cancer center’s associate director for cancer health disparities, Folakemi Odedina, Ph.D., has been working with a 69-member group of researchers at UF to develop cancer control and population sciences specialties.
“UF is uniquely positioned for an effective cancer control and population sciences program due to our collaborative environment, diverse expertise, talented clinicians, outstanding basic scientists and strong institutional support for team science and multidisciplinary approaches,” Odedina said. “Our goal is to do human-focused cancer prevention research that will improve quality of life.”
Linda Cottler, Ph.D., came to UF in 2011 as a professor and founding chair of the department of epidemiology.
She hit the ground running and never stopped. Despite her full plate — teaching, doing research and running the department — Cottler is focused on her passion: reducing disparities for health research and health services. Health care disparities matter, especially for minorities and those in rural communities, she said.