Abstract
Muin J. Khoury, MD, PhD, director of the Centers for Disease Control and Prevention's Office of Public Health Genomics and head of the National Cancer Institute's Epidemiology and Genomics Research Program in the Division of Cancer Control and Population Sciences, talks about challenges and opportunities in cancer epidemiology research and efforts in the epidemiology community to transform the field.
Epidemiology community aims to transform the field and extend its reach beyond etiologic research
Although epidemiology is viewed as inexact science in some quarters, it's a critical part of cancer research, says Muin J. Khoury, MD, PhD, director of the Centers for Disease Control and Prevention's Office of Public Health Genomics, and head of the National Cancer Institute's Epidemiology and Genomics Research Program in the Division of Cancer Control and Population Sciences. Simply put, “you need to validate basic science discoveries in human populations,” he says.
However, Khoury notes that the field is under tremendous pressure due to tightening budgets and a growing sea of data. In December 2012, Khoury and researchers from around the country gathered to discuss challenges and opportunities in cancer epidemiology and how to maximize research dollars; eight recommendations emerged (Cancer Epidemiol Biomarkers Prev 2013;22:508–16). Recently, he spoke with Cancer Discovery's Suzanne Rose about the community's vision for the field and progress that's been made.
The first recommendation is to “extend the reach of epidemiology.” How do you do that while maximizing resources?
We need to broaden the scope of epidemiology beyond etiology and discovery research to include all phases of translation, including developing and evaluating interventions, knowledge integration, implementation sciences, and outcomes research. There are several initiatives under way at NCI and NIH, including the Big Data to Knowledge initiative, called BD2K, whose goals are to develop data integration tools and software to analyze data, and to promote open access and data sharing. In the past, we would do an epidemiology study and then wait for someone to replicate it, and this would take another 10 years. Now we're moving into a time of data sharing and collaboration.
Can you combine epidemiologic research with other types of studies?
That's the crux of the matter here. Using a multidisciplinary framework and team science, we fund epidemiologic consortia that bring together scientists from different fields. Epidemiologists can focus on finding gene X for disease Y, and, at the same time, others can do experimental studies. We're trying to see how we can embed interventions into some of the ongoing cohort studies.
Is new technology being incorporated into studies?
Yes, we are experimenting with novel technologies to better measure what people are exposed to, for example, or to monitor the impact of an intervention or our messages about carcinogens. If we can use data from validated technologies, we can improve our measurements. Measuring exposures, genes, phenotypes, outcomes, and other factors will make epidemiology more robust and allow us to conduct analyses at multiple levels.
We now have a lot of data on genes and cancers. But genes don't work in a vacuum. They interact with the environment. We need to be able to measure in the same study not only the genes, but the environment, the exposures, the air that people breathe, the social and policy environment. All of these things become expensive if you measure them for a large number of people, so it's important to figure out what technology to use—and in what context—to get the biggest bang for the buck.
“Epidemiology is key in raising new hypotheses, developing and evaluating interventions, and monitoring what is happening at the population level,” says Muin J. Khoury, MD, PhD.
“Epidemiology is key in raising new hypotheses, developing and evaluating interventions, and monitoring what is happening at the population level,” says Muin J. Khoury, MD, PhD.
How do you begin to understand what all the data means?
This is the challenge of the 21st century! If you have millions or billions of data points, the challenge is to figure out what's biologically meaningful and applicable in medicine and public health. That's why one of the eight recommendations that came out of the meeting is to figure out how to integrate the new tools of data science. Methodologies are being developed to identify a robust and reproducible signal from complex datasets. Once we generate biologic, clinical, and population data, we can analyze them in an open-access fashion so that we can move more quickly to the right answer.
What other advances should cancer researchers expect?
We have created an online database of funded epidemiology projects and their scientific output. We will also create tools for knowledge integration to help researchers and practitioners see what we don't know. That way, they can decide whether a particular question is worth asking, rather than reinventing the wheel and possibly reaching the same dead end that others have reached. But at the end of the day, we're going to try to get people together because, especially for cancer, collaboration among and across scientific disciplines is going to be very important.
What other challenges is epidemiology facing?
A main challenge is to enrich epidemiology training, emphasizing team science, molecular analysis, translation, and knowledge integration. We need to modernize epidemiologic infrastructures such as epidemiology cohorts so that they can become more nimble and accelerate the discovery-to-translation cycle. We need the research community to have greater access to data, which would allow more replication, more translation, greater reproducibility.
Will the effort to transform the field be worthwhile?
Definitely. You can do your lab and mouse work, but you can't translate it into a reduced cancer burden without conducting epidemiologic studies. Epidemiology is key in raising new hypotheses, developing and evaluating interventions, and monitoring what is happening at the population level.
For more news on cancer research, visit Cancer Discovery online at http://CDnews.aacrjournals.org.