Weida Tong, director of the Division of Bioinformatics and Biostatistics, in his office at the National Center for Toxicological Research.
On this page:
The world is awash in data.
Researchers in many countries—in government, private industry and academia—are studying the human body and how it is affected by everything it experiences, including foods, drugs and other medical products, and substances in the environment.
At the Food and Drug Administration’s National Center for Toxicological Research (NCTR) in Jefferson, Arkansas, computational scientists (researchers who use computers and mathematical models to analyze and solve scientific problems) pull together strands from this global scientific tapestry to find solutions to specific health challenges. They don’t reinvent the wheel by repeating the research: They collect and analyze massive amounts of already available data to find the answers they need.
That’s the mission of NCTR’s Division of Bioinformatics and Biostatistics, led by computational chemist Weida Tong, Ph.D. Tong is a recognized leader in the development of computer-based research tools. Bioinformatics uses computer science and mathematics to process biological data.
This is the newest division at NCTR—created in 2012—and one that holds great promise, says Center Director William Slikker, Ph.D. “This type of research will help FDA move forward in all areas of our mission.”
“We collect a lot of data—a broad range of data. And the question we ask is: How do these data reflect the biological complexities of the human body?” says Tong. “How can we extract these data and convert them to information that can be used in the decisions that FDA has to make in evaluating new medical products?
Safe use of drugs is one of the priorities for this division, and its particular efforts have been focused on preventing drug-induced liver injury (DILI). Medications can damage the liver. Serious DILI is the most frequent cause of acute liver failure because of the widespread use of medications in the overall population.
People can vary in how they respond to the same dose of the same drug, notes John R. Senior, M.D., associate director for science in FDA’s Office of Pharmacovigilance and Epidemiology, who collaborates with the NCTR team. Thus, he says, it is not enough to characterize the chemical properties of the drug or substance alone; you have to understand how different patients respond to them.
In pursuing both avenues of research, the NCTR scientists study drug properties, genomic data, cellular responses and animal data. Among their findings:
- Patients who have a particular allele (an alternative form of a gene) experience DILI more frequently than other patients when exposed to certain medications.
- A high lipophilicity in a drug plus a daily dose higher than 100 mg suggests possible liver toxicity, a finding verified by tests on 179 drugs. (Lipophilicity is a drug property that plays a crucial role in determining absorption, distribution, metabolism, excretion and toxicity of a drug.)
These and other findings are used by FDA in drug review, risk assessment, and identification of the types and mechanisms of this kind of injury. NCTR maintains a Liver Toxicity Knowledge Base, a publicly available resource with data on marketed drugs and predictive models that can be used in risk assessment.
Another area of concentration is called drug repositioning, which Tong explains is finding a new use for an existing drug. He gives as an example thalidomide, a drug that was once notorious for its toxicity but is now used in FDA-approved treatments. Given to pregnant women to treat nausea in the 1950s and 1960s, thalidomide caused severe birth defects in thousands of children, primarily outside of the U.S. Decades later, it is effectively used to treat a number of diseases, including multiple myeloma and leprosy (also known as Hansen’s disease).
NCTR scientists are searching the universe of data on existing medications to find new treatments for rare diseases. There are approximately 7,000 diseases affecting humans that have few or no treatment options.
The premise underlying this work is that similar drugs could have the same therapeutic effects, Tong says, and similar diseases could be treated with similar medications. The key challenge is how to identify these similarities and, most importantly, how to assess them. At NCTR, this evaluation is done with diverse data using bioinformatics.
One of the diseases the NCTR team is studying is cystic fibrosis, an inherited disease that affects the lungs and digestive system.
The scientists are looking at a number of medications that, with slight alterations, could be used to treat this disease. They include simvastatin (Zocor), a popular cholesterol medication, and rogislitazone (Avandia), a diabetes treatment.
These projects just touch the surface of NCTR’s bioinformatics work which, Tong says, includes the maintenance of a biological database that includes research conducted by many universities worldwide.
The division also advances scientific progress in such areas as:
- Developing models to classify benefits and risks for regulatory decision making regarding personalized medicine.
- Further developing and refining the knowledge base for assessing the endocrine-disrupting potential of chemicals, including drugs and food additives. Interference with the body’s endocrine system can cause developmental, reproductive, neurological and immune-system damage.
- Studying sex differences in molecular biomarkers (signs of a condition or disease) for individualized treatment of diseases that are not themselves gender-specific.
“The data are often changing and evolving. But It’s not just a matter of collecting the science being discovered all over the world,” says Tong. “What we do is interpret that research to determine how it can help FDA protect public health.”
This article appears on FDA’s Consumer Updates page, which features the latest on all FDA-regulated products.
July 1, 2014