Science in the Triangle

Imagine a chemist, a bioethicist, a pharmacist and a physician come together to discuss personalized medicine. The last thing you expect them to talk about is the need to wear matching jackets like TV weather forecasters.

But that’s exactly what a panel of experts did Monday at the RTI Fellows Symposium at the University of North Carolina’s Friday Center in Chapel Hill.

Personalized medicine - the idea of using gene-based information to individualize medical care - was one of eight hot research areas that RTI International’s scientific advisors recommended for further scientific scrutiny. Other areas addressed at the two-day symposium, the fourth since the inception of the RTI fellows program in 2002, include biofuels and the role computers play in accelerating innovation.

On the first day, personalized medicine attracted more listeners among the about 400 people who had registered for the symposium than any other topic.

The four members on the panel agreed that gene-based research, also known as genomics, and its offshoots, proteomics and metabolomics, have advanced medicine. The pursuit of personalized medicine has improved the understanding of diseases, reduced side effects from certain drugs and brought about new treatments.

But after about 20 years of research, the ‘omics are still far from being a crystal ball to assess a person’s risk for disease, said Eric Juengst, director of the Center for Genetic Research Ethics and Law at Case Western Reserve University School of Medicine in Cleveland, Ohio.

Predictive genetic risk assessment is rather like a weather forecast, said Juengst (photo at left). “We all watch it but we take it with a grain of salt.”

The idea of consulting the genome to improve on health care is alluring, said Howard McLeod, director of the UNC Institute for Pharmacogenomics and Individualized Therapy in Chapel Hill.

Prescription medicines approved to treat a particular disease generally work in only about half of all patients, said McLeod (photo at right).

A good number of patients don’t respond to certain medicines - 25 percent cannot activate the blood thinner Plavix, 10 percent don’t respond to tamoxifen, which reduces the recurrence of breast cancer - because of genetic reasons, he said.

Genetic differences among patients also require significantly different doses of Warfarin, a blood thinner that is similar to rat poison and just as toxic if taken incorrectly.

“We don’t really know how medicines work to the degree we need to know,” McLeod said. “It’s Yogi Berra pharmacology: We know what we know, but we don’t know what we don’t know.”

The ‘omics are helping to unravel some of the unknowns, said Susan Sumner, a chemist who works in the RTI Biomarkers and Systems Biology group.

Researchers have found metabolites - products of chemical reactions in cells - that help predict drug-induced liver injury, the predominant reason why experimental drugs fail. The metabolites can be found in urine samples.

Similarly, biomarkers in urine tests are early warning signs during pregnancy for premature labor, said Sumner (photo at left).

The technology is becoming available to delve deeper into what triggers disease and understand individualized risk factors for expensive, chronic diseases such as Alzheimer’s and diabetes.

Research centers advancing genomics, proteomics and metabolomics have sprung up and federal research funding is flowing to companies in pursuit of the $1,000 genome analysis per person. There’s even a bill before Congress that could give personalized medicine a role in the debate about health care reform.

The promise of personalized medicine is to reduce health care costs as technological advances make the ‘omics cheaper and faster. But embracing genetics in medicine too fast has its pitfalls, said Dr. Jim Evans, an internist, professor of genetics at UNC-CH and the editor-in-chief of the journal Genetics in Medicine.

Very little is known about how genes interacting with genes or with the environment can affect disease, said Evans (photo at left).

Sometimes, tests to screen for disease risks turn out to do more harm than good. A test to measure the amount of a protein called prostate-specific antigen, or PSA, as a tumor predictor is one such example, he said.

Sometimes, it’s not clear what the significance of certain genetic risk factors is. Researchers have found 19 genetic risk factors for Type 2 diabetes, but each has such a miniscule odd to cause the disease it becomes irrelevant.

Evans is a big believer in data that proves personalized medicine goes beyond the promise and actually promotes an outcome.

“It’s so seductive to believe that a good idea improves health,” he said. “But it’s really hard to prove it does.”