One such team of statisticians at North Carolina State University has been recognized for its contributions, and has earned a five-year National Institutes of Health (NIH) award of $225,000 per year, or $1.125 million in total. The award is renewable for an additional five years, for a potential total of $2.25 million.
The Method to Extend Research in Time (MERIT) award by the NIH's National Institute of Allergy and Infectious Diseases is given to "outstanding investigators" and is based on "superior competence and productivity," according to the NIH. The award acknowledges the value of the statisticians' statistical methodology in designing and analyzing complex kinds of clinical trials for HIV-related research.
Drs. Anastasios "Butch" Tsiatis, Marie Davidian and Marc Genton, all faculty in the Department of Statistics at NC State, emphasize that their research is not focused on developing new therapies. "That's the job of physicians and clinical scientists," said Tsiatis. "Our research instead is focused on a framework, based in statistical theory, that allows any set of therapies to be compared and evaluated properly."
The MERIT award will allow the statisticians to develop sophisticated statistical designs for especially complicated clinical trials. Their challenge is to design such trials for treatments with many variables, unlike clinical trials with "simple" structures.
"With the advent of highly active anti-retroviral therapy (HAART), which involves giving HIV-infected patients 'cocktails' of potent anti-HIV agents," says Tsiatis, "great strides have been made in reducing mortality. But long-term use of these cocktails is expensive, burdensome, and may lead to adverse reactions. So HIV scientists are looking for new ways to use these drugs for fewer complications and better prognosis."
One promising approach, he said, is "structured treatment interruption," where patients cycle on and off treatment, a process that seems to "train" the patient's immune system to fight the HIV virus on its own. But patients' widely varying responses to the cycling – and the variations in the cycling itself – make standard clinical-trial designs inadequate. The NC State team hopes to design a statistical model that can handle such complexity.
"We hope to develop statistical designs where patients can be randomized at different points in time to the next step in the cycle," said Tsiatis, "with the goal of finding the optimal treatment strategy, and learn how to cycle patients on and off over time according to how they're doing at each stage."
The team believes their statistical work goes beyond HIV clinical research. "We think these statistical methods are broadly applicable," Tsiatis said. "They're relevant to any disease in which the treatment of patients would involve multiple decisions made over time according to the patient's response up to that point. The research in the MERIT award will extend greatly our current work, and help us deal with much more complex strategies."
According to Dr. Daniel Solomon, dean of the College of Physical and Mathematical Sciences at NC State, the statisticians' MERIT award shows the value of interdisciplinary endeavors. "We've made a strategic decision to build programs at the interfaces between the physical and mathematical sciences and the biological and biomedical sciences – areas of tremendous growth and discovery," said Solomon. "Receiving this award is a significant endorsement of our direction and program strength, as well as the stature of our faculty."
Media Contacts:
Dr. Anastasios Tsiatis, 919-515-1928
Paul K. Mueller, News Services, 919-515-3470