TMC PULSE

t m c » p u l s e | j u ly 2 0 1 6 35 disorder, and we can identify alterations caused by these genes before onset of the disorder." Walss-Bass was one of four experts providing insight for "Science First: The Future of Personalized Medicine in Psychiatry," an education event hosted by Johnson & Johnson Innovation June 1 at its JLABS @ TMC facility, and moderated by Guy Seabrook, Ph.D., vice president of neuroscience innovation for Johnson & Johnson Innovation. Finding a gene Jimmy Lloyd Holder Jr., M.D., Ph.D., assistant professor of pediatrics in the section of neurology at Baylor College of Medicine, started in this area looking at a specific gene called SHANK3, found in those with autism, through a collaboration with Huda Zoghbi, M.D., director of the Jan and Dan Duncan Neurological Research Institute at Texas Children's Hospital and profes- sor of molecular and human genetics, pediatrics, neurology and neuroscience at Baylor. "I was interested in the loss-of- function mutations that cause autism spectrum disorder," Holder said. "Collaborating with Dr. Zoghbi, she had a research project to understand what happens when you have too much of the gene." When studying the abnormalities of mice, the researchers found multiple behavior abnormalities that, when put together, looked like a manic pheno- type. Working with the Miraca Medical Genetics Lab at Baylor College of Medicine, they were able to identify individuals who make too much of the SHANK3 gene. In developing treatments for these individuals, they tried two different drugs usually given to humans for bipolar disorder. Lithium, a drug that has been used for 30 or 40 years, had no effect on the abnormality of the SHANK3 gene. The second, valproic acid, was found to rescue all of the behavior abnormalities, including increased sensitivity to amphetamines, Holder said. "That was unique, and it made us think more about how the treatment is to always try the standard medicine, but by studying the mice, we began to see the pharmacogenetics," he added. "Then we began to understand the genetic cause of their disorders could lead to very specific treatments. From our example, we tested drugs that had been on the market for a very long time, but a priori we wouldn't know if these patients are more likely to respond to Depakote than lithium. It was only with the sort-of basic science that it came to light." His team is also trying to under- stand what the genetic abnormities are that lead to neuropsychiatric disorders, like bipolar disorder. In the long term, that could be beneficial in diagnos- ing and tailoring treatment based on genetic changes or mutations of the SHANK3 gene to individuals that pre- dispose them to bipolar. That, he said, could not only lead to earlier diagno- ses, but also earlier treatment with the correct medication. Targeting a drug Finding the correct medication takes some work, especially when medica- tions for psychiatric disorders don't just target one receptor or neurotrans- mitter system. "Medicines ideally have a fairly clear single mechanism, but in psychiatry, most of the medications have a dozen mechanisms," said Thomas Kosten, M.D., Jay H. Waggoner Endowed Chair in the Menninger Department of Psychiatry and Behavioral Sciences, and professor of neuroscience, pharma- cology, immunology and pathology and co-director of the Dan Duncan Institute for Clinical and Translational Research at Baylor. "It is difficult to make an exact match of a medication to a brain target. Once you have a medication that is spe- cific, you then need a molecular target that the medicine hits," he added. When that does happen, Kosten said, the patient will say, "Thank you," and the patient will get better, as opposed to, "I took this medication, got a bunch of side effects and I'm no better than I was to begin with, so thank you very much, but I won't take any medica- tion for my disorder anymore." Kosten's research focuses on sub- stance abuse disorders in which there are dozens of mechanisms, but the disease is clearer in terms of knowing what the brain targets are, the receptor targets, what genetic polymorphisms are in the gene for the receptor, what the function of the polymorphism is and why the medication works on those who have that particular variant of the gene and the receptor. To him, personalized medicine means selecting subgroups of patients for a specific medication with a specific target in order for that medication to be most effective. Otherwise, the It is difficult to make an exact match of a medication to a brain target. Once you have a medication that is specific, you then need a molecular target that the medicine hits. — THOMAS KOSTEN M.D. Jay H. Waggoner Endowed Chair in the Menninger Department of Psychiatry and Behavioral Sciences and Professor of Neuroscience, Pharmacology, Immunology and Pathology and Co-Director of the Dan Duncan Institute for Clinical and Translational Research at Baylor College of Medicine We tested drugs that had been on the market for a very long time, but a priori we wouldn't know if these patients are more likely to respond to Depakote than lithium. It was only with the sort of basic science that it came to light. — JIMMY LLOYD HOLDER JR., M.D., PH.D. Assistant Professor of Pediatrics in the Section of Neurology at Baylor College of Medicine medication isn't likely to do anything. "Drug companies don't like that message," Kosten said. "No one will buy a new medication if it won't work for 70 percent of those who take it, but if it works for 30 percent of the whole population with that disorder, and you can ID that group ahead of time, those patients will be quite convinced that your medication is worthwhile and doing something." Putting a plan in action In order for all of this to come together, Seabrook said it will take tools, technol- ogy and buy-in from key stakeholders. In addition, it will take a lot of data and sophisticated analytics to under- stand how personalized medicine can work in a real-world setting, said Lynda Chin, M.D., vice chancellor and chief innovation officer of The University of Texas System. To bring all of that together, she is working on building a digital health infrastructure that includes a cloud- based data interchange on a dedicated health care communication service net- work with added security and privacy, integrated with a cognitive analytic cloud, so that data from diverse sources can be gathered, securely, for analyses. "From a care delivery perspective, we need to capture expertise and make sure that knowledge is shared," she said. "We also have to show patients that it will improve their outcome when they give us their data."

• Cover