The newest scientist to join the Institute for Informatics (I²) at Washington University in St. Louis, Fuhai Li, PhD, will leverage years of expertise in biomedical informatics to research potential treatments for rare pediatric tumors and other conditions.
“Dr. Li’s training in the applied mathematics domain of drug repositioning will help the institute incorporate advanced methods of analyzing data sets when examining genomic data and medical records,” says I² Director Philip R.O. Payne, PhD, who recruited Dr. Li from the Department of Biomedical Informatics at The Ohio State University College of Medicine.
“At Ohio State, and at Houston Methodist Hospital Research Institute before that, Dr. Li was a consummate team scientist, which is essential for a project that is a collaboration between informaticians and clinicians,” Dr. Payne says. “In addition to working with different, very intensive teams as they analyzed data to find impactful results, he engaged with patients and family members to discuss their preferences and risk tolerances. He not only understands the biological and medical requirements of research, he’s also very aware of the sociocultural environment in which it works — how data-centric information affects human-centric decisions.”
Groundbreaking Genetic Research
In previous research, Dr. Li investigated drugs that showed potential to treat cancerous tumors by analyzing their effects on genetic data biomarkers. At I², he will investigate drugs that could help treat conditions that lead to tumors or genetic defects in pediatric patients.
“The Rare Disease Programs including the Undiagnosed Diseases Network Clinical Site and the Genomics of Birth Defects Projects in the Department of Pediatrics are enthusiastic about Dr. Li joining I² and our department,” says F. Sessions Cole, MD, assistant vice chancellor for Children’s Health. “He brings interests and skill sets that are critical to the discovery of causes of rare diseases and, most importantly, to the discovery of prevention and treatment strategies for rare diseases.”
Some diseases, such as certain types of medulloblastoma, are rare enough that pharmaceutical companies haven’t researched better treatments, which means no targeted therapies have been approved yet to treat these young patients.
“In spite of the advances in the current treatments involving surgery, radiation and chemotherapy, about 25 to 30 percent of patients still die from medulloblastoma. And severe side effects in developing children, such as cognitive deficits and endocrine disorders, have been linked to these aggressive treatments,” Dr. Li says. “This could affect a child’s neurological development.” Some children treated with radiation have shown a 20- to 30-point decrease in IQ points, for example, and even a reduced dose of radiation has been linked to a decline of 10 to 15 points.
“We’re entering unchartered territory,” Dr. Payne says, “taking methodologies that have shown to be effective in cancer and high-profile diseases, and bringing them to bear to genetic diseases in children.”
The Informatics Advantage
Dr. Li will take a data mining approach to his research. He will examine a multisource mash-up of data derived from clinical electronic health records, publicly available medical and genomics data produced by the National Institutes of Health and other nonprofessional agencies, and other open scientific realms. The work involves trying to pinpoint not only the signal pathways of drug resistance or tumor development, but also the indications of thousands of drugs to determine whether they show promise to aid in treatment.
“There are 2,500 FDA-approved small molecule drugs and another few thousand experimental and investigational drugs that have been tested but not yet FDA-approved,” Dr. Li says. “With computational methodology, we can focus on the potential efficacy of not just individual drugs, but of millions and millions of potential combinations.”
The genetic signaling network is complicated. A tumor cell, for example, is not just one simple marker but a networked system in which one change often activates another change farther down the genetic stream.
“You may find a drug that targets and affects a mutation — but it may work for only a few months, as the body develops a resistance to that drug,” Dr. Li says. “If, based on the patient’s personal genomics data, we can discover the driver signaling what causes this drug resistance, then we can try to identify other drugs that have been shown to target that signal, and which other sets of genes might have been changed by those drugs.”
Potentially Faster, More Cost-Effective Treatments
“Traditional drug discovery can take 15 or more years and $30 billion to develop a new agent,” Dr. Payne says. “In computational approaches to repositioning drugs to augment front-line therapies, we can conduct them in 6 to 12 months for $50,000 to $100,000. The difference is remarkable.”
Most activity in the drug repositioning domain has been related to cancer research, but Dr. Payne says there has been little work done in pediatrics — particularly in rare genetics. “We see patients present with little or no treatment available to them, and currently doctors are put in the position of using off-label treatments that may or may not be efficacious,” he says.
Dr. Li hopes his work will change this. “In my position at I², I hope to work with pediatric doctors to identify not only tumors but other developmental defects and genetic mutations that present the same challenges,” he says. “This will be the future of medicine — working together with doctors and biologists to identify treatment solutions that will affect children’s lives for a long, long time. I like this opportunity.”
“With computational methodology, we can focus on the potential efficacy of not just individual drugs, but of millions and millions of potential combinations,” says Dr. Li.