“You can end up in the ER for a heart attack caused by factors like high cholesterol, high blood pressure or diabetes. But for things like depression and autism, there is not yet a causal equivalent of high blood pressure. Our first task is defining the genetics, which will lead to understanding the cause and mechanisms of these devastating diseases.”
Autism spectrum disorders
As noted above, the choice of Dr. Daniel Geschwind, Ph.D., to lead the institute was apt, given his pioneering work in understanding the genetic underpinnings of autism spectrum disorders. In 2001, Geschwind, then a UCLA professor serving as scientific advisor to Cure Autism Now (now known as Autism Speaks), helped launched a project called the Autism Genetic Resource Exchange, a biobank that over the next decade collected more than10,000 DNA samples for genome sequencing from children with autism spectrum disorders and their families.
“Since then, we and others have identified 200 bona fide autism risk genes, half of them strong candidates and the other half representing possibilities,” Geschwind said. “Ten years ago we knew none of them.” Geschwind’s lab at UCLA has played a major role in these discoveries and has done important work to take them forward to develop a new understanding of people with autism spectrum disorders.
These findings have transformed both the research and the clinical fields. Now, genetic testing, including chromosomal microarray and exome sequencing, is a first-line diagnostic in autism spectrum disorders. Further, we can make a genetic diagnosis in nearly 20 percent of patients, and this number grows every few months. Diagnosis using sequencing to identify the specific genetic basis of an autism spectrum disorder in an individual is a prime example of precision medicine.
To test the function of mutant forms of some genes associated with autism spectrum disorders, Geschwind’s lab has engineered animal models by creating corresponding mutations in mice. Mice with mutations that cause autism in humans also had epileptic seizures, were hyperactive and exhibited repetitive behavior and decreased sociability, behaviors reminiscent of ASD in humans. Moreover, some neurons in their developing brains failed to form stable cell-cell contacts, or synapses.
Using these animal models, Geschwind’s team has discovered drugs that may improve symptoms in humans. These studies are pioneering, in that they link a mutation seen in a subset of patients with behaviors associated with autism spectrum disorders, suggest potential medications to address the symptoms and begin to describe how brain circuitry might be disrupted at a cellular level in that particular cohort of patients.
“Findings in animal models show that some neurodevelopmental disorders associated with autism could be reversed and represent a paradigm shift in our concept of developmental disorders,” Geschwind said. “If they generalize to humans, therapeutics targeting a genetically identified pathway would become the most important area of future treatment research in ASD.”
The stress on pathways, not on individual gene candidates, is important because of the sheer number of genes associated with autism spectrum disorders and the fact that none accounts for a particularly large proportion of cases. The multitude of candidates raises a bewildering question relevant to how feasible tailored treatments would be: Does each mutation perturb behavior or neural development in an entirely different way? If so, treatments for children with autism spectrum disorders might require hundreds of different interventions.
Geschwind’s lab employed computational approaches to address this challenge in a landmark study. It revealed that many ASD-associated genes can be assigned to subgroups that add up to a more tractable (and, by implication, more treatable) number of common pathways. For example, a few mutations may perturb formation of cortical layers, while another cluster might block construction of connections between brain hemispheres. In short, knowing this means that a single therapy targeting a gene network — for example, one that controlled synapse formation — could potentially benefit a subset of patients harboring diverse mutations.
Researchers in Geschwind’s lab and other colleagues at UCLA are using this combination of genetic tools and neurobiological investigations with the goal of revolutionizing our understanding and treatment of neurological and psychiatric disorders.
UCLA’s Institute for Precision Health will put UCLA at the forefront of innovation. The institute provides infrastructure to support big data approaches and serves as a resource for researchers and physicians campus-wide who are developing their own precision health projects. The institute is focused on the collection, generation and integration of genomic information with clinical data from hundreds of thousands of patients across the UC Health system. A key element of this endeavor is providing researchers with access to bioinformatic tools, which are absolutely essential to analyzing these big data. The long-term goal will be to partner with other academic institutions in Southern California and provide these integrated services to a greater number of researchers and clinicians.
Four years ago the parents of a teenage girl sought help for their daughter at a UCLA clinic. Since she was a toddler, the girl had suffered from a movement disorder known as ataxia and from problems with swallowing. Previously, doctors diagnosed her with a juvenile form of the motor neuron disease ALS, but UCLA neurologists, questioning that conclusion, asked that the patient donate a blood sample for what is called whole exome DNA sequencing (which means sequencing all 23,000 human genes). That test revealed that rather than ALS, the girl carried a mutation in a gene that causes a different disease, an extremely rare syndrome called Triple-A (AAA), for Achalasia-Addisonianism-Alacrima.
There is no cure for AAA syndrome, which is marked by tightening of the esophageal sphincter (achalasia); the inability to produce tears (alacrima), which the girl also suffered; and adrenal insufficiency (Addison's disease). UCLA physicians then combed the scientific literature to discover that indeed, a few individuals with AAA syndrome also exhibit ataxia-like symptoms, putting to rest any remaining doubts about the diagnosis.
Today, the patient’s ataxia persists, but she has had surgery to relieve the achalasia and has not yet developed full-blown Addison's disease. If she does, there are steroid treatments available for the condition. Her parents are freed from endless searching for the right diagnosis and now focus their energy and resources on helping their daughter manage her condition, one confirmed by evidence. The patient, now 21, still sees a UCLA neurologist, who says that she and her colleagues had never seen a case quite like this and that without genomic testing, a diagnosis might never have been made.
This clinical case study, one of more than 1,000 cases of rare disorders seen at UCLA’s clinical genomic center over the last three years, is an example of precision medicine. As a whole, such rare disorders account for nearly 10 percent of all patients seen, so in essence such cases are common. UCLA is a leader in this area. The case is now taught to medical students as a lesson in how genetics-based approaches are revolutionizing the practice of medicine: Once the parents of a child like this would have no option but to spend years facing expensive and possibly minimally informative standard diagnostic tests; now there are ways to obtain rapid, accurate and cost-effective information relevant to disease cause in numerous difficult-to-diagnose conditions.
The story also is a reminder that reliance on technologies unavailable to previous generations of physicians, such as next-generation DNA sequencing or advanced computational tools, in no way precludes compassionate or personalized care. Quite the opposite, it contributes knowledge to a learning health care system with a network of molecular biologists, mathematicians, software engineers, statisticians, other physicians and patients (some across town, others across the globe) and maybe even patients' loved ones — all working to create a learning health care system in which the contributions of many inform the care of one person. Advocates of precision medicine, or “precision health” as we call it, know that complex health issues patients face today can be resolved only by teamwork.
Investigating the mystery of depression
“There is no chest X-ray for depression.”
Dr. Nelson Freimer, director of UCLA’s Center for Neurobehavioral Genetics and associate director for research programs of the Semel Institute for Neuroscience and Human Behavior, works on what he calls “the greatest health problem of our time”: depression. As a researcher, he has studied the genetics of bipolar disorder, a condition in which patients cycle between episodes of intense euphoria or mania and depression. Much of his work explores the effect of inherited mutations on depression-related behaviors and how neuroanatomical changes in the brain may correlate with both.
Freimer says individualized approaches to treat bipolar disorder, or any kind of depression, have advanced at a glacial pace. “The last significant advance in precision treatment of bipolar disorder came 60 years ago when scientists discovered that for some patients, administration of lithium acted as a mood stabilizer,” Freimer said. “Lithium really was a miracle of modern medicine at the time, but not everyone could tolerate it, and we have no idea how it works.”
Although people will often say that "depression runs in my family," few professionals consider depression a genetically complex disease, if treatment choices are any indicator. Do a Google search for drug treatments for post-partum depression or combat-related PTSD, and the same candidates emerge.
But Freimer thinks that this one-treatment-cures-all view of depression, so debunked in pathologies whose genetics are well understood, is on the way out. “At a genetic level, depression is heterogeneous, and several hundred genes contribute to it,” he said. “But right now, physicians have no way of deciding which treatment to prescribe. People usually go through three or four attempts by trial and error.”
Freimer wants to end trial and error treatment by first classifying different forms of depression based on genetics, then tailoring treatment based in part on those outcomes. “What has stopped us in depression is that our assessment diagnostics are entirely based on subjective information,” he said. “You see a doc who asks you questions and you say you feel depressed,” he said. “It’s entirely based on your recall rather than objective tools: There is no chest X-ray for depression.”
A bold initiative at UCLA aims to change this. In 2015, UCLA launched its Depression Grand Challenge, an effort that will recruit 100,000 subjects from UCLA’s health system willing to undergo genetic screening, among other biological tests, for genes potentially mutant or deregulated in depression. Freimer is the initiative’s director. The goal of the DGC, which expects to raise $500 million dollars in the first 10 years, is lofty: to decrease the health and economic impact of depression by half by 2050. Project organizers are buoyed by the 2013 launch of UCLA's equally ambitious Sustainable L.A. Grand Challenge, whose mission is to develop a blueprint to transition Los Angeles County to 100 percent renewable energy, 100 percent locally sourced water and enhanced ecosystem health by 2050.
DNA sequencing technology coupled with bioinformatics analysis will of course play a big part in the depression challenge, given the ambitious goal of initially sequencing 100,000 genomes, making this a globally unique project. Freimer said UCLA is on target to succeed as it has access to the required large patient population and the infrastructure to generate and analyze DNA sequence data. But the real goal of the project is to then match up those genomic results with depression-related behaviors, as reported by patients. Behavioral analysis could require data gathering by a device more commonplace than a DNA sequencing machine.
“Most people have a low-cost tool for detecting and analyzing depression-related behaviors in their pocket,” Freimer said, meaning, of course, mobile devices. “A cellphone can objectively measure your sleep patterns, physical activity, interpersonal interactions, voice tone, and numerous other indicators of depression, not just once but over long periods of time."
Why not? A plethora of cellphone apps is now available for “precision” management of insulin dosage in diabetes. And at UCLA, Alex Bui, a faculty member in the UCLA Medical Imaging Informatics group, is using mobile phone and wearable technology to monitor environmental factors that could alert children to the risk of an asthma episode.
Freimer said UCLA is well suited to execute innovative projects that take apart questions with a lot of moving parts and that engage the community. “Universities that bring together a medical school, a hospital and people capable of looking at the economic and social impact of these issues are really the only places you can take on grand challenges like this,” he said. “Being in California is also a plus: It's conducive to thinking outside the box.”
The Depression Grand Challenge has already funded seven pilot projects exploring the biological basis of depression in collaborative and inventive ways. One team is studying how glial cells in the brain influence some types of depression; another is evaluating effects of novel drug treatments for forms of reproductive or postpartum depression.
If this sounds audacious, that’s because it is. But then, half a century ago it might have been unthinkable to propose that the numerous anomalies collectively called “cancer” were actually quite different from one another at the molecular level. “The development of precision medicine strategies in the field of cancer has been transformative,” Freimer said. “We want them to be equally transformative in our field.”