Mental illnesses are qualitatively different from other diseases. Most of the time, psychiatrists can’t use biochemical tests or MRIs to diagnosis mental illnesses; instead, they observe how patients behave and ask questions about thoughts and feelings. Psychiatric medications are commonly developed, approved and prescribed without a precise knowledge of how they may work in individual cases. We simply don’t understand the underlying genetics and biology of mental illness in the same way as other medical conditions or even other brain disorders, such as multiple sclerosis or Parkinson’s disease, for example.
However, we can use what we do know about the genetics, neurobiology and behavioral science of mental illness to improve diagnosis and treatment. The National Institute of Mental Health has pursued this strategy since 2013, when then-director Tom Insel launched its Research Domain Criteria (RDoC) project. The idea is to move away from using the lists of symptoms describing specific mental disorders, as noted in the Diagnostic and Statistical Manual of Mental Disorders (DSM), and toward an approach that emphasizes what is known about the biological roots of mental illness.
This transition is essential if we are ever to bring mental illness into the precision medicine era. In its purest form, precision medicine relies on genetic data or biomarkers to develop and tailor treatments for individual patients. We’re still a long way from being able to do that for mental illness. But in the case of schizophrenia, an analysis we recently performed lays the groundwork for further steps in that direction.
Schizophrenia is an incredibly diverse disease, with a broad spectrum of symptoms that vary from one person to the next. One patient may hear voices and harbor delusions, while another may show disorganized behavior and have difficulty learning or comprehending tasks. Yet another may become extremely isolative and close him or herself off from the world.
But the reality is that antipsychotic drugs are only indicated for the overall treatment of schizophrenia, not for specific types of symptoms. As a result, psychiatrists rely on past experience and professional expertise when needing to treat specific types of symptoms. This may work well in many cases, but it’s not exactly precision medicine.
This problem is also encountered in developing new medicines to treat patients with schizophrenia. The most widely used assessment of schizophrenia severity, called the PANSS (Positive and Negative Syndrome Scale), is used in most schizophrenia clinical trials. However, the scale broadly encompasses many areas affected by the illness and, 30 years after its introduction, lags behind our growing understanding of the nature of schizophrenia symptoms.
Because the PANSS total score is a composite—like the often-criticized IQ measure of intelligence—it is difficult to understand which specific symptoms a treatment addresses. For researchers, this limitation makes it nearly impossible to assert that any improvements are in the areas of current unmet need we hope to address.
To overcome this, our team recently published a study that used a novel method to analyze the PANSS score. Using this new statistical analysis, we were able to build and improve upon PANSS to isolate different clusters of schizophrenia symptoms.
This is just one example of how improving psychiatric assessments could pave the way toward more precise evaluations of new treatments with more specific clinical endpoints. It also potentially allows regulatory bodies to include these updated scales and their results in treatment labels, so physicians have more information to guide prescribing.
Unlocking the full potential of precision medicine in mental illness will require not only refining our psychiatric assessment tools, but a better understanding of the myriad of factors that influence how these disorders emerge and develop.
For example, researchers at Massachusetts General Hospital and Harvard Medical School have developed a new method to extract symptom information from electronic health records that predicts length of hospital stays or readmission risk in neuropsychiatric illnesses. Researchers at Stanford University are using brain imaging, along with other clinical and social information, to identify “biotypes” of depression and anxiety that can be studied with different treatments in clinical trials.
Ultimately, these efforts will enhance the efficiency of clinical development and empower physicians, researchers and patients with more reliable and comprehensive information about the nature of serious mental illness. Through better diagnostic tools and deeper insights into the biological, psychological and social variables contributing to each person’s mental health, we can begin to bring precision medicine approaches to this area of great unmet need.
Antony Loebel, M.D., is a board-certified psychiatrist and fellow of the American Psychiatric Association, serves as executive vice president, chief medical officer and head of global clinical development for Sunovion Pharmaceuticals Inc.