Applying Precision Medicine to Neuropsychiatric Disorders

machinelearningCancers were once looked upon as specific diseases, often defined by their tissue of origin, and drug treatments primarily aimed to kill rapidly growing tumor cells before harming slower growing normal ones. The result was often limited drug efficacy, accompanied by debilitating side-effects. Today, however, the growing understanding of the biology and genetics of particular tumor types is accelerating the use of precision medicine approaches, with molecularly targeted treatments that offer greater efficacy with fewer side-effects for many cancer patients.

 

Scientists are now hoping to apply some of the lessons learned from applications of precision medicine in oncology to the understanding and treatment of neuropsychiatric disorders such as depression, schizophrenia, autism and other conditions based on their biology. Until recently, despite the well-observed heterogeneity of patient populations in these conditions, neurobehavioral disorders have typically been treated as individual disorders defined primarily by symptoms. This situation is beginning to change with the development of new tools and scientific knowledge that advance a more precise understanding of the diversity of many of these conditions.

 

Developing precision medicine therapeutics is particularly challenging in neurobehavioral conditions, as it requires defining discrete patient subtypes based on underlying common biology, molecular pathways and targetable biomarkers, which as yet remain largely unknown. Further complicating this research, brain and other neural tissue cannot be biopsied to study the specific pathways involved, making the detection of potential drug targets more difficult. Moreover, neuropsychiatric disorders often present highly variable symptoms, which can vary in intensity and presentation between patients and with age of onset. Inheritable genetic variants can also be a critical factor, but detecting such influences in conditions where there is still limited understanding of the underlying biology requires large population studies. 

 

One new company, BlackThorn Therapeutics, is using next-generation artificial intelligence (AI) and machine learning to advance the discovery of targeted therapeutics for neurobehavioral disorders, with an initial focus on mood disorders and social-emotional conditions like autism spectrum disorder (ASD). The company has assembled a large library of clinical neurobiological data and is applying its AI platform to understand specific pathways underlying these different conditions and to identify biomarkers that can help develop targeted drugs. The company’s lead program is studying a kappa opioid receptor antagonist for the potential treatment of mood disorders; the drug candidate is expected to enter Phase 2 clinical testing in early 2020.

 

New tools for approaches to detecting targetable disease subtypes are also being employed to help bring precision medicine to neurological conditions. For example, NeuroPointDX, the neurological division of Stemina Biomarker Discovery, has conducted the largest study to date of the metabolism of children with autism. Their ongoing metabolomic analysis has already led to the identification of several unique metabolic subtypes predictive of high risk of ASD. The company is now preparing to conduct a clinical trial of a potential dietary supplement for one of these patient subtypes that is characterized by dysregulation of the metabolism of certain amino acids important to typical neurodevelopment.

 

Similarly, Moleculera Labs has developed a blood panel for markers of neuroinflammation as an aid to the diagnosis and treatment of neuropsychiatric conditions, such as obsessive-compulsive disorder, tics, attention-deficit hyperactivity disorder, depression, and behaviors associated with ASD occurring as a result of post-infection autoimmune attacks on the brain.