Repetitive behavioral disorders such as obsessive-compulsive disorder (OCD), Tic Disorder and Tourette Syndrome have been challenging to treat because they involve considerable complexity. For example, the symptoms of one disorder can vary among individuals as does the coexistence or non-existence of other behavior disorders such as anxiety and depression. Further, the treatment of OCD, Tic Disorder and Tourette Syndrome varies depending on the disorders present in an individual patient, how long the disorder’s traits have been present, how individual patients respond to treatment dosages and the efficacy of treatment.
Historically and even modernly, medical professionals have found themselves faced with an array of potential treatments that may or may not work alone or in combination with other drugs. Compounding the frustration among healthcare providers is the poor level of R&D investments governments are making which stalls progress. Rather than inventing a new drug that would take years to hit the market, it’s possible to use artificial intelligence (AI), and specifically deep learning, to discover how existing drugs with regulatory approval can be repurposed for other applications.
One such example is clemastine, an antihistamine, which has proven to prevent and reduce the occurrence of repetitive behaviors.
Repetitive disorders have been difficult to treat
OCD is a disorder in which unwanted thoughts and behaviors are repeated. The compulsive behaviors mainly appear in the form of cleaning, confirmation, delay, ordering and other repetitive behaviors. Typically, serotonin uptake inhibitors such as clomipramine, fluoxetine, and fluvoxamine are used to treat OCD, but the side effects may include anxiety, diarrhea, insomnia, vomiting and weight gain. OCD-related disorders include, but are not limited to, storage disorders, hair pulling disorders and skin-picking disorders, for example.
Tic disorders include chronic motor or vocal tics, provisional tic disorders and Tourette Syndrome. They are characterized by vocal tics or sudden rapid, repetitive, rhythmic or sterotyped movements that persist for more than 1 year, while Public Tic Disorder is characterized by the presence of one or more motor or vocal tics that are present for less than 1 year.
Tourette Syndrome is a neurological abnormality that repeats unconscious and uncontrolled sounds or actions and is characterized by multiple motor and vocal tics for more than one year. Tics are uncontrolled and appear in a wide variety of forms and have repetitive characteristics. In this case, dopamine receptor antagonists such as haloperidol, fluphenazine and pimozide are used for treatment, but the side effects can include cognitive decline, discomfort, and extrapyramidal symptoms.
Practitioners often find themselves frustrated because they are unable to achieve a satisfactory therapeutic effect using individual drugs or even a combination of drugs that are available today. In addition, some of those drugs are restricted due to their side effects. Quite often, doctors find themselves in continuous trial-and-error mode, trying this treatment, that treatment, or a combination of treatments, or adjusting drug dosages to reduce a patient’s unwanted repetitive behavior.
Recently, many gene mutations involved in the generation and function of nerve cell synapses have been discovered in patients with neuropsychiatric diseases and synaptopathy. These gene mutations are being recognized as one of the key mechanisms of neuropsychiatric diseases.
Clemastine, an antihistamine also referred to as meclastine, is used to treat allergic rhinitis, sneezing, dermatitis and itching. However, its therapeutic effect on repetitive behavior disorders has not been reported previously. Using artificial intelligence (AI), and specifically deep learning, it is possible to discover that clemastine effectively reduces repetitive behavior patterns in mice. In humans, it could prove to be the missing piece necessary to prevent or reduce OCD, Tic Disorder and Tourette Syndrome.
Why clemastine holds promise as an active ingredient
A pharmaceutical composition comprising clemastine or a pharmaceutically acceptable salt thereof can be used as an active ingredient for the prevention or treatment of repetitive behavioral disorders.
Clemastine is a first-generation antihistamine of the amino alkyl ether class that binds to the histamine H1 receptor and blocks the action of endogenous histamine. It has been approved by the United States Food and Drug Administration (FDA) and is currently a generic drug (e.g., tabegil).
Clemastine is a pharmaceutically acceptable salt that includes inorganic and organic acids and bases that are acceptable physiologically. Specifically, AI has identified that the pharmaceutically acceptable clemastine sale is clemastine fumarate. The pharmaceutical composition is formulated in the form of aerosols, capsules, emulsions, external preparations, granules, sterile injection solutions, suppositories, suspensions, syrups and tablets. Clemastine or other pharmaceutically acceptable salts can also be used in food including beverages, gum, health supplements, tea and vitamin complexes.
The pharmaceutically effective clemastine dosage is preferably 0.1mg/day to 5 mg/day, with the latter administered once to several times per day by dividing the dose. How much an individual patient requires depends on the person’s administration route, age, concurrent drugs, condition, severity of symptoms and sex. The clemastine or a pharmaceutically acceptable salt thereof and one or more drugs, can be administered simultaneously, sequentially, or in reverse order, such as serotonin reuptake inhibitors for OCD and dopamine receptor antagonists for Tourette Syndrome, for example.
Lab experiments with mice have already proven that the that an SH3 and multiple ankyrin repeat domains (SHANK3) scaffolding protein deficiency is a factor that induces repetitive behavior and that clemastine alleviates the repetitive behavior increase caused by SHANK3 deficiency. It is our hope that AI-based discovery equates to a step-change in treatment that is beneficial to patients and their prescribing doctors.
Why AI-based drug repurposing is wise
Drug discovery is a long, drawn-out process because traditional research takes considerable time and so does regulatory approval. In the meantime, patients suffering from a disorder or malady may be suffering unnecessarily. Drug repurposing is a faster alternative. However, the chemical details of the drug, its potential interactions with other drugs and side effects must be considered when navigating the universe of possibilities. AI can do all that at scale.
For drug repurposing, AI analyzes the original purpose and chemical composition of a drug as well as a universe of diseases and/or disorders to find correlations and perhaps even causation. Deep learning helps identify the “unknown unknowns” such as clemastine can be used to prevent or treat OCD and other repetitive behavior disorders.
AI is a powerful tool that can be used for patentable drug discovery and repurposing. Given the scale and speed at which AI can operate, it has the potential to advance the state of the art of drug discovery and repurposing at a much faster pace than has been possible using traditional tools and humans. Moreover, this improved efficiency has the potential to shrink drug discovery and repurposing time frames required to treat patients in new and improved ways.
About Jin Han Kim
Jin Han Kim is the Co-Founder, Chairman, and CEO of Standigm, a workflow AI-driven drug discovery company headquartered in Seoul, South Korea and subsidiaries in Cambridge, U.K. Standigm has proprietary AI platforms encompassing novel target identification to compound design, to generate commercially valuable drug pipelines.. Kim is an expert in developing AI platforms to identify new drug candidates and has led the company since its inception in May 2015.
Before founding Standigm, Kim worked as a Senior Research Scientist at the Samsung Advanced Institute of Technology, where he developed an AI algorithm for DNA damage and recovery mechanisms. He previously worked as a software developer for NCSOFT and for Namo Interactive. He also served as the Director of Drug Informatics at the Korean Society of Medical Informatics from January 2017-January 2020. Kim earned his Ph.D. in Artificial Intelligence at the University of Edinburgh, his M.S. in Artificial Intelligence at Seoul National University, and his bachelor’s degree in Applied Biology and Chemistry from Seoul National University. Kim speaks Korean and English.