Using AI and Machine Learning for Repurposing Drugs as Potential COVID-19 Therapies
Recently, a news report suggested that a team of AI scientists have made a belligerent effort to identify hundreds of new drugs as a potential COVID-19 therapy. COVID-19 is caused by the novel coronavirus, SAR COV-2, whose origins have been speculated to be a wet market in Wuhan in China. AI engineers and pharmaceutical companies are working together to develop drug discovery pipeline using Computational models using AI and machine learning algorithms. However, in the race to find potential drug therapies for COVID-19 pandemic, repurposing old drugs and known vaccines have been found to be the safest, and probably also the cheapest, considering the economic downfall brought about by the Wuhan-originated virus.
In a recent announcement, leading data analytics and business intelligence firm GlobalData underlines the role of AI as a promising “enabler” for biopharmaceutical companies to expedite the drug repurposing process. Reason to pull AI in to the race to find drugs: By using data and analytics of the potency of various drugs, AI can be used to understand the chances of any adverse reactions to the susceptible age groups and populations, including in non-human populations where the SARS-COV-2 virus could be found to exist and spread.
Recently, on 4 July 2020, WHO accepted the recommendation from the Solidarity Trial’s International Steering Committee to discontinue the trial’s hydroxychloroquine and lopinavir/ritonavir arms.
At the time of this announcement, Venkata Naveen, Senior Disruptive Tech Analyst at GlobalData, commentes,
“Typically developing a new drug takes almost a decade and costs anywhere between US$2-US$3bn. But now biopharmaceutical companies are in dire need to accelerate the entire drug development process given that COVID-19 cases and deaths are mounting every day. Under these circumstances, AI technologies allow companies to significantly shorten the preclinical drug identification and design process from several years to a few days or months.”
Remdesivir was previously tested as an Ebola treatment. It has generated promising results in animal studies for Middle East Respiratory Syndrome (MERS-CoV) and severe acute respiratory syndrome (SARS), which are also caused by coronaviruses, suggesting it may have some effect in patients with COVID-19.
The Innovation Explorer database of GlobalData’s Disruptor Intelligence Center reveals how AI startups are assisting in the identification of suitable molecules that target COVID-19, this enabling the biopharmaceutical industry to repurpose existing and already approved drugs to treat COVID-19.
Here are the top AI companies helping pharma companies meet the current expectations of repurposing drugs for COVID-19 therapies.
London-based BenevolentAI used its AI-based drug discovery platform to identify drugs that could disrupt certain viral entry pathways of COVID-19, thus inhibiting the replication process. Researchers screened inhibitors of AP2-associated protein kinase 1 (AAK1).
AAK1 is a known regulator of endocytosis and its disruption may inhibit the entry of the virus into cells. In a process that took just three days to complete, over 370 AAK1 inhibitors were identified, 47 of which are approved for use, with six showing the most promise. This includes baricitinib, currently marketed by Eli Lilly as Olumiant. Olumiant is used to treat rheumatoid arthritis in adults, as a potential treatment candidate for COVID-19.
The startup extracts new hypothesis by employing AI to its knowledge graphs, which establishes billions of machine curated relationships between diseases and drugs. Eli Lilly initiated a Phase III randomized controlled trial of baricitinib in 400 hospitalized adults with COVID-19 in the US, Europe and Latin America. The National Institute of Allergy and Infectious Diseases (NIAID) is also running a study of baricitinib with remdesivir in the second phase of the Adaptive COVID-19 Treatment Trial (ACTT-2).
Japanese startup Elix harnessed a variety of neural networks to predict the chemical properties of molecules that can help to destroy COVID-19. The startup’s AI tool screened a library of 350 million drug-like molecules to find potential drugs to treat COVID-19. Among the drug candidates identified by Elix’s AI tool was remdevisir, which recently received emergency use authorization from the FDA to treat COVID-19 patients.
Singapore’s Gero leveraged AI to quickly screen already existing drug molecules to treat COVID-19. Some of the identified drugs that are potentially effective include niclosamide, nitazoxanide, afatinib and reserpine. The startup is currently exploring partnerships with pharma companies and research institutes for the commercial development and immediate start of clinical trials of the identified drug molecules.
In addition, startups such as Repurpose.AI and Atomwise are actively forging partnerships with global research institutes to use their AI-powered predictive models to find new drug molecules for the treatment of COVID-19.
Naveen concludes: “AI startups are helping pharma companies to digitally bridge the gap between thousands of repurposed drug molecules, clinical trials and final drug authorization. The COVID-19 crisis has propelled the interest of venture capital firms in AI startups due to their potential to accelerate the process of finding new therapies. For example, Atomwise, Gero, Insitro and Exscientia have all raised hefty sums in various stages of funding in the first two quarters of 2020.”
With AI’s effectiveness, efforts to repurpose known drugs such as Remdesivir in COVID-19 therapies could prove to be successful, even as vaccines are yet to make their presence felt in the community studies.
Earlier this month, a report was published titled, “Predicting novel drugs for SARS-CoV-2 using machine learning from a >10 million chemical space”. The open access report was published by Joel Kowalewski and Anandasankar Ray who suggested that repurposing drugs for COVID-19 therapies are best suited for short term approval, and efficacious drugs that come out in the future could be used for long-term COCID-19 treatment.
The report states, “We have developed a machine learning drug discovery pipeline to identify several drug candidates. First, we collect assay data for 65 target human proteins known to interact with the SARS-CoV-2 proteins, including the ACE2 receptor. Next, we train machine learning models to predict inhibitory activity and use them to screen FDA registered chemicals and approved drugs (~100,000) and ~14 million purchasable chemicals.”
The abstract continued, “We filter predictions according to estimated mammalian toxicity and vapor pressure. Prospective volatile candidates are proposed as novel inhaled therapeutics since the nasal cavity and respiratory tracts are early bottlenecks for infection. We also identify candidates that act across multiple targets as promising for future analyses. We anticipate that this theoretical study can accelerate testing of two categories of therapeutics: repurposed drugs suited for short-term approval, and novel efficacious drugs suitable for a long-term follow up.”
As we continue to battle the COVID-19 pandemic and the numerous other infectious diseases, the role of AI and machine learning in repurposing drugs could only enhance as the time passes.
To share your insights with us, please write to firstname.lastname@example.org