Apr 27 2020
In light of the growing crisis due to the COVID-19 pandemic, scientists across the world are trying to identify the most effective treatment to fight against the poorly understood coronavirus that is responsible for the spread of this disease.
A research team led by Professor Dean Ho has harnessed the power of artificial intelligence to dramatically accelerate the discovery of drug combination therapies. Image Credit: National University of Singapore
Conventionally, during the emergence of dangerous new viral and bacterial infections, the immediate response is to devise a treatment methodology that integrates many different drugs. But such a process is not only time-consuming but also difficult. While the combinations of drugs are selected sub-optimally, and dose selection simply becomes a matter of trial and error.
This inefficient and expensive method of creating a treatment methodology becomes a problem when a quick response is critical for dealing with a global pandemic emergency and when resources had to be conserved.
Keeping this aspect in mind, Professor Dean Ho, Director of the N.1 Institute for Health at National University of Singapore (NUS) and Head of NUS Biomedical Engineering, headed a multidisciplinary research team to develop a groundbreaking artificial intelligence (AI) platform called “IDentif.AI,” short for Identifying Infectious Disease Combination Therapy with Artificial Intelligence, to significantly boost the efficiency of this development.
The results of the research were published in the Advanced Therapeutics journal on April 16th, 2020.
Drawbacks of Traditional Drug Screening
Traditionally, when choosing drugs for a specific treatment, the growth of bacteria or viruses is examined in response to different types of promising candidates. Increasing dosages of the drugs are given to the viruses or bacteria until the highest prevention of their growth is noted.
More drugs are subsequently added simultaneously to intensify the effect. But such techniques become ineffective when many drugs are concurrently examined as candidates. Moreover, these methods usually lead to positive outcomes in the case of in vitro studies, but the same is not seen in human studies.
If 10 or more drugs are examined, it is virtually impossible to study the effects of all the possible drug combinations and dosages needed to identify the best possible combination using traditional methods.
Dean Ho, Professor and Director of N.1 Institute for Health and Institute for Digital Medicine, National University of Singapore
Besides, in conventional screening, if a drug obtained from a pool of candidate treatments appears to have no evident impact on the microorganism, then most usually this drug will not be considered anymore.
“However, if this drug is systematically combined with more drugs, each at the correct doses, this could very well result in the best possible combination. Unfortunately, this remarkable level of required precision cannot be arbitrarily derived,” Professor Ho added.
Using Artificial Intelligence to Optimize Drug Therapies
To prevent the disadvantages of the conventional development of drug combination therapy, Professor Ho and his research team, along with colleagues from Shanghai Jiao Tong University, leveraged the processing power of the AI platform.
Twelve drugs, which were carefully selected by the researchers, are all found to be promising candidates for combating infection in lung cells induced by the vesicular stomatitis virus (VSV). The researchers then applied the IDentif.AI to prominently decrease the number of experiments required to intercept the entire range of combinations as well as the optimal dosages of all the 12 drugs.
Using IDentif.AI, we took three days to identify multiple optimal drug regimens out of billions of possible combinations that reduced the VSV infection to 1.5 per cent with no apparent adverse impact. This speed and accuracy in discovering new drug combination therapies is completely unprecedented.
Dean Ho, Professor and Director of N.1 Institute for Health and Institute for Digital Medicine, National University of Singapore
Most significantly, the researchers observed that the optimal dose of the leading drug combination was seven times more effective when compared to that of the sub-optimal doses. This demonstrated the crucial significance of the optimal drug and dose detection.
In a similar way, when one drug was replaced from the leading drug combination, and this novel combination, when given at sub-optimal doses, was found to be 14 times less effective.
“There is a notion in drug discovery that if you discover the right molecule, the work is done. Our results with IDentif.AI prove that it is critically important to think about how the drug is developed into a combination and subsequently administered,” added Professor Ho.
He further continued, “How do you combine it with the right drugs? How do you dose this drug properly? Answering these questions can dramatically increase efficacy at the clinical stage of drug development.”
Apart from verifying the IDentif.AI platform, the research included the contributions of a group of experts in operations research and healthcare economics from NUS Business School and KPMG Global Health and Life Sciences Centre of Excellence, and also insights from global health security and surveillance experts from MRIGlobal and EpiPointe LLC.
The researchers surmised that the methods that employ AI, like IDentif.AI platform, which can quickly improve the repurposing of drugs under severe economic conditions amidst the ongoing pandemics, could have a crucial role to play in enhancing the patient outcomes when compared to that of traditional methods.
Using IDentif.AI Against COVID-19 and More
After demonstrating the effectiveness of the IDentif.AI platform to quickly offer treatments for infectious diseases, the researchers have now set their sights on the COVID-19 virus.
As the development of vaccines and antibody therapies for COVID-19 are ongoing, we will need a rapid therapeutic strategy that addresses the virus which may evolve over time. Our strength is that we can perform one experiment and come out with a list of drug combinations for treatment within days.
Dean Ho, Professor and Director of N.1 Institute for Health and Institute for Digital Medicine, National University of Singapore
Professor Ho continued, “And in time, if patients do not respond well to the first combinations of drugs, we can derive new combinations, we can derive new combinations within days to re-optimise their care. Our platform is useful to address the possibility that patients will need different drug combinations depending on when treatment was initiated, and if downstream infection with a different strain occurs,” concluded Professor Ho.
In addition, the IDentif.AI platform could be directly deployed to deal with any other infectious diseases in the days to come.
“When an aggressive pathogen hits, a rapid response is needed, and this response may need to evolve quickly as the pathogen evolves. Now, with IDentif.AI, we will be ready,” concluded Professor Ho.