SINGAPORE - A group of researchers in Singapore will tap their artificial intelligence (AI) platform to evaluate 12 locally available drugs to derive a combination of drugs that can be used to treat Covid-19.
The interactive digital platform, known as IDentif.AI, leverages AI to calculate the most effective combination of drugs - along with their respective doses - from more than 530,000 possibilities.
The researchers, from the Institute for Digital Medicine (WisDM) at the National University of Singapore (NUS)'s Yong Loo Lin School of Medicine, found in a study conducted in April that the most optimal drug combination comprises remdesivir, lopinavir and ritonavir, which are used to treat patients with human immunodeficiency virus (HIV).
Their findings were validated using live virus in cell culture, and it has been pre-emptively cleared for a clinical trial in Taiwan, should the need arise.
Professor Dean Ho, director of WisDM, told The Straits Times on Friday (Oct 16) that the "combination enabled near complete inhibition of the virus, but remdesivir isn't readily available, which is a challenge that we will address in follow-on studies".
Instead, the research team will be evaluating a set of 12 locally accessible drugs, ranging from anti-virals to targeted therapies and other agents.
Remdesivir, which is among the very few authorised treatments for Covid-19 thus far, inhibited the virus by 15 per cent as initially validated by IDentif.AI, so it is used as a benchmark for the new set of drug combinations, said Prof Ho.
A clinical trial by the World Health Organisation on Thursday (Oct 15) found that remdesivir had little or no effect on Covid-19 patients' length of hospital stay or chances of survival, although the results of the trial have not yet been reviewed.
"In addition, remdesivir has to be administered in hospitals through an intravenous infusion, which could make it challenging to deploy if there are many Covid-19 patients in the community. Therefore, we are looking to investigate drugs available in tablet form so they can be consumed orally," said Prof Ho.
This also makes it easier to dispense and can potentially be administered at home, he added.
The study will be conducted on the new set of drug combinations in November.
Aside from using AI to optimise drug combination for infectious diseases like Covid-19, the researchers were also able to leverage their technology to offer personalised treatments for cancer patients.
Using another platform known as Curate.AI, drug doses given to patients can be modulated to produce optimal results throughout their entire duration of care.
Assistant Professor Raghav Sundar, from the NUS Department of Medicine and WisDM and a consultant with the Department of Haematology-Oncology at the National University Cancer Institute, Singapore, said: "Drug dosing in cancer treatments are typically based on the degree of side effects experienced by the patient. With Curate.AI, each patient's recommended dose is calibrated using clinical data generated from their individual response to treatment."
For instance, in a pilot clinical study conducted in collaboration with a US-based hospital, a patient with advanced prostate cancer was recommended a 50 per cent reduction in dose of an investigational inhibitor drug for increased efficacy. The patient was able to resume an active lifestyle as the lower dose was found to be more tolerable.
Similarly, a patient in Singapore who had advanced cancer was prescribed a reduced dose of nab-paclitaxel - a type of chemotherapy drug - which managed to stop his cancer from progressing and reduced the size of his lung tumour. This allowed the patient to continue treatment for a much longer duration compared with other patients who were given the same drug.
These findings have led to a clinical pilot trial which is currently recruiting patients.
In another expanded study using Curate.AI, the team leveraged a software known as the Multi-Attribute Task Battery, which comes in the form of a game with varying levels of intensity.
It will be trialled on patients who have received radiation therapy for brain cancer to provide diagnostic information on each patient's responses, which can be used to personalise treatment.
The software could also potentially be used as therapy for patients facing declining cognitive and physical abilities, such as diabetes, cognitive decline and Alzheimer's disease.