A research team from the Korea Advanced Institute of Science and Technology has developed an AI model to predict adverse reactions between oral anti-COVID-19 medication and prescription drugs.

Researchers from KAIST's Department of Biochemical Engineering made a new version of the DeepDDI AI-based drug interaction prediction model to check how ritonavir and nirmatrelvir, two components of Paxlovid by pharmaceutical giant Pfizer, would interact with prescription drugs. 

The new model, DeepDDI2, can compute for and process a total of 113 drug-drug interaction types, a press release noted.

It was later found that Paxlovid interacts with approximately 2,248 prescription drugs: 1,403 drugs with ritonavir and 673 drugs with nirmatrelvir. 

The researchers then proposed alternative options for prescription drugs with high adverse reactions with Paxlovid: they found 124 drugs with low potential adverse reactions with ritonavir and 239 drugs with nirmatrelvir.

WHY IT MATTERS

COVID-19 patients with comorbidities, such as high blood pressure and diabetes, are likely to be taking antiviral medication with other drugs. However, drug-drug interactions and adverse drug reactions with Paxlovid "have not been sufficiently analysed," the KAIST researchers said. Utilising AI technology, they then set out to explore how the continued use of antiviral therapy with other drugs may lead to serious and unwanted complications. 

THE LARGER TREND

Pfizer is inching close to getting the US Food and Drug Administration's full approval for Paxlovid. This comes as an advisory panel last week voted to recommend the approval as it deems the drug safe and effective. The company received emergency use approval for Paxlovid from the regulatory body in December 2021. Following the advisers' vote, it is expected that the US FDA will make a final decision on its full approval by May. 

ON THE RECORD

"The results of this study are meaningful at times like when we would have to resort to using drugs that are developed in a hurry in the face of urgent situations like the COVID-19 pandemic. [With DeepDDI2], it is now possible to identify and take necessary actions against adverse drug reactions caused by drug-drug interactions very quickly," KAIST Professor Sang Yup Lee said in a statement.