A new study has utilized machine learning to discover the most effective combinations of drugs that prevent the recurrence of COVID-19 following an initial infection. However, the study indicates that these combinations vary among individuals, highlighting the need for personalized approaches. According to the researchers from UC Riverside, individual characteristics, such as age, weight, and pre-existing conditions, play a crucial role in determining the drug combinations that effectively reduce the rates of COVID-19 recurrence.
The study utilized real-world data obtained from a hospital in China. The researchers discovered that COVID-19 patients in China were mandated to undergo post-hospitalization quarantine in government-operated hotels; this measure provided an opportunity for a more systematic assessment of reinfection rates. The study included data from over 400 COVID-19 patients, with an average age of 45. Treatment involved various combinations of antiviral, anti-inflammatory, and immune-modulating drugs, like interferon or hydroxychloroquine.
The success of different combinations among various demographic groups can be attributed to the virus’s behavior. COVID-19 suppresses interferon, a protein produced by cells to impede invading viruses. According to the co-author of the study, Jiayu Liao, individuals who had weaker immune systems before contracting COVID-19 needed immune-boosting medications to effectively fight against the virus. Conversely, younger individuals typically exhibit hyperactive immune responses to the infection, which can result in excessive inflammation of tissues and, in severe cases, even mortality. As a result, younger patients necessitate the inclusion of immune suppressants in their treatment regimen.
Liao urges reconsidering age and medical conditions when choosing treatments, as current practices often overlook variations. Despite advancements in our understanding of COVID-19 and the effectiveness of vaccines in reducing mortality, there remains a significant knowledge gap regarding treatments and prevention of reinfections. Xinping Cui hopes that the findings of this study will be applied to address issues surrounding recurrence.
The study’s findings were published in the Journal of Frontiers in Artificial Intelligence.
