Based a machine learning approach to investigate the factors influencing nirmatrelvir/ritonavir exposure in human plasma: a multicenter, observational study

We identified a set of variables that affect the treatment of N/R through therapeutic drug monitoring and established a machine learning model capable of predicting N/R concentrations with satisfactory performance. These findings provide a basis for integrating TDM with multivariable prediction tools to personalize N/R dosing and improve medication safety.