Case details of enrolled victim drugs
A total of 33 drugs, approved by the FDA between 2011 and 2020, were enrolled for evaluation of predictive performance after meeting specific screening criteria. To be eligible, these drugs were required to have available pharmacokinetic profiles when administered alone or co-administered with ketoconazole, as well as available ADME properties. The structure of these 33 compounds is presented in Figure S3, and their characteristics and statistics are listed in Table 1. The DDI potentials of the drugs were divided into two categories based on the 90% confidence interval of AUCR, where 15 out of the 33 compounds had AUCR<2, accounting for 45% of the total. Table 1 shows that there were no significant differences between the two categories in terms of logP, Peff, BP, Vss,fu , and ClR. However, compounds with AUCR>2 tended to have higher Clliver(ClCYP3A4 + Clother) and in vivofm . A correlation analysis of Clliver, in vitro fm , and AUCR was performed using a 3-D plot as showed in the Figure 1, which indicated that compounds with higher Clliver andin vitro fm had higher AUCRs. Out of 33 compounds co-administered with ketoconazole, six victim drugs experienced an increase in exposure of more than three times. All six of these victim drugs have a high Clliver (>15 L h-1) and a high fm(>75%). Conversely, the 18 drugs with lower clearance rates (<15 L h-1) among the 33 compounds showed an increase in exposure of less than three times after being co-administered with ketoconazole, regardless offm .