Lack of correlation between oral midazolam (M) and dextromethorphan (DM) with administration of nefazodone (N) and ketoconazole (K)

DM (30 mg) and M (10 mg) were administered to 17 subjects at baseline and after 11 days of K 200 mg or N titrated to 200 mg BID. DM challenges (8 hr urine collection) were performed 2 days prior to M. Blood sampling and Symbol Digit Modalities Test (SDMT) were obtained before and for 24 hours after M. % Change from baseline (mean±SD and range) M CL DM/3MM SDMT AUEC_(0-8) K -83±11 81±96 -34±25 (-9610-64) (-44 to 261) (-74 to-2) N -76±15 55±104 -22±12 (-90 to -41 ) (-38 to 255) (-47 to -4) Compared to baseline, a significant decrease in M oral CL was found for K (p=0.005) and N (p=0.0004). No increase in DM/3MM was noted after treatment with K (p=0.11) or N (p=0.52). Consequently, there was no correlation between % change M oral CL and DM/3MM after K (r=0.28, p=0.46) or N (r=0.27, p=0.51). SDMT AUEC_0-8 was consistent with CYP3A4 inhibition of M CL with N and K. Due to the large intersubject variability in DM/3MM, esp in the presence of potent CYP3A4 inhibitors, DM may not be a specific probe for CYP3A4.