Lack of pharmacokinetic interaction between nelfinavir and nevirapine

We note the article by Merry et al. [1] in this issue, suggesting that nevirapine (NVP) reduces the plasma area under the concentration-time curve (AUC) of nelfinavir (NFV) by 50%, as measured in seven patients. We would like to address several concerns over the design of this study and the interpretation of the resultant data in light of our own pharmacokinetic study of 23 patients [2]. NFV is metabolized by several cytochrome P450 enzyme isoforms, including CYP3A4, CYP2C19, CYP2C9, and CYP2D6. NVP is also metabolized by the cytochrome P450 system, and activates the CYP2B6 isoform to a greater extent than the CYP3A4 isoform. The interaction of these two drugs in vivo is not intuitive, hence the need for a carefully designed and controlled pharmacokinetic study. NFV, when taken alone, reaches peak plasma trough concentrations by day 2. Over the ensuing days of therapy, plasma concentrations decline approximately 40–50%, suggesting that NFV induces its own metabolism. Plasma concentrations stabilize by day 6, indicating that a steady state has been achieved (Fig. 1). For this reason, we measured pre-NVP concentrations of NFV on day 7. After the addition of NVP, with the standard dose escalation, NFV concentrations were measured on day 36, after NVP had achieved its steady state. We found no significant difference in NFV plasma AUC over an 8 h dose interval in the absence versus presence of NVP (mean ± SD, 19.7 ±7.5 versus 19.1 ± 8.2 mg × h/l, respectively; Fig. 2). Likewise, NVP had no significant effect on NFV plasma maximum or minimum concentrations [2]. In contrast, Merry et al. [1] measured initial NFV levels after 2 days of therapy, at which point NFV levels would be predicted to be at or near their peak. The reduction in NFV levels after the addition of NVP in their study is thus likely to be due to the expected autoinduction of NFV metabolism that would have occurred even in the absence of NVP.Fig. 1: . Mean nelfinavir (NFV) predose trough concentrations in 12 healthy volunteers during 6 days of NFV at 750 mg every 8 h. Mean concentration (± SD) of NFV on day 6 (2.1 ± 0.8 mg/ml) was 48% of mean concentration on day 2 (3.9 ± 1.6 mg/ml). In another study of 11 healthy volunteers, trough concentrations were equivalent on days 6 and 10 of therapy with NFV 750 mg every 8 h (2.5 ± 1.0 mg/ml on each day).Fig. 2: . Median steady-state plasma concentrations of nelfinavir (NFV) on day 7 (NFV alone) and day 36 [NFV plus nevirapine (NVP)] for the 23 patients in our study.As noted by Merry and coworkers, in other studies [3,4] NVP reduced plasma concentrations of the protease inhibitors saquinavir and indinavir, two drugs predominantly metabolized by CYP3A4. However, those results cannot necessarily be extrapolated to NFV or any other drug for which metabolic clearance is not clearly dominated by CYP3A4. Our observation that NFV levels are not reduced by NVP is consistent with similar results reported for NFV plus efavirenz, which is also reported to be a modest cytochrome P450 (CYP3A4) inducer [5,6]. Our study also presented clinical data that demonstrated potent antiretroviral activity of NFV at the standard 750 mg three times daily dosage, in combination with stavudine and NVP. By week 9, 84% of patients were suppressed to below 400 HIV RNA copies/ml, as measured by the Roche Amplicor assay [2]. Although the estimation of in vivo minimum effective concentration (MEC) from in vitro inhibitory concentration data is problematic, these virologic response data support the conclusion of Merry et al. [1] that NFV minimum concentration values in combination with NVP are at or above the MEC for NFV. In light of the data from our study of 23 patients, which showed no pharmacokinetic interaction between NFV and NVP and the clinical activity of the combination when utilizing the standard NFV dosage of 750 mg three times daily, we conclude that an increase in NFV dosage when given with NVP is not warranted.