Background: Cellular P-gp effluxes HIV protease inhibitors (PIs) out of cells and some tissues (e.g., the brain). CD4+ and CD8+ T cells express P-gp. Although high PI concentrations inhibit P-gp activity in vitro, it is unclear whether this occurs in vivo. Nucleoside reverse transcriptase inhibitors (NRTIs) do not inhibit P-gp. We measured the effect of patient plasma (obtained during steady state nelfinavir (NFV) dosing) on healthy volunteer lymphocyte P-gp activity.

Methods: One plasma sample was obtained from each of 17 adults during clinic visits (population sampling approach). Subjects were receiving NFV (1250 mg q12-h) plus NRTIs, but no other PI or non-NRTI. Time of last dose was by patient report. Dye efflux inhibition (DEI) assays were performed by incubating HIV-negative donor whole blood cells with the P-gp substrate DiOC₂(3). Cells were washed, then incubated at 37^°C with 90% patient plasma (with 5% 1M HEPES, pH 7.4). Mean fluorescence intensity (MFI) was quantified by flow cytometry.

Results: The 17 patients included 5 (29%) women, 4 (22%) African Americans, HIV-1 RNA was below limits of quantitation in 8 (47%), CD4+ T cells ranged from 175 to 819 cells/mm³, and sampling times from 0.5-9 hours post-dose. Some samples modestly inhibited CD4+ and CD8+ T cell P-gp. Samples from 2-6 hours post-dose inhibited CD4+ T P-gp (MFI 18.7±0.52) more than those from other times (MFI 16.6±0.61, p = 0.02), and inhibited CD8+ P-gp (MFI 15.0 ± 0.65) more than other samples (MFI 12.6±0.40, p=0.009). Inhibitory activity did not correlate with CD4+ T cell count, viral load, or demographics.

Conclusions:  Twice-daily NFV is associated with time-dependent variability in plasma P-gp inhibitory activity. Times of increased activity likely correlate with peak NFV and/or M8 metabolite concentrations. The relevance of modest P-gp inhibition during NFV therapy is uncertain. Ongoing studies are comparing P-gp inhibitory activity with NFV and M8 concentrations in these and additional patients.