Role of ATP-binding cassette and solute carrier transporters in erlotinib CNS penetration and intracellular accumulation

Abstract

Purpose: To study the role of drug transporters in central nervous system (CNS) penetration and cellular accumulation of erlotinib and its metabolite, OSI-420.

Experimental design: After oral erlotinib administration to wild-type and ATP-binding cassette (ABC) transporter-knockout mice (Mdr1a/b(-/-), Abcg2(-/-), Mdr1a/b(-/-)Abcg2(-/-), and Abcc4(-/-)), plasma was collected and brain extracellular fluid (ECF) was sampled using intracerebral microdialysis. A pharmacokinetic model was fit to erlotinib and OSI-420 concentration-time data, and brain penetration (P(Brain)) was estimated by the ratio of ECF-to-unbound plasma area under concentration-time curves. Intracellular accumulation of erlotinib was assessed in cells overexpressing human ABC transporters or SLC22A solute carriers.

Results: P(Brain) in wild-type mice was 0.27 ± 0.11 and 0.07 ± 0.02 (mean ± SD) for erlotinib and OSI-420, respectively. Erlotinib and OSI-420 P(Brain) in Abcg2(-/-) and Mdr1a/b(-/-)Abcg2(-/-) mice were significantly higher than in wild-type mice. Mdr1a/b(-/-) mice showed similar brain ECF penetration as wild-type mice (0.49 ± 0.37 and 0.04 ± 0.02 for erlotinib and OSI-420, respectively). In vitro, erlotinib and OSI-420 accumulation was significantly lower in cells overexpressing breast cancer resistance protein (BCRP) than in control cells. Only OSI-420, not erlotinib, showed lower accumulation in cells overexpressing P-glycoprotein (P-gp) than in control cells. The P-gp/BCRP inhibitor elacridar increased erlotinib and OSI-420 accumulation in BCRP-overexpressing cells. Erlotinib uptake was higher in OAT3- and OCT2-transfected cells than in empty vector control cells.

Conclusion: Abcg2 is the main efflux transporter preventing erlotinib and OSI-420 penetration in mouse brain. Erlotinib and OSI-420 are substrates for SLC22A family members OAT3 and OCT2. Our findings provide a mechanistic basis for erlotinib CNS penetration, cellular uptake, and efflux mechanisms.

Fig. 1

Compartmental pharmacokinetic model for erlotinib and OSI-420. D1, duration of zero-order absorption; ka, first-order absorption rate constant; V_ERL/F, apparent volume of distribution of erlotinib; CL_ERL/F, apparent clearance of erlotinib; V_OSI/FE, apparent volume of distribution of OSI-420; CL_OSI/FE, apparent clearance of OSI-420; k₂₄/k₄₂, first-order rate constants of erlotinib in the brain extracellular fluid; k₃₅/k₅₃, first-order rate constants of OSI-420 in the brain extracellular fluid; V_Brain, apparent volume of distribution of the brain extracellular fluid.

Fig.2

Plasma concentration-time plots of erlotinib (■) and OSI-420 (○) after oral administration of a single dose of 20 mg/kg erlotinib. Model-fitted curves are represented for erlotinib (solid line) and OSI-420 (dashed line)

Fig. 3

Brain penetration of erlotinib and OSI-420 after a single oral dose of erlotinib. A, penetration of unbound erlotinib and OSI-420 to brain ECF expressed as P_Brain (mean ± SD from 4 to 7 mice). * P < 0.05 and ** P < 0.01, Mann Whitney test of each knockout model compared with wild-type mice. B, representative unbound erlotinib concentration-time plots in brain ECF (○) and plasma (?) in wild-type, Abcg2^−/−, Mdr1a/b^−/−, Mdr1a/b^−/−Abcg2^−/− and Abcc4^−/− mice. Model curves from individual predicted parameters are represented for plasma (dashed line) and brain ECF (solid line).

Fig. 4

Erlotinib and OSI-420 intracellular accumulation in vitro in cell lines expressing efflux transporters. Values are the percentage of the maximum accumulation (mean ± SD; n=4-6) in control cells (Saos2-pcDNA for BCRP and MRP4 or LLC-PK1 for P-gp). A, time-course of drug accumulation in cell lines. Control, cells lines transfected with an empty vector control; Transfected, cell lines transfected with a vector expressing the indicated transporter; Transfected + Elacridar, transporter-transfected cells treated with 4 μM of the P-gp/BCRP inhibitor elacridar. B, intracellular accumulation data from 15 min and 30 min time-points combined. *P < 0.001 as compared to accumulation in control cells (ANOVA with posthoc Dunnet's test).

Fig. 5

Transport of erlotinib, A and OSI-420, B by human organic ion transporters. Results are shown for drug accumulation in HEK293 cells after 5 minutes incubation in each cell line. Columns represent (mean ± SD) 8 to 18 observations per expressed transporter, and are expressed as percentage of their respective control (white bar). Only one control bar is shown for clarity purposes. The contribution of each transporter towards erlotinib or OSI-420 uptake was established by comparing data obtained in HEK293 cells overexpressing the transporter and HEK293 cells transfected with an empty vector.* P < 0.05; ** P < 0.01; *** P < 0.001 versus control, one-way ANOVA was performed followed by a post Dunnett test.

Fig. 6

Proposed model for the role of efflux and uptake transporters in erlotinib CNS penetration. A, schematic diagram for endothelial cells with transporters localized on either apical border (facing blood) or basolateral border (facing brain ECF). B, schematic diagram of the choroid plexus forming the BCSFB. Transporters are localized on either the apical border (facing CSF) or basolateral border (facing blood). The arrows represent the direction of drug transport.