Breast Cancer Resistance Protein and P-Glycoprotein Influence In Vivo Disposition of 11C-Erlotinib.

UNLABELLED (11)C-erlotinib is a PET tracer to distinguish responders from nonresponders to epidermal growth factor receptor-targeted tyrosine kinase inhibitors and may also be of interest to predict distribution of erlotinib to tissues targeted for treatment. The aim of this study was to investigate if the known interaction of erlotinib with the multidrug efflux transporters breast cancer resistance protein (humans, ABCG2; rodents, Abcg2) and P-glycoprotein (humans, ABCB1; rodents, Abcb1a/b) affects tissue distribution and excretion of (11)C-erlotinib and has an influence on the ability of (11)C-erlotinib PET to predict erlotinib tissue distribution at therapeutic doses. METHODS Wild-type and Abcb1a/b or Abcg2 knockout mice underwent (11)C-erlotinib PET/MR scans, with or without the coinjection of a pharmacologic dose of erlotinib (10 mg/kg) or after pretreatment with the ABCB1/ABCG2 inhibitor elacridar (10 mg/kg). Integration plot analysis was used to determine organ uptake (CLuptake) and biliary excretion (CLbile) clearances of radioactivity. RESULTS (11)C-erlotinib distribution to the brain was restricted by Abcb1a/b and Abcg2, and CLuptake into the brain was only significantly increased when both Abcb1a/b and Abcg2 were absent (wild-type mice, 0.017 ± 0.004 mL/min/g of tissue; Abcb1a/b((-/-))Abcg2((-/-)) mice, 0.079 ± 0.013 mL/min/g of tissue; P < 0.001). The pretreatment of wild-type mice with elacridar increased CLuptake into the brain to levels comparable to Abcb1a/b((-/-))Abcg2((-/-)) mice (0.090 ± 0.007 mL/min/g of tissue, P < 0.001). The absence of Abcb1a/b and Abcg2 led to a 2.6-fold decrease in CLbile (wild-type mice, 0.025 ± 0.005 mL/min/g of tissue; Abcb1a/b((-/-))Abcg2((-/-)) mice, 0.0095 ± 0.001 mL/min/g of tissue; P < 0.001). There were pronounced differences in distribution of (11)C-erlotinib to the brain, liver, kidney, and lung and hepatobiliary excretion into intestine between animals injected with a microdose and pharmacologic dose of erlotinib. CONCLUSION ABCG2, ABCB1, and possibly other transporters influence in vivo disposition of (11)C-erlotinib and thereby affect its distribution to normal and potentially also tumor tissue. Saturable transport of erlotinib leads to nonlinear pharmacokinetics, possibly compromising the prediction of erlotinib tissue distribution at therapeutic doses from PET with a microdose of (11)C-erlotinib. The inhibition of ABCB1 and ABCG2 is a promising approach to enhance brain distribution of erlotinib to increase its efficacy in the treatment of brain tumors.