Richard Ho

Transporters and drug therapy: implications for drug disposition and disease.

The efficacy of drug therapy results from the complex interplay of multiple processes that govern drug disposition and response. Most studies to date have focused on the contribution of drug‐metabolizing enzymes to the drug disposition process. However, over the past decade, it has become increasingly apparent that carrier‐mediated processes, or transporters, also play critical roles in the overall disposition of numerous drugs in clinical use. In addition to their roles in xenobiotic transport, drug transporters often mediate important physiologic functions via transport of endogenous substrates such as amino acids, bile acids, and hormones that are critical for maintenance of normal homeostasis. In this review we focus on the emerging field of transporter proteins in relation to the drug disposition process, with particular emphasis on clinical implications of transporters to drug‐drug interactions and subsequent development of adverse effects, interindividual variability in drug response, and human disease.

Effect of drug transporter genotypes on pravastatin disposition in European- and African-American participants.

Objective Our aims were to evaluate the effects of polymorphisms in the hepatic drug uptake transporter organic anion transporting polypeptide 1B1 (OATP1B1, SLCO1B1) and efflux transporters multidrug resistance-associated protein 2 (MRP2, ABCC2), bile salt export pump (BSEP, ABCB11), and breast cancer-related protein (BCRP, ABCG2) on single-dose pravastatin pharmacokinetics in healthy European- and African-American participants. Methods The pharmacokinetics of a single oral 40 mg dose of pravastatin was determined in 107 participants (69 European-Americans and 38 African-Americans). Participants were genotyped for known OATP1B1, MRP2, BSEP, and BCRP polymorphisms. Baseline serum total and unconjugated plasma bilirubin concentrations were also determined. Results OATP1B1 genotypes were ethnicity-dependent with a 521C allele frequency of ∼15% in European-Americans and ∼1% in African-Americans. SLCO1B1 521TC genotype was associated with significantly higher pravastatin area under the curve [AUC(0–5)] (P=0.01) and Cmax values (P<0.05). When analyzed by diplotype, SLCO1B1*1a/*15 (N=8) participants exhibited 45 and 80% higher AUC values than SLCO1B1*1a/*1a (N=29) (P=0.013) and SLCO1B1*1b/*1b (N=34) (P=0.001) carriers, respectively. SLCO1B1*15/*15 (N=2) participants exhibited 92 and 149% higher AUC values than SLCO1B1*1a/*1a (P=0.017) and SLCO1B1*1b/*1b (P=0.011) carriers, respectively. European-Americans had significantly higher plasma pravastatin AUC(0–5) (P=0.01) and Cmax values (P=0.009) than African-Americans. Neither ABCC2, ABCB11, nor ABCG2 genotypes were associated with differences in pravastatin pharmacokinetics. We did not observe an effect of SLCO1B1 genotype on baseline total or unconjugated bilirubin levels. Conclusion SLCO1B1 genotype, in particular the 521C allele, had a significant effect on the pharmacokinetics of pravastatin. Even when adjusted for the presence of the SLCO1B1 521C or 388G variant allele, European-Americans demonstrated significantly higher pravastatin AUC and Cmax values than African-Americans.

Breast cancer resistance protein (ABCG2) and drug disposition : intestinal expression, polymorphisms and sulfasalazine as an in vivo probe

Breast cancer resistance protein (BCRP) is an efflux transporter expressed in tissues that act as barriers to drug entry. Given that single nucleotide polymorphisms (SNPs) in the ABCG2 gene encoding BCRP are common, the possibility exists that these genetic variants may be a determinant of interindividual variability in drug response. The objective of this study is to confirm the human BCRP-mediated transport of sulfasalazine in vitro, evaluate interindividual variation in BCRP expression in human intestine and to determine the role of ABCG2 SNPs to drug disposition in healthy patients using sulfasalazine as a novel in vivo probe. To evaluate these objectives, pinch biopsies were obtained from 18 patients undergoing esophagogastro–duodenoscopy or colonoscopy for determination of BCRP expression in relation to genotype. Wild-type and variant BCRP were expressed in a heterologous expression system to evaluate the effect of SNPs on cell-surface trafficking. A total of 17 healthy individuals participated in a clinical investigation to determine the effect of BCRP SNPs on sulfasalazine pharmacokinetics. In vitro, the cell surface protein expression of the common BCRP 421 C>A variant was reduced in comparison with the wild-type control. Intestinal biopsy samples revealed that BCRP protein and mRNA expression did not significantly differ between patients with 34GG/421CC versus patients with 34GG/421CA genotypes. Remarkably, in subjects with 34GG/421CA genotype, sulfasalazine area under the concentration–time curve was 2.4-fold greater compared with 34GG/421CC subjects (P<0.05). This study links commonly occurring SNPs in BCRP with significantly increased oral sulfasalazine plasma exposure in humans. Accordingly, sulfasalazine may prove to have utility as in vivo probe for assessing the clinical impact of BCRP for the disposition and efficacy of drugs.

Interplay between the Nuclear Receptor Pregnane X Receptor and the Uptake Transporter Organic Anion Transporter Polypeptide 1A2 Selectively Enhances Estrogen Effects in Breast Cancer

The ligand-activated nuclear receptor pregnane X receptor (PXR) is known to play a role in the regulated expression of drug metabolizing enzymes and transporters. Recent studies suggest a potential clinically relevant role of PXR in breast cancer. However, the relevant pathway or target genes of PXR in breast cancer biology and progression have not yet been fully clarified. In this study, we show that mRNA expression of organic anion transporter polypeptide 1A2 (OATP1A2), a transporter capable of mediating the cellular uptake of estrogen metabolites, is nearly 10-fold greater in breast cancer compared with adjacent healthy breast tissues. Immunohistochemistry revealed exclusive expression of OATP1A2 in breast cancer tissue. Interestingly, treatment of breast cancer cells in vitro with the PXR agonist rifampin induced OATP1A2 expression in a time-dependent and concentration-dependent manner. Consistent with its role as a hormone uptake transporter, induction of OATP1A2 was associated with increased uptake of estrone 3-sulfate. The rifampin response was abrogated after small interfering RNA targeting of PXR. We then identified a PXR response element in the human OATP1A2 promoter, located approximately 5.7 kb upstream of the transcription initiation site. The specificity of PXR-OATP1A2 promoter interaction was confirmed using chromatin immunoprecipitation. Importantly, we used a novel potent and specific antagonist of PXR (A-792611) to show the reversal of the rifampin effect on the cellular uptake of E(1)S. These data provide important new insights into the interplay between a xenobiotic nuclear receptor PXR and OATP1A2 that could contribute to the pathogenesis of breast cancer and may also prove to be heretofore unrecognized targets for breast cancer treatment.

Phase I Pharmacokinetic and Pharmacodynamic Study of 17-N-Allylamino-17-Demethoxygeldanamycin in Pediatric Patients with Recurrent or Refractory Solid Tumors: A Pediatric Oncology Experimental Therapeutics Investigators Consortium Study

Purpose: Heat shock protein 90 (Hsp90) is essential for the posttranslational control of many regulators of cell growth, differentiation, and apoptosis. 17-N-Allylamino-17-demethoxygeldanamycin (17-AAG) binds to Hsp90 and alters levels of proteins regulated by Hsp90. We conducted a phase I trial of 17-AAG in pediatric patients with recurrent or refractory neuroblastoma, Ewings sarcoma, osteosarcoma, and desmoplastic small round cell tumor to determine the maximum tolerated dose, define toxicity and pharmacokinetic profiles, and generate data about molecular target modulation. Experimental Design: Escalating doses of 17-AAG were administered i.v. over 1 to 2 h twice weekly for 2 weeks every 21 days until patients experienced disease progression or toxicity. harmacokinetic and pharmacodynamic studies were done during cycle 1. Results: Fifteen patients were enrolled onto dose levels between 150 and 360 mg/m2; 13 patients were evaluable for toxicity. The maximum tolerated dose was 270 mg/m2. DLTs were grade 3 transaminitis and hypoxia. Two patients with osteosarcoma and bulky pulmonary metastases died during cycle 1 and were not evaluable for toxicity. No objective responses were observed. 17-AAG pharmacokinetics in pediatric patients were linear; clearance and half-life were 21.6 ± 6.21 (mean ± SD) L/h/m2 and 2.6 ± 0.95 h, respectively. Posttherapy increases in levels of the inducible isoform of Hsp70, a marker of target modulation, were detected in peripheral blood mononuclear cells at all dose levels. Conclusion: 17-AAG was well tolerated at a dose of 270 mg/m2 administered twice weekly for 2 of 3 weeks. Caution should be used in treatment of patients with bulky pulmonary disease.

Ritonavir, Saquinavir, and Efavirenz, but Not Nevirapine, Inhibit Bile Acid Transport in Human and Rat Hepatocytes

Human immunodeficiency virus-infected patients on antiretroviral drug therapy frequently experience hepatotoxicity, the underlying mechanism of which is poorly understood. Hepatotoxicity from other compounds such as bosentan and troglitazone has been attributed, in part, to inhibition of hepatocyte bile acid excretion. This work tested the hypothesis that antiretroviral drugs modulate hepatic bile acid transport. Ritonavir (28 μM), saquinavir (15 μM), and efavirenz (32 μM) inhibited [3H]taurocholate transport in bile salt export pump expressing Sf9-derived membrane vesicles by 90, 71, and 33%, respectively. In sandwich-cultured human hepatocytes, the biliary excretion index (BEI) of [3H]taurocholate was maximally decreased 59% by ritonavir, 39% by saquinavir, and 20% by efavirenz. Likewise, in sandwich-cultured rat hepatocytes, the BEI of [3H]taurocholate was decreased 100% by ritonavir and 94% by saquinavir. Sodium-dependent and -independent initial uptake rates of [3H]taurocholate in suspended rat hepatocytes were significantly decreased by ritonavir, saquinavir, and efavirenz. [3H]Taurocholate transport by recombinant NTCP and Ntcp was inhibited by ritonavir (IC50 = 2.1 and 6.4 μM in human and rat, respectively), saquinavir (IC50 = 6.7 and 20 μM, respectively), and efavirenz (IC50 = 43 and 97 μM, respectively). Nevirapine (75 μM) had no effect on bile acid transport in any model system. In conclusion, ritonavir, saquinavir, and efavirenz, but not nevirapine, inhibited both the hepatic uptake and biliary excretion of taurocholate.

Polymorphic variants in the human bile salt export pump (BSEP; ABCB11): functional characterization and interindividual variability.

Objectives Our aims were to identify and functionally characterize coding region nonsynonymous single nucleotide polymorphisms in the hepatic efflux transporter, bile salt export pump (BSEP; ABCB11), and to assess interindividual variability in BSEP expression. Methods We identified 24 single nucleotide polymorphisms, including nine nonsynonymous variants, in ABCB11 from genomic DNA of ∼250 ethnically diverse healthy individuals using denaturing high-performance liquid chromatography analysis and DNA sequencing. Wild type and variant BSEP were generated and functionally characterized for taurocholate transport activity in vitro in HeLa cells using a recombinant vaccinia-based method. BSEP expression was assessed by real-time mRNA analysis, western blot analysis, and immunofluorescence confocal microscopy. Results For the most part, polymorphisms were rare and ethnic-dependent. In vitro functional studies revealed several rare variants, including 616A>G, 1674G>C, 1772A>G, and 3556G>A, to be associated with significantly impaired taurocholate transport activity while the 890A>G variant trended towards impaired function but was not statistically significant. The 3556G>A variant was associated with reduced cell surface to total protein expression compared with wild-type BSEP. Expression of BSEP by mRNA and protein analysis was determined from a bank of human liver samples. Wide interindividual variability was noted in both mRNA (19-fold) and protein (31-fold) expression levels. The common variant 1331T>C was associated with significantly reduced hepatic BSEP mRNA levels. Conclusion Accordingly, our study indicates there are functionally relevant polymorphisms in ABCB11 which may be of potential relevance in the predisposition to acquired liver disorders such as drug-induced cholestasis.

Identification of novel functional organic anion-transporting polypeptide 1B3 polymorphisms and assessment of substrate specificity.

Objective The uptake carrier organic anion-transporting polypeptide 1B3 (OATP1B3, gene SLCO1B3) is involved in the hepatic clearance of xenobiotics including statins, taxanes, and mycophenolic acid. We thought to assess the SLCO1B3 coding region for yet unidentified polymorphisms and to analyze their functional relevance. Methods We used DNA of ethnically diverse individuals for polymerase chain reaction, and determined polymorphisms by sequencing or temperature-dependent capillary electrophoresis. We then created variant OATP1B3 expression plasmids by site-directed mutagenesis, which were transiently expressed and functionally characterized in human cervical carcinoma (HeLa) cells using radiolabeled substrates. Results We identified six nonsynonymous polymorphisms including novel variants such as 439A>G (Thr147Ala), 767G>C (Gly256Ala), 1559A>C (His520Pro), and 1679T>C (Val560Ala). Allelic frequencies occurred to be ethnicity-dependent, with the latter observed only in African–Americans (3.6%). After expression in HeLa cells, His520Pro, Val560Ala, and Met233Ile or Met233Ile_Ser112Ala haplotype variants showed decreased uptake activity compared with wild type for cholecystokinin-8 and rosuvastatin, but not for atorvastatin. Kinetic cholecystokinin-8 analysis showed reduced Vmax without altering Km. His520Pro and Val560Ala exhibited decreased total and plasma membrane protein expressions. Val560 mapped onto a structural model of OATP1B3 showed that this is a key region for substrate–transporter interaction. His520 resides in a predicted extracellular region thought to be critical to the pH-dependent component of OATP1B3 activity. Loss of activity at pH 7.4 and 8.0 relative to pH 6.5 was significantly greater for the Pro520 variant. Conclusion OATP1B3 polymorphisms that result in altered expression, substrate specificity, and pH-dependent activity may be of potential relevance to hepatic clearance of substrate drugs in vivo.

Patterns and severity of vincristine‐induced peripheral neuropathy in children with acute lymphoblastic leukemia

Vincristine, a critical component of combination chemotherapy treatment for pediatric acute lymphoblastic leukemia (ALL), can lead to vincristine‐induced peripheral neuropathy (VIPN). Longitudinal VIPN assessments were obtained over 12 months from newly diagnosed children with ALL (N = 128) aged 1–18 years who received vincristine at one of four academic childrens hospitals. VIPN assessments were obtained using the Total Neuropathy Score‐Pediatric Vincristine (TNS©‐PV), National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE©), Balis© grading scale, and Pediatric Neuropathic Pain Scale©–Five (PNPS©‐5). Of children who provided a full TNS©‐PV score, 85/109 (78%) developed VIPN (TNS©‐PV ≥4). Mean TNS©‐PV, grading scale, and pain scores were low. CTCAE©‐derived grades 3 and 4 sensory and motor VIPN occurred in 1.6%/0%, and 1.9%/0% of subjects, respectively. VIPN did not resolve in months 8–12 despite decreasing dose density. VIPN was worse in older children. Partition cluster analysis revealed 2–3 patient clusters; one cluster (n = 14) experienced severe VIPN. In this population, VIPN occurs more commonly than previous research suggests, persists throughout the first year of treatment, and can be severe.

Measuring vincristine-induced peripheral neuropathy in children with acute lymphoblastic leukemia

Background: Vincristine-induced peripheral neuropathy (VIPN) is difficult to quantify in children. Objective: The study objective was to examine the reliability, validity, and clinical feasibility of several VIPN measures for use in children with acute lymphoblastic leukemia. Interventions/Methods: Children (n = 65) aged 1 to 18 years receiving vincristine at 4 academic centers participated in the study. Baseline and pre–vincristine administration VIPN assessments were obtained using the Total Neuropathy Score–Pediatric Vincristine (TNS©-PV), the National Cancer Institute Common Terminology Criteria for Adverse Events, the Balis grading scale, and the FACES Pain Scale. The TNS-PV scores (n = 806) were obtained over 15 weeks. Blood was obtained at several time points to quantify pharmacokinetic parameters. Results: Cronbach’s &agr; for a reduced TNS-PV scale was .84. The TNS-PV scores correlated with cumulative vincristine dosage (r = 0.53, P = 0.01), pharmacokinetic parameters (r = 0.41, P = 0.05), and grading scale scores (r range = 0.46–0.52, P = .01). FACES scores correlated with the TNS-PV neuropathic pain item (r = 0.48; P = .01) and were attainable in all ages. A 2-item V-Rex score (vibration and reflex items) was the most responsive to change (effect size = 0.65, P < 0.001). The TNS-PV scores were attainable in 95% of children 6 years or older. Conclusions: The TNS-PV is reliable and valid for measuring VIPN. It is sensitive to change over time (15 weeks) and feasible for use in children 6 years or older. Implications for Practice: The TNS-PV may be a useful tool for assessing vincristine toxicity in children with acute lymphoblastic leukemia.