BACKGROUND: Pazopanib has shown clinical activity against multiple tumour types and is generally well tolerated. However, isolated elevations in transaminases and bilirubin have been observed. This study examined polymorphisms in molecules involved in pharmacokinetic and pharmacodynamic pathways of pazopanib and their association with hepatic dysfunction. METHODS: Twenty-eight polymorphisms in 11 genes were evaluated in pazopanib-treated renal cell carcinoma patients. An exploratory analysis was conducted in 116 patients from a phase II study; a replication study was conducted in 130 patients from a phase III study. RESULTS: No polymorphisms were associated with alanine aminotransferase elevation. The Gilbert's uridine-diphosphoglucuronate glucuronosyltransferase 1A1 (UGT1A1) TA-repeat polymorphism was significantly associated with pazopanib-induced hyperbilirubinemia in the phase II study. This association was replicated in the phase III study (Po0.01). Patients with TA6/TA6, TA6/TA7, and TA7/TA7 genotypes experienced median bilirubin increases of 0.31, 0.37, and 0.71 Â upper limit of the normal range (ULN), respectively. Of the 38 patients with hyperbilirubinemia (X1.5 Â ULN), 32 (84%) were either TA7 homozygotes (n ¼ 18) or TA7 heterozygotes (n ¼ 14). For TA7 homozygotes, the odds ratio (95% CI) for developing hyperbilirubinemia was 13.1 (5.3 -32.2) compared with other genotypes. CONCLUSIONS: The UGT1A1 polymorphism is frequently associated with pazopanib-induced hyperbilirubinemia. These data suggest that some instances of isolated hyperbilirubinemia in pazopanib-treated patients are benign manifestations of Gilbert's syndrome, thus supporting continuation of pazopanib monotherapy in this setting.

Pazopanib, an oral inhibitor of vascular endothelial growth factor receptor, platelet-derived growth factor receptor, and c-kit kinases, inhibits multiple cytochrome P450 (CYP450) enzymes in vitro. This study in patients with advanced cancer evaluated the effect of pazopanib on CYP450 function by comparing the pharmacokinetics of CYP-specific probe drugs in the presence and absence of pazopanib. The probes used included midazolam (CYP3A specific), warfarin (CYP2C9 specific), omeprazole (CYP2C19 specific), caffeine (CYP1A2 specific), and dextromethorphan (CYP2D6 specific). The estimated ratios of the geometric means (90% confidence interval (CI)) for the area under the curve to the last measurable point (AUC(0-t)) for these probe drugs with/without pazopanib were as follows: midazolam, 1.35 (1.18-1.54); omeprazole, 0.81 (0.59-1.12); caffeine, 1.00 (0.77-1.30); and S-warfarin, 0.93 (0.84-1.03). The geometric least-squares (LS) mean ratio of urine dextromethorphan:dextrorphan ranged from 1.33 (0-4-h interval) to 1.64 (4-8-h interval). The data suggest that pazopanib is a weak inhibitor of CYP3A4 and CYP2D6 and has no effect on CYP1A2, CYP2C9, and CYP2C19 in patients with advanced cancer.

No abstract

Introduction: Electrocardiographic (ECG) criteria for left ventricular hypertrophy (LVH), such as the Cornell and Sokolow-Lyon voltage criteria were derived from Western populations. However, their utility and accuracy for diagnosing echocardiographic LVH in Asian populations is unclear. The objective of this study was to assess the accuracy of ECG criteria for LVH in Asians and to determine if alternative gender-specific ECG cut-offs may improve its diagnostic accuracy. Materials and Methods: ECG and echocardiographic assessments were performed on 668 community-dwelling Asian adults (50.9% women; 57 ± 10 years) in Singapore. The accuracy of ECG voltage criteria was compared to echocardiographic LVH criteria based on the American Society of Echocardiography guidelines, and Asian ethnicity and gender-specific partition values. Results: Echocardiographic LVH was present in 93 (13.6%) adults. Cornell criteria had low sensitivity (5.5%) and high specificity (98.9%) for diagnosing LVH. Modified gender specific cut-offs (18 mm in women, 22 mm in men) improved sensitivity (8.8% to 17.5%, 0% to 14.7%, respectively) whilst preserving specificity (98.2% to 94.2%, 100% to 95.8%). Similarly, Sokolow-Lyon criteria had poor sensitivity (7.7%) and high specificity (96.1%) for diagnosing LVH. Lowering the cut-off value from 35 mm to 31 mm improved the sensitivity in women from 3.5% to 14% while preserving specificity at 94.2%. A cut-off of 36 mm was optimal in men (sensitivity of 14.7%, specificity of 95.5%). Conclusion: Current ECG criteria for LVH derived in Western cohorts have limited sensitivity in Asian populations. Our data suggests that ethnicity- and gender-specific ECG criteria may be needed. Key words: Cornell, Ethnicity, Sokolow-Lyon, Voltage