Login / Signup

PBPK Model of Coproporphyrin I: Evaluation of the Impact of SLCO1B1 Genotype, Ethnicity, and Sex on its Inter-Individual Variability.

Hiroyuki TakitaShelby BarnettYueping ZhangKarelle MénochetHong ShenKayode OgungbenroAleksandra Galetin
Published in: CPT: pharmacometrics & systems pharmacology (2021)
Coproporphyrin I (CPI) is an endogenous biomarker of OATP1B activity and associated drug-drug interactions. In this study, a minimal physiologically-based pharmacokinetic model was developed to investigate the impact of OATP1B1 genotype (c.521T>C), ethnicity, and sex on CPI pharmacokinetics and interindividual variability in its baseline. The model implemented mechanistic descriptions of CPI hepatic transport between liver blood and liver tissue and renal excretion. Key model parameters (e.g., endogenous CPI synthesis rate, and CPI hepatic uptake clearance) were estimated by fitting the model simultaneously to three independent CPI clinical datasets (plasma and urine data) obtained from white (n = 16, men and women) and Asian-Indian (n = 26, all men) subjects, with c.521 variants (TT, TC, and CC). The optimized CPI model successfully described the observed data using c.521T>C genotype, ethnicity, and sex as covariates. CPI hepatic active was 79% lower in 521CC relative to the wild type and 42% lower in Asian-Indians relative to white subjects, whereas CPI synthesis was 23% higher in male relative to female subjects. Parameter sensitivity analysis showed marginal impact of the assumption of CPI synthesis site (blood or liver), resulting in comparable recovery of plasma and urine CPI data. Lower magnitude of CPI-drug interaction was simulated in 521CC subjects, suggesting the risk of underestimation of CPI-drug interaction without prior OATP1B1 genotyping. The CPI model incorporates key covariates contributing to interindividual variability in its baseline and highlights the utility of the CPI modeling to facilitate the design of prospective clinical studies to maximize the sensitivity of this biomarker.
Keyphrases
  • machine learning
  • electronic health record
  • emergency department
  • big data
  • gene expression
  • drug induced
  • genome wide
  • rna seq
  • data analysis