Be Careful What You Ask for: Challenges of Predicting Human Clearance for a Low Metabolic Turnover Compound, ELND006

The discovery of a new chemical entity (NCE) suitable for CNS drug development is a challenging endeavor, as many different properties require simultaneous optimization.  One of these properties, metabolic stability, is a surrogate for in vivo clearance (CL).  A goal for many discovery project teams is to select an NCE with the greatest metabolic stability; however, very high metabolic stability can also pose challenges for development.  The prediction of human CL, with the objective of selecting compounds with half-lives suitable for once daily dosing, has become an important selection criterion for many discovery organizations. 

ELND006, a potent and APP selective gamma secretase inhibitor demonstrating CNS reduction of Beta Amyloid in rodents and non-human primates¹ was selected from a series of NCEs based, in part, on its metabolic stability in both NADPH and UDPGA supplemented liver microsomes.  Further experiments aimed at calculating the predicted human intrinsic CL (i.e., incubations in human liver microsomes and hepatocytes) revealed that this compound exhibited very low turnover (<5%).   In an attempt to improve turnover, incubations with increased microsomal protein and cell numbers were attempted; however, consistent with the lipophilicity (logD 3.9) of ELND006, high non specific binding leading to low free concentrations (Fu,microsomes = 0.06) likely obscured any benefit of increased enzyme levels.  While it was evident from these data that ELND006 would exhibit low CL, the actual magnitude of the CL could not be accurately determined.  Phase I clinical data revealed that the average apparent terminal half-life in healthy elderly subjects following oral dosing ranged from 220 to 306 hours², yielding a total body clearance (CL/F) of 1 mL/min/kg.   After estimations for bioavailability, this value is suggestive of a human CL of 0.2 to 0.3 mL/min/kg. 

The availability of micropatterned co-cultures of hepatocytes and fibroblasts in a 96-well plate format³ prompted a retrospective analysis to determine whether the human CL of ELND006 could be predicted.  This system mimics elements of in vivo liver physiology and increases cell viability and function of drug metabolizing enzymes on a scale of days rather than hours.  ELND006 was metabolized under these conditions resulting in a prediction of human clearance (0.9 mL/min/kg) that was within three to four-fold of that observed in vivo.  In addition, the CL values for the positive controls, warfarin and quinidine, agreed well with literature reports.   Our data suggests that while further investigation is warranted, micropatterned co-cultures of hepatocytes may offer an alternative to empirical methods for predicting the human CL low turnover compounds in drug discovery. 

1)    Brigham, et al.  2010.  Pharmacokinetic and pharmacodynamic investigation of ELND006, a novel APP-selective gamma-secretase inhibitor, on amyloid-β concentrations in the brain, CSF and plasma of multiple nonclinical species following oral administration. ICAD. Honolulu, Hawaii. S546

2)    Liang et al. 2011. A Phase I dose escalation study to evaluate the safety, tolerability, PK/PD of multiple oral dosing of D6 in healthy elderly subjects. ICAD. Paris, France. P2-476

3)    Khetani and Bhatia. 2008. Microscale culture of human liver cells for drug development. Nature Biotechnology 26, 120 - 126