Use of an exposure-response model to aid early drug development of an oral sphingosine 1-phosphate receptor modulator

• Published in: Journal of clinical pharmacology Vol. 49; no. 1; p. 50
• Main Authors: Rohatagi, Shashank, Zahir, Hamim, Moberly, James B, Truitt, Kenneth E, Inaba, Shin-ichi, Shimozato, Takaichi, Carrothers, T J
• Format: Journal Article
• Language: English
• Published: England 01.01.2009
• Subjects: Administration, Oral; Adolescent; Adult; Amino Alcohols - administration & dosage; Amino Alcohols - pharmacokinetics; Amino Alcohols - pharmacology; CD4-Positive T-Lymphocytes - cytology; CD4-Positive T-Lymphocytes - drug effects; Cohort Studies; Dose-Response Relationship, Drug; Double-Blind Method; Humans; Lymphocyte Count; Lymphocyte Subsets - cytology; Lymphocyte Subsets - drug effects; Lymphocytes - cytology; Lymphocytes - drug effects; Lysophospholipids - administration & dosage; Lysophospholipids - pharmacokinetics; Lysophospholipids - pharmacology; Male; Middle Aged; Models, Biological; Prodrugs - administration & dosage; Prodrugs - pharmacokinetics; Prodrugs - pharmacology; Pyrroles - administration & dosage; Pyrroles - pharmacokinetics; Pyrroles - pharmacology; Receptors, Lysosphingolipid - physiology; Sphingosine - administration & dosage; Sphingosine - analogs & derivatives; Sphingosine - pharmacokinetics; Sphingosine - pharmacology; Young Adult
• ISSN: 0091-2700
• DOI: 10.1177/0091270008325672

Pharmacokinetic (PK) and exposure-response modeling of a selective sphingosine 1-phosphate receptor-1 modulator (CS-0777) was conducted in an iterative process to guide early clinical development decisions. A model based on preclinical data from monkeys was extrapolated to humans to support a single ascending dose (SAD) study. The model was updated after each cohort, providing guidance on both maximal inhibition and time to recovery for lymphocyte counts. A 2-compartment PK model with first-order absorption and elimination was found to describe the monkey and human datasets. The relationship between lymphocyte counts and active metabolite (M-1) concentrations was modeled via an indirect response model, whereby M-1 inhibited the reentry of lymphocytes to the circulation. The indirect-response model based on SAD data had an Imax of approximately 85% and an IC50 of 0.24 ng/mL. Additionally, based on SAD data, similar models were developed for lymphocyte subsets, including CD4 cells. Subsequently, simulations were utilized to design a multiple ascending dose study with adaptive dosing regimens that would meet targeted pharmacodynamic (PD) response thresholds (eg, minimum 40% reduction in lymphocytes) while maintaining CD4 counts above a reasonable safety threshold. In conclusion, model-based development and use of adaptive designs for dose optimization can reduce the time and number of subjects needed in early clinical development.