Population PKPD modelling of biomarkers ANP and big ET-1 for two neutral endopeptidase inhibitors
M. Marchand (1), E. Fuseau (1), A.C. Heatherington (2), S. Sultana (2), P. R. Hidi (2), M. Boucher (2), P. Ellis (2), S.W. Martin (2)
(1) EMF Consulting, Aix-en-Provence, France ; (2) Pfizer Ltd, Sandwich, UK
Background/aims: Pfizer Ltd is exploring treatments for female sexual dysfunction (FSD) via novel mechanisms. Neutral endopeptidase (NEP) inhibitors may have utility in the treatment of FSD by improving vaginal blood flow and lubrication. Two NEP inhibitors UK-447,841 and UK-505,749 have undergone pharmacological evaluation to assess their effect on plasma big endothelin (Big ET-1) and atrial natriuretic peptide (ANP) levels. The objectives were to select a reliable soluble biomarker for the two NEP inhibitors, to build a suitable PKPD model to assist with dose selection in subsequent efficacy studies and to understand the translation from in vitro models to clinical biomarkers.
Methods: Healthy subjects (N = 57) received oral doses of placebo or UK-447,841 in a single escalating dose cross-over study (3 to 800 mg) or in a multiple dose parallel group study (100 to 800 mg). Healthy subjects (N = 27) received single escalating oral doses of placebo or UK-505,749 in a cross-over design (0.1 - 540 mg). Pharmacokinetic (PK) and pharmacodynamic (PD) (Big ET-1 and ANP) samples were collected during the 3 studies. Population PK and PKPD analyses were carried out using NONMEM, V. Age, sex, weight and baseline biomarker concentration were considered as covariates for PK and PKPD parameters. Model evaluation was performed using visual performance predictive checks.
Results: PK properties were similar for both molecules. For both molecules, a dose response was evident in concentration-time profiles for Big ET-1 and ANP, respectively. Biomarker response versus drug concentration plots revealed hysteresis. Initially, separate PKPD models (effect compartment) for Big ET-1 and ANP were fitted to the data simultaneously for both molecules. UK-505,749 showed 10-fold greater potency than UK-447,841 for both biomarkers. Dependency of ANP baseline on Big ET-1 baseline values was established leading to the development of a combined Big ET-1/ANP indirect response model. Relationships between biomarkers were assessed; the age effect on both ANP production and Emax was estimated.
Conclusions: Big ET-1 has ideal characteristics of a soluble biomarker: it demonstrates dose-concentration-response, time-linearity and reproducibility of effect with similar Emax for two NEP inhibitors; ANP was a reasonable biomarker. PKPD analysis confirms prior knowledge about the relationship between the selected biomarkers of NEP activity and the effect of age on ANP activity.