Safety, tolerability, pharmacokinetics, and efficacy of AMG 403, a human anti-nerve growth factor monoclonal antibody, in two phase I studies with healthy volunteers and knee osteoarthritis subjects

Abstract

Introduction: Nerve growth factor plays a key role in the pathology of osteoarthritis (OA) related chronic pain. The aim of these studies was to evaluate the safety, tolerability, pharmacokinetics, and clinical response of AMG 403, a human anti-nerve growth factor monoclonal antibody, in healthy volunteers and subjects with knee OA.

Methods: Two phase I, randomized, placebo-controlled, double-blind studies were conducted. The single-ascending dose study randomized healthy volunteers (n = 48) 3:1 to receive AMG 403 (1, 3, 10, or 30 mg intravenously; or 10 or 30 mg subcutaneously; n = 8 per group) or placebo. The multiple-ascending dose study randomized knee OA subjects (n = 18) 3:1 to receive AMG 403 (3, 10, or 20 mg subcutaneously once monthly for four doses) or placebo. Safety, tolerability, and pharmacokinetics (PK) were assessed for both studies. Patient's and physician's disease assessments and total WOMAC score were determined in knee OA subjects.

Results: AMG 403 appeared to be well-tolerated after single and multiple doses, except for subject-reported hyperesthesia, pain, and paresthesia (mild to moderate severity). These treatment-emergent neurosensory events showed evidence of reversibility and a possible dose-dependence. Three serious adverse events were reported in AMG 403 treated subjects, but were not considered treatment related. AMG 403 PK was linear with an estimated half-life of 19.6 to 25.8 days. After multiple doses, AMG 403 PK showed modest accumulation (≤2.4-fold increase) in systemic exposure. Knee OA diagnosis, body weight, and anti-drug antibody development did not appear to affect AMG 403 PK. Patient's and physician's disease assessments and total WOMAC score showed improvement in AMG 403 treated knee OA subjects compared with placebo.

Conclusions: AMG 403 was generally safe and well-tolerated in both healthy volunteers and knee OA patients, and exhibited linear pharmacokinetics. Preliminary clinical efficacy was observed in knee OA subjects.

Trial registration: ClinicalTrials.gov NCT02348879 . Registered 23 December 2014. Clintrials.gov NCT02318407 . Registered 2 December 2014.

Figures

Fig. 1

AMG 403 concentration-time profiles following single and repeat dose administration. AMG 403 was administered intravenously (IV) or subcutaneously (SC) and serum AMG 403 levels were measured pre-dose through end of study. Each point represents the mean (+SD) AMG 403 concentration. a Healthy volunteers received a single IV or SC dose of AMG 403 (1–30 mg, LLOQ = 0.0016 μg/mL). b Patients with knee osteoarthritis received a once-monthly SC dose of AMG 403 (3–20 mg, LLOQ = 0.0051 μg/mL) for up to four dose administrations. Arrows represent dose administration

Fig. 2

Visual predictive check of the AMG 403 population pharmokinetics (PK) model. A two-compartment model was fit simultaneously to individual AMG 403 PK data from healthy volunteers (HV) and patients with knee osteoarthritis (OA). The model parameter estimates were used to simulate 1000 trials for a visual predictive check. The predicted median PK (solid line) and 80 % prediction interval (shaded area) are shown with the observed individual concentrations (solid dot). Note the different x-axis scales for the single-dose and multiple-dose regimens. IV intravenous, SC subcutaneous

Fig. 3

Covariate effects on AMG 403 clearance (CL) and (Vc). Possible covariate effects were explored for individual CL and Vc parameters derived from simultaneous population pharmacokinetics modeling of data from healthy volunteers (HV) and patients with knee osteoarthritis (OA). Effect of body weight on CL (a) and Vc (b) is illustrated with a loess regression line. c The effect of anti-drug antibody (ADA) status (negative (n = 34) or positive (n = 14)) or disease status (HV (n = 30) or patients with knee OA (n = 18)) on CL is presented in box plots (median, 25th, and 75th percentiles; whiskers are 1.5 times the inner-quartile range and dots are outliers)

Fig. 4

Clinical effect of AMG 403 in patients with knee osteoarthritis. Mean change (+ standard error) from baseline values over time for each dose group were plotted for total Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) (a) score and visual analog score (VAS) (b) for patient disease assessment. Arrows represent subcutaneous dose administrations (SC)