Heterozygous familial hypercholesterolemia (HeFH) is a genetic disorder characterized by elevated levels of low-density lipoprotein cholesterol (LDL-C). This phase 2 dose-finding study (NCT02890992) evaluated the efficacy, safety, and dose selection of proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor alirocumab in pediatric HeFH patients. HeFH patients (n = 42) who were aged 8-17 years, had body weight (BW) ≥25 kg, and had LDL-C ≥130 mg/dL despite optimal statin/other lipid-modifying therapies were enrolled in 4 cohorts according to BW: cohort #1: 30 mg (<50 kg) or 50 mg (≥50 kg) every 2 weeks (Q2W), #2: 40 mg (<50 kg) or 75 mg (≥50 kg) Q2W, #3: 75 mg (<50 kg) or 150 mg (≥50 kg) every 4 weeks (Q4W), #4: 150 mg (<50 kg) or 300 mg (≥50 kg) Q4W. Primary endpoint was LDL-C % change from baseline to week 8. Mean age was 12.4 years and 95% of patients were on a statin. Baseline LDL-C levels were 160.0-188.9 mg/dL and free PCSK9 was 186.4-201.7 ng/mL across the cohorts. At week 8, the higher dose cohorts (2 and 4) demonstrated the greatest reductions in LDL-C (-46% and -45%, respectively). Free PCSK9 levels were lowest at week 8 in cohorts 2 and 4 (42.2 ng/mL and 8.6 ng/mL, respectively). Adverse events were reported in 50-90% of patients across the cohorts, and 2 patients discontinued due to adverse events. In pediatric HeFH patients, LDL-C reductions were greatest in the higher dose cohorts. Alirocumab was generally well tolerated at all doses.
Background and Objective Proprotein convertase subtilisin/kexin type 9 inhibition with monoclonal antibodies such as alirocumab significantly reduces low-density lipoprotein-cholesterol levels ± other lipid-lowering therapies. We aimed to develop and qualify a population pharmacokinetics (PopPK) model for alirocumab in healthy subjects and patients, taking into account the mechanistic target-mediated drug disposition (TMDD) process. Methods This TMDD model was developed using a subset of the alirocumab clinical trial database, including nine phase I/II/III studies (n = 527); the model was subsequently expanded to a larger data set of 13 studies (n = 2870). Potential model parameters and covariate relationships were explored, and predictive ability was qualified using a visual predictive check. Results The TMDD model was built using the quasi-steady-state approximation. The final TMDD–quasi-steady-state model included a significant relationship between distribution volume of the central compartment and disease state: distribution volume of the central compartment was 1.56-fold higher in patients vs. healthy subjects. Separately, application of the model to the expanded data set revealed a significant relationship between linear clearance and statin co-administration: linear clearance was 1.27-fold higher with statins. The good predictive performance of the TMDD model was assessed based on graphical and numerical quality criteria, together with the visual predictive check and comparison of the predictions to those from a PopPK model with parallel linear and Michaelis–Menten clearances (i.e., simplification of the TMDD PopPK model). Conclusions This mechanistic TMDD PopPK model integrates the interaction of alirocumab with its target and accurately predicts both alirocumab and total proprotein convertase subtilisin/kexin type 9 concentrations in healthy subjects and patients.