PB0033 - Apixaban population pharmacokinetic model in very elderly patients with atrial fibrillation (ADAGE study): a useful tool to predict plasma concentration decrease in case of urgent invasive procedure

Sunday, June 23, 2024

13:45 – 14:45 ICT

Room: Exhibition Hall

Co-Author(s)

TD

Tu-Nhi Duong

Pharm D, Msc
Université Paris Cité, Inserm UMR-S 1140, Hématologie biologique, Hôpital Lariboisière (AP-HP)
Paris, Ile-de-France, France
SB

Sharon Bouzaglou

MSc
Université Paris Cité, Inserm UMR-S 1140, Hématologie biologique, Hôpital Lariboisière (AP-HP)
Paris, Ile-de-France, France
GF

Geoffrey Foulon-Pinto

Pharm, PhD
Université Paris Cité, Inserm UMR-S 1140, Hématologie biologique, Hôpital Lariboisière (AP-HP)
Paris, Ile-de-France, France
Georges Jourdi, PharmD, PhD

Associate Professor in haematology
Inserm Research institute
Paris, Ile-de-France, France
EP

Eric Pautas

MD, PhD
Sorbonne Université, Unité de Gériatrie aiguë, Hôpital Charles Foix (AP-HP), INSERM UMR-S-1140, Paris
Ivry-sur-Seine, Ile-de-France, France

Presenting Author(s)

Virginie Siguret

PharmD, PhD
Université Paris Cité, Inserm UMR-S 1140, Hématologie biologique, Hôpital Lariboisière (AP-HP)
Paris, Ile-de-France, France

Co-Author(s)

EC

Emmanuel Curis, PhD PharmD

Pharm, PhD
Université Paris Cité, UR 7537 BioSTM (Biostatistics), Hématologie biologique, Hôpital Lariboisière (AP-HP)
Paris, Ile-de-France, France

Background: A significant proportion of very elderly patients receiving apixaban for non-valvular atrial fibrillation (NVAF) requires an urgent invasive procedure every year. These patients are often polymorbid and polymedicated, leading to substantial variability in apixaban plasma concentration. However, no study specifically investigated apixaban pharmacokinetics (PK) in these patients, particularly concentration decrease following drug intake, making patient management challenging in specific clinical settings (e.g. invasive procedure).

Aims: To build a population PK model in very elderly NVAF patients receiving apixaban, taking into account patient characteristics.

Methods: ADAGE-NCT02464488 was a prospective multicentre academic study, enrolling in-patients aged 80-years, receiving apixaban for NVAF for at least 4 days. Each patient had 1-5 samples taken at different time-points over a 20-day period. Apixaban concentration (ng/mL) was measured using Liquid-anti-Xa (Stago). Apixaban PK was modelled using a mono-compartment model with first-order absorption, at steady state. Model building (error model, inter-individual variabilities, covariate selection) used nested-models likelihood ratio tests (LRT), with asymptotic chi-square distribution, with a 5% Type I error (p < 0.05). Model quality was validated using standard methods, especially visual predictive check (VPC). Data were analysed using R and Monolix softwares.

Results: We collected 237 samples from 117 patients, mean age 87 years (80-100), 74% females, mean creatinine clearance (CrCl-Cockcroft-Gault 45 mL/min); 77.1% received a reduced-dose (2.5 mg bid). The final Pop-PK model included correlated inter-individual variability on clearance (CL) and distribution volume (Vd). Clearance variability was partly explained by sex, CrCl and intake of CYP3A4/5 inhibitors. All parameters had plausible values and were correctly estimated except the absorption rate constant (Table). Individual concentrations were accurately predicted (Figure).

Conclusion(s): The final model was consistent with known PK properties of apixaban. It shows that sex, CrCl and CYP3A4/5 inhibitor intake should be accounted for when predicting the apixaban clearance in very elderly patients.