Revisiting target-mediated elimination of therapeutic antibodies: the irreversible binding approximation
David ternant (1), Laurence Picon (2), Guillaume Cartron (3), Denis Mulleman (1), Mario Campone (4), Jean-Louis Merlin (5), Philippe Goupille (1), Matthias Büchler (6), Thierry Lecomte (1), Gilles Paintaud (1).
(1) EA 7501 GICC, University of Tours, France, (2), Department of gastroenterology, University Hospital of Tours, France, (3) University of Montpellier, CNRS UMR 5235, Montpellier, France, (4) CNRS UMR 7039 CRAN Université de Lorraine, Nancy, France, (5) Institut de Cancérologie de l’Ouest, Angers, France, (6) EA 4245 T2I, University of Tours, France.
Objectives: The pharmacokinetics (PK) of therapeutic monoclonal antibodies (mAbs) often present a nonlinear elimination shape due to a target-mediated drug disposition (TMDD). The TMDD model [1] is however rarely used to described target-dependent PK of mAbs because necessitates rich databases with dense sampling strategies and measurements of free mAb, free target and mAb-target complexes. Usually, only mAb concentrations are available and nonlinear elimination is often described using a Michaelis-Menten model which relies on the assumption of constant antigen mass remaining constant during follow-up. The irreversible binding approximation described previously [2] was rarely used. A model describing irreversible binding where antigen is treated as a variable (IBLV) but was previsouly used with success to describe the PK of rituximab in chronic lymphocytic leukemia [3] and therefore deserves further investigation. This work aimed to investigate the relevance of IBLV on real mAb PK databases.
Methods: The IBLV approximation includes a second-order elimination term involving both mAb concentration and antigen mass amount. Being unavailable, antigen target amount is treated as a latent variable. To investigate the relevance of iIBLV approximation, the following mAb PK databases were revisited:
SPAXIM (infliximab in spondylarthropathies, N=26) [4],
IFX-CD (infliximab in Crohn’s disease, N=133) [5],
LYSA-RTX (rituximab in non-hodgkin lymphoma, N=108) [6],
RTX-RA (rituximab in reumatoid arthritis, N=91) [7],
RADHER (trastuzumab in breast cancer, N=79) [8],
ORL-CTX (cetuximab in head and neck cancer, N=30) [9],
STIC-avastin (bevacizumab in colorectal cancer, N=130) [10],
LYMPHO (lymphoglobuline in kidney transplantation N=14) [11]
Several PK models were tested. They are made of a 2-compartment model with first-order transfer and elimination rates and target-mediated elimination (TLE) term as follows:
- 2C – linear 2-compartment model, with TLE = 0;
- 2C-MM – 2-compartment model with both linear and Michaelis-Menten elimination rates:
TLE = Vmax.Cc / (Km+Cc), where Vmax is maximum saturable elimination rate, Cc is mAb concentration in the central compartment and Km is Michaelis constant;
- 2C-IB-1 – 2 compartment model with second-order target-mediated elimination and latent target turnover:
TLE = - kon.Cc.L,
dL/dt = kin – kout.L – kon.Cc.L, L(0) = kin/kout
where kon is second-order mAb-target association and elimination rate constant, L is latent target amount variable, and kin and kout are zero-order input and first-order output of latent target amount, respectively;
- 2C-IB-2 – 2 compartment model with second-order target-mediated elimination and initial target amount :
TLE = - kon.Cc.L
dL/dt = - kon.Cc.L, L(0) = L0, where initial target amount is estimated
PK parameters were estimated using nonlinear mixed-effects modelling (Monolix Suite 2018R2, Lixoft, Antony, France).For each database the best model was determined using Akaike’s information criterion.
Results: In all but one [9] databases, the published PK model was linear 2-compartment model without target-mediated elimination. Values of AIC of each model were for 2C, 2C-MM, 2C-IB-1, 2C-IB-2. The value of AIC of the best model was in bold:
Study 2C 2C-MM 2C-IB-1 2C-IB-2
SPAXIM 4421.4 4428.3 4425.8 4427.9
IFX-CD 6685.0 6691.4 6672.4 6679.7
LYSA-RTX 8106.7 8111.5 8075.9 8099.4
RTX-RA 3409.9 3425.8 3409.0 3410.1
RADHER 5378.0 5309.9 5365.8 5323.2
ORL-CTX 13096.7 12697.6 12709.7 12700.4
STIC-avastin 3828.7 3804.9 3817.2 3806.6
LYMPHO 1524.5 1536.9 1480.0 1470.5
Target-mediated elimination was detected in 6 of 8 databases and IBLV model were superior to Michaelis-Menten in 3 of 6 of nonlinear PK databases. Of note, nonlinearity was detected for infliximab in Crohn’s disease for the first time.
Conclusions: Superiority of IBLV may be due to antigen mass which varies in time. This approximation should be considered in case of mAbs which present nonlinear PK.
References:
[1] Mager DE, Jusko WJ. General pharmacokinetic model for drugs exhibiting target-mediated drug disposition. J Pharmacokinet Pharmacodyn. 2001;28:507-32.
[2] Gibiansky L, Gibiansky E. Target-mediated drug disposition: new derivation of the Michaelis-Menten model, and why it is often sufficient for description of drugs with TMDD. 2010 PAGE 19; Abstr 1728. Available at http://www.page-meeting.org/?abstract = 1728 (last accessed 6 march 2019).
[3] Tout M, Gagez AL, Leprêtre S, Gouilleux-Gruart V, Azzopardi N, Delmer A, Mercier M, Ysebaert L, Laribi K, Gonzalez H, Paintaud G, Cartron G, Ternant D. Influence of FCGR3A-158V/F Genotype and Baseline CD20 Antigen Count on Target-Mediated Elimination of Rituximab in Patients with Chronic Lymphocytic Leukemia: A Study of FILO Group. Clin Pharmacokinet. 2017 Jun;56:635-47.
[4] Ternant D, Mulleman D, Lauféron F, Vignault C, Ducourau E, Wendling D, Goupille P, Paintaud G.Influence of methotrexate on infliximab pharmacokinetics and pharmacodynamics in ankylosing spondylitis. Br J Clin Pharmacol 2012 ;73:55-65.
[5] Aubourg A, Picon L, Lecomte T, Bejan-Angoulvant T, Paintaud G, Ternant D. A robust estimation of infliximab pharmacokinetic parameters in Crohn's disease. Eur J Clin Pharmacol. 2015 ;71:1541-2.
[6] Tout M, Casasnovas O, Meignan M, Lamy T, Morschhauser F, Salles G, Gyan E, Haioun C, Mercier M, Feugier P, Boussetta S, Paintaud G, Ternant D, Cartron G. Rituximab exposure is influenced by baseline metabolic tumor volume and predicts outcome of DLBCL patients: A LYSA study. Blood 2017;129:2616-23.
[7] Lioger B, Edupuganti SR, Mulleman D, Passot C, Desvignes C, Bejan-Angoulvant T, Thibault G, Gouilleux-Gruart V, Mélet J, Paintaud G, Ternant D. Antigenic burden and serum IgG concentrations influence rituximab pharmacokinetics in rheumatoid arthritis patients. Br J Clin Pharmacol 2017;83:1773-81.
[8] Bernadou G, Campone M, Merlin JL, Gouilleux-Gruart V, Bachelot T, Lokiec F, Rezai K, Arnedos M, Diéras V, Jimenez M, Paintaud G, Ternant D. Influence of tumour burden on trastuzumab pharmacokinetics in HER2 positive non-metastatic breast cancer. Br J Clin Pharmacol 2016 ;81:941-8.
[9] Cetuximab Pharmacokinetics Influences Overall Survival in Patients With Head and Neck Cancer. Pointreau Y, Azzopardi N, Ternant D, Calais G, Paintaud G. Ther Drug Monit. 2016;38:567-72.
[10] Caulet M, Lecomte T, Bouché O, Rollin J, Gouilleux-Gruart V, Azzopardi N, Léger J, Borg C, Douillard JY, Manfredi S, Smith D, Capitain O, Ferru A, Moussata D, Terrebone E, Paintaud G, Ternant D. Bevacizumab Pharmacokinetics Influence Overall and Progression-Free Survival in Metastatic Colorectal Cancer Patients. Clin Pharmacokinet. 2016;55:1381-94.
[11] Ternant D, Büchler M, Bénéton M, Alván G, Ohresser M, Touchard G, Hurault de Ligny B, Toupance O, Watier H, Lebranchu Y, Paintaud G. Interindividual variability in the concentration-effect relationship of antilymphocyte globulins – a possible influence of FcγRIIIa genetic polymorphism. Br J Clin Pharmacol 2008;65:60-8