Kim Dao(1), Paul Thoueille(1), Laurent Arthur Decosterd(1), Thomas Mercier(1), Monia Guidi(1,2), Carine Bardinet(1), Arnaud Castang(3), Catherine Guittet(3), Luc-André Granier(3), Thierry Buclin(1)
(1) Service of Clinical Pharmacology, Lausanne University Hospital, Lausanne, Switzerland. (2) School of Pharmaceutical Sciences, University of Geneva, University of Lausanne, Geneva, Switzerland. (3) Advicenne Pharma SA, Nîmes, France
Objectives:
Sultiame (Ospolot®), an inhibitor of carbonic anhydrase, is a first choice treatment in selected countries for benign epilepsy with centrotemporal spikes, an epileptic syndrome of childhood. Its pharmacokinetic (PK) profile was scarcely studied in humans. Linear disposition is reported by the manufacturer, with a half-life imprecisely described (“between 2 and 16 h”), while no values for volume of distribution (V) and clearance (CL) are published. It represents a suitable candidate for paediatric formulation optimization, as the current coated tablets of 50 or 200 mg allow neither precise and adapted dosing, nor convenient administration to young children. In that context, a pilot study aiming at specifying sultiame’s PK characteristics was conducted. Preliminary results indicated a marked affinity for erythrocytes, attributed to sultiame binding to erythrocytes carbonic anhydrase. A secondary aim was then defined to characterize sultiame exchanges between plasma and erythrocytes.
Methods:
Single oral doses of 50, 100 and 200 mg of sultiame (Ospolot®) were administered in open-label during periods 1, 2 and 3 respectively, at 3-4 weeks intervals in four healthy volunteers. On each period, serial plasma, whole blood and urine samples were collected. A validated high performance liquid chromatography with mass spectrometry method was used for the quantification of sultiame.
An in vitro spiking experiment was also performed to further characterize sultiame exchanges between plasma and erythrocytes observed in vivo.
PK parameters were evaluated using standard non-compartmental calculations as well as non-linear mixed effect modelling (NONMEM®) accounting for saturable uptake by red blood cells. A three-compartment model was implemented, incorporating a saturable ligand to receptor binding parametrized in terms of constants of association (kon) and dissociation (koff), and ligand maximal specific binding capacity (Btot), along with CL, plasma V (Vp), erythrocytes V (Very) and renal extraction fraction (QRen). Variability was set onto CL, Vp, Btot and QRen. Predicted amounts of sultiame (mg) Aa in the absorption compartment, Ap in plasma, Aery in erythrocytes and Aren in urine were respectively computed over time according to the following differential equations:
dAa/dt = – ka·Aa
dAp/dt = ka·Aa – ke·Ap – kon·Ap · (Btot – Aery ) + koff ·Aery
dAery/dt = kon·Ap ·(Btot – Aery ) – koff · Aery
dAren/dt = ke·Ap ·Qren
Results:
The plasma concentration results showed striking non-linear disposition of sultiame, with tenfold increases in concentrations while doses were only doubled. Conversely, whole blood concentrations increased less than dose-proportionally and remained much higher than plasma concentrations. Very quick uptake of sultiame into erythrocytes was observed both in vivo and in vitro. Minimal efflux from erythrocytes was confirmed in vitro. Non-compartmental calculations indicated non-linearity in apparent CL (CL/F) between doses, suggesting a saturation process. Geometric means (CV%) for half-life in plasma were 50.8 (CV: 62%), 90.9 (19%) and 40 (31%) h after a dose of 50, 100 and 200 mg respectively, whereas half-lives in whole blood were much longer with means of 313 (23%), 233 (16%) and 253 (21%) h.
The final population parameters and BSV were: plasma CL = 11 L/h (28%), Vc = 56.3 L (9.3%), Very = 2.93 L, kon = 0.949 mg/h, koff = 0.796 h-1, Btot = 97.1 mg (12%), QRen = 0.247 (15%). Additive error on erythrocytes concentrations was 0.012 mg/L, and proportional residual errors of 56%, 26% and 43 % respectively for plasma, erythrocytes and urine concentrations. The quality of model fitting was good, based on usual diagnostic tests, showing that our model adequately captured the nonlinear behavior of sultiame disposition.
Conclusions:
We described sultiame’s PK profile in detail for the first time, including estimations of CL and V. Plasma-to-blood concentration ratio revealed a strong affinity of the drug for erythrocytes. The remarkable ability of our simple saturable binding model to fit plasma, erythrocyte and urine concentrations is consistent with a strong affinity of sultiame for carbonic anhydrase, abundant in erythrocytes. Further studies should consider this peculiarity, which will affect the interpretation of therapeutic drug concentration monitoring, considered for sultiame dosage adjustment in patients.
Reference: PAGE 28 (2019) Abstr 8987 [www.page-meeting.org/?abstract=8987]
Poster: Drug/Disease Modelling - CNS