Sultiame pharmacokinetic profile in plasma and erythrocytes after single oral doses: A pilot study in healthy volunteers

Abstract A pilot study was conducted aiming at specifying sultiame's pharmacokinetic profile, completed by in vitro assays evaluating the intraerythrocytic transfer of sultiame and by a pharmacokinetic model assessing its distribution. Single oral doses of sultiame (Ospolot® 50, 100, and 200 mg) were administered in open‐label to four healthy volunteers. Serial plasma, whole blood, and urine samples were collected. A spiking experiment was also performed to characterize sultiame's exchanges between plasma and erythrocytes in vitro. Pharmacokinetic parameters were evaluated using standard noncompartmental calculations and nonlinear mixed‐effect modeling. The plasma maximal concentrations (C max) showed striking nonlinear disposition of sultiame, with a 10‐fold increase while doses were doubled. Conversely, whole blood Cmax increased less than dose proportionally while staying much higher than in plasma. Quick uptake of sultiame into erythrocytes observed in vivo was confirmed in vitro, with minimal efflux. A two‐compartment model with first‐order absorption, incorporating a saturable ligand to receptor binding, described the data remarkably well, indicating apparent plasma clearance of 10.0 L/h (BSV: 29%) and distribution volume of 64.8 L; saturable uptake into an intracellular compartment of 3.3 L with a maximum binding capacity of 111 mg accounted for nonlinearities observed in plasma and whole blood concentrations. Pharmacokinetic characteristics of sultiame are reported, including estimates of clearance and volume of distribution that were so far unpublished. The noticeable nonlinearity in sultiame disposition should be taken into account for the design of future studies and the interpretation of therapeutic drug monitoring results.

standard noncompartmental calculations and nonlinear mixed-effect modeling.
The plasma maximal concentrations (C max ) showed striking nonlinear disposition of sultiame, with a 10-fold increase while doses were doubled. Conversely, whole blood C max increased less than dose proportionally while staying much higher than in plasma. Quick uptake of sultiame into erythrocytes observed in vivo was confirmed in vitro, with minimal efflux. A two-compartment model with first-order absorption, incorporating a saturable ligand to receptor binding, described the data remarkably well, indicating apparent plasma clearance of 10.0 L/h (BSV: 29%) and distribution volume of 64.8 L; saturable uptake into an intracellular compartment of 3.3 L with a maximum binding capacity of 111 mg accounted for nonlinearities observed in plasma and whole blood concentrations. Pharmacokinetic characteristics of sultiame are reported, including estimates of clearance and volume of distribution that were so far unpublished. The noticeable nonlinearity in sultiame disposition should be taken into account for the design of future studies and the interpretation of therapeutic drug monitoring results.

K E Y W O R D S
clearance, healthy volunteers, pharmacokinetic, sultiame, volume of distribution

| INTRODUC TI ON
Sultiame (Bayer 1960, MW 290 g/mol) 1 is a cyclic sulfonamide, whose antiepileptic activity is thought to mainly result from the inhibition of various subtypes of carbonic anhydrase (CA). 2 In particular, the blockade of cytosolic CA II seems to produce a degree of intracellular acidification sufficient to stabilize seizure-eliciting neurons.
In addition, an inhibitory action on sodium channels possibly contributes to its antiepileptic efficacy. 2 Sultiame is a first choice treatment in some countries for benign childhood epilepsy with centrotemporal spikes, a common nonlesional epilepsy syndrome of childhood usually appearing between 4 and 12 years of age characterized by nocturnal oro-facial motor or sensory seizures, with or without bilateral propagation. 3 The overall prognosis is good, with spontaneous disappearance of epilepsy at adolescence.
However, some children may have mild neurocognitive deficits sometimes linked to interictal epileptiform discharges on electroencephalogram. Although not all cases deserve a treatment, patients with more frequent seizures benefit from a course of antiseizure drug.
Sultiame has been reported to be well tolerated and possibly more effective than other antiepileptic drugs commonly used in focal seizures, such as carbamazepine, gabapentine, or levetiracetam. [3][4][5] Despite a regular use over decades by pediatric neurologists, the pharmacokinetic (PK) profile of sultiame was only scarcely studied in humans. [6][7][8][9] Linear disposition is assumed by the manufacturer, with a half-life described to lie "between 2 and 16 hours", while no values have been published for bioavailability (F), apparent volume of distribution (V/F), and apparent clearance (CL/F). Plasma-free fraction amounts to 71%. Plasma levels are said to display a significant degree of between-and within-individual variability, even at steady state.
The manufacturer mentions a mixed elimination by metabolism (hydroxylation) and renal excretion, the latter concerning 30%-60% of an ingested dose. Two metabolites of sultiame have been identified in urines, the main one being hydroxy-sultiame (an inactive compound). 8 Our primary aim was to assess sultiame's PK parameters, in order to better characterize its concentration-time profile. Preliminary investigations of the stability of blood samples for PK measurement indicated a marked uptake into erythrocytes that we possibly attributed to sultiame's affinity for CA abundant in erythrocytes. 10 A secondary aim was thus defined to characterize sultiame's exchanges between plasma and erythrocytes through an in vitro spiking experiment.

| Subjects
Eligible subjects for the clinical study were healthy adult male volunteers aged 18-45 years, with a body weight ranging between 55 and 95 kg and a body mass index of 18-29 kg/m 2 . Subjects with a history or evidence of clinically significant diseases or interfering conditions were excluded. Other exclusion criteria included: history of any clinically significant laboratory value including serology for hepatitis and HIV, relevant alcohol or drug abuse, recent acute illness, or use of any medication the week prior to study.

| Analytical method
The quantification of sultiame in whole blood, plasma, and urine was performed using a high-performance liquid chromatography

| Pharmacokinetic parameters
Erythrocytic concentrations (C ery ) were deduced from whole blood and plasma measurements as: where Ht is the hematocrit of the sample. 11

| Noncompartmental analysis
Plasma, whole blood, and urine sultiame PK parameters were first The area under the curve for a single dose (AUC 0-inf ) was calculated using the log trapezoidal rule with extrapolation to infinity.
The terminal rate constant ( z ) was derived from a two-exponential model curve. CL/F was calculated as the dose divided by AUC 0-inf , the half-life (t 1/2 ) as ln(2)/ z , and V/F as (CL/F)/ z .

| Compartmental analysis
A population PK analysis was also performed using a nonlinear mixed-effect modeling approach (NONMEM version 7.4, ICON Development Solutions, Hanover MD, USA). Based on graphical exploration and in vitro experiments, a two-compartment model with first-order absorption incorporating a saturable ligand to receptor binding was devised, as illustrated in Figure 1, to account for sultiame's affinity for CA abundant in erythrocytes. The model was expressed in terms of differential equations and first-order rate constants of absorption (k a ), elimination (k e ), association with (k on ) and dissociation from (k off ) receptors, maximal binding capacity (B tot ), apparent central volume of distribution (V c /F), and erythrocytes' volume of distribution (V ery ). Note that V ery actually encompasses the volume of all types of cells containing receptors, assumed to tally with the sampled erythrocytes. Plasma protein binding was assumed constant over the whole range of observed concentrations. Renal extraction fraction (Q ren ) was added to characterize urine excretion with an additional compartment for urine data. Values of k on and k off characterized in vitro at different temperatures were used as initial estimates and the value of k on to 2018 µM −1 h −1 at 37°C fixed in the in vivo model. F was not evaluated, making the model estimate apparent values for V c /F and CL/F. The first-order elimination rate constant from the central compartment was calculated as k e = (CL/F)/ (V c /F). The first value below the LLOQ was equaled to LLOQ/2 and subsequent nonquantifiable values were discarded (M6 method). 12 Details concerning differential equations and model building steps and adjustment are available in Supplemental material.
A stepwise procedure was used to identify the model that best fitted the data, comparing two-, three-, and four-compartmental models (peripheral or separate intracellular compartments for free and bound sultiame), and other nonlinear models (B max sigmoid model).
Exponential errors were used for the description of between-subject variability (BSV) of PK parameters. Proportional, additive, and mixed error models were compared to describe the residual variability. The stability and internal validity of the final model were assessed by means of standard procedures, detailed in Supplemental material. Covariates were not analyzed due to the limited number of subjects included.

| Safety assessments
All adverse events were recorded with an evaluation of their severity and imputability to the study drug. Each volunteer received a full medical examination at screening and at the end of the study. Visual analogue scales (VAS) on paper support recorded the

| In vitro assays
A spiking experiment was performed to better characterize sultiame exchanges between plasma and erythrocytes. The influx into cells was investigated through spiking fresh human whole blood on EDTA at different target concentrations (1,5,10,20,40

| Noncompartmental pharmacokinetic parameters
Average sultiame plasma and whole blood concentrations after single oral doses of 50, 100, and 200 mg, administered at 4 weeks intervals, are represented in Figure 2. The plasma maximal concentration (C max ) revealed striking nonlinear disposition of sultiame, with 10-fold increases between each dose levels, while doses only doubled. Conversely, whole blood C max increased less than dose proportionally and were much higher than those in plasma. PK parameters of sultiame obtained by noncompartmental calculations are reported in Table 1   Conversely, classical linear models were unable to describe the data. The final base population parameters are listed in Table 2 and the average concentration predictions for each dose level are shown on Figure 2. Individual profiles of predicted concentrations

| In vitro experiments
The concentration-time profiles describing sultiame influx into erythrocytes at different concentrations are shown in Figure 3. Sultiame was taken up almost instantaneously into the erythrocytes and the erythrocyte/plasma ratio reached a steady state after about 10 minutes.
This erythrocyte/plasma ratio decreased from 800 at concentrations of 10 mg/L or less, to less than 30 for concentrations of 20 mg/L and above. These results indicate that the saturation of erythrocytes occurs around or slightly above 20 mg/L (of whole blood concentration).
At higher concentrations, the excess of sultiame increasingly distributes between the plasma and the erythrocytes (Figure 4). Decreasing the experiment temperature slowed down the rate of entry of sultiame into the cells, without significantly modifying the erythrocyte/plasma ratios.
It allowed the estimation of k on and k off using a similar ligand to receptor model at 4°C where the rate of entry was observable. Values at 37°C could be deduced using Arrhenius' law ( Figure 5)

| D ISCUSS I ON
To our knowledge, our preliminary clinical PK study is the first to provide a detailed description of sultiame's PK characteristics, since the patenting of this antiepileptic agent in 1955 and its commercialization in 1960. 9 The PK results obtained after single oral adminis- Competition studies indicated that sultiame had a stronger affinity for erythrocytes than zonisamide. Clinically, this suggests that displacement of sultiame from the erythrocytes compartment is unlikely to occur with topiramate or zonisamide, while sultiame addition may lead to an at least transient increase in free zonisamide, with a potential risk of increased effects or toxicity. Fluctuations in the plasma/erythrocyte ratio might also be associated with variations in efficacy when sultiame is given as an add-on antiepileptic drug in clinical practice.
Using noncompartmental approaches, plasma and especially whole blood t 1/2 appeared to be markedly longer than previously reported. 6  As our population model is based on a limited amount of data, between subject variability and estimation of errors are evaluated with poor precision, as indicated by their wide standard error and CI95%.
Globally, our results have mainly an orientation value and should essentially serve to design further clinical studies with sultiame, particularly in the pediatric population as part of therapy of benign childhood epilepsy with centrotemporal spikes. 3 They should also be taken into consideration to elaborate an adequate framework of interpretation for the therapeutic monitoring of sultiame circulating concentrations, not only in plasma but also in whole blood, which might present a clinical interest as for other recent antiepileptic agents. 19

CO N FLI C T O F I NTE R E S T
The authors declare that Advicenne Pharma, France, provided financial support for this study. No other conflict of interest is reported.

E TH I C A L S TATEM ENT
All procedures performed in studies involving human participants were in accordance with the ethical standards of the competent research ethics committee and regulatory authorities and with the 1964 Helsinki Declaration and its later amendments. Informed consent was obtained from all individual participants involved in the study.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available in DDMoRe Repository at http://repos itory.ddmore.found ation/ /, reference number DDMODEL00000298. These data were derived from real data and key dosage and concentration data were not altered.