The Journal of Clinical...

http://www.jclinpharm.orgPharmacology

The Journal of Clinical

1998; 38; 227 J. Clin. Pharmacol.MT Marino, BG Schuster, RP Brueckner, E Lin, A Kaminskis and KC Lasseter

against nerve agents in humansPopulation pharmacokinetics and pharmacodynamics of pyridostigmine bromide for prophylaxis

http://www.jclinpharm.org/cgi/content/abstract/38/3/227 The online version of this article can be found at:

Published by:

http://www.sagepublications.com

On behalf of: American College of Clinical Pharmacology

can be found at:The Journal of Clinical Pharmacology Additional services and information for

http://www.jclinpharm.org/cgi/alerts Email Alerts:

http://www.jclinpharm.org/subscriptions Subscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

© 1998 American College of Clinical Pharmacology. All rights reserved. Not for commercial use or unauthorized distribution. at ACCP Member on September 14, 2007 http://www.jclinpharm.orgDownloaded from

http://www.accp1.org

http://www.jclinpharm.org/cgi/alerts

http://www.jclinpharm.org/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://www.jclinpharm.org

PHARMACOKINETICS AND PHARMACODYNAMICS

J Clin Pharmacol 1998;38:227-235 227

Population Pharmacokinetics andPharmacodynamics of Pyridostigmine

Bromide for Prophylaxis against NerveAgents in Humans

Mark T. Marino, MD, Brian G. Schuster, MD, Raif P. Brueckner, MD, Emil Lin, PhD,

Andris Kaminskis, BS, and Kenneth C. Lasseter, MD

This study was conducted to determine the pharmacokinetics and pharmacodynamics

of pyridostigmine given as 30 mg of pyridostigmine bromide every 8 hours in healthy

subjects. Plasma pyridostigmine concentration and red blood cell acetyicholinesteraseactivity were measured in blood samples collected during a 3-week period. Population

analysis was performed using standard pharmacokinetic and pharmacodynamic mod-

els with the nonlinear mixed-effect modeling software (NONMEM). The pharmacoki-

netic model that best fit the pyridostigmine plasma levels was a two-compartment open

model with first-order absorption, a lag time, and first-order elimination from the central

compartment. The pharmacodynamic model that best fit red blood cell acetyicholines-

terase activity was an inhibitory Emx model with an effect compartment linked to the

central compartment. The results showed that the pharmacokinetics of pyridostigminebromide are both gender and weight dependent. The pharmacodynamic effect does not

lag significantly from the plasma concentration and returns to near normal within 8hours. With the present dosage regimen of 30 mg every 8 hours, 30% of individualsmay not have red blood cell acetylcholinesterase inhibition >10% at the time of the

trough. J Clin Pharmacol 1998;38:227-235.

P yridostigmine is a reversible inhibitor of acetyl-cholinesterase and is used for the symptomatic

treatment of myasthenia gravis.1 It has recently beenshown to be an adjunct to the therapy of organophos-phorus cholinesterase inhibitors in animal studies.2

From the Department of Pharmacology, Division of Experimental Thera-peutics, Walter Reed Army Institute of Research, Washington DC (Drs.Marino, Schuster, and Brueckner), the School of Pharmacy, Universityof California, San Francisco, California (Dr. Lin), the United States ArmyInstitute of Chemical Defense, Aberdeen Proving Ground, Maryland (Mr.Kaminskis), and Clinical Pharmacology Associates, Miami, Florida (Dr.Lasseter). This study was funded by the U.S. Army Medical MaterialDevelopment and Acquisition Command. The views and opinions in this

article are those solely of the authors and do not reflect official policyof the Department of the Army or the Department of Defense. Submittedfor publication August 4, 1997; accepted in revised form December 1,1997. Address for reprints: Mark 1. Marino, LTC MC, C, Department ofPharmacology, Division of Experimental Therapeutics, Walter Reed ArmyInstitute of Research, Washington, DC 20307-5100.

Pyridostigmine is thought to act by preventing theirreversible binding of organophosphorus agents toacetylcholmnesterase. The ability to prevent irrevers-ible binding of organophosphorus agents to acetyl-cholinesterase would make pyridostigmmne useful asa pretreatment to minimize the effects of “nerve gas”if used in conjunction with the standard treatment ofatropine and pralidoximine chloride (2-PAM).3 Theconcern for the use of nerve agents in modern war-fare was high enough that the coalition forces issuedand used pyridostigmmne bromide during the war

against Iraq.4This study was designed to determine the popula-

tion pharmacokinetics and pharmacodynamics ofpyridostigmine bromide when given as a 30-mg tab-let every 8 hours. Red blood cell acetylcholmnesteraseactivity was chosen as a pharmacodynamic endpointbecause it has been shown to correlate with survivalin nerve agent exposure in some animal models.5 Thestandard dose that was prescribed to United States



MARINO ET AL

228 S J Clin Pharmacol 1998;38:227-235

armed forces personnel was tested in order to de-velop an understanding of the population variabilityof the drug and explore any covariates that could beused to optimize the dosage regimen.

SUBJECTS AND METhODS

Clinical Protocol

The study was approved by the First Foundation forthe Protection of Human Subjects in Research, Inc.,of Miami, Florida. Ninety healthy male and femalevolunteers between 18 and 45 years of age were in-cluded in this placebo-controlled, randomized, dou-ble-blind study. All subjects signed the approved in-formed consent form.

In all, 60 subjects (30 men and 30 women) receivedthe study drug, with 30 subjects (15 men and 15

women) receiving placebo. The average age for thegroup receiving the study drug was 32 years (rangefor men and women, 19-44 years). Weight rangedfrom 50 to 105 kg for the men and 43 to 91 kg for

the women.Baseline examination of all subjects was con-

ducted within 21 days before admission to the exper-imental unit (Clinical Pharmacology Associates, Mi-ami, FL) and included a medical history, physicalexamination, and laboratory screening, includingelectrocardiogram, chest x-ray (not taken if normalwithin 12 months), urinalysis, complete blood countand differential, electrolytes, liver associated en-zymes, antibody to human immunodeficiency virus,and red blood cell acetylcholinesterase. Female sub-jects were admitted to the study if they had a negativeserum beta human chorionic gonadotrophin (/3-HCG)test result within 24 hours before admission. 3-HCGwas also measured on discharge from the experimen-tal unit.

All subjects were confined to the experimentalunit for the full 25 days of the study, which included1 day before administration and 3 days after the lastdose. The subjects were followed as outpatients for1 year after the treatment regimen.

Pyridostigmine Regimen

The regimen was designed to study the multiple dosepharmacokinetics and pharmacodynamics of pyrido-stigmine when given orally as a 30-mg tablet (Mesti-non; LaRoche, Nutley, NJ) every 8 hours for 21 days.Subjects were given the test medication or placebounder observation. All subjects fasted from midnightuntil 2 hours after the 8:00 AM dose and also fastedfor 1 hour before and after the 4:00 PM and 12:00 AM

doses. Vital signs (blood pressure, pulse, and temper-

ature) were taken daily throughout the study before9:00 AM. Physical examinations were performed ondays 1, 8, 15, 22, and before discharge on day 25.Laboratory examinations and electrocardiogramswere performed on days 8, 15, and 23.

Pyridostigmine Plasma Levels and Red Blood CellAcetylcholinesterase Activity

Blood samples (5 mL) were obtained for plasma pyri-dostigmine levels and red blood cell acetyicholines-terase activity before and 1, 1.5, 2, 2.5, 3, 4, 5, 6, 7,

and 8 hours after the first dose. Levels were drawn5 minutes before the daily 8:00 AM dose on days 4,

7, 9, 11, 14, 16, 18, and 21. During the last dose onday 22, samples were drawn 5 minutes before and1, 1.5, 2, 2.5, 3, 4, 5, 6, 7, 8, 10, 12, 14, 18, 24, 28,

36, 40, 48, 52, 56, 60, 66, and 72 hours after adminis-tration.

The samples were collected in heparinized Vacu-tamer tubes (Becton Dickinson, Franklin Lakes, NJ)for analysis of plasma pyridostigmine levels and inEDTA tubes for analysis of red blood cell acetylcho-linesterase activity. They were placed on wet ice be-fore transfer to the clinical laboratory. The bloodsamples used for the determination of red blood cellacetylcholinesterase were diluted 1:10 with 1% tn-mex 100 (Triton X-100; Sigma, St. Louis, MO) beforeanalysis. The blood samples used for the determina-tion of pyridostigmine were centrifuged at 4#{176}Cfor 10minutes before separation of the plasma from the redblood cells. All samples were stored at -80#{176}Cuntilanalysis was performed.

Plasma was analyzed for pyridostigmine usinghigh-performance liquid chromatography (HPLC).The assay required 0.5 mL of plasma for the determi-nation of pyridostigmine concentration. The HPLCsystem included an LC-600 pump with an SPD bAultraviolet detector with absorbance at 208 urn (Shi-madzu, Kyoto, Japan) and an Intelligent Sampleprocessor 710 B (Waters, Milford, MA). Separationwas carried out on an Axiom Silica column (4.6 X

250 mm, 5 mm particle size; Richard Scientific, No-vato, CA).

The mobile phase consisted of acetonitnile/water(1:1, v:v) with final concentrations of 0.05% tetra-methylammonium chloride and 5 mmol/L dibasicammonium phosphate (final apparent pH 7.2). Ace-tonitrile, phosphoric acid, and dibasic ammoniumphosphate were obtained from Fischer Scientific(Fairlawn, NJ); tetramethylammonium chloride wasobtained from Fluka Chernika (Buchs, Switzerland);pyridostigmine bromide (lot no. 8950426) was ob-tained from Sigma Chemical (St. Louis, MO); and



PK/PD OF PYRIDOSTIGMINE BROMIDE FOR PROPHYLAXIS AGAINST NERVE AGENTS

PHARMACOKINETICS AND PHARMACODYNAMICS 229

neostigmine bromide (lot no. KTO513OJ) was ob-tained from Aldrich Chemical (Milwaukee, WI).

Assay samples were prepared by spiking knownvolumes of plasma with a known amount of internalstandard (neostigmine bromide). Standard curvesamples were generated by spiking known amountsof interference-free human plasma (Irwin MemorialBlood Bank, San Francisco, CA) with knownamounts of pyridostigmine and internal standard.Samples were prepared by adding 18.75 mg of neo-stigmine bromide (internal standard) and 1 mL ofacetonitrile to each 0.5-mL plasma sample. The mix-ture was centrifuged and the supernatant was placedon a Varian C8 Bond Elut SPE cartridge (Harbor City,CA). The cartridge was rinsed with 2 mL of water, 4

mL of CH3CN/H20 (50:50), 2 mL of CH3CN, and 0.5mL of CH3CN/H20 (85:15) containing 1 mmol/L(NH4)2HP04 at pH 3.6. The fraction containing pyri-dostigmine was eluted by adding 2 mL of CH3CN/H20 (85:15) containing 1 mmol/L (NH4)2HP04 at pH3.6. The eluates were concentrated by evaporationto approximately 200 mL under nitrogen at roomtemperature, transferred to WISP vials (Waters, Mu-ford, MA) and 25- to 50-mL aliquots were injectedon column.

The peak height ratios of pyridostigmine (retentiontime 14 mm) and internal standard (retention time16 mm) were calculated for each sample. Standardcurve concentrations and peak height ratios were fitby weighted linear least squares regression to theequation y = mx + b (x equals known concentrationand y equals peak height ratio). Drug concentrationsin the unknown samples were calculated from thestandard curve for each sample run.

The assay was linear within the range of the stan-dard curve (1.53-76.3 ng/mL pynidostigmine freebase). The lower limit of quantitation was 1.53ng/mL, with an intraday coefficient of variation (CV)of 14.6% (n = 6) and an interday CV of 12% (n =

6).6 No significant degradation of standard sampleswere noted over 4 months when kept at -80#{176}C,andall samples were analyzed within that time period.The red blood cell acetylcholinesterase levels weremeasured by standard enzymatic techniques with alower limit of quantitation of 0.46 U/mL with a CVof 6.5% and linearity to 20 U/mL. The CV was lessthan 2% between 5 and 15 U/mL.7

PharmacokineticfPharmacodynamic PopulationAnalysis

The pharmacokinetic and pharmacodynamic analy-sis was performed with NONMEN IV, version 2.1on a Silicon Graphics Indigo 2 UNIX workstation(Mountain View, CA). A two compartment model

with first-order absorption with a lag time and first-order elimination from the central compartment wasused to fit the plasma pyridostigmine concentrations.A standard two-step procedure was followed to dothe population pharmacokinetic analysis.9 The phar-macokinetic model was first fitted to the pharmacoki-netic data for the entire population. A relationshipbetween the pharmacokmnetic parameters and covari-ates such as weight and gender was explored usingresidual analysis and linear regression.1#{176} Relation-ships between volume of the central compartmentand weight and clearance and gender were found tobe significant. Volume of the central compartmentand clearance were then both modeled as being de-pendent on weight and gender, respectively (equa-tions 1 and 2).

= O x Kg

Clearance = 02

#{176}wo,nen

(1)

(2)

Analysis on the models was done by comparingthe objective functions (differences in the full andreduced model’s objective function using x2 with thenumber of parameters in the reduced model as thedegrees of freedom.11 Parameter estimates for differ-ent covariates (gender and weight) were comparedusing the t test.

In the second step the Bayesian estimates of thepharmacokinetic parameters for each individualwere used in equation 3 to generate the effect siteconcentration. The effect compartment was linkedto the central compartment via a standard effect com-partment equation. The effect (red blood cell acetyl-cholinesterase activity) was modeled with an inhibi-tory Emax model:

Red blood cell acetylcholmnesterase activity

Emax X C,,= Baseline - __________

(ECSO + Ce)(3)

The NONMIEM model for the red blood cell acetyl-cholinesterase data is the same general model as usedfor the plasma concentration (equation 4). The NON-MEM model for the ith concentration in the jth mdi-vidual is given by equation 4:

C,, = f(p, t,,) + H(f, 9) x (4)

where f is the predicted concentration and p, arethe pharmacokinetic parameter estimates for the jthindividual. The second term in equation 4 representsthe departure of the model from the observationswhere e, was assumed to be a random gaussian van-



MARINO ET AL

230 #{149}J Clin Pharmacol 1998;38:227-235

TABLE I

Pharmacokinetic and Pharmacodynamic Parameter Estimates

Mean (SEM) Variability (%)

Pharmacokinetic Parameters

Intraindividual Error

Cl/F (men), L/hr 221 (13.4)Cl/F (women), L/hr 172 (11.4)Vc/F (L/kg) 2.09 (0.12)Vp/F(L) 4240 (570)Lag time (mm) 29 (3.4)Ka (hr’) 0.32 (0.027)8 (ng/mL) 1.04 (0.84)82 (%) 26 (3)

313151431

9.7

Pharmacodynamic Parameters

Intraindividual Error

Keo (hrs’) 1.32 (0.07)EC50 (ng/mL) 69 (2.7)Ema,JBaSeline (U/mL) 10.4 (0.14)8 (U/mL) 0.62 (0.15)

92 (%) 9.0 (0.20)

22.29.7

able with a mean of zero and variance cr2,, H(f, 9) is

given by

H(f,0)=012+022xf (5)

represents a combination of an additive (021) and pro-portional (02) error model.

Both the pharmacokinetic and pharmacodynamicparameters were modeled with the assumption of lognormal distribution as follows:

P1 = Ppop X exp (llpj)

where Ppop is the population mean and 1h is the dif-ference between the population mean and the param-eter of the jth subject. i, was assumed to be agaussian variable with a mean of zero and variancea2. The population parameters were assumed to beindependently distributed without any covariance.

RESULTS

Pharmacokinetics

The pharmacokinetic parameters estimated were ab-sorption rate constant (Ka), volume of the centralcompartment (Vc/F), volume of the peripheral com-partment (Vp/F), oral clearance (Cl/F), and lag time.Population means and standard errors as well as esti-mates of the intersubject variances were obtainedand are shown in Table I. The population and Bayes-ian predictions are plotted against the measured val-

ues in Figure 1. The log10 pyridostigmine plasmaconcentrations are plotted against time for the lastdose interval for each individual in Figure 2. Themajority of subjects demonstrate a distribution andelimination phase. Relationships between the indi-vidual Bayesian predicted pharmacokinetic parame-ters and the covariates of body weight and genderwere used in the model as previously stated (equa-tions 1 and 2). The relationship between Vc/F andweight shown in Figure 3 and clearance and genderis shown in Table I. The high variability in Vp/F is

(6) secondary to six individuals having values wellabove and below the mean value.

Pharmacodynainics

The pharmacodynamic parameters estimated wereelimination rate constant from the effect compart-ment (K,,,,), baseline (where baseline is measured inunits of RBC acetylcholinesterase activity), and con-centration at steady state giving half maximal effect(EC50). To simplify the model, maximal effect (Em,,,,)was set to baseline. This was done because inhibitioncould not exceed baseline red blood cell acetyicho-linesterase activity, and with increasing pyridostig-mine plasma levels high degrees of inhibition(>70%) are possible.9 Population means and theirstandard deviations, along with intersubject vari-ances, were obtained and are shown in Table I.Transfer rate constant into the effect compartment(Kie) was fixed at 0.001 of elimination rate constantfrom the central compartment (1K,,) so that the amountof pyridostigmine modeled in the effect compart-



0 Q- - in

E E

-S. 0 in 0

CeC0 0

) C’) C’)

o 0

Q. > 0

‘5

0. a ‘-

C C

Ce

0 10

A Pyridostigmine plasma levels (ng mL-1) B Pyridostigmine plasma levels (ng mL-1)

0 10 20 30 40 50 0 10 20 30 40 50

E0)C

C

0

C

C

8CeECe

(9a.9)C

E0)

0V

0 4 5 6 7 8

Time (hrs)


PK/PD OF PYPIDOSTIGMINE BROMIDE FOR PROPHYLAXIS AGAINST NERVE AGENTS

Figure 1. Population predictions (A) and Bayesian predictions (B) plotted versus the individual values.

Figure 2. Pyridostigmine plasmalevels for all individuals andpopulation predicted mean ver-sus time for the average malesubject of 77kg (dotted line) andthe average female subject of 68

kg (solid line) during the firstdosing interval.



C’)

0U,C..J

0((1

0U,

50 60 70 80 90 100

Weight (kg)

Figure 3. Volume of the centralcompartment/F plotted versusweight. Solid line is the popula-tion predicted volume of the cen-tral compartment/F and the dot-ted lines are the 95% confidencelimits for the population predic-tion.

232 #{149}J Clin Pharmacol 1998;38:227-235

-J

LL

0>

C

E(9

0.

8(9

C

01

0a)E

0>

MARINO ET AL

ment did not influence the pharmacokinetics. K,,,,was 1.32 hrs, and the half-life (t112) for the effectwas 32 minutes. This is consistent with the shortdelay between maximum observed concentration(Cmax) and the maximum effect for each individualas well as the near return to effect baseline just before

the next dose. The red blood cell acetylcholinester-ase activity during the first dose interval for eachindividual is shown in Figure 4. No relationship wasfound between the pharmacodynamic parametersK,,0, EC50, and baseline and the covariates of weightand gender.

Figure 4. Red blood cell acetyl-cholinesterase activity for all in-dividuals and population pre-dicted mean versus time for theaverage male subject of 77 kg(dotted line) and the average fe-male subject of 68 kg (solid line)during the first dosing interval.





DISCUSSION

This report describes the first population pharmaco-kinetic/pharmacodynamic analysis of pyridostig-mine bromide. It also evaluates the relationship ofthe covariates gender and weight with basic kineticand dynamic parameters. The kinetic model that bestfit our data was one with a first-order input and a lagtime into a two-compartment model with first-orderelimination from the central compartment.

A previous study of pyridostigmine has shown itto have a two-compartment kinetic profile with intra-venous infusion or intravenous bolus with t112s of 7

to 9 minutes and 97 to 112 minutes for the distribu-tion and elimination phases, respectively.10’2 Onereport describes a three-compartment model whengiven as an intravenous bolus to anesthetized pa-tients, although the method of analysis was based ona radioimmunoassay for pyridostigmine.’3 Previousstudies using oral administration demonstrated aone-compartment kinetic profile with an eliminationt112 of 200 minutes, which the authors felt was dueto absorption proceeding slower than elimination.12

In our study, we were unable to detect the rapiddistribution phase as described in the intravenousstudies because of the slower absorption phase afteroral administration. We were able to estimate thisslow elimination phase, because the assay used wasmore sensitive and plasma samples were measuredfor longer periods after administration than in previ-ous studies.12 This suggests that intravenously ad-ministered pyridostigmine would follow a three-compartment model if plasma levels were taken fora sufficiently long period of the elimination phase,as confirmed by Stone et al.’3

Bioavailability was not measured in this study.Our value for Vc/F suggests a bioavailability of 12%using published values for Vc (21 ± 7 L).’#{176}This valueis consistent with a previously published value of14%.12 The variability in the estimate of Vc/F maybe due to the intraindividual variability from doseto dose within subject. Our value for oral clearanceis compatible with those in previous reports and sup-ports the idea that either poor absorption or signifi-cant first-pass metabolism occurs.’4 Oral clearance,which was calculated using the area under the con-centration-time curve (AUC) by the trapezoidal rulefrom the average of the first and last dose for eachsubject, was 311 ± 120 L/hr, which is much higherthan clearance calculated from intravenous adminis-tration (36 ± 7 L/hr).” Results of human metabolicstudies using ‘4C-labeled pyridostigmine are consis-tent with first-pass metabolism.’5 Human metabolitesof pyridostigmine have been defined in small stud-ies,’415 but the enzyme(s) responsible for pyridostig-

mine metabolism is unknown. Whether metabolite(s)have any intrinsic ability to inhibit acetylcholines-terase is also not known.

The pharmacodynamic model that best fit the redblood cell acetylcholinesterase activity was an inhib-itory Emax model linked to an effect compartment. Kie

was fixed to a small value so as not to influencethe kinetic parameters.’6 The baseline red blood cellacetylcholinesterase activity was modeled as the Em,,,,because maximal inhibition could not reduce valuesbelow zero and 100% inhibition is pharmacologi-cally possible with several of the inhibitors.’7 Thisalso reduced the number of parameters that neededto be estimated. K,,,, was found to be 1.32 hrs1, givinga t,,2 effect of 32 minutes.

In conjunction with the kinetic parameters govern-ing absorption and distribution, this explains therapid approach to pharmacodynamic equilibriumand the return to baseline within several hours ofthe final dose. With this rather short Keo t,,2, a slow-release formulation or an infusion would be requiredto maintain a more consistent level of red blood cellacetylcholinesterase inhibition.

The EC50 is the steady-state plasma concentrationof pyridostigmine needed to produce 50% of thebaseline red blood cell acetylcholinesterase activity.In previously published studies of other acetylcho-

linesterase inhibitors, red blood cell acetylcholines-terase percent inhibition was used as the pharmaco-dynamic endpoint.’8 This is the difference betweenthe baseline red blood cell acetylcholinesterase ac-tivity and the measured red blood cell acetylcholin-esterase activity at any time point expressed as apercentage of the baseline red blood cell acetylcho-linesterase activity. We did not use this in our analy-sis, because our placebos showed a ±16% intrasub-

ject variation from their mean baseline measurementover the course of the study, and we had only onepredose red blood cell acetylcholinesterase activitymeasurement for each subject. Rather, we estimatedthe baseline red blood cell acetyicholinesterase ac-tivity (using NONMEM) in our model.

In animal studies, 10% red blood cell acetyicholin-esterase activity inhibition provides protectionagainst organophosphorus agents.9 Based on the pha-rmacodynamic model the mean plasma concentra-tion needed to obtain 10% inhibition from this studyis 7.7 ng/mL. With the dosage regimen tested in thisstudy, the mean level of inhibition was greater than10% throughout the study. The population variabil-ity, however, would suggest that 30% of individualsmay be below this level at the time of the pyridostig-mine trough (Figure 5).

Despite tremendous variability in both the kineticand dynamic effects of pyridostigmine, we have



0C’)

(aCa,E o

U)(9a,E0

0

0,- 0_

Ez

0

C.’)

0-

-20 40 60

Inhibition of RBC AChE at troughbased on individuals modeled baseline AChE

0 20

MARINO ET AL

234 #{149}.1 Clin Pharmacol 1998;38:227-235

Figure 5. Histogram of individualred blood cell acetylcholinester-use percent inhibition based oneach individual’s modeled base-line at each trough.

shown that it is possible to produce an effectivemodel to describe both the pharmacokinetics andpharmacodynamics of orally administered pyrido-stigmine bromide. In this study of pyridostigminepopulation pharmacokinetics and pharmacodynam-ics, we found that when pyridostigmine bromide isgiven orally as a 30-mg tablet every 8 hours for 21days the drug demonstrates an “intermediate” distri-bution phase and a “slow” elimination phase. Thispharmacokinetic profile has not been previously de-scribed for the oral administration of pyridostigmine,probably because most studies did not take plasmasamples into the terminal elimination phase.

We also found significant interindividual variationin Vc/F. The covariates of body weight and genderdid appear to have a significant relationship with thekinetic parameters Vc/F and Cl/F. With the possibil-ity of a significant first-pass effect for this compound,tissue esterases and perhaps cytochrome P450 statusmay be more important factors in determining clear-ance or bioavailability.

Population pharmacodynamic parameters govern-ing red blood cell acetylcholinesterase inhibitionhave not been previously described. The analysisshows that an Em,,,, model with an effect compartmentfits the data well. Previously published work on an-other acetylcholinesterase inhibitors has shown theutility of the Em,,,, model.’8 The majority of the van-

ability in the pharmacodynamic response is second-ary to the variability in the kinetic profile. The t1,2for the effect compartment is relatively “fast,” andthus the delay between peak plasma levels and peakeffect is brief, as is the delay between trough plasmalevels and effect. Although the mean response isgreater than 10% inhibition during any dose interval,30% of subjects had inhibition lower than this valuearound the plasma trough level (Figure 5). Based onthe results of this study, it appears that this dosageregimen has utility in the possible protection of sol-diers against organophosphorus nerve agents.

The authors thank Ms. Michelle Sperry, Ms. Ruby Day, andMs. Natille Lee for their assistance with the complex task of data

management.

REFERENCES

1. Taylor P: Anticholinesterase agents. In, Hardman JG, Limbird

LE, Molinoff PB, Ruddon RW, Gilman AG (eds.): The Pharmaco-logical Basis of Therapeutics. New York: McGraw-Hill, 1996; 131-

149.

2. Koplovitz I, Stewart JR: A comparison of the efficacy of H16

and 2-PAM against soman, tabun, sarin, and VX in the rabbit.

Toxicol Lett 1994;70:169-279.

3. Gordan JJ, Leadbeater L, Maidment MP: The protection of ani-

mals against organophosphate poisoning by pretreatment with a

carbamate. Toxicol Appl Pharmacol 1978;52:71-92.





4. Keeler JR, Hurst CG, Dunn MA: Pyridostigmine used as a nerve

agent pretreatment under wartime conditions. JAMA 1991;

266:693-695,

5. Kluwe WM: Efficacy of pyridostigmine against soman intoxica-

tion in a primate model. In, Proceedings of the Sixth MedicalChemical Defense Bioscience Review. Aberdeen Proving Ground,

MD: United States Army Medical Research Institute of Chemical

Defense, 1987; 227-234.

6. Lin ET, Gee WL: Quantitation of pyridostigmine (as free base)

in human plasma by high-performance liquid chromatography

with a silica gel column and aqueous mobile phase. United States

Army Medical Research and Material Command Contract No.

DAMD17-92-C-2028, 1995.

7. Ellin RI, Kaminskis A: Carbamoylated enzyme reversal as a

means of predicting pyridostigmine protection against somail. IPharm Pharmacol 1989;41:633-636.

8. Beal SL, Sheiner LB: NONMEN Project Group. San Francisco:

University of California, 1989.

9. Lennox WI, Harris LW, Talbot BG, Anderson DR: Relationship

between reversible acetylcholinesterase inhibition and efficacy

against soman lethality. Life Sci 1985;37:793-798.

10. Cronnelly R, Stanski DR, Miller RD, Sheiner LB: Pyridostig-

mine kinetics with and without renal function. Clin PharmacolTher 1980; 28:78-81.

11. Beal SL, Sheiner LB: Estimates, confidence intervals, and hy-

pothesis intervals. In, NONMEM Project Group Manual 5. SanFrancisco: University of California, 1989.

12. Breyer-Pfaff U, Maier U, Brinkmann AM, Schumm F: Pyrido-

stigmine kinetics in healthy subjects and patients with myasthenia

gravis. Gun Pharmacol Ther 1985;37:495-501.

13. Stone JG, Matteo RS, Orustein E, Schwartz AE, Ostapkovich

N, Jumdar SC, Diaz J: Aging alters the pharmacokinetics of pyrido-

stigmine. Anesth Analg 1995;81:773-776.

14. Somani SM, Roberts JB, Wilson A: Pyridostigmine metabo-

lism in man. Gun Pharmacol Ther 1972;13:393-399.

15. Komfield P, Samuels AJ, Wolf RL, Osserman KE: Metabolism

of 14C-labeled pyridostigmine in myasthenia gravis. Neurology

1970; 20:634-641.

16. Sheiner LB: Modeling pharmacodynamics: parametric and

nonparainetric approaches. In, Rowland M (ed.): Variability inDrug Therapy: Description, Estimation, and Control. New York:Raven Press, 1985;139-152.

17. Sidell FR, Groff WA: The reactivability of cholinesterase in-

hibited by VX and sam in man. Toxicol App) Pharmaco)1974; 27:241-252.

18. Cutler NR, Seifert RB, Schleman MM, Sramek JJ, Szylleyko OJ,

Howard DR, et al: Acetylcholinesterase inhibition by zifrosilone:

pharmacokinetics and pharmacodynamics. Clin Pharmacol Ther1995; 58:54-61.



The Journal of Clinical...

Documents

Transcript of The Journal of Clinical...