influence of the cyp1a2 inhibitor fluvoxamine on tacrine pharmacokinetics in humans*

PHARMACOKINETICS AND DRUG DISPOSITION

Influence of the CYPlA2 inhibitor fluvoxamine on tacrine pharmacokinetics in humans

Ob&ctire: Tacrine is extensively metaboiized by cytochrome P45OlA2 (CYPlA2). Fluvoxamine, a potent CYPlA2 inhibitor, may be wadministered with tacrine. The aim of this study was to examine the influence of fluvoxamine administration on the disposition kinetics of single-dose tacrine administration. Met&s& Thirteen healthy volunteers participated in this double-blind, randomized crossover study, which compared the effects of fluvoxamin e (100 mg/day during 6 days) and placebo on the pharmawkinetics of a single orai dose of tacrine (40 mg). Result: Fluvoxamin e caused a sign&ant increase in tacrine area under the plasma concentration versus time curve (AUC): arythmetic mean, 27 (95% confidence interval [CI], 19 to 38) ng * hr/mi versus 224 (95% CI, 166 to 302) ng - hr/mi. Fhrvoxamine caused a decrease in the apparent oral clearance of tacrine from 1683 f 802 to 200 f 106 L/hr (mean -C SD), which was explained by a decrease in its nonrenai clearance. Five subjects had gastrointesti.uai side tffects during fluvoxamine administration. Fhrvoxamine administration was associated with significant increases in the plasma AUC values of three monohydroxylated tacrine metaboiites and in the total urinary recovery measurements of tacrine and its metaboiites (9.1% -C 4.6% versus 24.0% 2 2.6% of recovery). These results may be attributable to fluvoxamine- dependent inhibition of CXPlA2, which is responsible of the biotransformation of tacrine into its monohydroxyiated metabolites and further into dihydroxylated and reactive metabolites. Conclusion: Fluvoxamin e inhibits the metabolism of tacrine. CYPlA2 may be the target of this inhibition. Fluvoxamine may modulate the hepatotoxicity of tacrine, depending on the relative contribution of tacrine and its reactive metabolites to this toxicity. (Clin Pharmacol Ther 1997;61:619-27.)

Laurent Becquemont, MD, Isabelle Raguen=u, MD, Marie Annick Le Bot, PhD, Christian Riche, MD, PhD, Christian Funck-Brentauo, MD, l?hD, and Patrice Jaillon, MD Paris and Brat, France

From the Clinical Pharmacology Unit, Saint Antoine University Hospital, Paris, and the Department of Pharmacology, Univer- sity of Brest-School of Medicine, Brest.

Supported by a grant-in-aid from the Agence Francaise du MC- dicament (Clinical Pharmacology network) and the BIOMED II research program (Hepatox network project).

Received for publication Aug. 7, 1996; accepted Jan. 5, 1997.

Reprint requests: Laurent Becquemont, MD, Unite de pharma- cologie clinique Hopital Saint Antoine, 184 rue du Faubourg Saint Antoine, 75012 Paris, France.

Copyright 0 1997 by Mosby-Year Book, Inc.

0009-9236/97/$5.00 + 0 13/UlUJ343

Alzheimer’s disease, the most common form of dementia, has become a major public health prob- lem with the progressive aging of populations in developed industrial countries. Tacrine (O-amino- 1,2,3,4-tetrahydroacridine), an anticholinesterase inhibitor, is currently the only drug approved for the treatment of Alzheimer’s disease.‘s2

Serum aminotransferase elevation occurs in up to 50% of patients receiving tacrine.3 The mechanism of this transaminitis remains unclear; tacrine has been shown to be directly responsible for its hepatotoxicity through protein synthesis inhibition,4 but

619

620 Becquemont et al. CLINICAL PHARMACOLOGY & THERAPEUTICS

JUNE 1997

the responsibility of tacrine metabolites (through reactive metabolites) is also well documented.5-7 In- deed, tacrine is extensively metabolized by the liver during first pass. 2 This leads to the production of different monohydroxylated metabolites, which are further metabolized into dihydroxylated and reactive metabolites.6-8 Hepatic cytochrome P45OlA2 (CYPlA2) has been shown to be the major CYP isoform involved in this metabolic process.5>779 Whether tacrine or its metabolites are responsible for hepatotoxicity observed in patients remains to be determined.

Depressive symptoms are frequently observed among patients with Alzheimer’s disease. Up to 25% of patients treated with tacrine receive concomitant antidepressive drugs, and fluvoxamine may be one of those prescribed. Fluvoxamine, a selective serotonin reuptake inhibitor (SSRI), is a potent inhibitor of CYPlA2 in vitro and in viva.‘-” Further- more, fluvoxamine has been shown to inhibit tacrine metabolism in vitro.’ Because tacrine metabolic clearance plays a crucial role in its elimination,2v’2 fluvoxamine may dramatically decrease its clearance and, as a consequence, modulate its hepatotoxicity.

The aim of this study was to investigate the influence of fluvoxamine on single-dose tacrine pharmacokinetics in healthy volunteers.

MATERIAL AND METHODS Study &sign. The study was a randomized, double-

blind, two-period cross-over trial that was performed in 14 healthy male volunteers. Clinical ex- amination and standard laboratory tests were performed before the subjects’ inclusion to ensure that they were healthy. Subjects had a normal diet before the study, but they were advised that coffee, tea, and chocolate were forbidden during the study period because these substances could interfere with the metabolism of tacrine. They were also advised that the presence of caffeine would be assayed in plasma during the two study periods. None of the subjects were smokers. The study protocol was approved by the Committee for the Protection of Hu- man Subjects in Biomedical Research of Paris, PitiC- Salpttribre. Subjects gave written informed consent to participate, and the study was performed according to French regulations.

The order of the two study periods was randomized according to a Latin-square design. During one study period, subjects received one capsule of placebo each morning for 6 days. On the morning of the sixth day, subjects were hospitalized in the Clin-

ical Pharmacology Unit at Saint Antoine University Hospital, the last dose of placebo was taken together with one tablet of 40 mg tacrine taken orally with 150 ml tap water after an overnight fast. Blood samples were collected from an antecubital vein immediately before tacrine administration (time zero) and 1/2, ?A, 1, l?$ 13/4, 2, 3, 4, 6, 8, 12, 16, 24, and 30 hours thereafter. Subjects emptied their bladders immediately before tacrine administration, and urine was collected during the following 24 hours. The blood samples collected just before tacrine administration were used to quantify plasma tacrine and its metabolites, as well as fluvoxamine and caffeine. The volume of urine collection from each volunteer was measured, and 20 ml samples were stored at -24” C until analysis. Subjects were hospitalized for 24 hours and remained in a fasting state until 2 hours after tacrine intake.

During the other study period subjects received 1 capsule of 100 mg fluvoxamine each morning for 6 days. On the morning of the sixth day, subjects were hospitalized, the last dose of fluvoxamine was taken together with 40 mg tacrine, and procedures followed during the other study period were repeated exactly in the same way.

Fluvoxamine plasma concentrations were determined during both periods on the sixth day of both periods at time zero (just before the intake of tacrine) and 24 hours later. To verify that the subjects were not taking coffee during the study, caffeine plasma concentrations were determined on the morning of the first day of each period and on the sixth day of both periods at time zero. Blood samples were collected in lithium-heparinized glass tubes for determination of tacrine and fluvoxamine plasma concentrations and in dry glass tubes for caffeine assay. The tubes were centrifuged at 400@ and +4” C within 10 minutes, and plasma was collected and stored at -24” C until further analysis.

Drugs, chemicals, and reagents. Tacrine and fluvoxamine tablets were purchased from the hospital pharmacy as the commercially available 40 mg tablets (Cognex, Parke-Davis, France) and 100 mg tablets (Floxyfral, Solvay Pharma, Duphar, France), respectively. To make the fluvoxamine and placebo pills indistinguishable, fluvoxamine tablets were pul- verized by the hospital pharmacist and then recon- stituted in capsules of the same appearance as the placebo capsules.

Standards for HPLC. Tacrine hydrochloride, 1-hydroxytacrine maleate, and caffeine were purchased from Sigma Chemical Co. (Saint Quentin

CLINICAL PHARMACOLOGY &THERAPEUTICS VOLUME 61, NUMBER6 Becquemont et al. 62 1

Fallavier, France). 2-Hydroxytacrine maleate, 4-hydroqacrine maleate, and fluvoxamine maleate were a gift from Parke Davis Pharmaceutical Re- search Division (Ann Arbor, Mich.), and Solvay Duphar (Weesp, Holland), respectively.

D&?l2?minationoftacrilleandits-i?lplasma and urine. Plasma levels of tacrine, 1-hydroxytacrine, 2-hydroxytacrine, and 4-hydroxytacrine were measured by HPLC. Tacrine extraction was performed according to the method of Haughey et ali3 with a double extraction procedure to improve the extraction yield. The double extraction procedure was performed after alkalinization of 1 ml plasma or urine with 250 ~1 of 1 mol/L sodium hydroxide by use of a mixture of chloroform/propanol-1 (9/l, vol/vol). The urine was not submitted to hydrolysis before extraction.

HPLC analyses were preformed with use of a Therm0 Separation Product system (P4000, AS3000 set at 4” C; San Jose, Calif.) on a 5 Frn Cis 250 X 4 mm end-capped Lichrospher column (Merck) with a 5 pm Cis 4 X 4 mm Lichrospher precolumn (Merck). Fluorescence detection was performed with a TSP FL2000 spectrofluorometer with an ex- citation wavelength of 240 nm and an emission wavelength of 356 mu. The solvent mixture was composed of 0.8% triethylamine in water adjusted to pH 5.5 with phosphoric acid and acetonitrile. The percentage of acetonitrile was maintained at 7% for 20 minutes and then linearly raised to 20% in 5 minutes and maintained at this value for 15 minutes. The flow rate was 0.9 ml * min-‘. The detection limit of the method was 0.1 r&ml for tacrine and its metabolites.

Determination of j’luvoxamine in pkwna. Fluvox- amine was extracted from plasma according to the method of Foglia et al. I4 A double extraction from plasma was carried out and fluvoxamine was subse- quently analyzed with HPLC. The elution solvent was a mixture of water/acetonitrile/triethylamine (55/45/l; vol/vol), with a flow rate of 1 ml * min-r. A 5 t.t.rn C,, 250 X 4 mm end-capped Lichrospher 100 RP column (Merck) was used. The material used was the same as that used for the measurement of tacrine. Detection was carried out at 254 nm on a TSP ultraviolet 2000 photometer. The limit of detection was 3 rig/ml.

Determination of caffeine in plasma. The measurement of caffeine was carried out with use of a Syva Bio MCrieux kit and an EMIT method with a Cobas Mira Plus photometer (Roche). Wavelength was set at 340 nm, and the detection limit of this method was 1 P&ml.

Phurmacokinetic analysis. Tacrine pharmacokinet- its were analyzed by using noncompartmental techniques. The area under the plasma concentration versus time curve (AUC) was calculated by use of the trapezoidal and log-trapezoidal rules for ascending and descending plasma concentrations, respectively. l5 Because plasma levels of 2-hydroxytacrine and 4-hydroxytacrine remained low during both study periods, AUC(O-24) was calculated instead of AUC(O-co) for these metabolites.

Tacrine apparent oral clearance (CL,,) was calculated as follows:

in which dose is the tacrine dose administered as the base (40 mg) and AUC(O-CQ)t,,i,, is the area under the tacrine plasma concentration versus time curve calculated from zero time to infinity. Tacrine apparent plasma terminal elimination half-life (t& was calculated as follows:

0.693k

in which k, is the slope of the log(tacrine plasma concentration) versus time line after least-square regression analysis of the terminal portion of this relationship.

Renal clearance (CL,) values of tacrine, 1-hydroxytacrine (1-OH-tacrine), 2-hydroxytacrine (ZOH-tacrine), and 4-hydroxytacrine (4-OH- tacrine) were calculated as AetacrineJAuC

WWtactine7 Ael-oH-tacrin~AUC(O-24)1-OH-tacrine, Ae2-oH-tacrin~AUC(O-24)2.0H.taciine, A%.OH-tactiJ AUC(O-24),oH.tactieY respectively, with use of data measured in urine collected over 24 hours where Ae represents the amount of tacrine or its metabolites excreted in urine over 24 hours and the subscripts 1, 2, and 4 represent l-, 2-, and 4-hydroxytacrine respectively. Tacrine apparent nonrenal clearance (Ch,) was calculated as follows:

CLraI - tacrine CL,

Statistical analysis. The results are reported as mean + SD. Comparisons between the two periods were made by use of the paired Wilcoxon test. Dif- ferences were considered to be statistically significant when the probability of erroneously rejecting the null hypothesis of no difference was less than 5%.

For the log-normally distributed pharmacokinetic parameters (i.e., AUC and peak plasma concentration [C&J), results are expressed as the arithmetic mean with 95% confidence interval. For these phar-

622 Becquemont et al. CLINICAL PHARMACOLOGY &THERAPEUTICS

JUNE 1997

30- Plasma concentratioll q glml

zo-

30-

Plasma conceatntlon nglml

20-

Tacrine fluvoxamine

, I

0 10 20 30

time (II)

I-OH-tacrine placebo

I-OH-tacrine fluvoxamine

b io 20 i0

time (h)

Fig. 1. A, Tacrine plasma concentration versus time curve during coadministration of placebo (open circles) and fluvoxamine (solid circles). B, l-Hydroxytacrine (l- OH-tacrine) plasma concentration versus time during coadministration of placebo (open circles) and fluvoxamine (solid circles). Data are expressed as mean values k SD of the 13 subjects.

macokinetic parameters, differences between study periods were calculated by geometric means on the basis of logarithmic transformation of the intraindi- vidual ratios of the log-normal parameters AUC and C mu The differences between study periods and the standard deviations of the 90% confidence limits were calculated by ANOVA and the Dunnett test.

RESULTS Subjects had a mean age of 27 If: 4 years (range,

21 to 33 years). Thirteen of them completed the study and one was excluded after the second dose of fluvoxamine because he had nausea and hypoten- sion. Compliance during both study periods was judged to be excellent on the basis of pill count.

Fluvoxamine and caffeine plasma concentrations. Fluvoxamine plasma concentrations immediately before administration of the tacrine dose and 24 hours later were 31 -C 24 and 31 + 24 q/ml, respectively, indicating that fluvoxamine steady state was reached. No fluvoxamine was detected in plasma during the placebo period. Traces of caffeine were found in plasma samples of only two subjects: one at time zero during both study periods, the other at time zero of the placebo period. Caffeine concentrations were at the lower limit of detection, and these subjects were not excluded from the study.

Pharmacokinetic parameters for tucrine and its metabolites. Fluvoxamine caused eightfold and fivefold increases in tacrine AUC(O-m) and C,,, respectively (Fig. 1, A), without any significant changes in the time to reach C,, (t,,) and t,,, of tacrine (Table I).

Three metabolites were assayed in plasma. Quan- titatively, l-hydroxytacrine was the major metabolite, followed by 2-hydroxytacrine, and Chydroxytacrine, respectively. The three metabolites were detectable in the plasma of all subjects during the placebo period, whereas Chydroxytacrine was undetectable in the plasma samples of two subjects during coadministration of fluvoxamine. Twofold to fivefold increases in the AUC values of the three metabolites were observed in the presence of fluvoxamine (Fig. 1, B, and Table I). 1-Hydroxytacrine and Zhydroxytacrine C,, values remained unchanged during both study periods, whereas 4-hydroxytacrine C,, increased significantly during fluvoxamine administration. Apparent trn of 1-hydroxytacrine increased from 3.7 + 0.8 hours to 7.6 k 3.0 hours during fluvoxamine administration. There was a significant correlation between the AUC values of the three metabolites during placebo (Fig. 2, A and B) and fluvoxamine administration (Fig. 2, C and D).

CIearances and urinary recoveries. There was a great inter-individual variability in tacrine apparent CLoral during the placebo period. Fluvoxamine caused an eightfold decrease in tacrine apparent CLoral that was associated with a large decrease of its variability (Fig. 3 and Table II). Subjects who had the highest tacrine apparent CL,,,, values during the placebo period had the greatest decrease in tacrine apparent CL,, values in the presence of fluvoxamine (Fig. 3). Because tacrine CL, remained unchanged during both periods, tacrine CI+, decreased significantly during fluvoxamine administration (Table II).

The CL, values of 1-hydroxytacrine and

CLINICAL PHARMACOLOGY & THERAPEUTICS VOLUME 61, NUMBER6 Becquemont et al. 623

Table I. Single-dose plasma pharmacokinetic parameters of tacrine and its metabolites in study subjects Placebo Fluvoxamine period period

Tacrine Cm, WW AUC(O-m)t,,i,e (ng : hr/ml) AUC(O-24)tacrine (ng - hr/ml) t,, W tljz (W

I-Hydroxytncrine Cm, @g/ml) AUCtO-O”)t-OH-tacrine (ng * h/ml) AUC(O-24)l-oH-tacrine (ng - hrhl) b, thd b2 W

2-Hydroxytcrine Cm, (rig/ml) AUC(@24)2-oH-tactine (ng - hr/ml) t,, W

4-Hydroxytacrine C,, (rig/ml) (n = 11) AUC(0-24)4-oH-tacrine (ng * hr/ml) (n = 11) t,, (hr) (n = 11)

7 (4-10) 39 (30-51)* 27 (19-38) 224 (166-302)* 27 (19-38) 222 (165-299)* 1.1 2 0.4 1.3 2 0.5 3.8 + 1.7 4.1 2 0.7

32 (24-38) 104 (82-132) 103 (81-131) 1.2 2 0.4 3.7 2 0.8

4 (3-5) 14 (11-18) 1.1 2 0.5

1.0 (0.8-1.2) 5 (3-7)

1.5 + 0.2

31 (26-38) 294 (246-352)* 273 (233-319)* 1.0 + 0.6 7.6 t 3.0*

4 (3-5) 28 (23-34)* 1.3 + 1.4

2.0 (1.8-2.3)* 25 (21-33)* 5.4 ? 3.7*

C ITlax? Peak plasma concentration; AUC, area under the plasma concentration-time curve; f-, time to reach C,,; t,,, apparent terminal elimination half-life.

Results are expressed as the mean t- SD (or arithmetic mean with 95% confidence interval for the log-normal pharmacokinetic parameters AUC and C,,,a of the 13 subjects for tacrine, 1-OH-tacrine and 2-OH-tacrine plasma data and of 11 subjects for 4-OH-tacrine plasma data. Statistical comparison between olacebo and fluvoxamine oeriods were made bv use of the uaired Wilcoxon test or by ANOVA and Dunnett test for AUC and C,,. L l p < 0.05.

4-hydroxytacrine remained unchanged during both periods, whereas 2-hydroxytacrine CL, was slightly but significantly increased. Tacrine, 1-hydroxytacrine, and 2-hydroxytacrine could be detected in the urine of all subjects during both periods. Because of technical reasons, there was an important degradation of the 4-hydroxytacrine in the urine of the first six subjects. Therefore urinary recovery of this metabolite could be quan- tified in only the last seven subjects.

There was a sevenfold increase in tacrine urinary recovery during fluvoxamine administration (Table II). The urinary recovery values of the three metabolites was also increased, but to a lesser extent (twofold to fivefold), in the presence of fluvoxamine.

Side e$ects. No side effect could be observed during the placebo period. Five subjects had side effects during administration of tacrine and fluvoxamine soon after tacrine intake. The side effects consisted in nausea, vomiting, sweating, and diarrhea. Inter- estingly, four of these subjects had the highest tacrine C,, and the side effects appeared 15 to 45 minutes after t,, Side effects remitted rapidly, and the 13 subjects could complete the study. No bio- logical abnormality-namely, serum aminotransfer-

ase elevation-could be detected during or after any of the two study periods.

DISCUSSION Singlfdose phanmcokinetics of tacrine in the

absence of Jluvommine. Tacrine’s major measured metabolite was 1-hydroxytacrine. The percentage of recovery in a 24-hour urinary collection was about 8% of the administered dose for the sum of tacrine and its metabolites. These results are in agreement with those observed in a previous study conducted in humans in which the percentage of urinary recovery of tacrine and its metabolites was 7%.16 In vitro studies have shown a fourth monohydroxylated metabolite (7-hydroxytacrine) and traces of dihydroxylated metabolites when tacrine was incubated with human liver microsomes.5P71’7 The existence of these additional metabolites and the incomplete gastroin- testinal absorption of tacrine may contribute to the explanation of the poor urinary recovery values of tacrine and its metabolites observed in the literature. We have no clear explanation for the absence of detection of these other metabolites in this study, but in vitro studies used isotopic detection of tacrine and its metabolites, a technique that is much more

624 Becquemont et al. CLINICAL PHARMACOLOGY &THERAPEUTICS

JUNE 1997

r = 0.6 p = 0.04

A AUC l-OH-tacrine hWml) B AUC I-OH-tacrine (nghhb

01 0 loo 200 300 400

C AUC I-OH-tacrine (&ml)

501 = r

$ 4o-p E

g 30- ‘Z Y

2 20-

4,

= =

0.6 0.04

a*

I’3’ 0 100 200 300 400

D AUC 1-OH-tacrine (q/ml)

Fig. 2. Regression analysis of 2-hydroxytacrine area under the plasma concentration versus time curve [AUC(O-24); A and C] and 4-hydroxytacrine AUC(O-24) (B and D) versus 1-hydroxytacrine AUC(O-24) during coadministration of placebo (open circles) and fluvoxamine (solid circles), respectively.

sensitive than the HPLC assay used in the this study. Nevertheless, to our knowledge, the detection of these additional metabolites in humans has never been published.

Interestingly, there was a significant correlation between the AUC of 1-hydroxytacrine and the two other metabolites, which suggests that these three metabolites are formed by the same isozyme. There is abundant literature that shows that the in vitro metabolism of tacrine involves one single human hepatic isozyme, namely, CYPlA2?-7*g~16~‘7 A correlation between the rate of each tacrine metabolite generation has been observed in vitro.” The rate of each tacrine metabolite generation was also corre- lated to the immunocontent of human hepatic CYPlA2.” Furthermore, tacrine inhibits in vivo theophylline metabolism, which is also mediated by CYP1A2.‘1 These different findings confirm that the

correlation between the AUC of the different tacrine metabolites observed in the present study in- dicates that these metabolites are produced by CYPlA2 in vivo.

Tacrine pharmacokinetics exhibited a high interindividual variability. There was a sevenfold interindividual variability in tacrine apparent CLoral. These results are in agreement with previous studies that reported such a high variability.2-18-‘g This high interindividual variability may be related to tacrine extensive hepatic first-pass metabolism, which may contribute to the nonlinearity of tacrine pharmacokinetics.520 Because tacrine is extensively metabolized by hepatic CYPlA2 and because this cytochrome P450 has a thirtyfold interindividual variability in its expression in human livers,21 the interindividual variability of pharmacokinetic parameters observed in this study is not surprising.

CLINICAL PHABMACOL4XY k THERAPEUTICS VOLUME 61. NUMBER 6 Becquemont et al. 625

Table II. Clearances and percentages of urinary recovery of tacrine and its metabolites

Placebo Fluvoxamine period period

Clearances (Llhr) axa Tacrine CL, 1-OH-tacrine CL, 2-OH-tacrine CL 4-OH-tacrine CL,

(n = 7) C&Z

Urinary recovery Total recovery (%)

(n = 13) Tacrine (%)

(n = 13) 1-OH-tacrine (%)



(n = 7)

1683 ? 802 200 -c 106*

34oz4 33 422 2 9 23 2 8 27 t 9*

722 11 2 11

1678 t- 801 197 t 106*

7.9 4 3.9 24.6 5 2.3*

0.3 ” 0.3 2.2 + 1.0*

6.8 2 3.5 20.5 + 1.6*

0.7 + 0.3 1.7 + 0.4*

0.1 f 0.1 0.5 ” 0.2*

CL.,,,, Apparent oral clearance; t&, renal clearance; Cb, nonrenal clearance.

Results are expressed as mean values t SD of the 13 subjects, except for the percentage of COH-tacrine urinary recovery, which was expressed as the mean -t SD of the last seven subjects because COH-tacrine urinary recovery was not available for the six first subjects (see Results section). Statistical comparison between placebo and fluvoxamine periods were made by use of the paired Wilcoxon test.

*p < 0.05.

Influence of jluvoxamine on tacrine single-dose pharmacokinetics. The results of this study show that in healthy volunteers, fluvoxamine interferes with tacrine pharmacokinetics, leading to a dramatic increase in tacrine plasma concentrations.

Fluvoxamine caused a significant increase in tacrine AUC and C,, The observed decrease in tacrine apparent CL,, may be caused by a decrease in tacrine intrinsic clearance, an increase in tacrine bioavailability (F), or a combination of changes in these two parameters, inasmuch as observed apparent CLoral is the ratio of intrinsic clearance divided by F. The increase in tacrine AUC in the presence of fluvoxamine may be related in part to an increase in tacrine intestinal absorption, because we observed increases in the urinary recovery values of tacrine and its metabolites during fluvoxamine administration. However, this hypothesis remains theoretical because no clinical study has yet indicated that fluvoxamine may increase intestinal absorption of other drugs. Therefore it seems to be likely that the increase in tacrine AUC and the decrease of its

4000-

3ooo-

ypc 2ooo-

(I/h)

lOoo-

O- tacrine + tacrine + placebo tluvoxamine

Fig. 3. Individual apparent oral clearance (CL,,,,) of tacrine during coadministration of placebo and fluvoxamine.

apparent CL,,, are mainly related to inhibition of tacrine hepatic metabolism inhibition in the presence of fluvoxamine. This hypothesis is supported by the fact that tacrine is extensively metabolized by human liver CYPlA2 and that fluvoxamine is a well-known CYPlA2 inhibitor in vivo.2,10-‘2

Fluvoxamine abolished the large interindividual variability of tacrine apparent CLoral observed in its absence. The greatest decrease in tacrine apparent CLoral in the presence of fluvoxamine was observed among the subjects who had the highest tacrine apparent CLoral in the absence of fluvoxamine. This may be the consequence of fluvoxamine-dependent CYPlA2 inhibition, because the large interindividual variability in tacrine CL,, in the absence of fluvoxamine may be related to the interindividual variability of CYPlA2 expression. These observa- tions reinforce the fact that in vivo tacrine clearance is mainly dependent on CYPlA2 activity and that fluvoxamine interferes with tacrine clearance mainly by CYPlA2 inhibition.

Injluence offluvoxamine on the disposition of &wine metuboliks. The AUC values of the tacrine metabolites also increased significantly but to a lesser extent than the AUC of tacrine. The increases in the AUC values of the tacrine metabolites in the presence of fluvoxamine were the result of decreases in their Cb, values, because their CL, values remained unchanged or increased during the fluvoxamine period. The concomitant increases in the, AUC values of the tacrine monohydroxylated

626 Becquemont et al.

metabolites may seem paradoxical because fluvoxamine-dependent CYPlA2 inhibition of tacrine metabolism should have inhibited their production. Results from previous studies may help to explain these results.

It has been observed in vitro that the monohydroxylated metabolites of tacrine (l-, 2-, 4-, and 7-hydroxytacrine) are further metabolized in part into dihydroxylated metabolites and reactive metabolites6-* CYPlA2 is involved in the biotransformation of tacrine monohydroxylated metabolites into dihydroxylated and reactive metabolites.7p8 Be- cause fluvoxamine is a potent inhibitor of CYPlA2, it may inhibit tacrine metabolism into monohydroxylated metabolites and also the metabolism of the latter into dihydroxylated and reactive metabolites, as shown previously when with other CYPlA2 inhibitors.5’8 The potential inhibition of tacrine monohydroxylated metabolites into dihydroxylated and reactive metabolites produced by fluvoxamine may lead to increases in the AUC values of tacrine monohydroxylated metabolites and to increases in their urinary recovery, as observed in this study. It is not possible to confirm this hypothesis because dihydroxylated and reactive metabolites remained undetectable in plasma or urine.

Whatever the precise mechanism of the interaction between fluvoxamine and tacrine, our data clearly show that the combination of both drugs is associated with dramatic increases in both tacrine plasma concentrations and its monohydroxylated metabolites.

Clinical impbtions. In this study, fluvoxamine administration increased tacrine plasma concentrations during single-dose administration of tacrine. Our results cannot necessarily be extrapolated to patients with Alzheimer’s disease who are receiving repeated doses of tacrine. The manipulation of fluvoxamine tablets performed in this study may have also influ- enced the present pharmacolcinetic interaction.

However, if such a drug interaction occurs during repeated dosing of tacrine, different clinical impli- cations could theoretically be observed. First, gas- trointestinal side effects, such as those observed in this study, may appear. They may lead to drug dis- continuation or to poor compliance by the patient. Such side effects may not be restricted to the interaction of fluvoxamine with tacrine metabolism but also with other well-known or potential CYPlA2 inhibitors such as antibacterial quinolones, cimetidine, and even grapefruit juice.22-24 With regard to tacrine hepatotoxicity, the consequences of the interaction will depend on the mechanism of this ad-

CLINICAL. PHARMACOLQGY &THERAPEUTICS IUNE 1997

verse effect, which remains uncertain. If tacrine is directly responsible for its hepatotoxicity, as sug- gested by Fariss et al., 4 it may be predicted that coadministration of fluvoxamine with tacrine will increase the proportion of patients with elevation of serum aminotransferase during tacrine treatment. In contrast, if reactive metabolites are responsible for hepatotoxicity, it is difhcult to predict from the results of this study what the effects of fluvoxamine on tacrine hepatotoxicity could be. The consequences of concomitant prescription of tacrine and CYPlA2 inhibitors in patients with Alzheimer’s disease need further clinical investigation.

References 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

Knapp MJ, Knopman DS, Solomon PR, Pendlelbury W, Davis, CS, Gracon SI. A 30-week randomized controlled trial of high-dose tacrine in patients with Alzheimer’s disease. JAMA 1994;271:985-91. Schneider LS. Clinical pharmacology of aminoacri- dines in Alzheimer’s disease. Neurology 1993; 43(suppl4):%4-79. Walkins PB, Zimmerman HJ, Knapp MJ, Gracon SI, Lewis KW. Hepatotoxic effects of tacrine administration in patients with Alzheimer’s disease. JAMA 1994; 271:992-g. Fariss MW, Johnsen SA, Walton LP, Mumaw VR, Ray SD. Tetrahydroaminoacridine-induced ribo- somal changes and inhibition of protein synthesis in rat hepatocyte suspension. Hepatology 1994;20: 240-6. Madden S, Woolf TF, Pool WF, Park BK. An investigation into the formation of stable, protein-reactive and cytotoxic metabolites from tacrine in vitro. Bio- them Pharmacol 1993;46:13-20. Madden S, Spaldin V, Hayes RN, Woolf TF, Pool WF, Park BK. Species variation in bioactivation of tacrine by hepatic microsomes. Xenobiotica 1995;25: 103-16. Woolf TF, Pool WF, Bjorge M, Chang T, Goel OP, Purchase CF II, et al. Bioactivation and irreversible binding of the cognition activator tacrine using human and rat liver microsomal preparations. Drug Metab Dispos 1993;21:874-82. Spaldin V, Madden S, Pool WF, Woolf TF, Park BK. The effect of enzyme inhibition on the metabolism and activation of tacrine by human liver microsomes. Br J Clin Pharmacol 1994;38:15-22. Becquemont L, Le Bot MA, Riche C, Beaune P. Influence of fluvoxamine on tacrine metabolism in vitro: potential implication for the hepatotoxicity in vivo. Fundam Clin Pharmacol 1996;10:156-7. Brasen K, Skjelbo E, Rasmussen BB, Poulsen HE, Loft S. Flwoxamine is a potent inhibitor of cytochrome P450 lA2. Biochem Pharmacol1993;45:1211-4.

CLINICAL PHARMACOLOGY &THERAPEUTICS VOLUME 61, NUMBER 6 Becquemont et al. 627

11. Sperber AD. Toxic interaction between fluvoxamine and sustained release theophylline in an 11-year-old boy. Drug Safety 1991;6:460-2.

12. Davis KL, Pow&& P. Tacrine. Lancet 1995;345:625-30. 13. Haughey DB, McNaney CA, Collis MS, Brown RR,

Siedlik PH, Balogh L, et al. Simultaneous determination of tacrine and 1-hydroxy-, 2-hydroxy-, and 4-hydroxytacrine in human plasma by high- performance liquid chromatography with fluorescence detection. J Pharm Sci 1994;83:1582-5.

14. Foglia JP, Birder LA, Perel JM. Determination of fluvoxamine in human plasma by high-performance liquid chromatography with ultraviolet detection. J Chromatogr 1989;495:295-302.

15. Chiou WL. Critical evaluation of the potential error in pharmacokinetic studies of using the linear trapezoidal rule method for the calculation of the area under the plasma level time curve. J Pharmacokinet Biopharm 1978;6:539-46.

16. Hooper WD, Pool WF, Woolf TF, Gal J. Stereose- lective hydroxylation of tacrine in rats and humans. Drug Metab Dispos 199422719-24.

17. Spaldin V, Madden S, Adams DA, Edwards RJ, Da- vies DS, Park BK. Determination of human hepatic cytochrome P45OlA2 activity in vitro: use of tacrine as an isoenzyme-specific probe Drug Metab Dispos 1995;23:929-34.

18. Selen A, Balogh L, Siedlik P. Pharmacokinetics of tacrine in healthy subjects [abstract]. Pharm Res 1988; 5:S218.

19. Hartvig P, Askmark H, Aquilonius SM, Wikhmd L, Lindstriim B. Clinical pharmacokinetics of intrave- nous and oral 9-amino-1,2,3,4-tetrahydroacridine, tacrine. Eur J Clin Pharmacol 1990;38:259-63.

20. Cutler NR, Sedman AJ, Prior P, Underwood BA, Selen A, Balogh L, et al. Steady-state pharmacokinet- its of tacrine in patients with Alzheimer’s disease. Psychopharmacol Bull 1990;26:231-4.

21. Belloc C, Baird S, Cosme J, Lecoeur S, Gautier JC, Challine D, et al. Human cytochromes P450 expressed in Escher&a coli: production of specific antibodies. Toxicology 1996;106:207-19.

22. Fuhr U, Anders EM, Mahr G, S&gel F, Staib AH. Inhibitory potency of quinolone antibacterial agents against cytochrome P45OlA2 activity in vivo and in vitro. Antimicrob Agents Chemother 1992;36:942-8.

23. Fuhr U, Klittich K, Staib AH. Inhibitory effect of grapefruit juice and its bitter principal naringenin, on CYPlA2 dependent metabolism in man. Br J Clin Pharmacol 1993;35:431-6.

24. Knodell RG, Browne DG, Gwozdz GP, Brian WR, Guengerich FP. Differential inhibition of individual human liver cytochromes P-450 by cimetidine. Gas- troenterology 1991;101:1680-91.

CORRECTION An error occurred in the article, “Effects oftobacco smoking on ocular smooth pursuit” (Domino EF, Ni LS,

Zhang H. Clin Pharmacol Ther 1997;61:349-59). The second sentence in the legend of Fig. 1 should read: “Between -5 and 0 minutes, each group either sham smoked an unlit placebo cigarette or smoked a favorite cigarette.”

influence of the cyp1a2 inhibitor fluvoxamine on tacrine pharmacokinetics in humans*

Documents