Higher Intracerebral Concentration of Tacrolimus AfterIntermittent Than Continuous Administration to Rats

Yoshihiro Sakamoto,* Masatoshi Makuuchi,* Yasushi Harihara,* Hiroshi Imamura,*and Hitoshi Sato†

Neurotoxicity associated with tacrolimus after liver trans-plantation is a serious problem. The optimal way toadminister tacrolimus to reduce neurotoxicity remains tobe clarified. Three groups of rats were administeredtacrolimus for 2 weeks: group C, continuous intravenousinfusion (0.25, 0.5, and 1.0 mg/kg/d); group I, intermit-tent intravenous bolus injection twice daily (0.25, 0.5, and1.0 mg/kg/d); and group O, oral administration twicedaily (5 mg/kg/d; n � 12 each). Rats were killed either day7 or 14 to measure whole-blood and intracerebral troughconcentrations of tacrolimus. The area under the whole-blood concentration-time curve (AUC) was determinedday 7. The relative risk ratio of neurotoxicity was evalu-ated on the basis of the brain to blood concentration ratio(Kp) and intracerebral concentration to AUC ratio (RAUC).The whole-blood concentration of tacrolimus and AUCvalue were greater in group C than group I. Conversely, theintracerebral concentration and Kp and RAUC values weresignificantly greater in group I than group C. The differencein Kp values between groups C and I significantly increasedwith the dose and duration of administration. Whole-bloodand intracerebral concentrations in group O were similarto those at the 0.25-mg/kg/d dose in group I. In conclu-sion, the intracerebral concentration of tacrolimus wasgreater after intermittent than continuous administrationof the drug. Continuous administration of tacrolimusmight be more advantageous than the intermittentmethod to reduce the intracerebral concentration andneurotoxicity. (Liver Transpl 2001;7:1071-1076.)

Tacrolimus is an immunosuppressant that has be-come indispensable to patient management after

organ transplantation.1,2 However, neurotoxic eventsassociated with this agent3,4 are one of the limitations toits administration. Major neurotoxic events, such asseizures, psychosis, encephalopathy, and movementdisorders, have been reported to occur in 21.3% of allpatients who undergo liver transplantation and areadministered tacrolimus.4 Because neurotoxic eventssometimes lead to respiratory disorders or death, dosereduction4 or conversion to another immunosuppres-sant5 may be required. Therefore, identifying the opti-mal method of tacrolimus administration to minimizethe incidence of neurotoxicity while maintaining itsimmunosuppressive effect is of great interest to clini-cians engaged in liver and other organ transplantation.

In two large randomized trials comparing tacrolimusand cyclosporine, neurotoxic events after liver transplan-tation were more frequent during the intravenous infusionof tacrolimus than oral administration.1,2 Based on this

observation, it was recommended that tacrolimus beadministered orally, and low-dose oral induction oftacrolimus after liver transplantation has been proposed bysome clinicians.6-9 However, in those early trials, tacroli-mus was administered by high-dose 4-hour bolus intrave-nous infusion, not continuous infusion, and oral admin-istration is easily affected by interindividual andintraindividual variability and may not be appropriate forprecise control of whole-blood trough concentrations inthe early postoperative period or the period of organ rejec-tion.10 To the best of our knowledge, no controlled study,either clinical or experimental, has shown an advantage oforal over intravenous administration.

The incidence of neurotoxicity increases withwhole-blood concentrations,11,12 probably as a result ofincreasing concentrations in the brain, which is sub-strate for neurotoxicity. In this connection, in an exper-imental study in rats, we recently showed that the neu-rotoxicity of tacrolimus directly correlated with itsintracerebral concentration.13 We therefore conductedthe current study in a search for the optimal method ofadministering tacrolimus that can minimize intracere-bral concentrations of the drug.

Materials and Methods

Animals and Experimental Design

Male Wistar rats aged 7 weeks and weighing 250 to 300 gwere used throughout the study. Animals were given free

From the *Division of Surgery, Artificial Organ and Transplanta-tion Surgery, Graduate School of Medicine, University of Tokyo; and the†Department of Clinical and Molecular Pharmacokinetics/Pharmacody-namics, Faculty of Pharmaceutical Sciences, Showa University, Tokyo,Japan.

Supported in part by grant-in-aid no. 11470256 for scientificresearch from the Ministry of Education, Science and Culture; and theMinistry of Health and Welfare of Japan.

Address reprint requests to Masatoshi Makuuchi, MD, PhD, Divi-sion of Artificial Organ and Transplantation Surgery, Graduate School ofMedicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. Telephone: 813-5800-8841; FAX: 813-5800-8843;E-mail: makuuchi-tky@umin.ac.jp

Copyright © 2001 by the American Association for the Study ofLiver Diseases

1527-6465/01/0712-0001$35.00/0doi:10.1053/jlts.2001.28964

1071Liver Transplantation, Vol 7, No 12 (December), 2001: pp 1071-1076

access to food and water. All experiments were conducted inaccordance with principles and procedures outlined in theGuidance for Care and Use of Experimental Animals pre-pared by the National Academy of Sciences. Rats were ran-domly assigned to three groups, and each group of rats wasadministered tacrolimus for 2 weeks by one of the followingmethods of administration: (1) continuous intravenous infu-sion (group C; 0.25, 0.5, and 1.0 mg/kg/d; n � 12 each), (2)intermittent intravenous bolus injection twice daily (group I;0.25, 0.5, and 1.0 mg/kg/d; n � 12 each), and (3) oraladministration twice daily (group O; 5.0 mg/kg/d; n � 12).Oral doses were selected on the basis of a previous report thatin terms of whole-blood concentrations, intravenous infusionof a 0.5-mg/kg/d dose corresponded to oral administration ofa 5.0-mg/kg/d dose in rats.14

Doses and Methods of TacrolimusAdministration

Tacrolimus used in this study was provided by Fujisawa Phar-maceutical Co Ltd (Osaka, Japan). A formulation containing10 mg of tacrolimus in 1 mL of polyoxyethelated hydroge-nated castor oil 60 dissolved in ethanol (Nikko Chemicals CoLtd, Osaka, Japan) diluted with 5% wt/vol glucose (intrave-nous solution) was used for intravenous infusion, and a solid-dispersion formulation containing 20% wt/wt of the activeingredient was used for oral administration.

In group C, under ether anesthesia, an osmotic pump(Alzet Osmotic Pump no. 2ML4; Alza Co Ltd, Palo Alto, CA)containing the previously mentioned intravenous solution oftacrolimus was implanted subcutaneously. Before implanta-tion, each pump was incubated in normal saline solution at37°C for 3 hours to ensure the outflow. The infusion catheter(PE-60; Fisher Scientific, Santa Clara, CA) was introducedinto the right external jugular vein. The infusion rate was 60�L/d, and rats were administered 0.25, 0.5, or 1.0 mg/kg/d oftacrolimus (n � 12 for each dose).

In group I, under ether anesthesia, each rat was adminis-tered a bolus injection of tacrolimus, 0.25, 0.5, or 1.0mg/kg/d, dissolved in 0.5 mL of 5% glucose solution (n � 12)twice daily (11:00 AM and 11:00 PM) for 2 weeks. Each dosewas injected within a few seconds through the dorsal penisvein.

In group O, each rat was administered tacrolimus, 5mg/kg/d, dissolved in 2.5 mL of 5% glucose suspensionthrough a stomach tube. Doses were administered withoutanesthesia twice daily (11:00 AM and 11:00 PM) for 2 weeks(n � 12).

Whole-Blood and Brain Sampling andTacrolimus Assay

In groups I and O, rats were killed immediately before thetime of the next scheduled dose (at 11:00 AM and 11:00 PM) toevaluate the whole-blood trough concentration days 7 or 14.In group C, timing of sampling was synchronized with theother two groups. Rats were anesthetized with ether and killed

by exsanguination from the abdominal aorta. Immediatelyafter blood sampling, the brain was removed and stored at–20°C until measurement of tacrolimus concentration. Con-centrations of tacrolimus in whole blood and brain were mea-sured by a two-step enzyme-linked immunosorbent assay, asdescribed previously.15

Time Profile of Whole-Blood Concentrationand Area Under the Whole BloodConcentration-Time Curve Value

The area under the whole-blood concentration-time curve(AUC) of tacrolimus was evaluated day 7 in another group ofrats. In groups I and O, blood samples (0.2 or 0.4 mL/rat)were collected from the external jugular vein at 0, 1, 2, 3, 6,and 12 hours after tacrolimus administration (n � 4 at eachpoint). In group C, sampling was performed at intervals of 0,4, 12, and 24 hours (n � 4 at each point). Blood samples werecollected twice from each rat. The daily profile of whole-bloodconcentration was evaluated in each group and at each dose.AUC values were calculated using a mean value of whole-blood concentration of tacrolimus on each point according tothe trapezoidal rule.

Data Analysis. Steady-state total-body clearance was esti-mated by means of a model-independent moment analy-sis.16,17 Relative risk ratios for neurotoxicity to immunosup-pressive effect were evaluated from the brain to bloodconcentration ratio (Kp) and intracerebral concentration toAUC value ratio (RAUC) in each group days 7 and 14, assum-ing that the AUC or trough blood concentration of tacroli-mus relates to an immunosuppressive effect. Data areexpressed as mean � SE and were analyzed by the ANOVAmethod or Mann-Whitney test. In the ANOVA, method ofadministration, tacrolimus dose, and duration of administra-tion were entered as the main effect, and method of adminis-tration by dose interaction and method of administration byduration interaction also were considered. Differences at Pless than .05 are considered statistically significant.

Results

Time Profile of Whole-Blood Concentrationsand AUC Values

Time profiles of whole-blood concentrations of tacroli-mus in the three groups day 7 are shown in Figure 1.Whole-blood concentrations of tacrolimus in group Cremained stable for 24 hours, whereas they peakedwithin 4 hours after administration in groups I and O.Total clearances at doses of 0.25, 0.5, and 1.0 mg/kg/dday 7 were 244, 383, and 329 mL/h, respectively. AUCvalues in group C were 308, 400, and 695 ng � h/mL atdoses of 0.25, 0.5, and 1.0 mg/kg/d, respectively. AUCvalues in group I were 195, 347, and 429 ng � h/mL atdoses of 0.25, 0.5, and 1.0 mg/kg/d, respectively. AUC

1072 Sakamoto et al

values were greater in group C than group I at everydose.

Whole-Blood Trough Concentrations ofTacrolimus in Groups C and I

Whole-blood trough concentrations of tacrolimus ingroups C and I at different doses days 7 and 14 are listedin Table 1. The whole-blood trough concentration oftacrolimus was significantly greater in group C thangroup I at every dose and duration (P � .05, Mann-Whitney), in addition to the overall difference (P �.0001, ANOVA). Moreover, an interaction existedbetween dose of tacrolimus and method of administra-tion (P � .0001), but not between duration of admin-istration and method of administration (P � .89). This

indicates that the difference in whole-blood concentra-tions of tacrolimus between groups C and I increaseswith the dose.

Intracerebral Concentration of Tacrolimus inGroups C and I

Intracerebral concentrations of tacrolimus in groups Cand I at different doses days 7 and 14 are listed in Table2. The overall intracerebral concentration of tacrolimuswas significantly greater in group I than group C (P �.0001, ANOVA). In addition, an interaction existedbetween the dose of tacrolimus and method of admin-istration (P � .024), but not between the duration andmethod of administration (P � .15). This indicates thatthe difference in intracerebral concentrations between

Figure 1. Time profiles of whole-blood trough concentrations in groups C, I, and O. In groups I and O, whole-bloodconcentrations of tacrolimus peaked within 4 hours after administration. (●), 1.0 mg/kg/d; (E) 0.5 mg/kg/d; (�), 0.25mg/kg/d. *Log scale.

Table 1. Whole-Blood Trough Concentrations ofTacrolimus Days 7 and 14

DayDose

(mg/kg/d)

Concentration (ng/mL)

PGroup C Group I

7 0.25 12.8 � 0.5 4.5 � 0.7 .00560.5 16.3 � 1.4 5.9 � 0.5 .00611.0 38.0 � 7.0 6.8 � 0.9 .0039

14 0.25 10.2 � 1.2 3.1 � 0.8 .00880.5 17.4 � 0.8 4.9 � 0.9 .00441.0 33.3 � 6.7 4.3 � 0.7 .034

Table 2. Intracerebral Concentration of TacrolimusDays 7 and 14

DayDose

(mg/kg/d)

Concentration (ng/g)

PGroup C Group I

7 0.25 54 � 8.3 92 � 26 .410.5 89 � 15.4 188 � 25 .0251.0 206 � 38 262 � 32 .15

14 0.25 58 � 1.2 89 � 18 .140.5 87 � 7.4 207 � 22 .00421.0 210 � 61 380 � 63 .049

1073Brain Concentration of Tacrolimus in Rats

groups C and I increases with the dose. The maximumtacrolimus concentration in brain was 470 ng/g in a ratin group I at the 1.0-mg/kg/d dose day 14.

Kp Values in Groups C and I

Kp values in groups C and I at different doses days 7 and14 are shown in Figure 2. The overall Kp value wassignificantly greater in group I than group C (P �.0001). In addition, interactions existed between thedose of tacrolimus and method of administration (P �.03) and between duration and method of administra-tion (P � .01). This indicates that the difference in Kpvalues between groups C and I increases with dose andduration of administration.

RAUC Values in Groups C and I

RAUC values in groups C and I at different doses days 7and 14 are shown in Figure 3. The overall RAUC valuewas significantly greater in group I than group C (P �.0001). The difference in RAUC values between groupsC and I increased with dose and duration of adminis-tration, although it did not reach statistical significance;i.e., no significant interactions existed between the doseof tacrolimus and method of administration (P � .19)or between duration and method of administration(P � .23).

Comparison of Whole-Blood and IntracerebralConcentrations of Tacrolimus in Groups Iand O

Whole-blood trough and intracerebral concentrationsof tacrolimus, Kp, and RAUC values day 14 and AUC

value day 7 at the 0.25-mg/kg/d dose in group I weresimilar to values in group O at the 5-mg/kg/d dose.There were no significant differences in these valuesbetween the two groups (Table 3). The absolute bio-availability in group O was estimated to be 5.1%.

Discussion

The neurotoxicity associated with tacrolimus stronglycorrelated with the intracerebral concentration oftacrolimus, and neurotoxic events in rats administeredoral doses of tacrolimus appeared when the intracere-bral concentration was greater than 700 ng/g.13 Con-versely, the immunosuppressive effect of tacrolimus isbelieved to be a function of the whole-blood troughconcentration or AUC value.18,19 In clinical settings,the whole-blood trough concentration of tacrolimususually is used to monitor the immunosuppressiveeffect of the drug. Consequently, in theory, the optimalmethod of tacrolimus administration from the stand-point of neurotoxicity should be the method that min-imizes its intracerebral concentration while maintain-ing its whole-blood trough concentration and/or AUCwithin treatment levels.

In the present study, Kp and RAUC values were usedto evaluate the relative risk ratio of neurotoxicity to theimmunosuppressive effect of tacrolimus. A greater Kpvalue represents a greater intracerebral concentration oftacrolimus relative to the whole-blood concentration,i.e., a greater risk ratio of neurotoxicity compared withits immunosuppressive effect. Kp and RAUC values weresignificantly greater in group I than group C, and dif-

Figure 3. RAUC values in groups C and I days 7 and 14.The overall RAUC value was significantly greater in groupI than group C (P � .0001). The difference in RAUC valuesbetween groups C and I increased with dose and durationof administration, but the increases were not statisticallysignificant (P � .19 and P � .23, respectively). (●) GroupI; (E), group C. ANOVA test.

Figure 2. Kp values in groups C and I days 7 and 14. Theoverall Kp value was significantly greater in group I thangroup C (P � .0001). Interactions exist between the doseof tacrolimus and both method (P � .03) and durationand method of administration (P � .01). Thus, the differ-ence in Kp values between groups C and I increases sig-nificantly with dose and duration of administration. (●)Group I; (E), group C. ANOVA test.

1074 Sakamoto et al

ferences in Kp and RAUC values between groups I and Cincreased with the dose of tacrolimus and duration ofadministration. Therefore, the relative risk ratio of neu-rotoxicity to immunosuppressive effect is greater ingroup I than group C, and the ratio increases with thedose and duration of administration. It was reportedthat in 238 liver transplant recipients, neurotoxic eventswere more common after 4-hour bolus injection ther-apy (10.3%) than continuous infusion therapy (3.3%),10

although there were no differences between rejectionrates and graft survival. The experimental results fromour study may explain these clinical observations ofneurotoxicity depending on mode of intravenous ad-ministration of tacrolimus.

Oral administration of tacrolimus at a dose of 5mg/kg/d was shown to correspond to intermittentintravenous injection at a dose of 0.25 mg/kg/d in termsof whole-blood trough concentration, pattern of thetime-concentration profile, and AUC, Kp, and RAUC

values (Table 3). Conversely, Kp and RAUC values weresignificantly greater in group C than group I. Theseresults suggest that the relative risk ratio of neurotoxic-ity will be greater after oral administration of tacrolimusthan continuous intravenous infusion. We also foundthat oral administration of greater doses of tacrolimus(10 and 20 mg/kg/d) were accompanied by increasingbioavailability (25% and 50%, respectively; data notshown). That is, a greater oral dose of tacrolimus willfurther increase the risk ratio of neurotoxicity. In ourprevious study,13 we observed that when rats wereadministered a 5-mg/kg dose of tacrolimus orally twicedaily (10 mg/kg/d), all surviving rats showed tremors orseizures during the second week. When administeredtacrolimus at a dose of 20 mg/kg/d, 40% of rats showedneurological disorders during the first week.

In this experiment, the whole-blood trough concen-tration in group I or AUC values of tacrolimus did notincrease linearly according to dose of the drug. Tacroli-mus exists in the whole blood binding to the immu-nophilin (FK506 binding protein [FKBP]) or red blood

cells.20 Bolus injection of a greater dose of tacrolimusincreases the unbound fraction of tacrolimus, whichleads to a greater clearance ratio of the drug. Therefore,greater doses of tacrolimus are associated with decreasedwhole-blood concentration of AUC values.

Extrapolating the present results to clinical settingssuggests that repeated oral administration of tacrolimusmay increase its intracerebral concentration to anextremely high level and cause neurological disorderswhen a high trough concentration in whole blood ismaintained, e.g., in intestinal transplantation. Con-versely, continuous intravenous infusion of tacrolimusduring the early postoperative period can minimize theintracerebral concentration of the drug while maintain-ing its whole-blood trough concentration. According tothis analogy, oral administration of tacrolimus individed doses may reduce its intracerebral concentra-tion and, in turn, reduce the risk for neurotoxicity.

The greater Kp and RAUC values observed afterintermittent than continuous administration need to beexplained on a pharmacokinetic basis. The most likelycause of this finding is saturation of the efflux pump atthe blood-brain barrier, P-glycoprotein,21-23 at rela-tively high peak concentrations of tacrolimus in wholeblood. The peak concentration of tacrolimus in wholeblood after a bolus dose of tacrolimus, either intrave-nous or oral, may reach such a high level that it saturatesthe efflux pump. Beyond this threshold concentration,the intracerebral concentration of tacrolimus willincrease rapidly as the whole-blood concentrationincreases. Conversely, the level of FKBP in the brain is10 times greater than that in other tissues,24 and theextensive intracerebral binding of the drug to FKBPmay result in high and sustained accumulation oftacrolimus in the brain.

Recently, it was reported that peak concentration orAUC may provide better markers than trough concen-tration to monitor cyclosporine-based immunosup-pression. The principal pharmacodynamic effect ofcyclosporine and tacrolimus is similar in the inhibition

Table 3. Comparison Between Groups I and O Day 14

Group I(0.25 mg/kg/d)

Group O(5 mg/kg/d) P

Whole-blood trough concentration (ng/mL) 3.1 � 0.8 4.0 � 0.7 .18Intracerebral concentration (ng/g) 89 � 18 121 � 13 .09AUC (ng � h/mL) (day 7) 195 204 Not evaluatedKp 36 � 11 35 � 8.1 .81RAUC 0.46 � 0.09 0.59 � 0.07 .22

1075Brain Concentration of Tacrolimus in Rats

of calcineurin in the T cell.25,26 If peak concentrationcan also be a better marker than trough concentrationto monitor tacrolimus-based immunosuppression, thetherapeutic window of tacrolimus that can minimizeneurotoxicity while maintaining the immunosuppres-sive effect can be determined by peak concentration.However, further investigation is necessary to deter-mine the best marker to monitor immunosuppression.

In conclusion, the pharmacokinetic evidence ob-tained in this study suggests that the relative risk ratio ofneurotoxicity to immunosuppressive effect of tacroli-mus may be greater after intermittent than continuousadministration of the drug. In addition, the ratio in-creases with the dose of the drug and duration of ad-ministration. Therefore, continuous intravenous ad-ministration of tacrolimus will be an effective methodto minimize the risk for neurotoxicity while maintain-ing its immunosuppressive effect. However, results ob-tained in this study require further investigation of theclinical evidence to show a correlation between neuro-toxicity and brain concentrations of tacrolimus.

Acknowledgment

The authors thank Dr Yutaka Matsuyama for statistical sup-port.

References1. European FK506 Multicentre Liver Study Group. Randomised

trial comparing tacrolimus (FK506) and cyclosporin in preven-tion of liver allograft rejection. Lancet 1994;344:423-428.

2. The US Multicenter FK506 Liver Study Group. A comparisonof tacrolimus (FK506) and cyclosporine for immunosuppressionin liver transplantation. N Engl J Med 1994:331:1110-1105.

3. Eidelman B, Abu-Elmagd K, Wilson J, Fung JJ, Alessiani M, JainA, et al. Neurologic complications of FK506. Transplant Proc1991;23:3175-3178.

4. Muller A, Platz K, Bechstein W, Schattenfroh N, Stoltenburg-Didinger G, Blumhardet G, et al. Neurotoxicity after orthotopicliver transplantation. Transplantation 1994;58:155-170.

5. Jain A, Brody D, Hamad I, Rishi N, Kanal E, Fung J. Conversionto Neoral for neurotoxicity after primary adult liver transplanta-tion under tacrolimus. Transplantation 2000;69:172-176.

6. Inomata Y, Tanaka K, Egawa H, Umemoto S, Ozaki N, Oka-jima H, et al. The evolution of immunosuppression with FK506in pediatric living-related liver transplantation. Transplantation1996;61:247-252.

7. McMaster P, Mirza DF, Ismail T, Vennarecci G, Patapis P,Mayer AD. Therapeutic drug monitoring of tacrolimus in clin-ical transplantation. Ther Drug Monit 1995;17:602-605.

8. Margarit C, Rimola A, Gonzalez-Pinto I, Cuervas-Mons V, EdoA, Andreu H, et al. Efficacy and safety of oral low-dose tacroli-mus treatment in liver transplantation. Transpl Int 1998;11(suppl 1):S260-S266.

9. Kokado Y, Takahara S, Ichimaru N, Jing-Ding W, Miki T,

Okuyama A. Low-dose tacrolimus (FK506)-based immunosup-pressive protocol in living donor renal transplantation. Trans-plant Int 1998;11(suppl 1):S60-S64.

10. Abu-Elmagd K, Fung J, Draviam R, Shannon W, Jain A, Ales-siani M, et al. Four-hour versus 24-hour intravenous infusion ofFK506 in liver transplantation. Transplant Proc 1991;23:2767-2770.

11. Backman L, Nicar M, Levy M, Distant D, Eisenstein C, RenardT, et al. FK506 trough levels in whole blood and plasma in livertransplant recipients. Correlation with clinical events and sideeffects. Transplantation 1994;57:519-525.

12. Bottiger Y, Brattstrom C, Tyden G, Sawe J, Groth CG. Tacroli-mus whole blood concentrations correlate closely to side-effectsin renal transplant recipients. Br J Clin Pharmacol 1999;48:445-448.

13. Sakamoto Y, Makuuchi M, Harihara Y, Imamura H, Sato H.Correlation between neurotoxic events and intracerebral concen-tration of tacrolimus in rats. Biol Pharm Bull 2000;23:1008-1010.

14. Iwasaki K, Shiraga T, Matsuda H, Teramura Y, Kawamura A,Hata T, et al. Absorption, distribution, metabolism and excre-tion of tacrolimus (FK506) in the rat. Xenobiotic Metab Dispos1998;13:259-265.

15. Tamura K, Kobayashi M, Hashimoto K, Kojima K, Nagase K,Iwasaki K, et al. A highly sensitive method to assay FK-506 levelsin plasma. Transplant Proc 1987;19:23-29.

16. Yamaoka K, Nakagawa T, Uno T. Statistical moments in phar-macokinetics. J Pharmacokinet Biopharm 1978;6:547-558.

17. Benets LZ, Galeazzi RL. Noncompartmental determination ofthe steady-state volume of distribution. J Pharm Sci 1979;68:1071-1074.

18. Kershner R, Fitzsimmons W. Relationship of FK506 wholeblood concentrations and efficacy and toxicity after liver andkidney transplantation. Transplantation 1996;62:920-926.

19. Ku YM, Min DI. An abbreviated area-under-the-curve monitor-ing for tacrolimus in patients with liver transplants. Ther DrugMonit 1998;20:219-223.

20. Nagase K, Iwasaki K, Nozaki K, Noda K. Distribution andprotein binding of FK506, a potent immunosuppressive macro-lide lactone, in human blood and its uptake by erythrocytes.J Pharm Pharmacol 1994;46:113-117.

21. Cordon-Cardo C, O’Brien J, Casals D, Rittman-Grauer L, Bie-dler JL, Melamed MR, et al. Multidrug-resistance gene (P-gly-coprotein) is expressed by endothelial cells at blood-brain barriersites. Proc Natl Acad Sci U S A 1989;86:695-698.

22. Schinkel A, Smit J, van Tellingen O, Beijnen JH, Wagenaar E,van Deemter L, et al. Disruption of the mouse mdr1a p-glycop-rotein gene leads to a deficiency in the blood-brain barrier and toincreased sensitivity to drugs. Cell 1994;77:491-502.

23. Yokogawa K, Takahashi M, Tamai I, Konishi H, Nomura M,Moritani S, et al. P-Glycoprotein-dependent disposition kineticsof tacrolimus: Studies in mdr1a knockout mice. Pharm Res1999;16:1213-1218.

24. Steiner JP, Dawson TM, Fotuhi M, Glatt CE, Snowman AM,Cohen N, et al. High brain densities of the immunophilin FKBPcolocalized with calcineurin. Nature 1992;358:584-587.

25. Grant D, Kneteman N, Tchervenkov J, Roy A, Murphy G, TanA, et al . Peak cyclosporin levels (Cmax) correlate with freedomfrom liver graft rejection. Transplantation 1999;67:1133-1137.

26. Levy GA. Relationship of pharmacokinetics of clinical outcomes.Transplant Proc 1999;31:1654-1658.

1076 Sakamoto et al