[CANCER RESEARCH 58, .1339-3343, August I, 1998

Regional Pharmacokinetics of Doxorubicin following Hepatic Arterial and PortalVenous Administration: Evaluation with Hepatic Venous Isolation andCharcoal Hemoperfusion1

Takeshi Iwasaki, Yonson Ku,2 Nobuya Kusunoki, Masahiro Tominaga, Takumi Fukumoto, Sanshiro Muramatsu, and

Yoshikazu KurodaFirst Department of Surgen: Kobe Universit\ School of Medicine. Chuo-ku, Kobe 650, Japan

ABSTRACT

We evaluated the regional pharmacokinetics of doxorubicin after hepatic arterial infusion (HAI) and portal venous infusion (PVI) using anovel system for hepatic venous isolation and charcoal hemoperfusion(HVI-CHP). The HVI-CHP system was used to determine directly the

doxorubicin plasma concentration in the hepatic vein and the hepaticvenous flow rate, and simultaneously, to eliminate hepatic re-entry of the

drug. Beagles received doxorubicin (1 mg/kg) through either the hepaticartery (HAI group, n = 6) or the portal vein (PVI group, n = 6). In both

groups, hepatic venous blood was completely isolated and directed to theCHP filter. The filtered blood was returned through the left jugular vein.During HVI-CHP, the hepatic venous flow rate was monitored and plasmadoxorubicin concentrations were serially measured in prefilter ( = hepatic

venous), postfilter, and systemic blood. The hepatic tissue uptake ofdoxorubicin was determined based on the blood flow rate and doxorubicinlevel in the hepatic vein. The hepatic extraction ratio of doxorubicin wasdefined as the percentage hepatic tissue uptake to the amount of drugadministered. During drug infusion, similarly in either group, HVI-CHPproduced a 66-87% reduction of the postfilter doxorubicin level as

compared with the prefilter level. The prefilter drug level was significantlylower in HAI group than in PVI group i/' < 0.01). Thus, the area under

the time concentration curve for the prefilter drug level in the HAI group(6.90 ±0.96 jug min/mi i was significantly lower than that in the PVI group(18.10 ±2.90 /ig min/ml. /' < 0.01). Conversely, the hepatic extraction

ratio in the HAI group (84.6 ±2.9%) was significantly higher than that inthe PVI group (58.1 ±3.4%, P < 0.01). We conclude that in the beagle,doxorubicin is more effectively extracted by the liver when administeredvia the hepatic artery than when administered via the portal vein. Theseresults indicate that HAI of doxorubicin is superior to PVI in terms ofreduction of systemic drug exposure and systemic toxicity.

INTRODUCTION

Regional chemotherapy has been used frequently for both primaryand secondary hepatic malignancies. The rationale of regional chemotherapy lies in attempts to increase drug delivery to the cancer-bearing region with reduction of systemic toxicity (1-3). The theo

retical basis is that the increase in local drug concentration afterregional infusion of any drug depends on the blood flow rate of theinfused vessel and the rate of drug elimination by the rest of the body,whereas the reduction of systemic drug exposure depends largely onthe ability of the infused organ to extract and metabolize the drug.Thus, the first-pass HER3 of the drug is one major factor that governs

the delivery advantage of regional chemotherapy of the liver.

Received 10/17/97: accepted 6/2/98.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1Supported by grants-in-aid for scientific research (04670781 and 06454386l from the

Ministry of Education. Science and Culture, Japan.~ To whom requests for reprints should be addressed, at First Department of Surgery,

Kobe University School of Medicine. 7-5-2 Kusunoki-cho, Chuo-ku. Kobe 650. Japan.Phone: (81) 78-341-7451: Fax: (81) 78-382-2110.

1The abbreviations used are: HER. hepatic extraction ratio; HVI-CHP, hepatic venous

isolation and charcoal hemoperfusion; HAI, hepatic arterial infusion: PVI. portal venousinfusion; IVC. inferior vena cava; HPLC, high-performance liquid chromatography: AUC.area under the time-concentration curve.

The dual blood supply to the liver by the hepatic artery and theportal vein raises the question of which of these two access routesshould be used in regional chemotherapy for hepatic tumors. Althoughinfrequently, PVI is employed in specific situations such as earlyadjuvant chemotherapy after resection of primary colorectal carcinomas (4, 5). However, based on a vast number of studies regardingtumor blood supply (6-8) and pharmacological studies in patients ortumor-bearing animals (9-12), HAI is predominantly employed in

clinical practice. Accordingly, doxorubicin has been administered viathe hepatic artery, exhibiting activity against hepatocellular carcinomaand a number of metastatic hepatic tumors (13-16). The general beliefis that this drug has a low first-pass HER. Some investigators have

reported that the HER of doxorubicin ranged between 5% and 50%(17) or was less than 24% (18), whereas others have reported that theratio is probably close to 60% (19). Thus, the reported data for HERvary widely, and little is known regarding the first-pass HER of

doxorubicin administered via various routes of administration.The novel system of HVI-CHP. described by us elsewhere, seems

well suited for the determination of the HER of anticancer drugs witha high charcoal affinity (20-23). The HVI-CHP system profoundlyreduces hepatic re-entry of the drug after the first-pass and, thus,

enables us to compare more precisely the influence of differentadministration routes. Moreover, the actual amount of drug recoveredfrom the hepatic vein can be quantified as the product of the bloodflow rate and the drug concentration in the completely isolated hepaticvein. The purpose of this study was to determine the effects of twodifferent administration routes on the first-pass HER of doxorubicin

and to provide a pharmacologie basis for their use in regional chemotherapy of the liver.

MATERIALS AND METHODS

Animals. A total of 14 beagles of both genders from the same large colonywere used. The use of animals in this study conformed to the guidelines of theNIH and was approved by our institutional Animal Care Committee. Food waswithheld for 12 h preoperatively. Each beagle was anesthetized with sodiumpentobarbital (25 mg/kg) and pancuronium bromide (O.I mg/kg) i.v. The dogwas then intubated and ventilated mechanically throughout the experiment.The left carotid artery was cannulated for blood sampling as well as forcontinuous monitoring of blood pressure and heart rate.

Experimental Groups and Surgical Procedure. After a midline laparot-

omy and complete freeing of the liver from all ligaments around it, the dogswere randomly assigned to two groups. In the HAI group (n = 7). the

gastroduodenal artery was divided and cannulated with a catheter (3F) forhepatic arterial administration of doxorubicin. In the PVI group (n = 7). the

pancreaticoduodenal vein was divided and a catheter (3F) was placed into theportal vein for portal venous administration of the drug. For one set ofexperiments the HVI-CHP system was established in six dogs each in the HAI

and PVI groups, weighing 8.5 ± 0.5 and 8.6 ± 0.6 kg (mean ± SE),respectively, according to the method described previously, with minor modifications (19). In brief, a 12F Trocar catheter was introduced into the retro-

hepatic IVC through the right femoral vein for collection of hepatic venousoutflow, and was connected to an extracorporeal unit containing the CHP filter(DHP-I; Kuraray Co., Ltd., Osaka, Japan) and a centrifugal pump (Biopump-

3339

on May 6, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HEPATIC EXTRACTION OF ARTERIAL VS. PORTAL DOXORUBICIN

B

Fig. 1. Experimental design of HVI-CHP system (A) and HVI alone (B). Catheters were introduced into the retrohepatic IVC and infrarenal IVCfrom the right and left femoral veins, respectively.A catheter placed in the retrohepatic IVC was connected to a CHP filter and pumped to the leftjugular vein together with infrahepatic IVC blood.HVI was achieved by clamping the suprahepaticand infrahepatic IVC. Hepatic venous blood flowrate was measured by an electromagnetic flowprobe placed in the extraeorporeal circuit. Bloodsamples were drawn at inlet side (<i)and outlet side(b) of the CHP filter in the extraeorporeal circuitand the carotid artery. The experimental design ofHVI alone (fi) is similar with A except that thecircuit contains no CHP filter. Blood samples werecollected from the hepatic venous line (a) and thecarotid artery.

Lt. jugular vein Lt. jugular vein

Bio-pump

Flow probe Flow probe

50; Bio-Medicus, Inc., Minneapolis, MN; Fig. 1A). The priming volume of the

CHP filter is about 60 ml, and the rotation of the pump was adjusted to deliverthe maximum blood return at the minimal revolutions/min. After completion ofthe extracoporeal circuit with introduction of catheters into the infrarenal IVCthrough the left femoral vein and the left jugular vein, the suprahepatic IVCwas encircled transthoracically as well as below the hepatic veins with Rum-

mei tourniquets. Subsequently, heparin sodium (150 mg/kg) was injected as ani.v. bolus for anticoagulation. After hepatic venous isolation, infrahepatic IVCblood together with filtered hepatic venous blood was pumped to the leftjugular vein. The systolic blood pressure was maintained at over 90 mmHgduring the procedure by appropriate volume loading with lactated Ringer's

solution (10-15 ml/min). No vasopressor or inotropic agent was used to

maintain blood pressure. All dogs were sacrificed by bolus injection ofpotassium chloride at the completion of each experiment.

In the second set of experiments, in one dog weighing 11 kg in each group,we evaluated the influence of hepatic re-entry of the drug on pharmacokinetic

indices including HER. In these experiments, the design was essentially thesame as that described above except that the CHP filter was excluded from theextraeorporeal circuit (Fig. IB). Therefore, the isolated hepatic venous bloodwas simply returned to the systemic circulation in both dogs.

Anticancer Drug and Routes of Administration. Doxorubicin (KyowaHakko Co., Ltd., Osaka, Japan) at a dose of 1 mg/kg was dissolved in 20 mlof sterile physiologic saline and administered continuously through either thehepatic artery (HAI group) or the portal vein (PVI group) for 10 min using asyringe infusion pump (Terufusion Model STC-523, Terumo Co., Ltd., Tokyo,Japan). The HVI-CHP system was maintained for 20 min after the initiation of

drug infusion.Doxorubicin Measurement. In the HVI-CHP setting, blood samples were

obtained from the left carotid artery (= systemic blood). pre-CHP filterextraeorporeal circuit line (= hepatic venous blood), and post-CHP filter line

just before and 2, 4, 6, 8, 10, 15 and 20 min after the initiation of drugadministration. In the setting of HVI alone, blood samples were collected fromthe hepatic venous line and the left carotid artery at the same interval as in theHVI-CHP setting. The plasma concentrations of doxorubicin were determined

by HPLC as described previously (24). In brief, aliquots of plasma were placedon minicolumns (Nucleosil 5C18; Chemco Co., Ltd., Takatsuki, Japan). Afterwashing, the drug was eluted and the eluent was evaporated in vacuo. Theresidual samples were redissolved in the HPLC mobile phase, and doxorubicinconcentrations were measured by routine HPLC. A standard curve was obtained with samples dissolved in control canine plasma.

Flow Measurement and Pharmacokinetic Evaluation. Hepatic venousblood flow rate was monitored continuously with an electromagnetic flow

probe (Bioprobe TX40; Bio-Medicus, Inc.) placed at the pre-CHP filter circuit

line (Fig. 1). Based on the blood flow rate (Q) and doxorubicin concentrationsin the hepatic vein (C), the amount of the drug in the hepatic effluent wasdetermined as the product of Q and C (20, 21, 25). The hepatic tissue uptakewas calculated using the following equation:

Hepatic tissue uptake/., = Dose^ - AUC,_/Q,y

where Dose,4 = the amount of drug administered between the times i and j,AUC,_y = area under the time-concentration curve between times ; and j, andQ,y(ml/min) = the average blood flow rate between the times i and/ The HER

of doxorubicin was defined as the percent hepatic tissue uptake to the amountof drug administered. The drug extraction ratio of the CHP filter at eachsampling time was calculated as the percent filter extraction (%) at samplingtime i as follows: (Ci-C'i)/Ci X 100, where Ci and C'i = the prefilter and the

postfilter drug concentration at the sampling time of i. The AUC was calculated by the trapezoidal method.

Statistical Analysis. All values are presented as mean ±SE. The significance of difference was analyzed using the unpaired Student's /-test, and a P

value of <0.05 was considered statistically significant.

RESULTS

Hemodynamic Effects of HVI-CHP. In both groups, the meanarterial pressure showed a slight decrease after the initiation of HVI-

CHP. However, all dogs had stable systolic blood pressures over 90mmHg throughout the course, and no significant difference was notedbetween the two groups. The two groups also had similar hepaticvenous flow rates in the range from 240-390 ml/min during HVI-

CHP, which were markedly stable in each dog. The hepatic venousblood flow rates of the HAI and PVI groups averaged 268.3 ±16.4and 267.5 ±16.5 ml/min, respectively.

Time Course of Drug Extraction Ratio of the CHP Filter.During a 10-min drug infusion, the drug extraction ratios by the CHP

filter were similar in the two groups: the mean values at 2 and 10 minwere 84.5 ±3.2 and 66.4 ±4.5% for the HAI group, and 87.8 ±2.6and 79.5 ±4.9% for the PVI group, respectively. Regardless of theroute of administration, HVI-CHP profoundly reduced the postfilter

and the systemic drug concentrations compared with the prefilter drugconcentration.

3340

on May 6, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HEPATIC EXTRACTION OF ARTERIAL VS. PORTAL DOXORUBICIN

2500

I2000 -

1500-

1000-

500 •¿�

PrefilterHAI

PVI

10 15 20Drug infusion

Time (min)

B

500

400 -

Postfilter and Systemic-•- HAI (Postfilter)-•- PVI (Postfilter)

t -O- HAI (Systemic)

PVI (Systemic)

10 15 20Druginfusion

Time (min)

Fig. 2. Plasma doxorubicin concentration in the prefilter (A) and postfilter and systemic(B) blood samples. These data represent the means ±SE. Bars are not shown when theyare smaller than the symbols. Significant differences between HAI and PVI are indicatedby asterisks (*P < 0.05, "*P < 0.01).

Fig. IB shows the time courses of the postfilter and systemicdoxorubicin concentrations. Despite the similar extraction ratios of theCHP filter in the two groups, the postfilter plasma doxorubicin levelsin the PVI group tended to be higher than those in the HAI group,resulting from the consistently higher prefilter drug concentrations inthe PVI group. The mean AUC was 1.94 ±0.28 in the HAI group and3.30 ±0.53 /Agmin/ml in the PVI group (P < 0.05; Table 1). On theother hand, the peak systemic concentration of doxorubicin in the HAIgroup was 140.8 ± 21.7 ng/ml, and that in the PVI group was249.2 ±49.4 ng/ml, showing no significant differences in the twogroups at any measurement time point. Consequently, there was nosignificant difference in the mean AUC of systemic plasma betweenthe two groups.

HER of Doxorubicin. Fig. 3 illustrates the HER of doxorubicin atthe discrete time intervals during drug infusion. The ratio for theinterval from 0-2 min after the start of drug infusion was 92.5 ±2.7

and 83.9 ±2.4% in the HAI and PVI groups, respectively. For all ofthe subsequent time intervals, the HER in the HAI group exceeded80%, whereas it decreased to as low as 47.5 ±6.7% in the PVI group.The overall (0- to 10-min) HER in the HAI group (84.6 ±2.9%) was

significantly higher than that in the PVI group (58.1 ± 3.4%,P < 0.01).

Pharmacokinetic Comparison with and without CHP. Fig. 4illustrates the time courses of hepatic venous and systemic doxorubicin concentrations after each HAI and PVI with HVI-CHP orHVI alone. After HAI, the mean hepatic venous (= prefilter) and

systemic concentrations of doxorubicin in the six dogs with CHPwere markedly lower at all measurement time points than those inthe dog without CHP (Fig. 4A). Exclusion of the CHP filter fromthe extracorporeal circuit produced a 2- and 7-fold increase in the

AUC of doxorubicin for hepatic venous and for systemic plasma,respectively, as compared with the corresponding mean values inthe six dogs with CHP. This difference was similarly observedafter portal venous administration (Fig. 4B). The dog receivingintraportal doxorubicin without CHP exhibited a 1.7- and 10-fold

increase in the AUC for the hepatic venous and for the systemicplasma, respectively, compared with the corresponding mean values in the six dogs with CHP. The HER during the 10-min infusion

with and without CHP was 84.6% (mean value for six dogs) and41.9%, respectively, after HAI, and 58.1% (mean value for sixdogs) and 19.2%, respectively, after PVI.

Table 1 AUC of plasma doxorubicin and HER during the 10-min drug infusion

AUC (ng min/ml)"

Group Prefilter Postfilter Systemic HER (0-10 min)"

HAIPVI

6.90 ±0.96"

18.10 ±1.72

1.94 ±0.28r

3.30 ±0.531.71 ±0.242.76 ±0.54

84.6 ±2.9%"

58.1 ±3.4%" Each value represents mean ±SE.* P < 0.01 compared with the PVI group.c P < 0.05 compared with the PVI group.dP < 0.01 compared with the PVI group.

Time Course of Plasma Doxorubicin Concentrations duringHVI-CHP. The peak hepatic venous concentration (prefilter level) of

doxorubicin in the HAI group (600.6 ±88.1 ng/ml) was significantlylower than that in the PVI group (1880.0 ±145.4 ng/ml; P < 0.01;Fig. 2A). In addition, at all measurement time points during the10-min drug infusion, the prefilter concentrations in the six dogs in the

PVI group were consistently higher than those in the HAI group.Consequently, as listed in Table 1, the mean AUC for the prefilterplasma in the HAI group was significantly lower than that in the PVIgroup (6.90 ±0.96 and 18.10 ± 1.72 p.g min/ml, respectively,P < 0.01).

100

4-6 6-8

Time (min)

0-10

Fig. 3. HER was calculated as percentage hepatic tissue uptake to the amount of drugadministered. Each column represents the mean ±SE of each time interval. Significantdifferences between the HAI group and PVI group are indicated by asterisks (*P < 0.05,**/> < 0.01).

3341

on May 6, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HEPATIC EXTRACTION OF ARTERIAL VS. PORTAL DOXORUBICIN

A

2000 -

1500 -

1000 -

IS 500 -

Prefilter-HVI alone

Prefilter-HVI-CHP

Systemic-HVI alone

Systemic-HVI-CHP

10

Time (min)

15 20

Prefilter-HVI alone

Prefilter-HVI-CHP

Systemic-HVI alone

Systemic-HVI-CHP

10

Time (min)

15 20

Fig. 4. Plasma doxorubicin concentrations of the HVI-CHP or HV1 alone (withoutCHP) under HAI; (A) and PVI; (B). Data of HVI-CHP (•and •¿�)in panels A and Brepresent the mean values of six dogs with HAI and PVI. respectively.

DISCUSSION

Direct measurement of HER has been technically elusive, since theconcentration and flow rate must be measured at three sites: thehepatic artery, portal vein, and hepatic vein (25, 26). We used theHVI-CHP system for determining the HER of doxorubicin because

the system allows us to quantify the amount of drug escaping hepaticmetabolism and degradation directly as the product of drug concentrations and blood flow rates in the hepatic circulation. In addition, ithas been shown that HVI-CHP profoundly reduces hepatic re-entry

after HAI of a drug which, like doxorubicin, has high charcoal affinity(20-23). In this study, we compared the HER of doxorubicin after

regional administration via two different routes. We also evaluated theinfluence of drug recirculation in regional pharmacokinetics usingHVI-CHP and using HVI alone.

The first set of experiments with HVI-CHP revealed that the systemprovided 66-85% and 80-88% reduction in systemic exposure to

doxorubicin in HAI and in PVI, respectively, as assessed by the filterextraction ratios. Thus HVI-CHP in this canine model was highlyeffective in preventing hepatic re-entry of doxorubicin after the first-

pass irrespective of the route of hepatic infusion. Under conditions of

drug elimination by HVI-CHP, the HER of doxorubicin averaged 84%

with HAI and 58% with PVI. These findings indicate that doxorubicinis more efficiently extracted by the normal liver when administeredvia the hepatic artery compared with the portal vein. Thus, HAI ofdoxorubicin seems to be more beneficial than PVI for reducingsystemic drug exposure. However, since HER, per se, does not predicttumor drug uptake, it should be noted that tumor blood supply is stillthe primary determinant of the vessel to be selected for hepatic druginfusion.

The published data for the HER of doxorubicin vary greatly,ranging from 10-60% (17-19, 28). In the rabbit model, Harris and

Gross (19) found that the HER of doxorubicin was 60% by comparingthe AUC for lung concentrations following 30-min PVI at 3 mg/kg

and after i.v. infusion at the same dose rate. In human subjects. Balletet al. (28) found a very low HER of less than 10% using temporaryplacement of hepatic venous catheters with i.v. administration ofdoxorubicin. On the other hand, Garnick et al. (17) reported that theHER ranged from 5-50% in five patients with metastatic liver tumors

and two patients with bile duct tumors during i.v. infusion of the drugat a dose of 40 mg/m2. Although the species, dose rates, and routes of

infusions used were different in these studies, these published valuesfor the HER of doxorubicin are somewhat lower compared with thedata obtained under HVI-CHP. There are at least two possible reasons

for this. First, in the previous studies by others, HER was calculatedaccording to the following equation: HER = (hepatic arterial concentration (CA)-hepatic venous concentration (CV)/CA. The CA and Cvwere determined at a steady-state during drug infusion at a constantrate. As pointed out by some investigators in the studies on 5-fluorou-racil or 5-fluoro-2'-deoxyuridine (26, 27), the calculated hepatic ex

traction is flawed by failure of the conventional equation to incorporate recirculation of the drug to the liver through the hepatic artery andthe portal vein irrespective of the route of drug infusion. This isparticularly notable with drugs exhibiting low HER because theamount of hepatic re-entry may be greater after the first-pass. Con

sequently, the lower the HER of any drug, the greater the underestimation. In addition, there is technical difficulty in applying thisequation to regional infusions. In the studies using this equation in amodel with peripheral venous infusion, CA can be represented by thedrug level in the carotid artery, but this equation is not applicable forregional infusion without direct measurement of CA or flow measurement, which is technically difficult in usual pharmacokinetic modelsbut feasible in our HVI-CHP model.

Recently, August et al. (29) studied the HER of doxorubicin afterHAI for 90 min with use of the HVI-CHP in a swine model. Although

they carried out HVI without laparotomy using a specially designeddouble-balloon catheter, their model and ours are based on the same

concept that the system aims at circumventing an extensive operationwhile providing an appropriate efficacy in eliminating superfluousdrugs after hepatic arterial chemotherapy. In their study with HAI ofdoxorubicin at dose ranges of 0.5-9.0 mg/kg, HER varied from74-91%. Of note, our results with HAI were almost consistent withtheir data: they also found that the values were not dose-related and

were higher than previously estimated. Unfortunately, they did notdetermine the HER with PVI because of an inherent technical limitation on portal venous access in the nonlaparotomized model. However, the consistent findings with HAI in the two studies stronglysuggest that hepatic recirculation may profoundly affect the HER ofdoxorubicin measured in conventional pharmacokinetic models.

The underestimation of the HER of doxorubicin was also found inour second set of experiments with exclusion of the CHP filter fromthe extracorporeal circuit. In the absence of the filter, profoundhepatic re-entry of the drug via the dual inflow vessels of the liver

ensued after the first pass, as assumed by the systemic AUC. HVI

3342

on May 6, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

HEPATIC EXTRACTION OF ARTERIAL VS. PORTAL DOXORUBK IN

alone showed 7- and 10-fold increases in systemic exposure to doxo-rubicin with HAI and PVI, respectively, compared with HVI-CHP.

Thus, we consider that mainly hepatic recirculation of the drug accounts for the marked decrease in HER with HVI alone comparedwith that with HVI-CHP with either HAI or PVI.

Another factor that might cause overestimation of the HER ofdoxorubicin with HVI-CHP is the delayed washout of the drug from

the liver after the end of drug infusion. As shown in the prefilter timeconcentration curve, the AUC fraction between 10 and 20 min constituted approximately one-fourth of the AUC during 20 min. This

was also supported by the observation by August et al. (29) in a swinemodel of HVI-CHP that unmetabolized doxorubicin is washed out of

the liver after the end of drug infusion at a higher concentration thandrug entering the liver. Thus, they noted, the liver acts as a net sourceof doxorubicin during this washout phase. Indeed, previous studieshave shown that doxorubicin is highly tissue bound and its systemicpharmacokinetics are characterized by a rapid distribution phase (half-life of 8-11 min) followed by a prolonged elimination phase (half-lifeof 25-30 h) (30, 31 ). The major portion of the total AUC is under the

terminal portion of the elimination curve, which was not quantified inour study. Moreover, a portion of the hepatic tissue uptake estimatedwith HVI-CHP may simply be tissue-bound rather than metabolized

or excreted in the bile, and this bound drug may return to thecirculation at a later time. This would contribute to overestimation ofthe HER in the present experiments.

We are not able to account for the significantly higher HER ofdoxorubicin in HAI than in PVI. One possible explanation is the largervascular bed of the hepatic artery compared with the portal vein.Either portal or hepatic arterial blood ultimately enters the hepaticsinusoids, where the drug is extracted by the hepatocytes. However,anatomic studies by others have shown that the vascular bed of thehepatic artery expands to the bile ducts, forming the peribiliaryplexus, portal structures, and the capsule of the liver before enteringthe hepatic sinusoids (32). Thus it seems reasonable to assume thatpresinusoidal structures of the hepatic artery contribute to the higherHER of doxorubicin with HAI.

In summary, this pharmacokinetic study clearly demonstrated thatthe hepatic extraction of doxorubicin is significantly higher with HAIcompared with PVI, thereby confirming the advantage of HAI overPVI in terms of reduction of systemic drug exposure and systemictoxicities. We believe that this canine model of HVI-CHP may be

useful in determining the hepatic pharmacokinetics of other drugswith high affinity to CHP filters.

REFERENCES

1. Dedrick. R. L. Arterial drug infusion: pharmacokinetic problems and pitfalls. J. Nati.Cancer Inst.. 80: 84-89. 1988.

2. Collins, J. M. Pharmacologie rationale for regional drug delivery. J. Clin. Oncol., 2:498-504. 1984.

3. Stephens, F. O. Pharmacokinetics of intra-arterial chemotherapy. Recent ResultsCancer Res., 86: 1-12, 1983.

4. Taylor, I., Brooman. P., and Rowling. J. T. Adjuvant liver perfusion in colorectalcancer: initial results of a clinical trial. Br. Med. J. 2: 1320-1322, 1977.

5. Fielding, L. P., Hittinger, R.. Grace, R. H., and Fry. J. S. Randomised controlled trialof adjuvant chemotherapy by portal-vein perfusion after curative resection for colorectal adenocarcinoma. Lancet, 340: 502-506, 1992.

6. Ackerman. N. B., Lien. W. M., Kondi, E. B., and Silverman. N. A. The blood supplyof experimental liver métastases:I. The distribution of hepatic artery and portal veinblood to "small" and "large" tumors. Surgery. 66: 1067-1072. 1969.

7. Suzuki. T.. Sarumaru. S.. Kawabe. K.. and Honjo. I. Study of vascularity of tumorsof the liver. Surg. Gynecol. Obstet., 134: 27-34, 1972.

8. Archer. S. G.. and Gray, B. N. Vascularization of small liver métastases.Br. J. Surg.,76: 545-548, 1989.

9. Sigurdson. E. R.. Ridge. J. A.. Kemeny. N.. and Daly. J. M. Tumor and liver druguptake following hepatic artery and portal vein infusion. J. Clin. Oncol.. //: 1836-

1840, 1987.10. Ridge. J. A.. Bading. A. R.. Gelbard. A. S.. Benua. R. S.. and Daly. J. M. Perfusion

of colorectal hepatic métastases.Relative distribution of flow from the hepatic arteryand portal vein. Cancer (Phila.), 59: 1547-1553, 1987.

11. Erichsen, C., Christenson. P.. Eriksson. G., Yngner. T.. Jonsson. P., and Stenram. U.Effects of administration routes on the uptake of uridine and 5-fluorouridine into anadenocarcinoma transplanted to rat liver. J. Surg. Oncol.. 33: 76-80. 1986.

12. Archer, S. G., and Gray. B. N. Comparison of portal vein chemotherapy with hepaticartery chemotherapy in the treatment of liver micrometastases. Am. J. Surg.. 159:325-329, 1990.

13. Bern, M. M.. McDermott. W.. Cady. B.. Oberfield. R. A.. Trey, C., Clouse, M. E.,Tullis, J. L.. and Parker. L. M. Intra-arterial hepatic infusion and intravenous adria-

mycin for treatment of hepatocellular carcinoma: a clinical and pharmacology repon.Cancer (Phila.), 42: 399-405. 1978.

14. Bismuth, H., Morino, M., Sherlock, D., Castaing. D., Miglietta, C.. Cauquil. P.. andRoche, A. Primary treatment of hepatocellular carcinoma by arterial chemoemboli-zation. Am. J. Surg., 163: 387-394. 1992.

15. Ku, Y., Iwasaki, T.. Fukumoto, T., Tominaga. M., Muramatsu, S., Kusunoki. N.,Sugimoto, T., Suzuki, Y., Kuroda. Y.. Saitoh. Y.. Sako. M.. Matsumoto. S.. Hirota.S., and Obara. H. Induction of long-term remission in advanced hepatocellularcarcinoma with percutaneous isolated liver chemoperfusion. Ann. Surg.. 227: 519-

526. 1998.16. Ku, Y.. Tominaga. M., Iwasaki, T.. Kitagawa. T., Maeda. I., Shiotani. M., Kusunoki.

N.. Maekawa, Y., Samizo, M.. Fukumoto, T., Kuroda, Y., Hirota, S.. and Saitoh. Y.Percutaneous hepatic venous isolation and extracorporeal charcoal hemoperfusion forhigh-dose intra-arterial chemotherapy in patients with colorectal hepatic métastases.Surg. Today (Tokyo), 26: 305-313. 1996.

17. Garnick. M. B., Ensminger. W. D., and Israel. M. A clinical-pharmacological evaluation of hepatic arterial infusion of adriamycin. Cancer Res.. 39: 4105-4110. 1979.

18. Ballet. F., Vrignaud. P., Robert. J., Rey. C., and Poupon. R. Hepatic extraction,metabolism and biliary excretion of doxorubicin in the isolated perfused rat liver.Cancer Chemother. Pharmacol.. 19: 240-245, 1987.

19. Harris, P. A., and Gross, J. F. Preliminary pharmacokinetic model for adriamycin(NSC-123127). Cancer Chemother. Rep., 59: 819-825, 1975.

20. Ku. Y., Saitoh. M., Nishiyama, H.. Fujiwara, S., Iwasaki, T., Tominaga, M.,Maekawa, Y., Ohyanagi. H.. and Saitoh, Y. Extracorporeal removal of anticancerdrugs in hepatic artery infusion: the effect of direct hemoperfusion combined withvenous bypass. Surgery. 107: 273-281, 1990.

21. Ku. Y., Fukumoto, T., Iwasaki. T.. Tominaga. M., Samizo. M., Nishida. T., Kuroda.Y., Hirota, S., Sako, M., Obara, H., and Saitoh, Y. Clinical pilot study on high-doseintra-arterial chemotherapy with direct hemoperfusion under hepatic venous isolationin patients with advanced hepatocellular carcinoma. Surgery. 117: 510-519, 1995.

22. Ku, Y., Fukumoto. T.. Tominaga. M.. Iwasaki. T., Maeda. 1., Kusunoki, N.. Obara.H.. Sako, M.. Suzuki. Y.. Kuroda. Y.. and Saitoh. Y. Single catheter technique ofhepatic venous isolation and extracorporeal charcoal hemoperfusion for malignantliver tumors. Am. J. Surg., 173: 103-109, 1997.

23. Tominaga, M., Ku, Y., Ohyanagi, H., and Saitoh, Y. The direct hemodynamic effectsof dopamine on hepatic blood flow in the dog—use of direct hemoperfusion (DHP)under hepatic venous isolation (HVI). Jpn. J. Surg., 93: 26-35, 1992.

24. Robert, J. Extraction of anthracycline from biologic fluids for HPLC evaluation. J.Liq. Chromatogr., 3: 1561-1572. 1980.

25. Ensminger. W. D., Rosowsky, A., Raso, V., Levin, D. C., Glode, M.. Come. S..Steele, G., and Frei, E., HI. A clinical-pharmacological evaluation of hepatic arterialinfusions of 5-fluoro-2'-deoxyuridine and 5-fluorouracil. Cancer Res., 38: 3784-

3792, 1978.26. Collins. J. M., Dedrick. R. L., King. F. G.. Speyer. J. L., and Myers, C. E. Nonlinear

pharmacokinetic models for 5-fluorouracil in man: intravenous and intraperitonealroutes. Clin. Pharmacol. Ther., 28: 235-246, 1980.

27. Kuan, H. Y.. Smith, D. E., Ensminger. W. D.. Knol. J. A.. DeRemer. S. }.. Yang. Z..and Stetson. P. L. Regional pharmacokinetics of 5-bromo-2'-deoxyuridine and 5-

fluorouracil in dogs: hepatic arterial versus portal venous infusions. Cancer Res.. 56:4724-4727, 1996.

28. Ballet. F., Barbare, J. C., and Poupon. R. Hepatic extraction of adriamycin in patientswith hepatocellular carcinoma. Eur. J. Cancer Clin. Oncol.. 20: 761-764. 1984.

29. August, D. A., Verma. N., Vaertan. M. A.. Shah, R., and Brenner, D. E. An evaluationof hepatic extraction and clearance of doxorubicin. Br. J. Cancer, 72: 65-71, 1995.

30. Greene, R. F., Collins, J. M., Jenkins. J. F.. Speyer, J. L., and Myers, C. E. Plasmapharmacokinetics of adriamycin and adriamycinol: implications for the design of invitro experiments and treatment protocols. Cancer Res., 43: 3417-3421. 1983.

31. Mross. K.. Maessen. P.. van der Vijgh. W. J. F., Gall, H., Boven. E.. and Pinedo.H. M. Pharmacokinetics and metabolism of epidoxorubicin and doxorubicin inhumans. J. Clin. Oncol.. 6: 517-526, 1988.

32. McCuskey, R. S. The hepatic microvascular system. In: I. M. Arias, J. L. Boyer. N.Fausto, W. B. Jakoby, D. A. Schachter, and D. A. Shafritz (eds.). The Liver: Biologyand Pathobiology, pp. 1089-1106. New York: Raven Press, Ltd., 1994.

3343

on May 6, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1998;58:3339-3343. Cancer Res   Takeshi Iwasaki, Yonson Ku, Nobuya Kusunoki, et al.   Hepatic Venous Isolation and Charcoal HemoperfusionArterial and Portal Venous Administration: Evaluation with Regional Pharmacokinetics of Doxorubicin following Hepatic

  Updated version

  http://cancerres.aacrjournals.org/content/58/15/3339

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/58/15/3339To request permission to re-use all or part of this article, use this link

on May 6, 2021. © 1998 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from