Pharmaceutical and Pharmacokinetic Research Laboratories,
Fujisawa Pharmaceutical Co. LTD., Osaka, Japan (M.K., Y.T., T.H.);
Department of Pharmacy, The University of Tokyo Hospital, Faculty of
Medicine, The University of Tokyo, Tokyo, Japan (K.Y., T.I.) and
Faculty of Pharmaceutical Sciences, Kyushu University, Fukuoka, Japan
(Y.S.)
The pharmacokinetic and pharmacodynamic behaviors of
4-[3-[3-[Bis(4-isobutylphenyl) methylamino]
benzoyl]-1H-indol-1-yl]-butyric acid (FK143), a new
nonsteroidal steroid 5
-reductase inhibitor, in the ventral prostate
were investigated after i.v. administration to rats. The relationship
between blood concentrations at 24 hr and doses was linear in the range
of 0.1 to 20 mg/kg. However, the levels of FK143 in the prostate were
saturated over the dose of over 5 mg/kg. The dissociation constant
(Kd) and maximum amount of binding
substances (Bmax), calculated according to
nonlinear kinetic analysis including a specific binding pool, was
0.0553 ± 0.0117 µg/ml (92 nM, estimated value ± S.D.) and
0.908 ± 0.092 µg/g tissue, respectively. A combined
pharmacokinetic/pharmacodynamic (PK/PD) model was constructed using
change in dihydrotestosterone (DHT) levels in the prostate after
i.v. administration of FK143 as an index for its pharmacological effect
and blood concentration as an input function. The apparent reaction
rate constant of drug and enzyme (K) was 39.7 ± 25.1 g
tissue/µg/hr (estimated value ± S.D.), the apparent turn-over
rate constant of enzyme (k) was 0.140 ± 0.107 hr
1, the elimination rate constant of DHT
(kel, DHT) was 1.13 ± 0.94 hr1 and the fraction of FK143-insensitive DHT
synthesis (F) was 0.461 ± 0.037. The PK/PD analysis suggested
that the duration of the effect of FK143 was related to its
accumulation in the binding pool of the prostate. After i.v.
administration of FK143 in the range of 0.1 to 20 mg/kg, the DHT levels
in the prostate decreased to about 40% of control value, after which
despite the rapid decline of blood FK143 concentration, slowly
recovered according to the elimination rate of FK143 in the prostate.
Moreover, the PK/PD profiles of FK143 after repeated i.v.
administration were predictable by using the PK/PD parameters obtained
after single administration of FK143.
 |
Introduction |
The
active androgen in the prostate is hypothesized to be DHT, which is
converted from testosterone by steroid 5-reductase, a hormone
secreted in the testis (Anderson and Liao, 1968
; Bruchovsky and Wilson,
1968a, b; Hammond, 1978; Sandberg, 1980). In support of this concept,
the level of DHT and steroid 5-reductase activity in the prostate in
patients with benign prostatic hyperplasia are higher than those in
normal men (Gloyna et al., 1970a, b; Hudson et
al., 1982a). Only finasteride
N-tert-butyl-3-oxo-4-aza-5-androst-1-ene-17
-carboxamide, Merck, Rahway, NJ), a 4-azasteroid, is presently in clinical use as a
steroid 5-reductase inhibitor. It reduces the prostate DHT level by
approximately 85% and about 20% of the volume of the prostate,
respectively, although more than 6 months are needed for the appearance
of its effect (Peters and Sorkin, 1993; The Finasteride Study Group,
1994; Rittmaster, 1994; Steiner, 1996). Therefore, the prediction of
its effect in the prostate from its blood concentration profiles would
be useful for effective dosage planning in clinical studies.
FK143, a novel compound with a non-steroidal structure, which inhibits
steroid 5-reductase in a noncompetitive inhibition manner, decreases
the level of DHT in the prostate (Hirosumi et al., 1995a)
and reduces the size of the prostate in rats and dogs (Hirosumi
et al., 1995b). We have previously investigated the disposition of FK143 in rats and reported that FK143 was transported from the blood to the prostate by a membrane-limited process and eliminated very slowly from the prostate as compared with from the
blood (Katashima et al., 1997). In this study, we
investigated the pharmacokinetic and pharmacodynamic behaviors of FK143
after a single i.v. administration to rats, using DHT levels in the prostate as an index for pharmacological effect due to the delay in
change of size of the prostate. The effect of FK143 on steroid 5-reductase activity after repeated administration was also
quantitatively evaluated using the PK/PD model.
| |
Materials and Methods |
Chemicals
FK143 and FR130976
(4-[1-[3-[Bis(4-isobutylphenyl)methylamino]benzoyl]-1H-indol-3-yl]-butyric
acid) were synthesized in Fujisawa Pharmaceutical Co. Ltd.(Osaka,
Japan). Acetonitrile and n-hexane were purchased from Wako Pure
Chemicals (Osaka, Japan) and were of HPLC grade. All other chemicals
used were of analytical grade.
Animals
Male Sprague-Dawley rats aged 8 weeks (Nippon Bio-Supp, Center,
Tokyo, Japan) were used in the experiments. Rats were maintained on a
12-hr light/dark cycle with food and water provided ad
libitum.
Pharmacokinetic Study
FK143 was dissolved in PEG-400 and i.v. bolus administered from
tail vein at a dose of 1 mg/kg to rats. At 0.05, 0.083, 0.25, 0.5, 1, 2, 4, 6, 8, 10 and 24 hr after administration, blood samples were
collected from the aorta using a heparinized syringe under light ether
anesthesia (just before sample collection) and were immediately
centrifuged at 2000 × g for 10 min. After collection of the blood samples, the ventral prostate was excised, weighed and
homogenized on ice with 3 volumes of ice-cold 20 mM phosphate buffer
(pH 7.0) using polytron homogenizer. In the same way, the blood and
prostate samples were collected at 24 hr after i.v. administration of
FK143 at doses of .1, .2, .5, 5, 10 and 20 mg/kg to rats. The samples
were stored at -20°C until analysis using HPLC (Katashima et
al., 1997). In the repeated dosing study, FK143 was administered
at 1 mg/kg every 24 hr for 3 or 7 days, via a PE-10 polyethylene tube
(Becton Dickinson & Co., sparks, MD) cannulated in the external jugular
vein. The blood and prostate samples were collected at 24 hr after the
first, third and seventh administration of FK143. Rats were killed at
each point for blood or prostate sampling.
Blood concentrations and prostate levels of FK143 after single or
repeated administration of 1 mg/kg have already been reported (Katashima et al., 1997).
Pharmacodynamic Study
FK143 was dissolved in PEG-400 and i.v. as a bolus
administered to rats at 0.1, 0.2, 0.5, 1, 5, 10 and 20 mg/kg. The
prostate samples were excised under light ether anesthesia (just before sample collection) at 1, 4, 10, 24, 30, 48, 72 and 240 hr after administration to rats (only 24-hr samples were collected in the case
of 10 and 20 mg/kg dosing). In the repeated dosing study, FK143 was
administered at 0.1, 1 and 20 mg/kg every 24 hr for 3 or 7 days and the
prostate samples were excised at the same time points as in the
"Pharmacokinetic study." The vehicle was administered to rats in
the control group, which were treated in the same way as
FK143-administered rats. The prostate samples were weighed and
homogenized on ice with 9 volumes of ice-cold distilled water. The
prostate homogenate samples were stored at -20°C until assay of DHT
levels with Radioimmunoassay (3H radioimmunoassay
kit, Amersham, Buckinghamshire, UK).
Data Analysis
Pharmacokinetic analysis of the prostate levels of FK143.
The blood concentrations of FK143 (Cb = Cp · RB) were calculated
from plasma concentrations (Cp) using blood to
plasma concentration ratio (Rb = 0.68) (Katashima
et al., 1997) and fitted to the following equation by the
nonlinear least squares method using a WinNonlin program (Version 1.1, Scientific Consulting Inc., Apex, NC)
|
(1)
|
The prostate levels of FK143 were analyzed by a compartment
model as shown in figure 1. FK143 was
very slowly transfered into the prostate, indicating that the
rate-limiting process of tissue distribution of FK143 is membrane
transport and eliminated very slowly from the prostate caused by the
binding pool (Katashima et al., 1997). In this model, the
following propositions were assumed. The drug is transported with the
uptake clearance of CL1 (ml/hr/g tissue) from the
blood compartment to the precursor pool (A1,
µg/g tissue), where the drug is unbound or nonspecifically bound, and
returns from the precusor pool to the blood compartment with efflux
rate constant of k2
(hr1). The binding pool
(A2, µg/g tissue), where the drug is
specifically bound, was connected with A1 by
bimolecular association rate constant (kon, g
tissue/µg/hr) and dissociation rate constant
(koff, hr1). The
mass balance of the drug in the prostate can therefore be expressed as
follows:
|
(2)
|
|
(3)
|
|
(4)
|
where B (µg/g tissue) is the amount of substances which
specifically bind to FK143 and Vd (ml/g tissue)
is the physical volume of the prostate. Assuming the rapid association
and dissociation,
|
(5)
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1.
A pharmacokinetic/pharmacodynamic model for FK143,
a steroid 5 -reductase inhibitor, in rats. Cb, Blood
concentration of FK143; CLtot, total body clearance of
FK143; A1, amount of drug in the precursor pool;
A2, amount of drug in the binding pool; CL1,
tissue uptake clearance from blood compartment to precursor pool;
k2, efflux rate constant from precursor pool to blood
compartment; Kd, dissociation constant of drug-binding
substances complex; Bmax, maximum amount of specific
binding substances of drug; E, amount of active steroid 5-reductase;
Ec, amount of inactive steroid 5-reductase; K, apparent
reaction rate constant of FK143 and steroid 5-reductase;
Ks, biosynthesis rate of steroid 5-reductase; k3, elimination rate constant of steroid 5-reductase;
k4, recovery rate constant of inactive steroid
5-reductase; k, apparent turn-over rate constant of enzyme
(k3+k4); v0, biosynthesis rate of
DHT; kel, DHT, elimination rate constant of DHT;  , ratio
of active and total steroid 5-reductase; F, fraction of DHT
synthesis, which is not inhibited by the drug.
|
|
Therefore, A2 is given by the the following
equation (6).
|
(6)
|
where Bmax (µg/g tissue) is the maximum
amount of substances that specifically bind to FK143 and
Kd (µg/ml) is the dissociation constant
of drug-binding substances complex (Kd = koff/kon); because At = A1 + A2, At is represented by
equation (7).
|
(7)
|
Accordingly, the differential equation for At is
as follows:
|
(8)
|
Furthermore, the following equations can be derived from equations (4)
and (8).
|
(9)
|
|
(10)
|
The value of CL1 was previously estimated by
integration plot (Katashima et al., 1997). The physical
volume of the prostate, Vd, was assumed to be 1 ml/g. The prostate levels of FK143 after i.v. administration in the
range of 0.1 to 20 mg/kg were fitted to the equations (4) and (9) by
the nonlinear least squares method using a WinNonlin program and the
kinetic parameters k2,
Kd and Bmax were
estimated. In the simulation study, the binding characteristics of
FK143 in the prostate was assumed not to be changed during repeated
administration.
PK/PD analysis of FK143.
In this PK/PD analysis (fig.
1), the changes of DHT level in the
prostate after i.v. administration of FK143 were used as an index of
pharmacological effect. The inhibitory effects on steroid
5-reductase were analyzed by the PK/PD model with inhibition parameters for the enzyme (Katashima et al., 1995; Sugiura
et al., 1992; Yamamoto et al., 1996), assuming
that the FK143 in the precusor pool can react with steroid
5-reductase. It was assumed that steroid 5-reductase was
synthesized at a constant rate, Ks, and
eliminated with first order rate constant, k3,
and DHT was synthesized at a constant rate, v0,
by steroid 5-reductase and eliminated with first order rate
constant, kel, DHT (Ko and Jusko, 1995). The
amount of active enzyme E and DHT levels in the prostate are given by
the following equations.
|
(11)
|
|
(12)
|
In the steady state, the amounts of steroid 5-reductase and DHT are
maintained at constant values of E0 and
DHT0, respectively.
|
(13)
|
|
(14)
|
After administration of FK143, drug in the precusor pool is assumed to
react with steroid 5-reductase with second-order rate constant, K,
and active form of enzyme E is assumed to be transformed to inactive
form Ec. If the elimination rate of E and
Ec are the same, the total amount of enzyme is
kept constant, E0 (= E + Ec). Assuming that Ec
recovered to E with first order rate constant, k4, the amount of E is expressed by the following
equation:
|
(15)
|
Assuming steroid 5-reductase activity () is linearly related to
the ratio of active enzyme to total enzyme
(E/E0), the following equation is derived from
equations (13) and (15).
|
(16)
|
where k (apparent turn-over rate constant of enzyme) is
(k3 + k4).
The synthesis rate of DHT is assumed to be related to
, therefore DHT levels after administration of FK143 are given by
the following equation (17):
|
(17)
|
where F is the DHT synthesis fraction which is not apparently
inhibited by FK143. The DHT level change in the prostate after administration of FK143 is derived from equations (14) and (17):
|
(18)
|
FK143 levels in the precursor pool (A1) were
calculated from blood concentrations and FK143 levels in the prostate
by equation (9). Then, equations (16) and (18) were fitted to the
changes of DHT levels after administration of FK143 in the range of
0.1-20 mg/kg by nonlinear least squares method using a WinNonlin
program to estimate the PK/PD parameters K, k, kel,
DHT and F.
| |
Results |
Pharmacokinetics of FK143.
The blood concentrations and
prostate levels of FK143 after i.v. administration of 1 mg/kg to rats
are shown in figure 2. The
pharmacokinetic parameters from equation 1 were 7.62 ± 4.49 µg/ml (estimated value ± S.D.) for A, 0.692 ± 0.370 µg/ml, 0.105 ± 0.021 µg/ml for C, 17.5 ± 10.0 hr1 for , 1.77 ± 0.78 hr1 for and 0.0672 ± 0.0157 hr1 for 
. FK143 was very slowly
eliminated from the prostate (t1/2 = 113 hr)
compared with that from the blood (t1/2 = 10.3 hr). The concentrations of FK143 in the blood and prostate at 24 hr after i.v. administration are shown in figure
3. The blood concentrations were linear
up to 20 mg/kg, although FK143 level profile in the prostate was
saturable with the doses of 5 mg/kg or more. FK143 levels in the blood
and prostate at 0.1 mg/kg were under determination limit (blood, 10 ng/ml; prostate, 40 ng/g).
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2.
Blood ( ) concentrations or prostate ( ) levels
of FK143 after i.v. administration of 1 mg/kg to rats. Each
point represents mean ± S.D. (n = 3).
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 3.
Relationship between dose and blood concentrations
(A) or prostate levels (B) of FK143 at 24 hr after i.v.
administration to rats. Each point represents mean ± S.D.
(n = 3-5). The line in (B) is simulation curve by
the PK model.
|
|
We carried out our pharmacokinetic analysis using the model with
binding pool in the prostate (fig. 1). The estimated parameters of
binding were 0.0553 ± 0.0117 µg/ml (92.0 nM, estimated
value ± S.D.) for Kd, 0.908 ± 0.092 µg/g for Bmax and 0.214 ± 0.022 hr1 for k2, assuming
Vd equals 1 ml/g. A good agreement was observed between the fitted curves and the observed values (fig.
4). Furthermore, the simulation curve
based on the above kinetic parameters showed good agreement with the
observed data after repeated administration of FK143 at 1 mg/kg once a
day (fig. 5).
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 4.
Prostate levels of FK143 after i.v.
administration to rats. Each point represents mean ± S.D.
(n = 3-5). Curves are the fitting lines by the PK
model.  , 0.2 mg/kg; , 0.5 mg/kg; , 1 mg/kg;  , 5 mg/kg;  ,
10 mg/kg;  ; 20 mg/kg.
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 5.
Prostate levels of FK143 after repeated
i.v. administration at 1 mg/kg once a day to rats. Each
point represents mean ± S.D. (n = 3). The
curve is the simulation line calculated by the PK model.
|
|
Pharmacokinetic/pharmacodynamic analysis of FK143.
DHT levels
in the prostate at 24 hr after administration of FK143 decreased with a
dose-dependent manner in the range of 0.1 to 20 mg/kg and appeared to
be nonlinear at more than 5 mg/kg. The estimated dynamic parameters
were 39.7 ± 25.1 g tissue/µg/hr (estimated value ± S.D.) for K, 0.140 ± 0.107 hr1, for
k 1.13 ± 0.94 hr1 and for kel,
DHT 0.461 ± 0.037 for F. A good agreement was noted between the fitted curves and observed values for prostate DHT level in
the range of 0.1 to 20 mg/kg (fig. 6).
Also, the simulation curves, for the change of DHT level after repeated
administration of FK143 at 0.1, 1 and 20 mg/kg, based on dynamic
parameters after single administration of FK143 were in good agreement
with the observed values (fig. 7).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 6.
DHT levels in the prostate after i.v.
administration of FK143 to rats. , individual data points. Curves
are the fitting lines by the PK/PD model.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 7.
DHT levels in the prostate after repeated i.v.
administration of FK143 once a day to rats. Each point represents
mean ± S.D. (n = 3-4). Curves are simulation
lines calculated by the PK/PD model.  , 0.1 mg/kg; , 1 mg/kg; ,
20 mg/kg.
|
|
| |
Discussion |
We have already suggested the existence of a binding pool in the
prostate, a part of which might be related to steroid 5-reductase or
its associated substances (Katashima et al., 1997). In this study, blood concentrations of FK143 at 24 hr after administration were
linear with dose escalation (0.1-20 mg/kg).However, the prostate levels
of FK143 were nonlinearly increased (fig. 3) with
Kd value of 92.0 nM. The
Kd values of 4-methyl-aza-steroid, which
is another steroid 5-reductase inhibitor, were 7 nM for the prostate
microsome and 6.5 nM for the liver microsome, respectively, and were
comparable with Ki value (5 nM) for
steroid 5-reductase (Liang et al., 1983). In addition,
NADPH-dependent binding capacities of 4-methyl-aza-steroid correlated
with the steroid 5-reductase activity in several tissues (Liang
et al., 1983). On the other hand, the affinity of FK143 for
binding pool in the rat prostate (Kd, 92.0 nM) in vivo was more than 10 times larger than
IC50 (=Ki, 4.2 nM)
obtained from the inhibitory effect on steroid 5-reductase in
vitro. Further study will be required to relate binding parameters
in vivo to those in vitro.
In the developed PK/PD model (fig. 1), FK143 in the precursor pool
(A1) directly reacts with steroid 5-reductase
to transform the enzyme to inactive form. The apparent turn-over rate
constant of enzyme, k, defined as k3 + k4, was estimated to be 0.140 hr1 and the apparent half life of the
enzyme, calculated from k, was 5.0 hr. It is reported that the
turn-over half lives of NADPH P-450 reductase, cytochrome
b5 and drug metabolic enzymes, P-450 (PB) and
P-450 (MC), in rat liver are 35, 50, 25 and 15 hr, respectively (Sadano
and Omura, 1982). Assuming that the turn-over of steroid 5-reductase
also takes as long as scores of hours, k4 should be much larger than k3, resulting in k 
k4. Accordingly, the half life of recovery from
inactive form to active form of enzyme was estimated to be 5.0 hr.
Then, the apparent Ki value in
vivo (k/K) was estimated to be 6 nM, which is similar to the
IC50 (4.2 nM) (Hirosumi et al., 1995a)
in vitro.
The recovery half-life of enzyme (5.0 hr) was shorter than the terminal
phase elimination half life of FK143 from the blood (10.3 hr, fig. 2).
However, the levels of DHT in the prostate after administration of
FK143 decreased gradually, reaching minimum value at about 10 to 24 hr
after administration, and then recovered very slowly (fig. 6). The
elimination of FK143 from the prostate was slow with a half-life of 113 hr (fig. 2). This finding suggested that the duration of the
pharmacological effect of FK143 is influenced by the existence of the
binding pool, which characterized the tissue distribution profile of
the drug in the prostate.
It is reported that the time course of the recovery of vitamin K
epoxide reductase activity is parallel with that of the level of
microsome-free warfarin binding sites in the liver after administration of warfarin to rats and that the half life of recovery of enzyme activity is about 7 days (Thijssen and Janssen, 1994). Similarly to
warfarin, the tissue binding of FK143 was closely related to its
inhibitory effect on the enzyme. We have previously analyzed the
inhibitory effect of proton pump inhibitors, omeprazole and lansoprazole, on acid secretion (Sugiura et al., 1992;
Katashima et al., 1995) and the antiplatelet effect of
aspirin (Yamamoto et al., 1996), using a PK/PD model similar
to that used in this study. The long duration of inhibitory effect of
these drugs, compared with elimination from the plasma was proved to be
controlled by the irreversible binding of drugs and enzymes. It seems
that our PK/PD model is useful for evaluating the quantitative
relationship between plasma drug concentration and enzyme inhibition
with various mechanisms.
In our PK/PD model, DHT is assumed to be synthesized at a constant rate
and eliminated with first-order process. DHT in the prostate is mainly
metabolized by 3-HSOR, 3-HSOR and 17-HSOR. Among these
enzymes, 3-HSOR plays an important role in controlling DHT levels in
the prostate (Isaacs and Coffey, 1981; Sandberg, 1980; Hudson, 1982b).
The elimination rate constant of DHT (kel, DHT) was estimated to be 1.13 hr1 in the
present PK/PD analysis. Intrinsic clearance of DHT metabolism (CLint = kel, DHT · Vd) in the prostate can be calculated to be 1.13 ml/hr/g tissue. CLint of DHT for the metabolism
by 3-HSOR in rat prostate microsome in vitro is estimated
to be 0.0509 ml/hr/mg protein based on the report by Fukabori et
al. (1992), which is further calculated to be 3.33 ml/hr/g tissue
based on the report by Haaparanta et al. (1983).
CLint estimated from the in vitro study was only about three times larger than that estimated from kel, DHT obtained by PK/PD analysis in
vivo. The difference between in vivo and in
vitro may be related to the location of 3-HSOR in the prostatic
cells, because 3-HSOR exists not only in microsomes but also in
cytosol and nuclear fraction (Fukabori et al., 1992; Abalain
et al., 1989). Furthermore, the CLint
in vivo may be underestimated because 3-androstanediol produced
from DHT by 3-HSOR is partly returned to DHT in vivo
(Isaacs and Coffey, 1981; Sandberg, 1980; Hudson, 1982b).
Although FK143 almost completely inhibited steroid 5-reductase
in vitro (Hirosumi, 1995a), DHT levels in the rat prostates were not decreased to less than 40% of control by FK143 (figs. 6 and
7), which is similar to the results reported by Hirosumi et
al., 1995b. In this PK/PD model, a route of DHT synthesis was assumed which is not inhibited by testosterone, but other possibilities cannot be excluded. For example, testosterone levels in the prostate may be increased by the inhibition of steroid 5-reductase by FK143,
resulting in the increase in DHT synthesis rate. But, it is reported
that the levels of testosterone do not influence the inhibitory effect
of FK143 on the enzyme (Hirosumi et al., 1995a). Another
possibility is that the concentrations of FK143 may not be sufficient
for substrate of steroid 5-reductase in the prostate in
vivo. The reason for incomplete reduction of DHT level remains unclear.
In conclusion, the PK/PD behavior of FK143 in the prostate was
clarified by nonlinear distribution data of FK143 in the prostate and
the change in DHT levels in the prostate as index for pharmacological effect. The PK/PD profiles of FK143 after repeated administration could
be successfully predicted from the profiles after single administration.
The authors thank Dr. Toshitaka Manda for his critical comments
on this paper, and Dr. Susumu Tsujimoto and Ms. Sanae Matsumoto for
technical assistance.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
-Reductase Inhibitor,
4-[3-[3-[Bis(4-Isobutylphenyl)Methylamino]Benzoyl]-1H-Indol-1-yl]-Butyric
Acid, in Rats
Top
Abstract
Introduction
Finasteride.
N Engl J Med
330:
120-125
