リスペリドンおよびパリペリドン薬物動態に対する妊娠の影響を検討した生理学的薬物動態モデル：妊婦とその出生児への適用

概要

Kobe University Repository : Kernel
PDF issue: 2024-05-02

Physiologically-based pharmacokinetic model to
investigate the effect of pregnancy on
risperidone and paliperidone pharmacokinetics:
Application to a pregnant woman and her neonate

Mahdy, Walaa Yousef Bassyouni
(Degree)
博士（医学）

(Date of Degree)
2023-03-25

(Resource Type)
doctoral thesis

(Report Number)
甲第8615号

(URL)
https://hdl.handle.net/20.500.14094/0100482363
※ 当コンテンツは神戸大学の学術成果です。無断複製・不正使用等を禁じます。著作権法で認められている範囲内で、適切にご利用ください。

（課程博士関係）

学 位 論 文 の 内 容 要 旨

Physiologically-based pharmacokinetic model to investigate the
effect of pregnancy on risperidone and paliperidone
pharmacokinetics: Application to a pregnant woman and her
neonate

リスペリドンおよびパリペリドン薬物動態に対する妊娠の影響を検討した
生理学的薬物動態モデル：妊婦とその出生児への適用

Kobe University, Graduate School of Medicine
Department of Pharmaceutics
(Supervisor: Professor Ikuko Yano)

WALAA YOUSEF BASSYOUNI MAHDY
ワラー ヨセフ バッシオニ マディ

Introduction
Risperidone (RIS) is one of the most commonly prescribed antipsychotic drugs for
managing schizophrenia and bipolar disorder. The safety profile for RIS during pregnancy is
still unknown. Although animal studies have demonstrated that RIS does not cause direct
reproductive toxicity or teratogenic effects, human case studies have shown that RIS may
cause significant unwanted effects, ranging from self-limiting side effects to newborns
requiring intensive care and prolonged hospitalization. Therefore, RIS prescriptions for
pregnant women are restricted to circumstances in which the benefits outweigh the risks to
the fetus. Pregnant women and newborns are often excluded from scientific studies due to
ethical and legal concerns. Consequently, current knowledge on optimal dosing regimens,
pharmacokinetics, and safety characteristics of various drugs during pregnancy, fetal, and
neonatal periods is inadequate. One possible solution is using physiologically-based
pharmacokinetic (PBPK) modeling, which incorporates drug-specific parameters (e.g.,
physicochemical and disposition characteristics), physiological parameters relevant to
pharmacokinetic processes, and clinical trial designs to generate a quantitative predictive
model, might be a feasible approach for optimizing dosing regimens in pregnant and pediatric
populations.
The objectives of this study were to develop, verify, and personalize pediatric and
pregnant PBPK models for RIS and its active metabolite paliperidone (PAL). Consequently,
we applied this personalized PBPK model to the assessment of RIS and PAL
pharmacokinetics in a Japanese pregnant woman and her newborn by utilizing the “virtual
twin” approach to estimate the risk of toxicity in pregnant and neonatal populations. Then,
we investigated which physiological changes with pregnancy affects RIS and PAL
pharmacokinetics by using a fixed-parameter approach.

Materials and Methods
PBPK models for RIS and PAL in adults, pediatric, and pregnant populations were
developed and verified using the Simcyp simulator. The following criteria were
predetermined to assess model performance. First, visual predictive check for predicted

(Cpred) and observed (Cobs) serum concentrations was used. The model was verified when the
observed values were within the virtual population’s 90% prediction interval (5th–95th
percentile range), and the ratio of the predicted value to the observed value was within a twofold difference. Moreover, the Cpred and the respective Cobs were compared in goodness-offit plots. Then, to assess the prediction bias and precision of each analyte concentration, the
mean error percentage (ME) ± standard error (SE) and root mean square error (RMSE) of the
Cpred compared with the Cobs were calculated.
RIS and PAL serum concentrations were determined in a pregnant woman and her
newborn by liquid chromatography-tandem mass spectrometry. PBPK models were then
applied to our two subjects, generating their “virtual twins.” Effects of pregnancy on RIS and
PAL were examined using models with fixed pharmacokinetic parameters. A fixedparameter approach was used to assess the extent to which physiological changes associated
with pregnancy affect RIS and PAL pharmacokinetics. In the neonatal PBPK simulation, ten
methods, comprising four maturation-based models and six serum creatinine-based equations,
were selected from glomerular filtration rate (GFR) estimates approaches published in the
literature.

Results and Discussion
We developed a PBPK model using five previous studies in healthy adults (internal
dataset), then we confirmed its accuracy by comparing its results to data from another seven
studies (external dataset). The verified model was consequently extrapolated to pediatric
population and validated by comparing the simulated pharmacokinetic data to the observed
clinical data. Additionally, by parameterizing and expanding the model structure to
pregnancy-induced physiological changes, the adult (non-pregnant) PBPK model for RIS and
PAL was extrapolated to the pregnancy population. Then, it was utilized to estimate active
moiety (RIS plus PAL) plasma concentrations at baseline (pregnancy week 0) and by
trimester during pregnancy. The unique final step was to individualize the final pediatric and
pregnant PBPK models and use them to predict the serum disposition of RIS and PAL in a
Japanese woman and her newborn. The developed PBPK models accurately predicted PAL’s

pharmacokinetics, as shown by minimal bias and acceptable precision across populations.
The individualized maternal model predicted all observed PAL concentrations within the
90% prediction interval.
Our patient gave birth vaginally to a male child after 40 weeks and five days of gestation.
At birth, there were no signs of toxicity related to the use of RIS, and the baby was born
healthy without any congenital malformations or neonatal abstinence syndrome. In our case,
the maternal RIS concentrations were not detected in all samples (below 0.5 ng/mL), and the
serum PAL concentrations were between 2.05−3.80 ng/mL before childbirth and 3.80–9.90
ng/mL postpartum. The PAL concentrations seemed to increase after birth, although the
dosage of RIS was only changed from 0.5 mg twice a day to 1 mg once a day.
During pregnancy, changes in oral bioavailability, tissue volumes, blood protein
bindings, enzyme activity, and cardiac output can theoretically cause this observed decrease
in the AM concentration throughout pregnancy. In our PBPK analysis, at the end of
pregnancy, simulated intrinsic clearance (CL) of RIS and PAL mediated by CYP2D6 were
three times higher than in the nonpregnant state, and simulated total CL of RIS and PAL were
increased 1.61 and 1.95-times, respectively. Although hepatic CYP2D6 induction during
pregnancy promotes PAL production through RIS 9-hydroxylation, the increased total CL of
PAL nullified the increase in PAL concentrations. Actually, in the fixed-parameter combined
model, where all the parameters of interest were held constant, the predicted PAL maximum
concentration as well as RIS total CL were almost equal to the values observed in the nonpregnant state. Namely, the decreased serum albumin and the increased fetoplacental volume,
GFR, and CYP2D6 activity, with a concomitant increase in PAL production from RIS hepatic
metabolism, influenced RIS and PAL pharmacokinetics during pregnancy.
The RIS concentrations in the umbilical artery and vein were undetectable, whereas the
PAL concentrations were 1.10 and 1.05 ng/mL in the umbilical artery and vein, respectively.
This assessment of the fetal drug exposure may aid in directing treatment selection and
providing a greater insight into the extent to which intrauterine drug exposure is primarily
responsible for neonatal potential toxicity. The median PAL maternal-to-cord concentration
ratio was determined to be 2.83 at steady-state (typically after 4-5 days for PAL), which

corresponds reasonably well with previous findings. In another previous study, the maternalto-cord plasma ratio for AM was determined to be 2.44 ± 2.80 [standard deviation], which
coincides to the ratio for PAL alone.
RIS was not detected in the neonatal serum samples (below 0.5 ng/mL). The neonate
serum PAL concentrations were 0.99, 1.03, 0.83, and 0.82 ng/mL on 0, 1, 2, and 3 days after
birth, respectively, and became undetectable afterward. Due to the expected poor clearance
in neonates, we evaluated numerous GFR estimating approaches to identify the optimal
ontogeny model for this vulnerable population using our personalized PBPK model. The
Flanders metadata equation consisting of serum creatinine, body height, and age showed the
lowest absolute bias (ME: 22.3% ± 6.0%) and the greatest precision (RMSE: 23.8%) in
predicting PAL plasma concentration in the neonatal population.

Conclusion
Our PBPK model elucidated the mechanisms underlying the observed variations in RIS
and PAL serum concentrations and clearances in pregnant women and newborns. Our study
has three main implications: (1) laying out a strategy for using PBPK to predict drug
disposition in an individual patient with limited serum samples; (2) identifying key
parameters that influence drug pharmacokinetic parameters in pregnant women; and (3)
comparing various maturation models and formulas for calculating eGFR in our newborn.
As part of model-informed precision dosage, this model may help optimize RIS dosing
regimens across populations, moving away from a "one-size-fits-all" approach and avoiding
subtherapeutic or toxic doses that could expose the mother and fetus to both the medicine
and the disease.

神戸大学大学院医学（系）研究科（博士課程）

論 文 審 査 の 結 果 の 要 旨
甲 第 3280号

氏

名

受 付番 号

WalaaYousef
BassyouniMahdy

P
h
y
s
i
o
l
o
g
i
c
a
l
l
ybasedpharmacokineticmodelt
oi
n
v
e
s
t
i
g
a
t
et
h
e
e
f
f
e
c
to
fpregnancyonr
i
s
p
e
r
i
d
o
n
eandp
a
l
i
p
e
r
i
d
o
n
e
論文題目 p
h
a
r
m
a
c
o
k
i
n
e
t
i
c
s
:A
p
p
l
i
c
a
t
i
o
nt
oapregnantwomanandher
T
i
t
l
eo
f
I
neonate
D
i
s
s
e
r
t
a
t
i
o
n
リスペリドンおよびパリペリドン薬物動態に対する妊娠の影孵を検
討した生理学的薬物動態モデル：妊婦とその出生児への適用

主 査

ChiefExamin
e
r
審査委員

Examiner

副

査

副

査

v
1
c
e
-exa
m1ner
v
1
c
eexa
m1ner

エ［蚊汽 之
乳素畠ぃ
岱、？叙 A

（
要旨は 1, 000字 ∼ 2
, 000字程度）

I
n
t
r
o
d
u
c
t
i
o
n
R
i
s
p
e
r
i
d
o
n
ei
soneo
ft
h
emostcommonlyp
r
e
s
c
r
i
b
e
da
n
t
i
p
s
y
c
h
o
t
i
cdrugsf
o
rmanaging
s
c
h
i
z
o
p
h
r
e
n
i
aandb
i
p
o
l
a
rd
i
s
o
r
d
e
r
.I
ntheAustralianc
l
a
s
s
i
f
i
c
a
t
i
o
nsystemf
o
rp
r
e
s
c
r
i
b
i
n
g
m
e
d
i
c
a
t
i
o
n
s during p
r
e
g
nancy
,r
i
s
p
e
r
i
d
o
n
ei
sc
a
t
e
g
o
r
i
z
e
da
sac
a
t
e
g
o
r
y C drug
,and
r
i
s
p
e
r
i
d
o
n
ep
r
e
s
c
r
i
p
t
i
o
n
sf
o
rpregnantwomena
r
er
e
s
t
r
i
c
t
e
dt
oc
i
r
c
u
m
s
t
a
n
c
e
si
nwhicht
h
e
b
e
n
e
f
i
t
soutweight
h
er
i
s
k
st
ot
h
ef
e
t
u
s
.C
o
n
s
e
q
u
e
n
t
l
y
,c
u
r
r
e
n
tknowl
e
dgeono
p
t
i
m
a
ld
o
s
i
n
g
r
e
g
i
m
e
n
s
,p
h
a
r
m
a
c
o
k
i
n
e
t
i
c
s,ands
a
f
e
t
yc
h
a
r
a
c
t
e
r
i
s
t
i
c
so
fv
a
r
i
o
u
sd
r
u
g
sd
u
r
i
n
gpregnancy
,
f
e
t
a
l
,andn
e
o
n
a
t
a
lp
e
r
i
o
d
si
sinadequate.
Theo
b
j
e
c
t
i
v
e
so
ft
h
i
sstudyweret
od
e
v
e
l
o
pp
e
d
i
a
t
r
i
candpregnantph
y
s
i
o
l
o
g
i
c
a
l
l
y
b
a
s
e
d
pharmacokinetic(PBPK)modelsf
o
rr
i
s
p
e
r
i
d
o
n
eandi
t
sa
c
t
i
v
em
e
t
a
b
o
l
i
t
ep
a
l
i
p
e
r
i
d
o
n
e
.The
d
e
v
e
l
o
p
e
d PBPK model was a
p
p
l
i
e
d t
o e
v
a
l
u
a
t
e r
i
s
p
e
r
i
d
o
n
e and p
a
l
i
p
e
r
i
d
o
n
e
pharmacokinetics i
n a Japanese pregnant woman and her newborn, and c
l
a
r
i
f
i
e
d
p
h
y
s
i
o
l
o
g
i
c
a
lchangeswithpregnancyandmaturationa
f
f
e
c
t
i
n
gr
i
s
p
e
r
i
d
o
n
eandp
a
l
i
p
e
r
i
d
o
n
e
p
h
a
r
m
a
c
o
k
i
n
e
t
i
c
s
.
Methods
APBPKmodelus
i
n
gt
h
eSimcyps
i
m
u
l
a
t
o
rwasd
e
v
e
l
o
p
e
dusingf
i
v
ep
r
e
v
i
o
u
ss
t
u
d
i
e
si
n
hea
l
t
h
yadu
l
t
s
,andi
t
sa
c
c
u
r
a
c
ywasc
o
n
f
i
r
m
e
dbycomparingi
t
sr
e
s
u
l
t
swithdatafrom
anothersevens
t
u
d
i
e
s
.Themodelwasthene
x
t
r
a
p
o
l
a
t
e
dt
op
e
d
i
a
t
r
i
cp
o
p
u
l
a
t
i
o
nandv
e
r
i
f
i
e
d
by comparing t
h
es
i
m
u
l
a
t
e
d pharmacokine
t
i
c data t
ot
h
er
e
p
o
r
t
e
dc
l
i
n
i
c
a
ld
a
t
a
. By
paramete
r
i
z
i
n
g and expanding t
h
e model s
t
r
u
c
t
u
r
et
op
r
e
g
n
a
n
c
y
i
n
d
u
c
ed p
h
y
s
i
o
l
o
g
i
c
a
l
changes,t
h
e adu
l
t(
n
o
n
p
r
e
g
n
a
n
t
) PBPKmod
e
lf
o
rr
i
s
p
e
r
i
d
o
n
e and p
a
l
i
p
e
r
i
d
o
n
ewas
e
x
t
r
a
p
o
l
a
t
e
dt
ot
h
epregnancyp
o
p
u
l
a
t
i
o
n
.
R
i
s
p
e
r
i
d
o
n
eandp
a
l
i
p
e
r
i
d
o
n
eserumc
o
n
c
e
n
t
r
a
t
i
o
n
sweredeterminedi
napregnant
womanandhernewbornbyl
i
q
u
i
dchromatography-tandemmasss
p
e
c
t
r
o
m
e
t
r
y
.Developed
PBPKmodelswerea
p
p
l
i
e
dt
ot
h
e
s
es
u
b
j
e
c
t
s,g
e
n
e
r
a
t
i
n
gt
h
e
i
r"
v
i
r
t
ua
lt
w
i
n
s
."E
f
f
e
c
t
so
f
pregnancyonr
i
s
p
e
r
i
d
o
n
eandp
a
l
i
pe
r
i
d
o
n
ep
h
a
r
m
a
c
o
k
i
n
e
t
i
c
sweree
xaminedu
s
i
n
gaf
i
x
e
d
parameterapproacht
oa
s
s
e
s
st
h
ee
x
t
e
n
tt
owhichp
h
y
s
i
o
l
o
g
i
c
a
lchangesi
nt
hepregnancy
werea
s
s
o
c
i
a
t
e
dwitht
h
e
i
rp
h
a
r
m
a
c
o
k
i
n
e
t
i
c
s
.I
nt
h
en
e
o
n
a
t
a
lPBPKsimu
l
a
t
i
o
n
,t
e
nme
t
h
o
d
s,
comprisingf
o
u
rmaturationbas
edmod
e
l
sands
i
xserumc
r
e
a
t
i
n
i
n
e
b
a
s
e
de
q
u
a
t
i
o
n
s
,were
s
e
l
e
c
t
e
d from g
l
o
m
e
r
u
l
a
rf
i
l
t
r
a
t
i
o
nr
a
t
e (GFR) e
s
t
i
m
a
t
e
s approa
che
sp
u
b
l
is
hed i
nt
h
e
l
i
t
e
r
a
t
u
r
e
.
R
e
s
u
l
t
sandD
i
s
c
u
s
s
i
o
n
Developed PBPKmodels a
c
c
u
r
a
t
e
l
yp
r
e
d
i
c
t
e
d r
i
s
p
e
r
i
d
o
n
ea
sw
e
l
la
sp
a
l
i
p
e
r
i
d
o
n
e
pharmacokinetic
s
,a
s shown by minimalb
i
a
s and a
c
c
e
p
t
a
b
l
ep
r
e
c
i
s
i
o
na
c
r
o
s
sr
e
p
o
r
t
ed
p
o
p
u
l
a
t
i
o
n
s(
h
ea
l
t
h
ya
d
u
l
t
s
,p
e
d
i
a
t
r
i
c
s
,andp
r
eg
nantwomen
)
.Th
ep
a
t
i
e
n
tgaveb
i
r
t
ht
oa
malec
h
i
l
da
f
t
e
r40weeksandf
i
v
edayso
fg
e
s
t
a
t
i
o
n.
Thematernalr
i
s
p
e
r
i
d
o
n
ec
o
n
c
e
n
t
r
a
t
i
o
n
s
werenotd
e
t
e
ct
e
di
nal
ls
amples(below0
.
5ng/mL),
andt
h
eserump
a
l
i
p
e
r
i
d
o
n
ec
o
n
c
e
n
t
r
a
t
i
o
n
s
werebetwee
n 2.
05-3
.
8
0ng/mLb
e
f
o
r
ec
h
i
l
d
b
i
r
t
hand3
.
8
0
-9
.
9
0ng/mLp
o
s
t
p
a
r
t
u
m
. The
p
a
l
i
p
e
r
i
d
o
n
ec
o
n
c
e
n
t
r
a
t
i
o
n
ss
e
e
med t
oi
n
c
r
e
a
s
ea
f
t
e
rb
i
r
t
h.Thei
n
d
i
v
i
d
u
a
l
i
z
e
dmaternal
mode
lp
r
ed
i
c
t
e
da
l
lo
b
s
e
r
v
e
dp
a
l
i
p
e
r
i
d
o
n
ec
o
n
c
e
n
t
r
a
t
i
onsw
i
t
h
i
nt
h
e90% p
r
ed
i
c
t
i
o
ni
n
t
e
r
v
a
l
.
Thestudyr
e
v
e
a
l
e
dt
h
a
tchangesi
no
r
a
lb
i
o
a
v
a
i
l
a
b
i
l
i
t
y
,t
i
s
s
u
ev
o
l
u
m
e
s
,b
l
o
o
dp
r
o
t
e
i
n
b
i
n
d
i
ng
s
, e
nzyme a
c
t
i
v
i
t
y
, and c
a
r
d
i
a
c ou
t
pu
t during p
r
egna
ncy can a
f
f
e
c
tt
h
e
p
h
a
r
m
a
c
o
k
i
n
e
t
i
c
sofd
r
u
g
sandt
h
i
ss
houldbec
o
n
s
i
d
e
r
e
dwhenpre
s
c
r
i
b
i
n
gdrug
sduring
p
r
e
g
n
a
n
c
y
. Fixed-parameter simu
l
a
t
i
o
n
ss
howed t
h
a
t CYP2D6 a
c
t
i
v
i
t
yl
a
r
g
e
l
ya
f
f
e
c
t
s
r
i
s
p
e
r
i
d
o
n
eandp
a
l
i
p
e
r
i
d
o
n
ep
h
a
r
m
a
c
o
k
i
n
e
t
i
c
sduringp
r
e
g
n
a
n
c
y
.
Int
h
en e w b o r n ~ wasnotd
e
t
e
c
t
edi
nt
h
es
e
r
umsamples也elow0
.
5n
g
/
m
L
,
.

＼

Theneonateserump
a
l
i
p
e
r
i
d
o
n
ec
o
n
c
e
n
t
r
a
t
i
o
n
srangedbetween0
.
8
2ng/mLi
nt
h
el
a
s
tday
and 0
.
9
9ng/mLi
nt
hef
i
r
s
td
a
y
. Th
eFl
anders metadatae
q
u
a
t
i
o
nc
o
n
s
i
s
t
i
n
go
fserum
c
r
e
a
t
i
n
i
n
e
,bodyh
e
i
g
h
t
,andageshowedt
h
el
owestabso
l
u
t
eb
i
a
s(meane
r
r
o
r
:22.3%士 6
.
0
%
)
andtheg
r
e
a
t
e
s
tp
r
e
c
i
s
i
o
n(
r
o
o
tmea
ns
q
u
a
r
ee
r
r
o
r
:2
3
.
8
%
)i
np
r
e
d
i
c
t
i
n
gp
a
l
i
p
e
r
i
d
o
n
eplasma
c
o
n
c
e
n
t
r
a
t
i
o
ni
nt
h
ene
o
n
a
t
a
lp
o
p
u
l
a
t
i
o
n
.
C
o
n
c
l
u
s
i
o
n
Dev
e
l
o
pedPBPKmodele
l
u
c
i
d
a
t
e
dtheme
c
hanismsu
n
d
e
r
l
y
i
n
gt
h
eo
b
se
r
v
edvar
i
a
t
i
ons
i
nr
i
s
p
e
r
i
d
o
n
eandpal
i
p
e
r
i
d
o
n
es
e
r
umc
o
n
c
e
n
t
r
a
t
i
o
nsandc
l
e
a
r
a
n
c
e
si
np
reg
nantwomenand
newborns.Thisstudyha
st
hre
ema
i
ni
m
p
l
i
c
a
t
i
o
n
s
:(
1
)l
ay
i
ngo
u
tas
t
r
a
t
e
g
yf
o
ru
s
i
n
gPBPK
t
op
r
e
d
i
c
tdrugd
i
s
p
o
s
i
t
i
o
ni
nani
n
d
i
vi
d
u
a
lp
a
t
i
e
n
twithl
i
m
i
t
e
ds
e
r
um samp
l
e
s
;(
2
)i
d
e
n
t
i
f
y
i
n
g
keyparameterst
ha
ti
n
f
l
u
e
n
c
edru
gpharmacokineticparametersi
npregnantwomen;and(
3
)
comparingv
a
r
i
o
u
smaturationmode
l
sandf
o
rmulasf
o
rc
a
l
c
u
l
a
t
i
ngeGFRi
nt
h
enewborn.
Thecand
i
da
t
e
,havingc
o
mpl
e
t
eds
t
u
d
i
e
sonp
h
y
s
i
o
l
o
g
i
c
a
l
l
y
b
a
s
e
dpharmacokineticmod
e
l
ana
l
y
s
i
so
fr
i
s
p
e
r
i
d
o
n
eandp
a
l
i
p
e
r
i
d
o
n
ei
np
r
e
g
n
a
n
c
y
,wi
t
has
p
e
c
i
a
l
t
yi
npha
rmacok
i
n
e
t
i
c
c
hangesi
nt
h
epregnantwomenandn
e
o
n
a
t
e
s,an
dhavingadvancedt
h
ef
i
e
l
do
fknowledgei
n
t
hea
r
e
ao
fp
e
r
s
o
n
a
l
i
z
i
n
ga
n
t
i
p
s
y
c
ho
t
i
cdrugdosageregimensi
nthes
p
e
c
i
a
lpop
u
l
a
t
i
o
ns
,i
s
herebyr
e
c
o
g
n
i
z
eda
shav
i
n
gq
u
a
l
i
f
i
e
df
o
rt
h
ed
e
g
r
e
eo
fP
h
.
D
.(
M
e
d
i
c
i
n
e
).

r
ー