30 BioPharm International eBook March 2021 www.biopharminternational.com
working group used PBPK models
in Simcyp to compare observed
and predicted outcomes for renal
and hepatic impairment studies
for 26 dr ugs eliminated princi-
pally by metabolism (15). For renal
impairment, all predictions were
considered acceptable using the
predefined two-fold criteria. For
hepatic impairment, the same con-
clusion was reached with excep-
tion of three cases for which a
modest over prediction was noted.
Moving forward, these models are
f inding utilit y in rationalizing
drug development in these special
populations and contributing to
a totality of evidence approach to
safety assessment.
IMPACT OF PBPK MODELING
ON ASSESSING FOOD EFFECT
Over the past 10 years, approxi-
mately 40% of approved dr ugs
intended for oral administration
report some change in pharmaco-
kinetics when dosed with a meal.
T hese cha nges in pha r macok i-
netics are commonly referred to
as food effects and can be large
enough where they warrant special
instructions—such as taking the
dose on an empty stomach—for
physicians and patients to ensure
safe and effective dosing.
Accordingly, the study of food
ef fec t in clinica l development
is expected by major regulatory
agencies around the world and
represents an area where virtu-
al-human approaches could have
an impact in streamlining new
drug development. The physical,
chemical, and biological processes
involved in drug absorption and
food effects have been integrated
into such an approach, namely
PBPK biopharmaceutics models, and
were evaluated by scientists work-
ing with the IQ Consortium (7).
The authors generated de novo
mechanistic absorption models
for 30 drugs, using a pre-specified
modeling approach to prospectively
evaluate the predictive performance
of the PBPK models and to establish
best practices intended to help drive
consistency, rigor, and acceptance of
the approach. Notably, the authors
had human pharmacokinetic data
following intravenous adminis-
tration, in addition to oral dosing
with and without food, for all drugs
in the study, which allowed them
to better study the changes in PK
that arise from changes in absorp-
tion, as opposed to any potential
effect of food on either elimination
or distribution. A thorough anal-
ysis of the data revealed no clear
trend in prediction accuracy with
Biopharmaceutics Classification
System (BCS) designation. However,
the authors concluded that when
the effect of food administration
involves changes in gastrointestinal
physiology such as fluid volume,
motility, pH, and bile salts, the food
effects were generally predictable. A
decision tree was formulated based
upon the analysis to guide the cat-
egorization of application of PBPK
models of food effects into those
of high, medium, and low confi-
dence (7). Such categorization will
be helpful in the interpretation of
PBPK food effect models as well as
in identifying areas where addi-
tional research is needed (i.e., those
models considered moderate and
low confidence).
CONCLUSION
PBPK modeling has been estab-
lished as an important predictive
science in pharmaceutical dr ug
development over the past t wo
decades; the aforementioned case
studies from the IQ Consortium
demonstrate the significant prog-
ress that the industry is making
toward enabling 'virtual' human
e x per i ments a nd st rea m l i n i ng
drug development.
REFERENCES
1. C. Perry, et al., Curr. Pharmacol. Rep., 6
71–84 (2020).
2. M. Jamei, G. L. Dickinson, A. Rostami-
Hodjegan, Drug Metab. Pharmacokinet.,
24 (1) 53–75 (2009).
3. L. Kuepfer L, et al., CPT
Pharmacometrics Syst. Pharmacol., 5
(10) 516–531 (2016).
4. M. Shebley, et al., Clin. Pharmacol. Ther.,
104 (1) 88-110 (2018).
5. M. Jamei, et al., Expert Opinion on Drug
Metabolism & Toxicology. 5 (2) 211–223
(2009).
6. N. Miller, et al., Clinical Pharmaco-
kinetics 58 (6) 727–746 (June 2019).
7. A.E. Riedmaier, et al., AAPS J. 22(6) 123
(2020).
8. H.M. Jones, et al., Clin. Pharmacol. Ther.
97 (3) 247–262 (2015).
9. H.J. Einolf, et al., Clin. Pharmacol. Ther.
95 (2) 179–188 (2014).
10. M.L. Vieira, et al., Clin. Pharmacol. Ther.
95 (2) 189–198 (2014).
11. FDA, In Vitro Drug Interaction Studies
— Cytochrome P450 Enzyme- and
Transporter-Mediated Drug Interactions
Guidance for Industry (CDER, January
2020).
12. I.E. Templeton, et al., CPT
Pharmacometrics Syst. Pharmacol. 5
(10) 505–515 (2016).
13. Y. Chen, et al., CPT Pharmacometrics
Syst. Pharmacol. 8 (9) 685–695 (2019).
14. O.A. Fahmi, et al., Drug Metab. Dispos.
44 (10) 1720–1730 (2016).
15. T. Heimbach, et al., Clin. Pharmacol.
Ther. in press.
BP
ABOUT THE AUTHORS
Heidi J. Einolf is director, mod-
eling and simulation, phar ma-
c o k i n e t i c s c i e n c e s , N o v a r t i s ;
Stephen D. Hall is senior research
fe l low, d r ug d i s p o sit ion, a nd
Tracy Williams is senior direc-
t o r A D M E / To x i c o l o g y/ P K P D,
both with Eli Lilly and Co; Aarti
Patel is director, DMPK modeling,
Gla xoSm it hK line; Chr istopher
Gibson is d ist ing uished scien-
tist and Nancy G.B. Agrawal is
v ice-president, pha r macok inet-
ics, pharmacodynamics, and drug
metabolism, bot h w it h Merck;
and Jens Sydor,* jens.x.sydor @
gsk.com, is vice-president, DMPK,
GlaxoSmithKline; all authors are
members of the IQ Consortium.
*To whom all correspondence should
be addressed
Regulatory Sourcebook Quality Collaboration
