Back to the Top
From the draft FDA draft Guidance on exploratory IND studies a
microdose is defined as "less than 1/100th of the dose calculated to
yield a pharmacological effect of a test substance and a maximum dose
of <=100 micrograms."
The microdosing approach in humans allows pharmacokinetic information
to be obtained, while minimising the exposure of subjects to the drug.
For this approach to be useful clearly the human pharmacokinetics at
the microdose level should mimic the pharmacokinetics observed for
the drug at the pharmacologically relevant dose; otherwise quite
misleading information could be obtained.
Could I solicit views on how often pharmacokinetic information
obtained following microdosing has been shown to be representative of
the phamacokinetics at the pharmacologically relevant dose.
Thank you
Angus McLean Ph.D,
8125 Langport Terrace,
Suite 100,
Gaithersburg,
MD 20877
tel 301-869-1009
fax 301-869-5737
BioPharm Global Inc.
Back to the Top
Hi Angus
We have literally just published a paper (Drug Discovery Today 10 (13)
890- 895 (July 2005)) on the role of microdosing in drug development
(attached [- sorry not attached db]) which addresses your concerns.
As you indicate one of the legitimate anxieties levied at microdosing
is the potential for the PK at a microdose to be misleading compared to
that achievable at a pharmacological dose. As a consequence the CREAM
(Consortium for Resourcing and Evaluating AMS Microdosing) trial was
sponsored by Eli Lilly, Roche, Schering AG and Servier. The results
were recently presented by Professor Malcolm Rowland at the 2005
American Society of Clinical Pharmaceutical & Therapeutics (ASCPT)
Conference in Orlando. The CREAM trial evaluated the microdose PK of
five molecules following oral and IV dosing compared to PK data
obtained at historical pharmacological doses. The compounds were
selected on the basis there would be difficulties in predicting the
therapeutic dose PK from a microdose.
For example, midazolam has a high first pass metabolism mediated by
CYP3A4 and diazepam is a low clearance molecule eliminated via CYP2C19.
On a priori grounds it had been anticipated that the predictive pK
capability of a microdose for a pharmacological dose for these types of
drugs would be poor due to small quantities of drug not saturating the
metabolism or clearance mechanisms. However, in both cases, the dose
normalised pK profiles were 'superimposable' providing real evidence of
the utility of microdosing even for difficult molecules. Another of
the CREAM molecules was a proprietary Schering AG compound, ZK253 for
which animal models had failed to predict poor human bioavailability at
a pharmacological dose. The drug was dropped after Phase I for these
bioavailability reasons but not before significant investment.
However, would microdosing have provided an equally clear outcome but
at a fraction of the cost or would it generate a false positive due to
the low drug dose? Following oral microdose administration, the plasma
levels of ZK253 were below the limit of quantification (LOQ) for a
highly sensitive AMS assay confirming the poor oral bioavailability of
ZK253 thereby building confidence in the decision-making potential of
microdosing. Less encouraging was the data for warfarin which is
characterised by extensive, albeit slow, in vivo metabolism principally
via CYP2C9. The microdose was not truly predictive of a
pharmacological dose, although the long half-life (due probably to slow
metabolism) was manifested in the microdose pK data. Therefore, in
general, the trends with microdose pK would have supported
decision-making for a pharmacological dose with this molecule. The
results with erythromycin were eagerly awaited as the molecule is a
substrate for both CYP3A4 and p-glycoprotein and again saturation of
processes with a microdose was likely to be questionable. However,
unfortunately, the design of the study led to unprotected delivery of
erythromycin which lacks stomach stability and therefore no result was
obtained.
In summary, the outcome of the CREAM study is extremely encouraging for
the microdosing concept with truly predictive PK for three 'difficult'
candidates and in all cases positive trends in the microdose PK,
compared to pharmacological dose.
If you need any more information then please let me know.
Best regards
Ian Wilding
Pharmaceutical Profiles
Mere Way
Ruddington Fields
Ruddington
Nottingham
UK
NG11 6JS
Tel: + 44 115 974 9000
Fax: + 44 115 974 8000
E-mail : iwilding.-at-.pharmprofiles.co.uk
www.pharmprofiles.com
www.microdosing.co.uk
Back to the Top
Dear Angus,
Another point to be taken into account when considering microdoses is
the specific binding of the drug to receptors. With conventional drugs
the amounts dosed are far in excess of the molar amount of the specific
target. This may not be so with microdosing. As soon as the molar
amounts of drug and binding target become similar, then the
distribution of the drug will become dependent upon these specific
binding events. It would be prudent to have knowledge of the expression
levels of the target binding entity or entities.
Consider the reaction A + B <=> C. Note also that drug A has inputs and
elimination, as has the binding target, B. When A and B bind they form
a complex, C, which itself can be eliminated. When more A is given, the
equilibrium reaction is pushed to the right, forming C and thereby
depleting free B. It can then be seen that, unless A is in excess, the
kinetics of A become very dependent upon the interaction with and
kinetics of B and C.
This is not so common with conventional drugs, but more so with
monoclonal antibodies, which can, in effect, be given as microdoses
when molar quantities are considered. At low doses, the distribution
and elimination of the monoclonal will be dependent upon what it binds
to. Hence, for example, if it binds to a circulating cell which has a
short half-life, then the monoclonal will exhibit a short half-life
when low doses are administered. At higher doses, keeping
concentrations in excess of the Kd for binding, then we observe the
long half-life and distribution typical of IgG class antibodies.
These effects are nicely summarised in the excellent paper by Lobo,
Hansen & Balthasar (2004 J Pharm Sci 93 2645)
http://www3.interscience.wiley.com/cgi-bin/abstract/109596510/ABSTRACT.
Best regards, Phil.
Philip J. Lowe PhD
Modelling & Simulation, Clinical Development & Medical Affairs
Novartis Pharmaceuticals AG
4002 Basel
Switzerland
Back to the Top
The following message was posted to: PharmPK
Hi All
If anyone would like a personal copy of our microdosing article from
the July edition of Drug Discovery Today then please send an email to
iwilding.aaa.pharmprofiles.co.uk.
Thanks and best regards
Ian Wilding
Pharmaceutical Profiles
Mere Way
Ruddington Fields
Ruddington
Nottingham
UK
NG11 6JS
Tel: + 44 115 974 9000
Fax: + 44 115 974 8000
E-mail : iwilding.-a-.pharmprofiles.co.uk
www.pharmprofiles.com
www.microdosing.co.uk
Back to the Top
The following message was posted to: PharmPK
phil.lowe.aaa.novartis.com wrote:
>
> Another point to be taken into account when considering microdoses is
> the specific binding of the drug to receptors. With conventional
drugs
> the amounts dosed are far in excess of the molar amount of the
specific
> target. This may not be so with microdosing.
I agree with Phil that one must be aware of the potential for binding
to specific receptors. This may be the reason why the CREAM study
found differences between conventional and microdosing PK for
warfarin. If you consider warfarin from the point of view of its
unbound concentration then it has a large volume of distribution
(around 1000 L/70 kg with conventional doses) which implies
extensive tissue binding (or partition). It would not therefore not
be surprising if microdosing had differerent PK due to differences in
volume of distribution.
IMHO the conventional classification of warfarin as a 'small' volume
of distribution drug (based on total plasma conc) is unhelpful in
making one aware of the magnitude of tissue binding.
Nick
Back to the Top
The following message was posted to: PharmPK
One reason why MICRODOSING has not been recognized earlier is the
conventional
use of assay methods that are selected for expediency and convenience,
frequently, however, lacking sensitivity and resolution.
For the same reason, studies of HORMESIS, low-dose stimulatory versus
high-dose toxic effects of the same compound, have been few. For too
long,
the value of microdosing and hormesis has remained unrecognized or
played
down. Related experiments require attention to detail and patience
that may
be considered inexpedient and are therefore easily dismissed.
For studies of microdose pharmacokinetics and hormesis, adequate
technology is
required. Sensitive techniques and attitudes need to be developed,
particularly for preclinical ADME procedures. Even for the testing of
relatively high pharmacological doses, current approaches may be
inadequate
(Monro, Drug Metab.Dispos. 22(1994):341-342).
Although it is widely acknowledged that the predictive value of
results from
routine ADMET studies is low and failure rate in drug development is
inexpediently high, regulatory requirements remain permissive.
Others and I (reviewed in J Pharm Tox Meth, 51(2005):25-40) argued
for a need
of in vivo detection with identification and characterization of high
specificity-low capacity tissue and cellular sites - against the current
listing of predominant low specificity-high capacity sites (?sites of
loss?)
with 14C-labeled compounds. 14C-labeled compounds with their low
specific
activity and administered at high dose, compared to tritium, may miss
recognition of target tissues and related pharmacokinetics.
Conventional ADME
radio-assays with organs or chunks of organs and whole body
autoradiography
frequently fail to provide the needed data, as is the case with the
elegant
non-invasive scanning procedures.
In preclinical studies especially, sensitive methods need to be added
to the
current ?expedient? low-resolution approaches. This is a topic to be
dealt
with in future articles and conferences.
Receptor microscopic autoradiography is one of the tested methods
that can
provide valuable information. For instance, tritium-labeled vitamin D
routine radioassay and whole body autoradiography identified about 10
tissues
that concentrated radioactivity ?brain was negative?, while receptor
microscopic autoradiography detected and characterized about 60 sites of
deposition as target tissues with maps provided for brain and spinal
cord
target neuron circuits. Details on the receptor micoautoradiography are
provided in: Drug Localization in Tissues and Cells
(http://www.unc.edu./~stumpfwe;
also: www.unc.edu/~stumpfwe/bio).
Walter E. Stumpf, Chapel Hill, NC
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)