Back to the Top
I have been struggling for some time with a compound that shows very
low cMax values at 5' after IV injection. I actually wrote in some
time ago (see http://www.boomer.org/pkin/PK04/PK2004356.html). I have
tried some of the suggestions I received (looked in tissues, looked
earlier, looked at partitioning into RBC), but none of these explain my
low value. I had ignored the possibility of precipitation because I
was dealing with a water soluble salt. I'm wondering now if I should
reconsider. I also should report that I see a very acute (within
seconds) reaction in the mice I dose IV - trembling, skin flushes (I'm
using nude mice), and lethargy for up to 20-30' post-injection. The
mice then recover with no obvious long-term effects, even after
repeated dosing. Could this be do to some effect of complement
fixation of a precipitate? I don't see anything at the site of
injection.
Any suggestions as to how I can follow up on this idea (if it is a
possibility) would be greatly appreciated. Also are there any
toxicology sites where I could post this question as well?
Thanks,
Noelle
Back to the Top
The following message was posted to: PharmPK
Noelle
It very much depends on the following:
What is the salt? and consequently, what is the native pH of the
compound "formulated" in H2O? If it is low e.g. an HCl, you could
expect precipitation to occur from bases at physicological pH,
particularly at higher dose concs. In my experience, if people manage
to obtain a stable salt, they tend to push the dosing envelope a bit
in the belief that the compound will stay in solution ;)!
The transient ataxia you describe with these animals does imply a
degree of test item precipitation as does the lethargy. It doesn't
explain your low 5' data however. Have you lost any animals
immediately post-dose? This usually means blockage of the pulmonary
vasculature through cpd precipitation.
What is this compounds stability profile (t0.5 and Cint) on murine
liver microsomes? It may have a v. high first pass metabolism.
There is also assay considerations. Is the compound highly plasma
protein bound and does the assay extraction recover all (or most) of
this? Does the assay ask for thie preparation of stds/QCs in murine
plasma also? I'm sure it will but one must ask!
PS : Sorry but I can't remember where I saw it but I do remember
something about differences in ADMET between nudes (xenograft?) and
CD-1 mice. Does anyone else know anymore?
Hope this helps.
Iain
Back to the Top
The following message was posted to: PharmPK
Hi Noelle,
I have a few questions :
what is the solubility profile of your drug ?
which kind of salt do you use ?
which kind of formulation do you use to dose mice ?
Thanks for your reply.
Regards,
Frederic
Back to the Top
Thanks for the rapid feedback. In reply to some of the questions I
have received...
The compound is formulated as an HCL salt.
The halflife in murine hepatocytes in vitro is just over 7 hours
It is 90% bound to plasma protein.
My extraction procedure (acetonitrile precipitation followed by SPE)
allows for recovery of 85-90% of compound.
My assay does use a standard curve in murine plasma, but I don't yet
have QCs built into the system - I'm a bit of a novice at all of
this, so if someone can suggest how to establish QC samples and what
exactly they control for, I would appreciate it.
At higher doses (15 mg/kg), I have lost animals immediately (within
1-2 minutes). I do not have histology back on these animals. My PK
data have been obtained with doses of 5mg/kg which does not cause
death, only tremors, flushing, and lethargy.
I'm not entirely sure how to do a solubility profile of the compound
- I will look into it, but if someone has a protocol of how this is
normally determined. I would be grateful. I know that its maximum
solubility in water is around 10 mg/ml.
The compound is dosed in mice in either PBS or 5% dextrose. The same
phenotype is observed with either diluent.
On a related note - would you expect nonlinear PK if you were
observing precipitation, especially at higher doses?
Noelle
Back to the Top
The following message was posted to: PharmPK
Hi Noelle,
I addition to the other feedback that you received I'd wonder about the
pH of your injection solution. Is your injection solution (brightly)
coloured and, if so, do you see changes of/in colour at the injection
site (e.g., the tail, if that's where you inject)? Is there any sign of
a reaction at the injection site? What's your injection volume and
where do you inject (tail vein)? Do you see any other products (e.g.,
metabolites) besides the parent compound? What's your detection method?
Is your compound a base (formulated as an HCl salt) and, if so, what
are its pKa and logP?
Regards,
Frederik Pruijn
--
Frederik B. Pruijn PhD MSc (Senior Research Fellow)
Experimental Oncology Group
Auckland Cancer Society Research Centre
Faculty of Medical and Health Sciences
The University of Auckland
Private Bag 92019
Auckland
New Zealand
Phone: +64-9-3737 599 x86939 or x86090
Fax: +64-9-3737 571
E-mail: f.pruijn.-a-.auckland.ac.nz
Back to the Top
Hi Noelle,
Quality control samples are very important to check precision and
accuracy of the method. For the quality control samples a minimum of
three cocentration in the calibration standard range is recommended.
One at 2 to 3 times of loq one any where in the middle and one
towards the higher side may be 80 to 85 % of uloq. If these quality
control samples will come within +/-15 % in precision and accuracy
then only you can conclude that for your samples you are getting
right back calculated values, at least you would know that the values
are within +/-15 %.
Arpana
B. A. Research
India
Back to the Top
Hi Noell
The idea of placing QC samples is to keep the Assay under control .
QC samples determines the accuracy, precision and reliability of the
unknown extracted samples which are predicted from the standard curve
(CC). The predicted concentrations from the CC is then compared to
the theoretical concentrations as % Bias and as % relative standard
deviation (RSD). these terms are very standard terms that could be
found any any of the bioanalytical journals. When an assay method is
developed the first thing one should do is to validate the assay. To
do this the analyst should analyze Quality control (QC) samples whose
concentrations are known. The best way to create them would be to
spike them at three concentrations of the calibration curve that
would be low, medium and high concentrations in replicates (at least
2 but better to have 3 in place). The lowest QC should be just above
the lowest limit of quantification (LLOQ). For example if the
calibration range is from 10 to 500 ng/mL, the LLOQ QC could be
placed around 12.5 ng/mL or as per the convenience of the spiking
procedure. The QC samples should be spiked independent of the spiking
done for the calibration curve samples. The validation should be done
on three different days with calibration curve in the beginning
followed by QC samples then end by placing another set of calibration
curve (CC) at the end of the run. The second CC is more often used if
the assay method is developed using an LC-MS/MS. The choice of
placing the 2nd CC is to keep the assay more under vigilance.
To add to the in-vivo precipitaion problem of the salt, have you
tried administering the lowest dose of the compound to the animals
that could be detected by the assay? How did the animals react to
that? Just curious to know?
Hope this would help you.
Good Luck
Manish Issar, Ph.D
Manager Pharmacokinetics
Eon labs Inc., 4700 Eon Drive
Wilson, NC-27614
m.issar.aaa.eonlabs.com , (252)-234-2340
Back to the Top
Frederik,
The pH of my injection solution (by pH paper) is close to 7. The
solution is not brightly colored, and no I do not see any changes in
color or reaction at the injection site. I inject in the tail vein
in 200 microliters. We have had an outside firm do the metabolic
stability and metabolite ID (in murine hepatocytes), so I do not
routinely look for the metabolites in blood samples. I guess it is
possible that there is an non-hepatic route of metabolism that is
much more active than hepatocytes. What do you think? My detection
method is LC/MS (single quad - I look only for parent ion species).
I think it is possible to do some fragmentation with our instrument -
we just haven't had any luck yet; therefore, the parent ion is all I
can detect. Yes, the compound is a base and is formulated as an HCL
salt.
I don't know the pKa and logP. I would love to figure out how to
determine these parameters. I have looked some in the literature but
have not worked out the best way. If you have suggestions, I would
be grateful.
Thanks
Noelle
Back to the Top
The following message was posted to: PharmPK
Hi Noelle,
in a previous post you mentioned that the half-life "in murine
hepatocytes in vitro is just over 7 hours". Although it seems that your
compound is quite stable in this model this result is very much
dependent on the experimental conditions (e.g., cell density, cell
viability, drug concentration, etc.). Where these studies done at
realistic drug concentrations? It seems to me that a large first-pass
effect is unlikely to explain the low plasma concentrations that you
observed (although you may want to try earlier sampling times) but as
you suggested yourself there are tissues and sites other than liver
that could be responsible for rapid metabolism of your compound. NB you
cannot rule out liver metabolism completely based on your hepatocyte
studies. However, I'd really encourage you to do a full scan on the
LC-MS on one or more plasma samples in order to look for
products/metabolites. You can easily do this without having to change
any of the settings (e.g., fragmentor voltage) but you'll need to a
plasma blank (plasma control) to be able to do a comparison. I think
you could be missing so much (information) if you only look for the
parent molecular ion. Have you ever looked for other possible
pseudo-molecular ions? For example, if you are looking for [M+H]+ but
the dominant ion is [M+NH4]+ you will never get a good picture of
what's going on. However, you did mention that you do calibration
curves in mouse plasma so this shouldn't be a problem (but you must run
QCs each time). To do fragmentation on a single quad should be really
easy and straightforward: bypass the column and set up a Flow Injection
Analysis (FIA) Series where you change the fragmentor voltage (or
whatever it is called on your machine) incrementally and you should get
some sort of bell-shape curve when plotting ion intensity (abundance)
of the parent molecular ion against fragmentor voltage. You are using
Electrospray Ionisation (ESI) in positive mode, aren't you?
There are some resources available on the Web that you can use to
calculate logP of your compound:
http://www.syrres.com/esc/est_kowdemo.htm, which also lists other logP
calculators on the Internet (bottom of page). I am not aware of similar
sites for calculation of pKa but by comparing with similar structures
with known pKa you should be able to make a reasonable guess.
Alternatively, ask a friendly chemist for assistance. If you are doing
LC-MS, metabolism studies, and PK, I think it serves to know at least
some of the physicochemical properties of your compound (before you
start!). This will definitely become more important if/when you are
going to determine solubility profiles.
HTH
Frederik Pruijn
--
Frederik B. Pruijn PhD MSc (Senior Research Fellow)
Experimental Oncology Group
Auckland Cancer Society Research Centre
Faculty of Medical and Health Sciences
The University of Auckland
Private Bag 92019
Auckland
New Zealand
Phone: +64-9-3737 599 x86939 or x86090
Fax: +64-9-3737 571
E-mail: f.pruijn.at.auckland.ac.nz
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)