Back to the Top
The following message was posted to: PharmPK
Dear all,
I am new to LCMS quantification analysis, as I am trying to quantify
the drugs in rat plasma using MRM, I face the follwing problem:
When i added the same amount Internal Standard (Carbamazepine) to the
spiked plasma sample (for calibration curve) and the real sample
(collected from the rats which were administered the drug), the
response (peak area) of IS in spiked sample was around 10 times higher
than that in real sample. I used liquid-liquid extraction to do sample
preparation. Although i did very well calibration curve, i can't use
it to calculate the unknown sample. If i use it, the results will be
very unreasonable.
Now i am wondering how come like this? Should i change the IS or
change the extraction method?
Any respons will be welcom.
Regards,
Dr Grace
Back to the Top
The following message was posted to: PharmPK
Dr. Grace,
Do you think if it is possible that in your real
samples, one endogenous component or the metabolite
has the same MRM transition as that of IS you are
using?
I would try another compound as an IS to see how it
looks like.
Xiaodong
Back to the Top
Hi Xin
There could be something in the Plasma sample that could be coeluting
in the region of the IS and unfortunately having the same mass as
that of the IS. it could be a metabolite or something else that could
be varying from animal to animal. Did you try using MS/MS mode that
seems to be more specific? if you cant monitor the daughter ion mode
then you probably have to work on the extraction procedure. How
different is the IS from the Parent being monitored chemically?
The other phenomenon that could be suspected is matrix effect. Have
you checked different lots (atleast 6-8) of plasma for matrix
effect? The presence of metabolites or decomposition products or the
presence of other components present in the formulation could lead to
this effect. Normally there is suppression of ionization but
enhancements in the signal could also occur. The unknown component(s)
could form an adduct with the IS which might be more volatile giving
a greater response.
The ways of checking this problem could be as follows :
(A) try taking the blank time zero sample from the animal, extract
without addition of the IS. reconstitute in the mobile phase having
the same concentration of the IS and inject it.
(B) inject a non extracted mobile phase containing the IS in the same
concentration. See if any differences occur in the peak areas of the IS.
if peak area of (A) is greater than that of (B) then there is
possibility of positive matrix effect. the other way is to perform a
post column infusion of the IS (neat standard solution) that will
produce a steady baseline followed by an injection of an extracted
blank matrix through the column. Compare the two signals. If there
is something coeluting in the region of the IS the baseline signal
will drift from the steady state level in that region.
Good Luck
Manish Issar, Ph.D
Manager Pharmacokinetics
Eon Labs Inc.,
4700 Eon drive
Wilson, NC-27893, USA
m.issar.-a-.eonlabs.com
(252)-234-2340
Back to the Top
Hello Dr. Grace,
I hope your source of plasma matrix for spiking of Calibration
standard is also rat. You might have encountered an unique problem of
matrix effect generally encountered in LC-MS/MS analysis. If this
happens, it can either cause 'ion supression' or 'ion enhancement'.
Accordingly, you may observe more or less peak area response than you
are actually supposed to observe in the absence of any effect of the
matrix. There are many reasons for this matrix effect like co-eluting
endogenous materials (salts, amines, fatty acids, etc). If you are
using ESI, there are more chances of encountering matrix effect than
APCI, because ESI is usually more susceptible to matrix effect. So,
changing of the ionisation mode also may help you, make sure that you
are not compromising with the response for your test compound and
comfortably determining the desired LOQ.
I hope multi-step sample clean-up may help you. Also shifting to SPE
may give you better results as the variabilities are less in SPE
compared to LLE. If you are keeping very short time, you can see the
effect on the overall data by increasing the run time say for 5-6
minutes per sample. In many instances, if you resolve the metabolite
peak from the analyte peak by playing with the LC variables it helps in
getting better consistent results. You can also get rid of the matrix
effect problem if you use stable-isotope-labeled internal standards.
I think in your case, ion supression is playing a major part. That's
why you are getting ten times higher response in the CC standards. This
clearly suggests that something is happening with the real study
samples. I recommend you to do the following exercises:
1) Have you done the recovery exercise? Do both absolute and
post-extraction recovery to see how much differnce in recovery you are
getting in both the experiments.
2) There can also be storage effect. Are you getting reproducible peak
area response between days? You can compare the response of the
standards/QC's between a freshly spiked CC/QC and a stored CC/QC for 24
hours at sub-zero temperature. Generally most of the drugs don't
degrade in 24 hours time at lower temperature. So, this will give you a
good idea whether there is really some effect of the samples on
storage.
3) Check out the storage vials. Select the same brand of plastic tubes
for processing and storing plasma samples and spiked plasma standards.
Are you using two different anticoagulants for your standards/QC's and
real samples?
4) Try to modify the LC conditions, increase the chromatographic run
time. Particularly so, if you are quantifying more than one analyte
with more than one IS. Sometimes, resolution between the drug and
metabolite helps a lot.
5) Look out for a deuterated IS if possible or otherwise try some
other IS.
6) Shift to SPE and check for the changes in response variability.
7) Do you get some other prominent transition for your test drug? if
so, try changing the MRM transition.
8) Lastly, if you are still having trouble, just give a try for APCI.
I hope this will help you in sorting out the current problem.
Regards
Neel Kamal Mohan
Pharmacokinetics Div.
Glenmark Research Center,
Mumbai-India
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)