Back to the Top
The following message was posted to: PharmPK
Dear group,
I have some quick questions re. LC/MS/MS assay validation for
pre-clinical PK/pharmacology studies in pursuit of an IND, and I hope
someone here can help me out.
1) Our pre-clinical PK/pharm. work is being done in a non-GLP
environment (tox. is being done via GLP, however) - will this be an
issue in the IND application?
2) We have detected an interfering peak at practically the same mass
(less than 1 AMU difference) in matrix of one species, and we cannot
get adequate separation between endogenous substance and the agent,
limiting our LLOQ to a conc. 10X higher than likely necessary for
adequate PK characterization. In my opinion, the assay doesn't have
sufficient selectivity per the FDA Guidance for Industry:
Bioanalytical Method Validation
(http://www.fda.gov/cder/guidance/4252fnl.htm).
a. Is it absolutely necessary to use this guidance for pre-clinical
assay development in pursuit of an IND? My opinion - it's best
practice, so use the guidance.
b. Obvious solution - improve separation or better mass
selectivity/sensitivity. Assuming that can't be done, we plan on
characterizing the interfering peak (the chemical composition of which
has been identified) in an untreated population of the species, and at
baseline for each treated subject (matrix "pre" level) and subtracting
it away from the peak area of agent. Perhaps some sort of statistical
adjustment would be needed in the validation, as this should
significantly affect the precision and accuracy. Would this
analytical approach be acceptable for pre-clinical data in an IND? My
feeling is that it would not, but I would like some sort of
verification from others.
c. Derivatize the agent using an adduction or reduction, thus
providing a different mass and chemical characteristic that may allow
better separation - issue with efficiency of this reaction could add
unacceptable variability. Would this be an acceptable analytical
approach for an IND?
d. Utilize 14C labeled agent - should provide sufficiently different
mass in AMU, can do beta radiodetection as well, and will give us the
advantage of performing ADMET as well as blood PK studies. With
limited resources (i.e., lack of radiochemistry support), we would
want to avoid this if possible until absolutely necessary. I don't
believe ADMET studies are necessary for an IND per se, correct?
Any insight or opinions would be appreciated.
Thanks,
Burgess
Back to the Top
Dear Burgess,
There is a trivial but important thing to rule out. Some time back I
faced this issue.
I noticed a interfering peak in my assay, and could not get rid of it
by any analytical method. When I investigated systematically, PP tube
in which plasma was stored was leaching out some chemical, which was
giving signal with the analyte MRM. When fresh plasma was collected I
did not have this interfering peak.
Please rule out the possibility.
Thanks,
Vinayak
Vinayak Nadiger
Manager , Bioanalytical Chemistry
11 Biopolis Way, Helios #08-05
Singapore 138667
E Mail: vnadiger.aaa.combinatorx.com
Back to the Top
The following message was posted to: PharmPK
There is a statement indicating that if the tox is being done GLP, the
analytical support for TK/PKshould also be GLP see Section 3 A
Guidelines
for Industry Toxicokinetics. Also Page 12 of guidance for IND allows
submission with draft data but does not indicate acceptance of non GLP
data.
--
Ed O'Connor, Ph.D.
Laboratory Director
Matrix BioAnalytical Laboratories
25 Science Park at Yale
New Haven, CT 06511
Web: www.matrixbioanalytical.com
Email: eoconnor.-a-.matrixbioanalytical.com
Back to the Top
The following message was posted to: PharmPK
Dear Burgess-
Regarding (1) analysis of TK samples from a GLP tox study by a
non-GLP-compliant facility, and (2) endogenous interferent in MS-based
method:
1. Analyzing samples from a GLP tox study in a non-GLP compliant
environment is definitely a problem. At best you will need to note that
fact in the protocol and study report, but the data may still not be
considered adequate by regulators. Any decisions based on the
concentration data and subsequent TK calculated parameters will be
questioned. Failure to conduct the analyses in a GLP compliant facility
risks invalidation of the TK data, and possibly all or part of the tox
study, by regulatory authorities.
2a. Unless you can justify other alternative practice, the FDA
Guidance is the norm for such work (bioanalysis of plasma samples from
GLP tox studies).
2b. A difference of 1 AMU or less means that you need chromatographic
separation, rather than mass separation (though consider other MS/MS
capabilities, different fragments, precursors, + vs -, etc). For this
reason and #1, you might consider a contract lab with GLP capability and
a broader range of instrumentation and solid experience that might
enable successful method generation. If you document your many efforts
to eliminate the interference, then you have justification for pursuing
alternative solutions, such as characterizing the interference in blanks
and subtracting the expected amount of interferent from the observed
concentration value. Prepare for lots of regulatory questions if you
choose this path.
2c. Derivatization is an acceptable alternative, and you may try to
derivatize the analyte or the interferent to enable separation. We
recently validated a GC-MS method for scyllo-inositol in human CSF,
which requires derivatization for separation from other isomeric
endogenous inositols with the exact same mass. (Tenth International
Symposium on Hyphenated Techniques in Chromatography and Hyphenated
Chromatographic Analyzers; Brugge Belgium; Jan 30-Feb 1, 2008).
Derivatization methods can potentially be very precise and accurate;
though if not, you may have reason to expand acceptance criteria a bit
from the criteria recommended in the Guidance.
2d. Unless you have an extremely potent drug dosed at the microgram or
lower amounts, for 14C-labeled drug you will still dilute it into 12C
drug to get doses, not use pure 14-C drug, so the mass detection
problems will still be unchanged. With 14C-labeled tracer in your
mainly 12C drug, you also won't know that the 14C in the sample is due
only to the drug - it could be a co-eluting metabolite. I don't see
that 14C tracer will address the problem. Besides, it would be very
challenging (and probably expensive) for a tox facility to try to run a
GLP tox study with radiolabeled drug. You are correct that in general,
definitive ADMET studies are not required for a first-in human studies,
although at least some preliminary work is the norm.
Tom
Thomas L. Tarnowski, Ph. D.
Bioanalytical Development
Elan Pharmaceuticals
800 Gateway Blvd.
South San Francisco, CA 94080
thomas.tarnowski.aaa.elan.com
Back to the Top
The following message was posted to: PharmPK
In addition to Tom's crystal-clear reply and regarding point 2d:
Why not consider stable isotopes/nuclides?
The spec. act. of most radiolabelled compounds is way too low to be
useful for mass spec. purposes anyway (and scint. counting is not
compound-specific).
IMHO 2d should be used as last resort.
Frederik Pruijn
Back to the Top
The following message was posted to: PharmPK
Burgess,
Using for your detector a triple quad mass spec that has moderately high
resolution capability might give you the specificity you need to
overcome
your LLOQ problem, (for example the Thermo Quantum). Are you doing the
analytical in-house? If not, it would be no problem to find a
contract lab
to do this for you. A day of exploratory work would let you know if
it's a
go or no-go.
Good luck,
Stuart
--
Stuart Coleman, Manager
Matrix BioAnalytical Laboratories
25 Science Park at Yale
New Haven, CT 06511
Web: www.matrixbioanalytical.com
Email: stuart.coleman.-a-.matrixbioanalytical.com
Back to the Top
The following message was posted to: PharmPK
To the list, and those who graciously responded to my inquiry:
Thanks for the input regarding this dilemma. I just wanted to address
each of the responses I had. They were all very helpful and exactly
what I was looking for.
Vinayak - interfering substances from the tube had been considered,
and this still may be the issue. We basically protein ppt the plasma
with ACN, and store that in PP tubes at -80, so perhaps the ACN is
leeching out a plasticizer or something. Will try using glass
apparatus and see if it resolves.
Ed - Your comment regarding GLP TK/PK support for GLP tox makes sense,
and is appreciated. I will refer to the guidance you provided. I do
have to stress that this data we are generating is preliminary, and
will probably not be used for an IND. When and if we do go for an
IND, we will need GLP bioanalytical TK/PK.
Thomas - Good insight. Agree that achieving chromatographic
separation is the 1st key. We are currently using an API 3200 triple
quad MS/MS. We do have access to a Thermo LTQ Orbitrap MS, which could
possibly differentiate between these subtle mass differences. We will
try that, and then possibly derivatize the analyte. 14C isn't
suitable or practical at this point.
Stuart - Will have to keep this in mind as we develop this non-GLP
method. Perhaps our method will be a good starting point for any
external lab when it comes time for the GLP bioanalytical. However, I
don't think we are ready yet to consider outsourcing.
Thanks again,
Burgess
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Non-GLP pre-clinical bioanalytical assay development for PK" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)