Back to the Top
Dear all
I was wondering if some one could consider helping in this issue:
If we prepared and validated a method with a calibration curve for a
standard bio-analysis and it turned out during the actual routine
analysis we should go to a lower limit of quantitation. The normal
procedure is to do cross validation. We have prepared long term
stability samples based on the old range and of course for the new
range the same was done. Due to deadlines we are submit this data but
the problem with the long term stability is that it has to take some
time before being analyzed and reported. Is it valid from a
regulatory point of view whither FDA or European to submit the data
for the long term QC samples that were based on the old range?
Raja' Sammour
Back to the Top
The following message was posted to: PharmPK
You can submit whatever you want to.
When there are concentration dependent chemical stability issues it's
usually at the high end. Example - Ampicillin, if it's frozen at -20 oC
it forms pockets of highly concentrated solution that speed up
degradation. The higher the concentration the more freezing point
depression and the greater the problem.
Physical stability such as adsorption are more significant at the low
end.
Back to the Top
Hello Raja,
Yes, you can submit data to the FDA even if you don't have presently
your long term stability data. However, upon review of your data/
dossier the FDA will ask for the long term stability data.
hopefully, by that time you will have carried the long term stability
data at the low end.
My other concern (this is my personal opinion and experience with the
FDA) is that with your cross validation; did you performed the freeze/
thaw, recovery, etc. (i.e. all the stabilities required in the
validation) with a lower concentration? Otherwise, you may have to
prove that your validation is as robust with a lower LOQ than you
previous validation.
I hope I made my explanation/concerns are clear enough.
Good luck and best regards,
Sylvain Mandeville, Ph.D.
LAB Research
sylvain.mandeville.at.labinc.hu
Back to the Top
Raja',
Regarding Long term stability submission,
First of all, if you now have a validated improved method, you must
use it.
Secondly, if you wish to quantify impurities for a particular time
point, you may do it at any time. That is, if you are willing to
attribute the impurity level you determine today to the time point
that happen in the past, you can do so.
Note: this approach has a negative side. I know a company that
submitted an impurity profile that was attributed to the end of a
clinical trial, by analyzing reserve samples that were years old. The
impurity levels were high. The FDA accepted the data, and said that
the impurity levels were too high for the product to be approved. As
a consequence, the company had to re-manufacture material according
to the batch records of the clinical trial, put that material on
stability for the appropriate real time of that clinical trial, and
then analyze it using the validated method. Only then was the product
approved.
Regards,
Frank Bales, Ph.D.
frankbales.at.msn.com
Back to the Top
The following message was posted to: PharmPK
Dear Frank,
What is considered acceptable level of impurities in particular for a
protein based product?
Best regards,
Avi Gordon
Regulatory and Clinical Affairs Manager
BiondVax Pharmaceuticals Ltd.
Science Park, Building No. 14,
Rehovot 76326, Israel
Telephone: +972-8-9302529
Mobile: +972-54-8058155
gordon.at.biondvax.com
www.BiondVax.com
Back to the Top
The following message was posted to: PharmPK
The acceptable level of impurities depends upon what the impurities
are, and what the exposure to the patient will be (amount and
frequency of dosing).
It is possible to fall-back on calling the impurity an unknown, but
then the regulators are likely force assuming it is a worst-case
compound. So try hard identify the impurities. This is why one goes
thru all the work of Extractables/Leachables for container/closures
and stress studies for API's.
To be more specific: are the impurities from the processing, or from
degradation of the API, or from the container/closure, or from
excipients?
Once the indentity, (or at least the class), of the impurity is
known, and the exposure has been determined; then there are general
regulatory rules, and more importantly, specific published toxocology
experience, that can be used to determine acceptable levels.
Hope this helps,
Frank
Frank Bales, Ph.D.
3548 Aycock Court
frankbales.-at-.msn.com
Back to the Top
The following message was posted to: PharmPK
Dear Colleagues and Frank,
Say you have a simple isocratic method [Methanol:
water or Acetonitrile: water (30:70)] for
determination of your API. When you are looking for
related substances/impurities, how hard you need to
look? i.e., a gradient ranging from 5% to 95% organic
or 25% to 75%, etc? do you need to justify your method
and based on what?
Rostam
Back to the Top
The following message was posted to: PharmPK
Rostam,
From Forced Degradation/Stress studies you should know where the
degradants come out. You should also know where your process
impurities come out, because they should be available to test.
The Foced Degradation/Stress studies are done using UV, about 40 C
heat, 10% Hydrogen Peroxide, and moderate changes in pH up and down,
until the API peak drops by about 10% to 15% (i.e. it is 85% to 90%
of what it was at the start) for at least one of these conditions.
Many degradants of non-polar materials are more polar, and so come
out early, perhaps in the void volume. To check for this, and
potential co-elutions, mass balance is required of the Forced
Degradation/Stress studies. The sum of the areas of all of the peaks
found should add to about 98% to 102% of the area of the API to
start, (when the API area has dropped by 10% to 15%).
For an isocratic method, late eluters usually show up as "ghost"
peaks on the next run, and can be noticed there as being very broad,
compared to what would be expected at that indicated RT. In that
case, the run time of the method needs to lengthened. To avoid this,
early in the development, run times can be set at 10 times the RT of
the API, (this is ususally, but not always, enough to avoid "ghosting").
Gradient methods are unusual for impurity methods, because they are
hard to validate: the precise proportions of the mobile phase may be
different on different instruments at the same run time, so the RRT's
of the impurities may vary.
Regards,
Frank
Frank Bales, Ph.D.
frankbales.-a-.msn.com
Back to the Top
The following message was posted to: PharmPK
Dear Frank,
Thanks you for your reply and useful comments. Is it
fair to say that the preferred methods for detection
of degradation product after forced/stress studies are
isocratic methods not gradient? If yes, how can you be
sure your are pulling everything out. I am not an
analytical chemist but is not it true that gradient
methods are more suitable for this type of detective
work? Do we need to characterize impurities before Ph
I or this is a post Ph I activity?
Rostam
Back to the Top
Dear Frank and Rostam,
I do not agree with your statement,frank, that Gradient methods are
unusual for impurity methods. I do agree with you that they are hard
to validate as each run show variation from other.
Many of the official method of USP and BP for the degradation product
and related substances are based on the gradient flow.
Regards
Vineet Singh
Back to the Top
The following message was posted to: PharmPK
Rostam,
The idea is to start with what we know, and build on that.
First of all, if we already know that there may an especially toxic
impurity present, the method should be designed to show if it is
present.
The degree of mass balance that can be demonstrated in the initial
Force Degradation/Stress studies, gives us the initial level of
confidence we can have in the method. By identifying the peaks from
those studies, we can learn about the major degradation pathways.
When those peaks show up in a stability run, we learn which of those
the pathways are important in the formulated product.
This knowledge allows us to qualify the various impurities. Over
time, our knowledge will improve, and we will learn which impurities
are more toxic and which ones have a significantly different UV
absorption (making detection more or less difficult). Then the method
may need to be modified, based on the knowledge of the probable
degradation chemistry and detection properties. Consequently, it's
common to repeat Force Degradation/Stress studies as the method changes.
Phase I studies are for safety. Since impurities can influence the
safety profile, they are important to know. Similar logic applies to
pre-clinical tox studies.
However, even after a drug has been on the market for a long time, it
sometimes happens that something changes which results in the
presence of an objectionable impurity that was not detected by the
methods then in place. We can never be sure that "we are pulling
everything out". In those cases, we learn how the methods need to be
improved.
Vineet, Yes, I agree; there are gradient impurity methods. Because
they offer the possibility of saving time, they are especially useful
in a QC production release environment.
Regards,
Frank
Frank Bales, Ph.D.
frankbales.at.msn.com
Back to the Top
The following message was posted to: PharmPK
Dear Rostam,
The analysis using forced/degradation sample means "stability-indicating
assay', which is most popular in the stability study of API as well as
formulation. The system condition prefer to choose the gradient and the
specificity item is specially needed to validate for the method.
Best regards!
Fred Li
Roche R&D Center (China)
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)