Back to the Top
Dear Group,
=20
This is to know from the bioanalytical & QA experts/regulatory auditors =
about the
acceptance of a stability batch in the below explained situation:
=20
For one of the stability experiments during bioanalytical method =
validation, the ULOQ
sample didn't show up any response for both the analyte and ISTD due to =
which it could not
be quantified. However, even without this standard point in the =
calibration curve, the
stability and comparision QCs are within acceptance limit for Precision, =
%Nominal and
%Change. So, can the stability batch/data be accepted? I am looking for =
an answer more
from a quality and regulatory than from a scientific perspective.
=20
Thanks for your comments in advance.
=20
Neel K Mohan=
Back to the Top
Dear Mr. Neel Kamal Mohan,
=20
If your SOP suggests that calibration curve must consist of LLOQ and =
ULOQ points
then you can not consider this batch.
I'm not sure about your SOP, but typically in bioanalytical method =
validation,
HQC should be around 80 to 85% of your ULOQ value. If you are not =
considering
your ULOQ then this point is failing to meet.
=20
If your SOP does not explain all these, you can consider this batch with =
proper
justification. The main aim of stability batch is to show the stability =
of QCs
and I'm sure that you have chosen QCs below ULOQ level. Don't forget to =
explain
the reason for not considering the ULOQ.
=20
Regards,
Sudipta Basu
Principal Scientist
Sai Life Sciences, Pune
Back to the Top
Dear Neel,=20
Does your batch has any other standard that is above the stability QC =
measured
in this batch? If yes then you can accept the batch if not then your =
high
stability QC is outside the calibration range and batch should be =
rejected. =20
Muhammed Ayub=20=
Back to the Top
Yes, the stability QC is bracketed by the second last calibration standard.
There are two standards above the HQC sample.
Regards
Back to the Top
Dear Neel,=20
Having standards above the high QC makes it possible to accept the =
batch. You
have at least two options:=20
a) If SOP allows the failure of a certain percentage of the standards =
and you
meet that requirement, then provide justification with respect to what =
was the
probable cause of the failure and also state that all reported results =
were
within the effective calibration range implying that the results to be =
accepted
are not compromised.=20
b) If such a failure is not allowed in your SOP, and you still want to =
accept
the results then a deviation report will be required where you will need =
to
provide the description of the event, plausible cause of why that could =
have
happened and why should you still accept the results (provide the =
justification
as in "a"). In this case I would also suggest revision of the relevant =
SOP to
insert this flexibility in your procedures.=20
=20
Muhammed Ayub=
Back to the Top
The following message was posted to: PharmPK
Dear Neel,
Check slope of Calibration Curve where ULOQ CS is removed. If slope
value is close to other batches, It will justify if you accept batch
wo ULOQ and HQCs are covered with CS.
Regards,Ravi
Back to the Top
Hi Neel,
We came across the similar situation which is in a study sample analysis and QA
accepted the batch, However we repeated all the samples which are between STD-H
and STD-G (ULOQ) conc.
=46rom a QA perspective in the below mentioned situation it is not accepted we
have to repeat the experiment. Stability is a part of validation.
Nagesh
Back to the Top
The following message was posted to: PharmPK
The Uloq and lloq are quite different from the other stds. Their =
performance
has been tested during accuracy and precision runs unlike the other =
curve
points. They are also the limits beyond which samples must be either =
diluted
or assigned < LLOQ. As you diminish the value of the curve by lopping =
off
failed points, you create untold difficulty by now requiring to re =
analyze
samples which fall below the false lloq or above the equally false Uloq. =
If
either of these specific points fail repeatedly it is an indication that =
your
curve range selection was arrived at in error and no amount of pruning =
will
help. Bite the bullet and fix the assay. The quick fix will return to =
bite you
one way or another.=
|
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 "Exclusion of ULOQ" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)