Back to the Top
Hi everyone!
I know there has been some discussion on LLOQ and its implications on
this forum. Still, I would greatly appreciate your view on the
following issue:
In a study, I have an assay that has been validated down to 100 pg/ml.
Therefore the formal LLOQ was set to 100 pg/ml. Drug concentration
levels of the lowest dose administered hardly go higher than 100 pg/ml
at all. Talking to the assay people, they are reasonably confident that
the assay measures rather accurately down to 70 pg/ml but during
validation it was not deemed necessary to validate to such low levels.
What do I do with the data below 100 pg/ml? Shall I set every value to
0 knowing that I introduce great inaccuracy with this?
How do regulatory authorities look at this?
Any input is tremendously welcome!
Best regards
Hanns
Dr. med. Hanns-Christian Tillmann
Medical Science Leader
Biomit Inc.
Hochstrasse 31
CH-4053 Basel
Switzerland
Phone: +41 (0)61 206 12 71
Fax: +41 (0)61 206 12 22
Email: hanns-christian.tillmann.at.biomit.com
Homepage: www.biomit.com
[0 can't be the right value. Missing value or not determined is more
appropriate. Maybe with proper(?) validation values between 70 and 100
could be reported as such - db]
Back to the Top
Typically, such values are reported as "Below the Quantification Limit"
(BQL). You should have a standard way (either SOP or stated in the
protocol) about how values designated 'BQL' will be treated in PK
calculations. Treating them as 0 is common, but other things are also
possible.
By FDA guidelines, you cannot extrapolate below the LLOQ. At the time
samples were assayed, the assay was not validated below 100 pg/mL, and
no QCs were included at the low range to assure that the assay was
functioning adequately below 100 pg/mL. You could validate the range
down to 70 pg/mL by collecting precision and accuracy data in the lower
range and then reassay samples in the lower range (with QC samples as
required by guidelines). If you choose to validate and reassay, you
should also reassay a few samples with values above 100 pg/mL as a
cross-validation to show that results from the new assay are consistent
with the previous one.
Thomas Tarnowski
Thomas L. Tarnowski, Ph. D.
Project Team Leader
Dept. Head, (Dep) Drug Metabolism and Pharmacokinetics
Roche Palo Alto
3431 Hillview Avenue, Palo Alto, CA 94304
email tom.tarnowski.-a-.roche.com
Back to the Top
Hans,
Yes, it is fine to revalidate to the lower level. If it exacly the same
assay, then it's just a matter of documenting the need and the
associated acceptance criteria to be met, and then proceeding. If the
assay is to be modified, for example, with additional standards, as is
likely, then it is still quite acceptable to revalidate to a lower
LLOQ, but the choices are more complicated because you should consider
whether you will benefit by considering this to be a new, low level
assay to be validated separately, from the beginning. To help to sort
out the alternatives, you should run a test of the precision and
accuracy at the desired LLOQ, using various standards. This test can
also be used if necessary to document the need for somewhat relaxed
acceptance criteria at the LLOQ.
Frank
Frank Bales, Ph.D.
Senior Regulatory Consultant
Worldwide Regulatory Affairs
PAREXEL Intl.
2520 Meridian Pkwy, Suite 200
Durham, NC 27713
(919) 294-5297 Phone
Email: frankbales.at.msn.com
Back to the Top
Dear Dr. Tillmann,
The way to proceed is the following:
A) Ask the lab to perform a limited validation trial to lower the LLOQ
to 70 pg/mL (or lower) this is a standard procedure and it is not much
work.
B) Providing you have enough of the samples left and suitable
Freeze/Thaw and overall stability data for the analyte, repeat the
analyses of the samples containing the analyte at concentrations below
100 pg/mL. As a precaution I would also consider re-analysing the
samples according to the standard procedure for re-analysis (i.e.
perform the analyses in duplicate) just in case you obtain strange
results.
C) Do not set those concentrations to zero if your colleagues in the
lab believe that the method was not properly validated. It is accepted
that a bioanalytical method is completely validated only once it has
been applied for the analysis of "real" samples from a clinical trial.
The Authorities see this as the best validation procedure.
As a final comment, not knowing the details of the analyte and of the
method, and since the concentrations are quite low, I would suggest to
perform experiments to exclude interference from metabolites and/or
degradation products that are present in the real samples but that were
not present in the spiked samples used for the validation experiments.
This is also welcomed by the Authorities.
I hope this helps.
Best Regards,
Stefano Persiani, PhD
Director, Drug Metabolism and Pharmacokinetics
Rotta Research Laboratorium, spa
Via Valosa di Sopra, 7-9
20052 Monza (MI) ITALY
Tel +39 039 7390396
Fax +39 039 7390371
Back to the Top
Dear Hans
I am afraid that they will ask for validation of the lowest point you
can estimate concentrations.
Gilberto De Nucci
Back to the Top
Hi!
Why not validate the system to a lower limit? If the people in the
analytical group are so confident that their LOQ is lower than the 100
pg/ml, maybe they are willing to find out how low they can go?
Toufigh Gordi
Back to the Top
Hai Hanns
As per as my knowledge your analytical method has to
detect or estimate upto atleast 3 half lifes of the
drug and the main purpose of collecting the blood
samples upto 3 half lifes in clinical phase is for
that purpose, other wise you may face problem after
regulatory submission.
If you are unable to increase the sensitivity of the
method and by seeing the safety of drug you can
increase the dose (FDA permits this process).
Best regards
tn rao
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)