Dear TPIC,
I am looking for some input on this situation.
An assay has a very low screen CPF (<1.1), the LPC calculated based on sensitivity (<5ng/mL) is also low. During validation, LPC sometimes failed along the broadline. Can LPC be slightly increased (eg. 2x ) without re-evaluating the CPF? The sensitivity changed, but LPC is still below 100 ng/mL.
Thanks
Hi Jason,
First of all, no need to re-evaluate the CPF. That would open a big can of worms that might result in essentially re-doing most of the validation experiments, complicate reporting, and add to timelines and costs. Moreover, cut point and LPC are not really related to each other: one describes signal distribution in a population of potential study subjects, and the other is an animal antibody spiked into pooled human serum or plasma. LPC failure is typically interpreted as falling below the cut point i.e. classifying as a false negative.
Note that 2019 FDA guidance on validation of ADA assays never specifies what LPC failure is; it only says "For the low-positive QC sample, we recommend that a concentration be selected that, upon statistical analysis, would lead to the rejection of an assay run 1% of the time." You could raise the LPC concentration to a sensible level that can be used to monitor assay sensitivity and set the acceptance criteria using Mean ± 2.807×SD. Mean and SD refer to LPC results from all validation runs. You could use t-distribution if you want to be more conservative, but for a validation with a large number of runs it won't make much of a difference. Using this range of responses to accept/reject LPC would result in approximately 1% failure rate; 0.5% LPC giving responses that are too high and 0.5% being too low. Alternatively, you could define your acceptance criteria to focus on 1% of the left tail of the LPC signal distribution and use Mean – 2.576×SD. There, I just did "statistical analysis" for you 😊.
Typically, cut point and LPC concentrations are determined using less than a week-worth of validation data which tends to underestimate variability of the assay over the duration of the validation and even less so, during sample testing. Also, LPC concentration is determined using fresh spikes titered to the cut point, while LPC is used after being frozen at -80C. These and other factors I can't think of right now may lead to trouble with LPC not performing as desired. If your new LPC is at, say 10 ng/mL, it should be quite reasonable and defensible.
Best regards,
Robert
------------------------------
Robert J. Kubiak, PhD
Director, Head of Bioanalytical Science
Third Arc Bio
[email protected]Disclaimer: Opinions expressed are solely my own and do not express the views or opinions of my employer.
------------------------------
Original Message:
Sent: 10-17-2025 23:55
From: Hao Wang
Subject: Question on LPC increase
Dear TPIC,
I am looking for some input on this situation.
An assay has a very low screen CPF (<1.1), the LPC calculated based on sensitivity (<5ng/mL) is also low. During validation, LPC sometimes failed along the broadline. Can LPC be slightly increased (eg. 2x ) without re-evaluating the CPF? The sensitivity changed, but LPC is still below 100 ng/mL.
Thanks,
------------------------------
Jason (Hao) Wang, Ph.D.
Resolian Bioanaytics,
Brisbane, Australia
------------------------------