BA/BE study - internal standard response variability [Bioanalytics]

posted by ElMaestro  – Belgium?, 2018-07-11 06:14 (874 d 17:21 ago) – Posting: # 19029
Views: 1,637

Hello Avinash,

there is no particular requirement specified, but you need to have an objectiv rule.

You can for example check if more than X percent of the samples deviate more than Y percent from the mean calibrator and QC IS response, and reinject the run when it happens. Pick X and Y wisely, as in, according to your own opinion, experience and performance ambitions/needs.

A good question is what to do -if anything- with individual values falling outside and when they are too few to merit a complete run re-injection. Again, there are no rules, but I think it is ok to repeat those individuals as well, possibly with a root cause analysis. They never show much, do they, those investigations?

Note that, unfortunately, if you take FDA's guideline verbatim then there can be no repeats until (before) a root cause has been established: "Repeat analysis is acceptable only for assignable causes (e.g., the samples are above the ULOQ, there are sample processing errors, there is an equipment failure, the chromatography is poor)", and you might wish to make a deviating IS response as an assignable cause in its own right. I have not heard anyone from FDA express an opinion on this.

You can do trending but note that it may be a bit questionable to use statistical objectivity for it (the correct one would be Spearman's rank, and note you may not just go by a p-value) and your mileage will vary if the number of samples differ between runs (and it will do when you take missings, repeats and ISR into consideration). I personally think it is better if you have a reliable to lab employee who can eyeball the data and check trend or no trend.

I could be wrong, but...

Best regards,

No, of course you do not need to audit your CRO if it was inspected in 1968 by the agency of Crabongostan.

Complete thread:

 Admin contact
21,225 posts in 4,427 threads, 1,482 registered users;
online 2 (0 registered, 2 guests [including 1 identified bots]).
Forum time: Tuesday 23:35 UTC (Europe/Vienna)

A big data-analyst is an expert
in producing misleading conclusions from huge datasets.
It is much more efficient to use a statistician,
who can do the same with small ones.    Stephen Senn

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz