USP <621> General chapter

[Melissa PV]

Good day to everyone!
My name is Melissa and I'm writing from Italy.
I am a new entry and I wish I could give my support to anyone who needs it.

But, at the moment, I need a small support: in USP <621>, in the section Sistem suitability, it is said that "Unless otherwise specified in the individual monograph, data from 5 replicated injections are used to calculate the RSD, if the requirement is <=2.0%; data from 6 replicated injections are used if the RSD requirements is >2.0%.

My question is: where does the requirement of 2.0% come from?

Going on reading, it is said that for the assay of a drug substance monograph, if not specified, for RSD rerquirement there is a specific table (equal to that of Eur Ph 2.2.46)... but I cannot find the requirement of 2.0%...

And which is the RSD requirement for the assay of a drug substance without monograph?

I don't understand...

Thanks to everyone who will answer to my question!

Melissa

[tom jupille]

And which is the RSD requirement for the assay of a drug substance without monograph?

There is no single answer. In general, the reproducibility of a measurement must be significantly greater than the allowed variability of the product or process being measured. Look at it this way: if you are trying to control the API in your formulation to +/- 5% RSD, and your measurement has itself an RSD of 5%, then your measurement is useless.

There is an entire subfield of statistics called "Gauge Theory" that deals with this, but to a very crude approximation, the RSD of the measurement should be no more than about 1/3 the allowed RSD of the product.