Sample for HPLC Method Precision Validation

[Terry]

Dear all,

According to ICH Q2(R1) as outlined below:

The precision of an analytical procedure expresses the closeness of agreement (degree of scatter) between a series of measurements obtained from multiple sampling of the same homogeneous sample under the prescribed conditions.

Precision should be investigated using homogeneous, authentic samples.

Does this mean that only one lot of sample could be used in the method precision study? Am I ICH compliant if I use 3 (three) different lots of sample, 2 (two) preparation for each lot to demonstrate the repeatability (precision)?

Regards,

Terry

[Mattias]

The method precision is best evaluated by analysing one homogeneous sample ( e.g. one lot of tablets, solution, nasal spray etc)

The sample is then taken through sample preparation and analysis:
By one person, one instrument etc = repeatability
By different persons, instruments, column lots etc = intermediate precision

If you bring in several batches of product you will also intoduce the batch variation into your precision results. You can do that if you want, since you will only make it harder for yourself.

The purpose of the precision validation is to find the variation during analysis. But again if you pass anyhow, then you have both a good product and good method

[Terry]

Dear Mattias,

Thanks for your information.

In some dinasour method validation protocol at my end, multiple lots of samples have been used in repeatbility study, we are concerned with regulatory risks. Currently, just one lot of sample is used.

Terry

[Consumer Products Guy]

We also only use one lot. The precision part is to evaluate the preciseness of the assay, and among separate analysts, not necessarily to show that different lots can be manufactured similarly.

Well - finally - a validation topic for which most will have agreement - "maybe" .

[HPLCaddict]

We also only use one lot. The precision part is to evaluate the preciseness of the assay, and among separate analysts, not necessarily to show that different lots can be manufactured similarly.

Well - finally - a validation topic for which most will have agreement - "maybe" .

Yes, maybe !
Full ACK with your comments, but one addition. Even when using only one batch of tablets, strictly speaking this is not an homogeneous sample. You'll have the variation from tablet to tablet. If the production process is fine, this variation will be small, but it's there. When doing precision studies, you'll measure the precision of your method PLUS the precision of the tablets themselves.
So sometimes, e.g. for dissolution validations, it might be wise to use a really homogeneous sample, i. e. a tablet aliqout from a trituration of tablets...

[Consumer Products Guy]

Or pulverize, mix well, take sample from that.