solution stability and method validation

[ydna6969]

Hi ,

How would you carry out validation procedures following ICH guidelines if solution stability was poor (<12 hrs) and run time was 30-45 mins?
As in accuracy(usually 9 not including sst injections) linearity (50-150% in duplicate plus sst injections) robustness testing
Due to poor solution stability only a number of injections can be performed, so analyses and validation is tricky.

How is it possible to validate such a method due to the injections involved?
Thanks

[HPLCaddict]

Dumb question: Why do you want to validate that method? Just for formal reasons (i.e. for authorities, auditors and the like)? Or to make sure it runs smoothly in routine work? This may sound silly or provoking, but it's a honest question and important to answer.

If you want to validate that method for routine work, then I'd reflect a bit over the question if such a combination of solution instability and long run time ever can be valid in routine work. With "valid" not meaning passing linearity, accuracy and the like in an artificial setting like a formal method validation, but producing valid data in a robust and precise way.

[KM-USA]

I think it's time to shorten the HPLC runs, then do your validation.

[ydna6969]

Hi

I agree with your points ,, but no way around parameters given to us.
Is the method valid for routine work ,,,interesting point.

But no scope to change.. so how best to validate method ?

[Peter Apps]

Is there anything that stops you making up each vial just before it needs to be injected ?

Peter

[ydna6969]

Hi

Just time restraints really.

Typical system suitability before samples would be approx. 10 so that is 5 hours gone

Think this maybe our only option

[HPLCaddict]

I'd too separate system repeatability and actual analyses into separate runs. As long as you don't change anything to the system (fresh batch of mobile phase for example) and the time gap between SST and analyses is adequately short, this should not pose a problem.

[Peter Apps]

Hi

Just time restraints really.

Typical system suitability before samples would be approx. 10 so that is 5 hours gone

Think this maybe our only option

It's only five hours gone if you sit around doing nothing while the HPLC runs each sample. Make like a fancy sample prep robot and start extracting and diluting etc just long enough before the injection is due to have the sample / std ready in time. Inject and then do something else until you need to start on the next one.

Believe it or not this was how we worked before everything was automated.

Peter

[gtma]

Believe it or not, I've had an experience validating an impurities method (at least 6 known impurities) with much longer run time (60 mins) and limited sample solution stability. The method used an external std approach which makes it even more challenging! It took a lot of planning and long hrs in the lab, esp for the chrom robustness. Would have been nice to have a UPLC method. In the end, the method was validated and good for its intended use. Soon after, I suffered from PTSD....crazy!