residual solvent validation in drug substance

[VHB]

Hi,

In the development study, the robustness was not studied. So if we propose robustness in an additional validation study, can we do ONLY 3 injections of the drug substance sample solution (spiked with the residual solvent at 100% specification level such as a repeatability sample solution) and calculate mean and RSD? Is it acceptable to inject 3 injections at each varying condition after confirming the system suitability requirements (6 injections with the std)?

[chromatographer1]

Tell the forum how you propose to perform your 'robustness' test and what information it will provide to the government regulator approving your method.

What you have discussed here so far does not match my experience in method validation with a major pharmaceutical company. Of course, it is certainly possible that what I learned concerning the requirements of 'robustness' is not that which is required today.

Please share more of your validation plan concerning this aspect of the analytical method if you may.

Rodney George
consultant

[DR]

I would be inclined to expect an intermediate precision (robustness) triplicate injections of triplicate final preparations (1 set from each of 3 different person/GC combinations) of solvent spiked samples at limit, at LOQ and at LOD (when I say "at", I mean pretty close to...).

[chromatographer1]

DR has given a solid proposal. I would consider it a bare bones minimum, but a good minimum. I would suggest a bit more rigor might impress the regulator more favorably. Realistically, why do all this work and then, to save a few hours of effort, ruin the approval of months of effort?

While upper level supervision may dislike wasteful efforts, "a penny saved is a dollar wasted" is not an unknown expression when it comes to method validation.

Do it right the first time and you can save a fortune in NOT doing it a second time.

But that is my opinion.

Rodney George

[krickos]

Agreed

As residual solvents are mentioned I assume it is pharmaceutical ICHQ2R1 guideline might help you here:
In the case of gas-chromatography, examples of typical variations are:
- different columns (different lots and/or suppliers);
- temperature;
- flow rate.

Test such as intermediate precision mentioned above and development usually provide "additional information" to robustness.

In essence, an additional RSD/precision study only seems not good enough in this case.