Updated: Jun 26
The emphasis is on the QC - as I see, there is a difference between a good analytical method and a good QC method.
One purpose and one purpose only: be able to make a decision if the batch fails or complies the specification, having a minimal risk to make first or second order error.
Naturally, like all other routine departments, cost angle is also to be considered.
So how can we think about the ideal QC-method to develop?
As robust as possible
'Having a minimal risk to make...' - it is always a red flag if you have to second-guess yourself, either with failing system suitability tests (SSTs), or lab-error OOSs. What you need is to be able to focus on the batch's quality. You have to make sure that by the time QC implements the method, all uncertainties are removed from the method.
Easy to carry out, easy to process
Still staying with the 'minimal risk' approach - you want to make sure that your QC analysts will be able to deliver the results in a planned manner. How can you help them? With omitting complex sample preparations, solution sequences, extremely low weighings etc. Carrying out a test should not be art, it should be a routine process. Don't make your analysts' life harder, allow them to focus on the results.
The same applies for the data processings: if it takes much time and effort to get your final data for your CoA, you lose your advantage of being able to investigate properly if you run into an undesired result.
Easy to carry out, easy to process - part 2
Automate whatever you can (and worth it) - beyond the traditional sample preparations and tests, there are so many solutions not yet implemented in pharma QC.
Let me add that automating a certain QC method step can be an engagement with your QC personnel - if you work on it together with them.
When someone develops a method from scratch (especially for GC and LC methods), that person wants to know every performance attribute - and that is totally OK. However, when transferring the developed method to QC, there should be a really rigorous decision about which SSTs should stay, and which one are to be thrown out.
When you have this kind of decision, you should consider your method's roboustness, your exposure to 'mother nature', and...yes, also maybe the direct and indirect price of having those SSTs (cost of reagents/standards, time to prepare them, etc.).
Here we need to consider some angles that don't necessarily come up during R&D phase. Let's go through them one by one.
The energy cost is far from the production's, but it still should be considered when finalizing a QC-method. Any solution that cuts down a method's consumption should be considered before finalizing the method.
Reagents to be used - sustainability
This is not fairly new, but let me emphasize to consider this - when you want to introduce a QC method to test hundreds or thousand of batches, you want to have a sustainable QC method. What does this mean?
Let me come up with an example: when the automobile industry was in crisis, there was a really fascinating consequence to the QC: because the production of acrilyc nitrile fell down, the byproduct production also got under: acetonitrile. There was a huge price increase on acetonitrile from one minute to another.
So before finalizing your QC method, think through what you need to carry out them.
Please don't take me as an inhuman person: pharma has a society responsibility of giving people jobs, also in QC. But the best approach to have engaged people in your company is when they feel appreciated and not doing stuff that could be automated. So I would urge you to go out and ask your personnel (especially your Y and Z generation people :) ) what they think about the QC-methods they perform.
These are just some angles to consider when you get the responsibility of having a QC-method to be transferred to your department - hope this helps :)
Let us maybe have a revisit this topic when we already have a not-the-best QC-method submitted to the authorities - what can we do then?