Newsletter December 2013: When regulatory guidelines cause trouble

Can you trust your supplier’s methods?

Setting the “right” specifications from the beginning is of immense importance – especially when it comes to particle sizes:  Different instruments and analytical methods usually leads to different results – and further, small changes in production might lead to large changes in particle size. Thus, if you base your particle size specification on the “first round” of experiments performed by, for instance, your API supplier’s results during development, you might run into big problems later on.


The method they use for determining particle sizes might have been validated internally: This validation is based on their production experience so far, their instrumentation, their special sample preparation etc. Though the validation is performed according to all current GMP rules one should always consider the risk if you change to a new supplier: They might not be able to obtain the same results – even though the observed change in particle size and morphology does not influence either processing or bioavailability, i.e. changes that have no effect on product properties. It is quite annoying having to discard a batch from a supplier just because the initial tests and acceptance criteria were not evaluated sufficiently. You might end up with an unnecessarily tight design space.



It is important to remember that only one (and the same!!) method should be applied at supplier(s) and internally. Thus, before accepting the suppliers methods for particle size determination it might be a good idea to perform a verification of the method using an independent laboratory in order to highlight potential risks.

Contact us if you want to hear more about the possibilities for method verification.


Outdated method – but it cannot be changed!

What do I do when I discover that my analytical equipment is no longer the best suited for the purpose? The most obvious answer is of course to change to the better equipment – but it is not as easy as it seems.


An example: You use Malvern Mastersizer 2000 for determining particle sizes. You know that the particle sizes are close to the limit of the instrument performance, so you are happy to learn that a new instrument (Mastersizer 3000) with a wider measuring range is on the market. From a scientific point of view, changing equipment is the most reasonable thing to do, as you want to make sure that you have your particles under control. However, as you are now able to determine the sizes “more accurate” –knowing that a different instrument will give different results – you might have to adjust you specifications.

Here the trouble starts! In order to change specifications you have to ask for permission in all the markets where the drug is approved – and it is not hard to imagine that it is not easily done: The individual regulatory authorities might have a quite differing point of view on this issue and they might not all agree. So how do you solve this?

Maybe you were hoping to read the answer to this tricky question here – but unfortunately we do not know. However, we do have a few suggestions – but are also very interested in knowing how you cope with this issue. Please join the discussion on our linkedin profile to let us hear your experiences.

Not needed -but still required?

Is identification of polymorphic crystal form in the final product necessary? In the ICH guideline it is stated that it is “very difficult” to identify the polymorphic form in the final product – which means that you do not have to show. Still, some countries are beginning to require documentation for the crystalline purity in the final product.


If you use the stable polymorphic crystal form in the final product it is -from a scientific point of view- easier to argue that this form will not transform into a more unstable form – as this is how the law of thermodynamics work. (read more about crystalline stability)

In other cases, a metastable form is used in the final product because – either it has favorable properties in relation to the stable form – or because of patent issues – and in these cases the burden of proof that it has not transformed into the more stable form has increased, as the risk that it will happen is somewhat larger. Thus, a thorough risk assessment is required in order to argue that crystalline purity check is not necessary.

What is your approach to this issue? Do you check polymorphic identity in final product?