CMC in Preclinical Development – is Polymorph Screening Appropriate?

I was fortunate to be invited to participate in a recent MPI Research sponsored event at the WBBA facility in Seattle, Washington. Titled “Chemical Manufacturing Controls: What’s Driving the Entire Drug Development Process?” it addressed those scientific and regulatory aspects of drug characterization needed at the pre-clinical stage of development.

SpaceneedleThe meeting was excellently hosted by WBBA – many thanks to Chris Rivera and his colleagues at their impressive facility. This and the tasty breakfast (it was a 7.30am start) undoubtedly contributed to the very solid turnout of about 60 attendees.

Steve Frantz of MPI Research did a great job of organizing the meeting, the speakers being experts in their fields and the topics very relevant to the subject at hand:

  • Steve Frantz of MPI Research who gave an overview and introduction
  • Peter Korytko, a preclinical safety consultant provided a consultant’s view of CMC considerations
  • Gustavo Mahler of CMC Biologics spoke about the clinical and commercial bioprocess development of biologics and
  • Jeff Gautschi from Agere Pharmaceuticals described small molecule issues and challenges

Following these presentations there was a panel discussion responding to questions from the attendees. My presence on the discussion panel was based on Velesco’s experience in formulating compounds for pre-clinical studies, frequently a very different exercise to formulating for clinical trials.

This being Seattle many questions focused on the large molecules and especially vaccines, however there were a number of questions on small molecules and their formulation. Early in the panel discussion the subject of polymorph choice was raised. This is a perennial concern to companies developing small molecules.

All panelists agreed that the polymorph used in the preclinical studies should be used in the clinic and with modern technologies such as pXRD this can easily be determined. The discussion then moved on (as it usually does) to question the need for polymorph screening in very early stage development.

We get asked about this a lot. So, my answer to the oft asked question “should we do a polymorph screen early on?” is an unequivocal “well, it depends”

And what it depends on is how far your company intends to develop the drug.

Large pharma companies have a big fear of discovering problems later in development so they focus a lot of attention on screening promising molecules early on. To them discovering that a compound has a large number of possible polymorphs should occur whilst they have time to figure out how to control the situation i.e. pre-IND as they do not want a big surprise with a previously unknown polymorph making an appearance halfway through their Phase 3 program.

In contrast, a polymorph screen does not help a small company whose aim is to get some Phase 1 clinical data and sell the molecule. Whatever data is gathered in a polymorph screen it will not speed the progress to a Phase 1 study and if they discover a lot of polymorphs it will greatly devalue their compound in the eyes of potential purchasers.

As most of our clients are small companies I always advise against doing a polymorph screen early on – the chance of getting data that can hurt the client far outweighs any theoretical gains that may (but probably do not) exist.


  1. Regarding your opinion on the need for polymorph screening in early development I see your point – but I do not fully agree: Even in Phase I, you have a potentially high risk of polymorphic transitions – a transition that might for instance lead to crystallisation/precipitation – either in the formulation or in vivo. In this case you might end up misinterpreting the clinical results. What should be discussed is the EXTENT of the screening at this time point: Maybe you can do with a slim screening where you examine the most commonly “risks” that the compound might experience (temperature,different solvents, different concentration gradients, pressure). Such a screening does not have to be very expensive and can be done in a few weeks – and it gives you an “assurance” that your compound it sufficiently stable with regard to crystalline form.

Leave a Reply