Too many co-crystal programs start with a coformer list pulled from a textbook and end with a form nobody can scale or patent. The gap between ‘we found a co-crystal’ and ‘we have a developable, protectable solid form’ is where most timelines and budgets quietly disappear.
1. Lock the Target Product Profile Before Screening Anything
Before selecting a single coformer, write down what the co-crystal actually needs to fix: solubility, dissolution rate, hygroscopicity, mechanical properties, or chemical stability. This defines your screening criteria and your go/no-go thresholds from day one.
Skipping this step is the most common and most expensive mistake we see. Teams generate a library of interesting new crystal forms with no way to rank them against the original problem, because nobody defined what ‘better’ meant before the work started.
- Do: Write quantitative or semi-quantitative success criteria tied to the clinical or formulation need.
- Common mistake: Treating ‘a new co-crystal exists’ as success, instead of ‘a co-crystal that solves our specific liability exists.’
2. Build a Rational Coformer Shortlist, Not a Random One
Use the API’s hydrogen-bonding functional groups, pKa, and known supramolecular synthons to shortlist coformers with genuine complementarity to the molecule. Cross-reference GRAS status and prior regulatory precedent if the compound is intended for clinical development.
A shortlist of 15–25 structurally rational candidates, informed by crystal engineering principles, consistently outperforms a brute-force screen of 50+ coformers chosen mainly because they were on hand in the lab.
- Do: Map hydrogen-bond donors/acceptors and known synthons before selecting coformers.
- Common mistake: Running a large, unstructured coformer panel without a hypothesis for why each candidate might work.
3. Match the Screening Method to the Molecule and the Timeline
Solvent-drop grinding, slurry conversion, solution crystallization, and melt methods each reveal different parts of the co-crystal landscape. For high-potency or controlled substance APIs, method choice also has direct consequences for containment and handling, which should be planned in parallel with the chemistry.
At PCL Research & Manufacturing, screening protocols are built around the specific compound rather than applied as a fixed template, because a rigid one-size-fits-all method frequently misses the thermodynamically relevant form for a given molecule.
- Do: Use at least two orthogonal screening methods to avoid method-dependent blind spots.
- Common mistake: Relying on a single technique and declaring ‘no co-crystal found’ when the method simply wasn’t suited to the compound.
4. Confirm the Form Rigorously Before You Call It a Co-Crystal
A new XRPD pattern is a hypothesis, not a conclusion. Confirm true co-crystal formation with a combination of XRPD, DSC, TGA, and, where needed, solid-state NMR or Raman spectroscopy, and verify stoichiometry and phase purity before moving forward.
This stage is where premature reporting creates downstream risk: a form later reclassified as a physical mixture, salt, or solvate can undermine both development timelines and any IP position already built around it.
- Do: Confirm stoichiometry and phase purity with orthogonal analytical techniques before advancing a lead form.
- Common mistake: Advancing a candidate on XRPD evidence alone, without thermal and spectroscopic confirmation.
5. Build the IP Strategy in Parallel, Not After the Screen
Patent counsel should be looped in once a shortlist of confirmed forms exists, not after a single lead has already been chosen. Novel co-crystals with demonstrated functional advantages (stability, solubility, manufacturability) are patentable assets in their own right, independent of the parent API patent.
Running the IP conversation in parallel with characterization lets you prioritize forms with both technical merit and claim strength, rather than discovering late that your preferred form has weak or contested novelty.
- Do: Share full characterization data with patent counsel as soon as a form is confirmed, not just at the end of the program.
- Common mistake: Treating IP strategy as a final administrative step instead of a parallel workstream.
6. Decide What to Scale, and Set a Clear Stop Point
Not every confirmed co-crystal deserves scale-up. Apply your original target product profile criteria to rank candidates, then confirm scalability with stability screening under ICH-relevant conditions and a small, representative crystallization run.
Programs without a defined stop point tend to keep generating ‘interesting’ forms indefinitely. A disciplined screening program ends with a documented decision, not an open-ended list of candidates still awaiting a final call.
- Do: Set an explicit decision gate with defined criteria before scale-up resources are committed.
- Common mistake: Continuing to screen ‘just in case’ after a form already meets the predefined success criteria.
A well-designed co-crystal screening program is as much about discipline and sequencing as it is about chemistry — the molecules rarely fail, but the process around them often does. If your team is scoping a co-crystal program, or reassessing one that has stalled, we are happy to discuss how a structured, data-driven approach could apply to your specific molecule. Reach out to the team at PCL Research & Manufacturing to start the conversation.
