Property Improvement
Different solid forms of an API can have very different properties. Those differences can impact bioavailability, solubility, dissolution rate, first pass metabolism, side effect incidence, stability, API and drug product manufacturability, and other important parameters. The vast majority of APIs can exist in multiple solid forms, including polymorphs, hydrates, solvates, salts, cocrystals, and non-crystalline forms.
Triclinic has extensive experience in solid form screening & selection and in devising and conducting studies to improve drug properties.
- If you are searching for a developable form of your API we can determine polymorphism, the most stable form, as well as develop salts, cocrystals, and non-crystalline materials
- If your API has poor aqueous solubility, there are types of solid forms that can be found and patented that will lead to improved solubility.
- If your drug has poor stability or handling properties, Triclinic can recommend approaches to improve those properties.
- If you are a generic company or innovator looking for an improved version of a generic drug to patent and market for increased profitability and franchise extension, Triclinic can suggest strategies to produce the improved product through use of novel, patentable solid forms.
- If you need lifecycle management approaches, searching for and patenting new solid forms several years before patent expiration of a drug is a valuable technique. Triclinic can carry out studies that will result in novel forms and
lead to development of improved, patentable product extensions.
Solid form screening & selection
Target forms can be polymorphs, hydrates, solvates, salts, cocrystals, and non-crystalline forms. The current inability to predict crystallization outcomes means that solid form screening is an empirical exercise, and experience has proven that the crystallization techniques used can be critical. In all cases the form selected should have the desired properties regardless of its species (salt, cocrystal, etc).
Triclinic has a vast amount of experience in finding new solid forms, evaluating their properties, and selecting the best form for commercialization. We are familiar with state-of-the-art crystallization techniques (solvent-based, thermal, sonication, grinding, high-throughput, etc.) and the analytical techniques necessary to initially sort and fully characterize forms found. In addition, Dr. Childs has developed new high throughput solid-form screening and characterization techniques that are the most advanced in the world and have demonstrated a higher success rate at finding usable forms than any other techniques available today.
Pattern Recognition and Cluster analysis: Rapid prototyping of solid form screening data
When performing a solid state screen - whether it be for new crystalline forms (polymorphs, hydrates, solvates, salts, cocrystals) or disordered forms - there is usually a plethora of data collected on very small non-ideal samples. The use of knowledge in the form of indexed unit cells can help make form identification more robust for lower-quality XRPD screening spectra. The same is true with respect to the use of knowledge in the form of known reference patterns and known crystal structure. Polymorph predictors can also be used to add more knowledge to the new form identification process. The unique pattern matching and clustering capabilities at Triclinic Labs are discussed below. These approaches can improve downstream decision making and help suggest which form to move forward in development.
When starting a solid form screen, very often little is known about the molecular entity or the crystalline/disordered solid forms that might exist. When faced with a large data set with many potential unknown analytical signals, the optimal approach is usually to cluster the data sets into groups of ‘similar’ patterns. If the clustering method is reliable then this reduces the subsequent number of data sets that need to be considered to one reference pattern per cluster. In this way many hundreds of initial analytical data files may be reduced to a few 10’s of reference patterns. The reference patterns should be generated from the data files within each cluster rather than just selecting a single representative pattern. In this way the signal to noise level of the reference pattern can be significantly higher than the individual data sets (depending on the number of data sets in each cluster). But even with a few 10’s of patterns it can be a challenge to sort out the relationships between the clusters and which clusters represent single phase solid forms and which clusters are actually mixtures.
The first step is usually to cross compare reference patterns from different clusters to identify the obvious mixtures and obvious double clusters. At this point, an experienced solid state chemist would bring to bear all the available knowledge of the molecular entity such as known forms, polymorph predictions, and experimental conditions. Attempts could also be made to index the reference patterns to add to the knowledge base. If cluster results were available for different analytical techniques, the two result sets could be combined to hopefully reduce the final number of primary clusters.
One of the main reasons why taking the initial cluster analysis towards identification of the new solid forms can be such a challenge is that much of the information related to inter-cluster relationships developed during the pattern matching is lost when the clusters are presented as a flat file result. Even by re-slicing the dendrogram the inter-cluster relationships are essentially thrown away.
At the heart of any pattern recognition or cluster algorithm is the distance measure between the different analytical data files being compared. The distance measure is derived by comparing each analytical pattern to every other pattern in the ensemble and determines how far apart (how dissimilar) any two analytical patterns are. The distance measure can be as simple as a correlation function but most are multi component and involve complex heuristics. The most successful algorithms have been written to work much like an expert Analytical Chemist and mimic the decision tree the Expert would go through to decide whether any two analytical traces were similar. These types of distance metrics are relative and the absolute distance returned has little meaning until compared to all the other distances.
At Triclinic Labs, we use a unique distance metric which provides an absolute measure of the distance between any two analytical data sets. The distance metric runs from 0 to 1 where 0 is a perfect match and 1 is no matching characteristics at all. The advantage of an absolute distance measure is that now the individual clusters can be plotted as vertices of a multidimensional polygon. The dimension of the polygon is shrunk until the smallest number of vertices are found that allows all the absolute distances to be self-consistent. Typically, the smallest number of vertices that allow a self-consistent solution represents the number of unique single phase solid forms. Mixtures tend to fall between the vertices on edges of the polygon if they are binary or on surface or inside volume elements if they are complex mixtures. Using the absolute distance metric provides a much more robust initial clustering result and usually focuses the new form characterization to just a few clusters. This approach ultimately provides confidence in the existence of multiple forms and mixtures and their unique characteristics.
Polymorphs
In the pharmaceutical industry, regulations require that polymorph screening be carried out during API development to determine if the API can exist in polymorphic forms and, if so, whether the choice of polymorph will affect the drug's efficacy or safety. Beyond that, a search for solid forms of an API is often conducted to overcome production, formulation, or drug product problems caused by the form produced initially. Triclinic Labs' scientists have performed hundreds of polymorph screens and problem solving exercises to control solid form during API manufacture, formulation, storage, and use. We can develop a screen suited to your compound's individual requirements and stage of development. Whether you seek initial indication of polymorphism or complete screening for intellectual property use, we can customize experiments and levels of characterization to meet your goals.
Salts
Salts of APIs are often used when the API itself is not sufficiently soluble or stable, or is difficult to formulate or manufacture. In addition, different salt forms can have different bioavailability profiles or organoleptic properties.
When considering making salts of an API, one needs to select appropriate counter ions based on a variety of factors such as, but not limited to, frequency of use in approved drugs, dissociation constants, expected salt solubilities, and so forth. Triclinic scientists can plan and carry out an effective salt selection project considering those factors, the particular properties of an API, and its intended use.
Current thinking is that salts represent one end of a structural continuum, with cocrystals at the opposite end (see Childs, S. L.; Stahly, G. P.; Park, A. Molecular Pharmaceutics 2007, 4, 323-338 here for a discussion).
Cocrystals
Cocrystals (or co-crystals), which are unique crystalline structures containing multiple components, have been known since 1844. The use of cocrystals containing pharmaceutical components was reported as early as 1895. Cocrystals have unique properties and have been shown to be stable and useful in pharmaceutical development. Advances have been made recently in methods of finding cocrystals and in generating them reproducibly using standard crystallization conditions.
Pharmaceutical companies are developing cocrystalline forms of new APIs, and some companies are searching for patentable cocrystals of generic or soon-to-be-off-patent APIs in order to establish intellectual property positions. Triclinic is uniquely qualified to carry out cocrystal screening and aid in the patenting and development of those found. Dr. Childs is a globally recognized expert in the cocrystal field and carries out cutting-edge research to improve understanding of the structure and properties of cocrystals and to develop new cocrystallization methods.
Figure 1. Establishing phase relationships can lead to a more intelligent approach to solid-form screening and development as well as prediction of undiscovered solid-forms or instability. Important tools in establishing phase relationships are thermal analysis techniques such as diff erential scanning calorimetry (DSC), thermogravimetric analysis (TGA), and increasingly, temperature-dependent XRPD. In-depth knowledge of the phase relations often leads to further re finement in synthetic procedures in an iterative fashion. New phases are characterized by their melting points and their stoichiometric domains. The latter is important for the many solids that are non-stoichiometric compounds. The cell parameters obtained from XRPD are particularly helpful to characterize the homogeneity ranges of the latter.
For a complete discussion on cocrystal screening, selection, and formulation please CLICK HERE
For a discussion of cocrystals, as well as polymorph and cocrystal screening, see Dr. Stahly's article in Crystal Growth & Design 2007, 7, 1007-1026 here, and an article by Dr. Childs et al. in JACS 2004, 126(41), 13335-42 here.
Non-crystalline forms
Non-crystalline forms of compounds dissolve more rapidly than crystalline forms, and can significantly increase bioavailability of poorly-water soluble APIs. However, the use of non-crystalline materials requires confidence that crystallization will not occur during product lifetimes. In addition, organic compounds do not necessarily exist simply in 'amorphous' and 'crystalline' states. Situations are known where x-ray amorphous forms of an API have different physical properties that depend on their method of generation.
When considering the use of a non-crystalline API, it is important to know...
- The crystalline states attainable by the API
- The character of the non-crystalline form to be manufactured
- How to manufacture the selected non-crystalline form consistently
- The physical stability of the non-crystalline form
The staff at Triclinic have experience in the procedures necessary to evaluate non-crystalline materials. For a material that has never been obtained in a crystalline form, attempts should be made to crystallize it. Characterization of a non-crystalline form should include analyses that can be used to differentiate it from other non-crystalline forms and guide development of a consistent preparation procedure. If a non-crystalline form exhibits poor physical stability (readily crystallizes), there are methods that can be used to render it more stable, such as the use of stabilizers or molecular dispersions.
API-Excipient Compatibility Screening
Understanding the interactions between the active and the excipients in drug products is crucial. Excipients may affect the dissolution, decomposition, hydroscopicity, mechanism of action, and other properties of the API. In some cases excipients may even cause a change in the polymorphic form of the active resulting in complete alteration of the desired and selected properties. Triclinic Labs Compatibility screens can determine API-excipient interaction, undesired excipient-excipient interactions, and identify actual and potential drug product failure issues.
