Understanding the solid state nature of an API is key, but where should this be integrated into the development pipeline? Julian Northen and Adam Moorhouse, Onyx Scientific, discuss how best to deploy screening to mitigate the risk of late stage attrition
Julian Northen in the Solid State lab
Solid form is becoming increasingly important within drug development, driven in particular by the fact that a growing number of APIs being developed fall within class II/IV of the BCS classification system, indicating that they are likely to prove problematic when it comes to pre-formulation and formulation development. It may then prove challenging to develop a formulation strategy that delivers sufficient in vivo exposure to elicit the desired therapeutic response.
This emphasis also manifests within process development (pR+D), where a thorough understanding of solid forms of an API can expedite crystallisation development/purification work throughout development. A good understanding of solid form variability and the potential for generating new versions of the API (salts or cocrystals) can bolster the IP position.
Upon lead selection, the shared goal of outsourcing managers and suppliers is to expedite the lead candidate through chemical development and ultimately to the first-in-man batch. A common phrase heard on both sides of the table these days is that said development has to be ‘fit for purpose’. This means that to mitigate the risk of clinical failure at such an early stage, any chemical development work needs to be carried out efficiently, providing chemical solutions that are scaleable and a robust (preferably the most stable thermodynamically ) solid form candidate, and leaving non-essential development work until the programme has been derisked somewhat by a successful Phase I trial.
Onyx Scientific has experience in expediting medicinal chemistry processes through to the first-in-man batch and, following the integration of a Solid State laboratory, appreciates the importance of integrating Solid State into any robust, ‘fit for purpose’ development programme. The company has seen many examples where Solid State has not been given appropriate consideration, and projects ‘muscled through’ with little understanding of form. Seeing the result of such endeavours, it would argue that this approach (although involving least expense up front), poses a much greater risk of late stage attrition, and ultimately delayed timelines and increased overall expenditure.
Solid form selection is best deployed when there is a good sense that this is the route to be taken forward, and hence an awareness of the nature of the impurities likely to be present
Solid form selection is best deployed towards the latter end of pR+D – whether the candidate is challenging or more straightforward – at a point when the first gram scale batches of the lead candidate have been produced via a route that is scaleable and optimised from the medicinal chemistry original. In other words, it may not be set in stone, but there is a good sense that this is the route to be taken forward, and hence an awareness of the nature of the impurities likely to be present.
There are other options for the integration of solid form development that may differentiate what type of screen may be applied but, particularly where a challenging candidate is suspected, the most favoured approach is to screen thoroughly slightly later in development after the first 10–20 grams have been made.
Two case examples are given that exemplify the challenges, risks and also the approaches that may be employed.
Case 1. Progression at risk:
The risks that this strategy aims to avoid are the significant chance of discovering a new form on scale-up if a project is pushed through with no consideration for solid form. For example, a customer had in hand one of the more typical BCS Type II compounds (a more challenging candidate). Previously the route the project took was three months pR+D ‘muscled through’ to get to a process that had some purification issues, and crucially no real form development. This was followed by a four-month period of scale-up through an eight-stage process that produced a half kilo of material.
Retrospective solid form analysis showed this to be a material of very low relative crystallinity and was consistent with a mixed phase product (>1 poorly defined crystalline form present). That immediately led to a three-week GMP production campaign, producing material that was taken through to formulation development.
Some of the issues relating to purification and ultimately form, which were intrinsically linked, were not really considered. This lack of consideration for solid form ultimately resulted in a 6kg clinical batch failing specification and led to poor behaviour during the initial phases of formulation development involving gross change of form and composition in terms of impurity content.
This is an example where some form development and consideration of form selection early on by means of a thorough screen would not only have aided the development process through the non-GMP and GMP stages, but would also have resulted in avoidance of a clinical batch failing spec and requiring re-work.
In this case, working through the above with chemical development would have saved approximately one month of resource during development where purity was a key issue and poor quality solvates and filtration candidates were struggled with.
Case 2. Fit for purpose mini screen:
An interesting and contrasting example demonstrates that with a little consideration for material behaviour from the start and the right mindset, a pragmatic approach to solid form development can be applied. In this case, a preliminary review of materials was completed for a customer who had engaged Onyx to perform chemical development to enable GMP production.
As part of this review solid form development was to be integrated. In this instance, there was sufficient evidence from microscopy images to suggest that the API was crystalline, even when fast-tracked direct from medicinal chemistry, which when combined with in silico modelled properties suggested a relatively well-behaved candiate.
On this basis, a pragmatic approach was suggested in the form of material characterisation and application of a mini screen, with the working hypothesis that a robust form of the material was already in hand. This hypothesis proved to be correct and with very little effort, a suitable form was identified and characterised. However, as part of this strategy, given the additional risk associated with performing only a mini-screen, solid form was closely monitored throughout development, scale-up and in particular throughout crystallisation development, such that any change of form could be rapidly addressed.
It is recognised that, while this is a highly efficient and relevant approach to form selection, in reality the number of cases where progression is so smooth, simply based on the percentage of challenging candidates entering into development, is in the minority.
A more thorough screen is required to understand fully the potential issues surrounding an API
Solid form screens, whether salt, polymorph or cocrystal, come in different sizes. Onyx Scientific’s offerings are based around a ‘mini’ screen, a ‘standard’ screen and a ‘late phase’ screen; the first two options are by far the most popular and relevant choices and are typically completed within a fixed period of 2–3 and 4–6 weeks respectively. These may also be developed into tailored programmes subsequently as studies may need to be extended if the API or salts of the API are exceptionally polymorphic.
The first option of a smaller screen is often applied earlier in chemical development, has smaller up-front costs and can give a flavour of (in terms of polymorphs) how polymorphic the compound may be, and (in terms of salt screens) whether the salts can be formed easily and whether they present enhanced solubility. These types of limited screens are often used to deselect lead candidates when presented with more than one option with similar biological properties.
However, taking a mini screen in its own right and considering that it produces sufficient form data to progress through pR+D and into the first GMP batch is inadvisable. This strategy presents an increased risk of late stage attrition and it could be argued that a poor understanding of form at the later stages will ultimately end up slowing down pR+D. It does not always, as exemplified by the second case study presented, but there is an increased risk of this being the case.
Ultimately at some point a thorough evaluation is going to be necessary. In Onyx Scientific’s experience, there are few examples where a mini screen has been applied very early on in the lifetime of the process and has been used to find one relatively stable form, and then has not seen some sort of form change when scaling up to kilo quantities and applied crystallisation development.
A more advisable approach, as demonstrated, would be the application of a mini screen where the intention is to readdress form throughout subsequent chemical development, i.e. it is integrated and form becomes tracked as a function of chemical development. In the simplest sense, form and purity profile can be correlated and monitored as a function of scale-up and process change.
In the current development environment, a more thorough screen is required to understand fully the potential issues surrounding an API. Onyx considers that this is where its ‘standard screen’ is most relevant, and that this screen is best deployed towards the back end of pre-clinical pR+D for a given lead candidate. It is here that a more thorough form investigation is required – a more robust work-up on a compound to supply all the data required to engage in pre-formulation risk assessment and facilitate form selection, be that salt, polymorph or co-crystal.
This screen should allow companies to understand fully what the associated challenges are with the chosen API. So within a more thorough screen (of the API and salts of the API) the aims should include a firm understanding of form relationships; these should be well defined and the control elements leading to change well understood. This should naturally include an investigation of both thermal manipulation and solvent-based experimentation under a variety of conditions. This delivers reliability and allows ‘quality by design’ to feed back into the development process.
Any robust selection would also include a study of relative solubility and stability in bio-relevant media and within a buffered pH range
Any robust selection would also include a study of relative solubility and stability in bio-relevant media and within a buffered pH range to assess not only form but also chemical stability. A similar series of assessments under stressed storage conditions is also relevant and simple to complete, to give an understanding of not only the chemical stability but also the solid form stability, under such conditions.
These screens typically involve a range of scale-up experiments, not simply to feed characterisation and selection experimentation, but also to give an early indication of process applicability. Internally, this feeds back directly into pR+D and chemical development, and as is often the case an indication from the early screening work that a change can be applied to chemical development is captured in real time, enabling the benefits of integrated development from an overall project point of view to be capitalised on right from the start.
When starting to map out what may be important in terms of salts or poly-morphs, it will be possible to see which systems give improved impurity profiles for the batch. This is especially relevant for APIs currently being taken into development where it is estimated that up to 70% of all new chemical entities can be classified as BCS class type II /IV. These molecules create an issue for not only formulation development, but also chemical development because solubility is so poor and the entrapment of impurities and solvents can be a real problem.
A screen of this nature is best placed within the development pipeline ahead of providing low kilo amounts, i.e. the first non-GMP batch. Ideally, a study like this would take place 4–6 weeks ahead of the completion of that non-GMP batch, so that at that point it would be known what form and what salts are going to be produced ahead of crystallisation develop-ment and GMP manufacture. This also allows for more robust use-tests of GMP typical materials to be applied in real time with a wealth of form and composition-related characterisation data in place.
Another significant benefit of an integrated approach is the numerous ways in which understanding and mapping solid form alongside chemical development can reduce project timelines and bottlenecks, especially when looking towards formulation.
The key is to understand the subtleties of material behaviour and to be able to transfer that data effectively and efficiently between different chemists. Having pR+D and solid form under one roof and through each step of the development pathway allows the chemists to discuss and observe these subtleties with a hands-on approach without the need for a tech transfer, enabling form to be mapped as a function of development, process and impurities, thereby simplifying project management.
Applying the right screen at the right time typically reduces the effort, risk and time to get the first batch delivered on time
Taking this approach and applying the right screen at the right time typically reduces the effort, risk and time to get the first batch delivered on time. Also Onyx Scientific would advise integration with formulation development as early as possible because, again, there may be more than one choice of form. For example, there may be a choice between a robust polymorphic form and two or three salt forms.
The process might not define which of those is the best to take forward. From a purity perspective there may be no difference, but there may be an indication early on of what the route of formulation might be and this could influence which salt or form is selected as the leading candidate and which becomes a reserve option. Some salt versions, for instance, may provide a range of solubility or different kinetic dissolution profiles that benefit certain formulations such as slow release.