There are many reasons why it might be desirable to create a drug formulation using the salt form of an active pharmaceutical ingredient (API). Most commonly, it is because the solid-state properties of the API are not suitable for further development, which can lead to an increase in the cost, complexity or timeline for the project. It may be that the API is not crystalline, not thermally stable or insoluble, or it might be hygroscopic, so by making a salt, the API’s chemical makeup is altered in such a way that that all of these properties might be changed. For the most part, these changes will be beneficial, however, sometimes the opposite is true, which is why it is important to use a range of experiments and screening procedures to determine a salt form, and not just choose one.
For basic drugs, the most common salt form is the hydrochloride; about 60% of all basic drug salt forms are hydrochlorides. Hydrochloric acid (HCl) is safe, and chlorine is abundant in the body. It is a very cheap choice, and with it being a strong acid, it will form a salt with most basic compounds. Conversely, for acidic drugs, the sodium salt predominates, as it is also abundant and safe, and the salts are made using cheap sodium hydroxide (NaOH). However, there is a wide range of other options if HCl or NaOH prove unsuccessful and these should not be excluded from a salt screening program.
When selecting a salt, it is important to consider the molecular weight of the counterion. If the API itself is small in terms of molecular weight, such as a drug designed to cross the blood–brain barrier to treat neurodegenerative disorders, the cost of goods could increase significantly when the API is diluted with a very heavy counterion. This is why sodium and potassium salts, as well as hydrochlorides are commonly used, as they add very little bulk to the API. Conversely, a large quantity of a common counterion, such as a bis-sulfate, could potentially be half as heavy as the drug itself, and while this could be unsuitable for some applications, it can be advantageous in formulation design in certain circumstances, such as if the API is highly potent. In this instance, a salt form which is 50% by mass of a salt former has little impact when the dose goes from 2 mg for a free API, to 3 mg for a salt form; but if a dose increases from 400 mg to 600 mg, this does.
These most prolific counterions will always be included in a saltscreening process, as they are the most likely to succeed, however, before a salt reaches the formulation team, its processing behavior may indicate that it is unlikely to be successful. An initial list of salt forms may be reduced down to two or three promising salts after an initial, broad set of salt screening experiments, which may include the sodium or hydrochloride salt, but it may be that their solid-state properties are not optimal. If a drug is marketed as a citrate or a maleate salt for example, this is usually a signpost that the simpler salt had been found unsuitable for further development.
One issue that is increasingly facing process chemists developing salt forms is the concept of potential genotoxic impurities (PGIs). Regulators are now taking great interest in whether a genotoxic impurity might be present in a drug product as a residue from processing, and sulfonate salts – and especially mesylate salts – are being flagged up for PGIs. In 2007, the antiretroviral drug Viracept (nefinavir mesilate) was recalled from European markets over concerns about elevated levels of ethyl methanesulfonate (EMS), which led to calls from European regulators to assess risk mitigation strategies for all marketed products employing a sulfonic acid counterion.1 Mitigating controls can, and should be put in place to avoid their formation. Extremely sensitive analytical techniques are now available that are able to prove that PGIs are not present in the final drug substance, and therefore fear of PGIs is not a good reason to avoid including sulfonate salts in screening. While the PGI tests add an extra step to the development process, if a mesylate salt performs far better than any of the alternatives that have been tried, there is no reason not to develop a drug as that salt form.
Assessing Developability
Developability will always be assessed as part of a salt screen, and from the processing standpoint of an API in its manufacture, hygroscopicity is important as the amount of moisture the salt form absorbs will affect the flow behavior of the material. When deciding on the best salt form from a processing point of view, it is crucial to try and balance all the different solid-state properties, such as flow, melting point, crystallinity and so on, as well as the potential for polymorphic forms to be located.
When screening potential salt forms, any potential risks should always be incorporated in the screening process and mitigated against. If screening shows that the theoretical “best salt” has potential issues, these need to be flagged up so that appropriate controls can be applied as it moves through the development pipeline.
It is also important to consider the longer-term implications of selecting a salt form, in assessing whether it is feasible that a chosen salt form is appropriate for commercial applications. For example, if a degree of salt disproportionation is observed at the screening stage, it may be that wet granulation will not be an appropriate process to use as that might cause the free base to precipitate out ahead of the desired salt.
Any potential risks that remain should be highlighted, and de-risking strategies developed. For example, if particle size could cause an issue, is it feasible to make larger particles? If there are concerns over polymorphism, how might the manufacturing process be controlled so that the desired polymorph can be reliably and consistently produced? Additionally, if it is impossible to stop impurities being formed during API synthesis, can a strategy be implemented by the development chemists and engineers so that they can be removed via processing, or process optimization?
There are different approaches to salt screening, with varying degrees of automation which impacts upon the speed of throughput. Adopting a medium throughput approach allows the experimental scientist to have greater control of the process, as automation often misses many of the critical observations that the scientist will make when carrying out laboratory procedures. The most efficient manner to carry out experiments is in parallel, and the first step is to draw up a list of the counterions that at first sight seem most suitable, based on the dose and route of administration, and any other relevant considerations such as molecular weight. Table 1 shows some commonly used salt formers.
The list of potential salt formers will be combined with a handful of different solvents, as the nature of the solvent can also have an impact on success. Typically, an initial screen may look at 24 counterions and six solvents, giving a total of 144 experiments to perform. The results of these initial experiments will allow assessment and reduction of the options down to two or three that will be scaled up for a deeper investigation of the solid-state properties and developability. Solubility and disproportionation potential will be assessed here, and very quick accelerated stability tests performed to allow a better evaluation on the manufacturability of a salt to be made.
Solvent toxicity must be considered and International Council for Harmonisation (ICH) guidelines2 categorize solvents into three classes (see Table 2). For the most part, Class 3 solvents will be used, but it is not inappropriate to include some Class 2 solvents at the screening stage. At this point, the priority is to identify the best salt in terms of solid-state properties - how it is best made is a different question, and with seed crystals in hand, different solvents that are more process friendly can then be investigated.
Other factors that might cause a salt form to be discounted include the potential for multiple hydrated states to form, or a propensity to form polymorphs. Either of these would represent a real manufacturing risk which can be mitigated against by working with experts who are experienced in the solid form area.
It is at this stage that the results need to be honed down, and all aspects of developability of the final form need to be assessed. However much cheaper a hydrochloride salt might be to prepare, if a succinate or citrate salt result in superior crystals than the hydrochloride, then this becomes the overriding factor. Screening should give a prioritized list of potential salt forms in terms of manufacturing risk, and also identify those which are not recommended.
Although each molecule needs to be judged individually, experience gained by a scientist over the years is invaluable. For example, if a salt is left out on a laboratory bench in a rainy or humid climate and there is the slightest indication that it is becoming a little sticky, that would be a pointer that on the manufacturing floor there might be problems. If, bearing in mind all the other properties of the API, this salt is still deemed the best option, it will need to be handled under dry conditions, and desiccants used on storage.
It is unusual to choose to make a solid dosage form with an amorphous form unless there really is no suitable crystal or cocrystal. Amorphous forms are, of course, appropriate for solution formulations, such as those designed for intravenous administration, where rapid dissolution in the clinic is desirable. But from a development and manufacturing point of view, crystalline materials are still preferred as they are easier to isolate and dry, and are likely to have better stability and shelf life.
Of course, there will always be instances when a salt form is simply not available for a particular API. This is usually driven by the strength of the acid (in terms of the measurement of its disassociation, pKa), and it is frequently clear at the outset that a salt is likely to be inaccessible. A salt might be formed in solution, but will not crystallize out, or it might dissociate or degrade under normal processing conditions. Typical indicators of degradation include color changes which flag up that further analysis is required, which will likely show that the intended species is not present. If a suitable salt cannot be made, then a cocrystal may be possible, or if solubility is an issue, then an amorphous solid dispersion could be successful.
The salt screening process is a vital component of the early stage development of any new drug. With care and sometimes creativity, it is possible to avoid many additional problems further down the development program. By choosing the best salt option at the outset, much time and money can be saved as the API moves closer to patients.
References
- D.P.Elder et al “The utility of sulfonate salts in drug development”, J Pharm Sci 99:2948–2961, 2010
- https://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Quality/Q3C/Q3C__R6___Step_4.pdf