formulation design
A quick tool to assess the feasibility of an amorphous solid dispersion
While generally considered as the most enabling formulation approach to tackle solubility issues, not all compounds benefit from formulation as an amorphous solid dispersion.
Amorphous solid dispersions (ASDs) are among the most widely used formulations to enhance the oral bioavailability of poorly water-soluble compounds. While generally considered as the most enabling formulation approach to tackle solubility issues, not all compounds benefit from formulation as an ASD. The practical utility of the ASD approach is critically dependent on the ability of the formulation to withstand drug crystallization during manufacturing and storage, as well as after administration. Understanding the crystallization risk of a poorly water-soluble drug candidate is therefore essential when assessing the feasibility of developing an ASD.
At Harpago, we utilize a thermal analysis approach, developed by researchers at Purdue University [1], to assess this crystallization risk. Below we outline the experimental protocol and illustrate how its results help in screening out compounds that are unsuitable for formulation as an ASD.
The procedure consists of subjecting the active compound to a heat-cool-heat experiment in a differential scanning calorimeter (DSC), where the compound is first heated above its melting point to create a melt, then cooled down at a controlled cooling rate of 20 °C/min to – 75 °C and lastly heated above its melting point again at a heating rate of 10 °C/min.
In such an experiment, a compound can either:
1. Crystallize upon cooling (Class 1 compounds)
2. Crystallize upon reheating (Class 2 compounds)
3. Not crystallize upon either cooling or reheating (Class 3 compounds)
At Harpago, we utilize a thermal analysis approach, developed by researchers at Purdue University [1], to assess this crystallization risk. Below we outline the experimental protocol and illustrate how its results help in screening out compounds that are unsuitable for formulation as an ASD.
The procedure consists of subjecting the active compound to a heat-cool-heat experiment in a differential scanning calorimeter (DSC), where the compound is first heated above its melting point to create a melt, then cooled down at a controlled cooling rate of 20 °C/min to – 75 °C and lastly heated above its melting point again at a heating rate of 10 °C/min.
In such an experiment, a compound can either:
1. Crystallize upon cooling (Class 1 compounds)
2. Crystallize upon reheating (Class 2 compounds)
3. Not crystallize upon either cooling or reheating (Class 3 compounds)
Typical picture observed for Class 1 (rapid crystallizers), Class 2 (intermediately rapid crystallizers) and Class 3 compounds (slow crystallizers) observed in the heat-cool-heat DSC experiment.
In addition to posing a physical instability liability, Class 1 compounds are also unlikely to provide for a meaningful absorption improvement when formulated as an ASD.
In terms of physical stability, Class 1 compounds will be much more challenging to maintain as the amorphous form compared to Class 3 compounds. Therefore, they are not appropriate candidates for the ASD technology. Conversely, Class 3 compounds are good candidates for formulation as an ASD. Class two compounds are moderately fast crystallizers, and additional tests (amorphous solubility measurements; physical stability profiling of API-polymer combinations) are required to assess ASD feasibility for these compounds more definitively. Although the process of crystallization is complex and consideration of a single descriptor cannot be used to predict crystallization behavior upon cooling from the undercooled melt, class I molecules generally tend to have lower molecular weight, fewer rotatable bonds and higher entropy and enthalpy of fusion compared to class III compounds [1].
In addition to posing a physical instability liability, Class 1 compounds are also unlikely to provide for a meaningful absorption improvement when formulated as an ASD. In recent years, it has been widely demonstrated that the biopharmaceutical performance of ASDs is critically dependent on the ability of the formulation to generate amorphous drug nanodroplets on contact with the gastrointestinal fluids. For Class 1 compounds, this may be technically very challenging or even impossible given rapid drug crystallization on hydration of the ASD. Thus, even in a context where long-term physical stability of a formulation is not required (for instance in discovery or preclinical studies), administration of a Class 1 compound formulated as an ASD is unlikely to deliver useful exposure benefits over crystalline drug substance. We therefore recommend other formulation approaches (e.g., nanosuspension technology) for this class of compound.
In addition to posing a physical instability liability, Class 1 compounds are also unlikely to provide for a meaningful absorption improvement when formulated as an ASD. In recent years, it has been widely demonstrated that the biopharmaceutical performance of ASDs is critically dependent on the ability of the formulation to generate amorphous drug nanodroplets on contact with the gastrointestinal fluids. For Class 1 compounds, this may be technically very challenging or even impossible given rapid drug crystallization on hydration of the ASD. Thus, even in a context where long-term physical stability of a formulation is not required (for instance in discovery or preclinical studies), administration of a Class 1 compound formulated as an ASD is unlikely to deliver useful exposure benefits over crystalline drug substance. We therefore recommend other formulation approaches (e.g., nanosuspension technology) for this class of compound.
This DSC-based risk assessment allows to screen out unsuitable ASD candidates early on.
While this DSC-based experimental procedure assesses crystallization tendency via cooling from the undercooled melt state, good correlation with crystallization behavior from rapid solvent evaporation techniques has been demonstrated [2]. Thus, the results obtained from this heat-cool-heat experiment can be used to assess ASD feasibility at large, regardless of whether the ASD is produced via melt techniques (e.g., melt extrusion) or solvent evaporation techniques (e.g., spray drying).
The development of a performant ASD clearly requires much more experimentation than one single DSC experiment, but the procedure outlined above at least allows to screen out unsuitable ASD candidates (Class 1 molecules) early on, greatly reducing unneeded experiments and thus allowing more well-informed decisions on formulation strategies.
1.) Baird et al., J Pharm Sci, 2010, 99, 3787-806 A Classification System to Assess the Crystallization Tendency of Organic Molecules from Undercooled Melts
2.) Van Eerdenbrugh et al., J Pharm Sci, 2010, 99, 3826-87 Crystallization Tendency of Active Pharmaceutical Ingredients Following Rapid Solvent Evaporation—Classification and Comparison with Crystallization Tendency from Undercooled Melts
The development of a performant ASD clearly requires much more experimentation than one single DSC experiment, but the procedure outlined above at least allows to screen out unsuitable ASD candidates (Class 1 molecules) early on, greatly reducing unneeded experiments and thus allowing more well-informed decisions on formulation strategies.
1.) Baird et al., J Pharm Sci, 2010, 99, 3787-806 A Classification System to Assess the Crystallization Tendency of Organic Molecules from Undercooled Melts
2.) Van Eerdenbrugh et al., J Pharm Sci, 2010, 99, 3826-87 Crystallization Tendency of Active Pharmaceutical Ingredients Following Rapid Solvent Evaporation—Classification and Comparison with Crystallization Tendency from Undercooled Melts
If you want to learn more about the utility of an amorphous solid dispersion for your compound, feel free to contact us