Introduction

The largest hurdle in the development of new pharmaceutical entities revolves around the increasing prevalence of poor solubility. This barricade is now affecting approximately 40% of the top 200 commercial pharmaceuticals and approx. 90% of drugs in the pharmaceutical pipeline [1, 2]. Poor solubility leads to a less pronounced concentration gradient, slow passive diffusion and in vivo failure. Many strategies have been employed to overcome this hurdle, all of which present their own unique advantages and disadvantages. Salt formation, which accounts for approx. 50% of drugs currently on the market, is the most popular method for solubility enhancement, but is however limited by the presence of ionizable groups [3]. Cocrystals are another multicomponent solid form allowing for improvements in solubility and dissolution rate, but require an API and coformer with adequate molecular complementarity and a high degree of hydrogen bonding propensity. Amorphization is another method for improving saturation solubility, but the trade-off includes often large reductions in stability. Particle size reduction to the nanoscale often leads to enhanced dissolution rate, leading to a more poignant concentration gradient promoting rapid passive diffusion, but these particles can sometimes be unstable in solid form or in suspension, and can often be difficult to isolate during manufacture [4].

A single strategy to enhance the solubility/dissolution rate of a poorly soluble drug may not be feasible, due to the vast diversity of pharmaceutical entities currently available. As such, it is necessary to develop adequate knowledge on the simultaneous solid-state and particle size control of pharmaceutical compounds to provide pharmaceutical chemists a multitude of avenues to improve drug physiochemical properties. This study seeks to further the knowledge on solid-state and particle size control of multicomponent solid pharmaceuticals, using scalable and continuous methodologies, attractive at commercial scales.

Method

The most recently approved pharmaceutical cocrystal, Seglentis® (celecoxib + tramadol hydrochloride), was employed as a model cocrystal to demonstrate the ability of the supercritical CO₂-assisted nano-spray drying (SASD) process, described previously [5], to control both solid-state and particle size. A design of experiments (DoE) approach was employed here, highlighted in figure 1, which varied three critical process parameters: solution flow rate, solution concentration, and pressure of the atomizing gas. Briefly, a heated, high-pressure coaxial nozzle with a small mixing volume is employed to atomize a mixture of the drug solution and the supercritical CO₂, into a heated drying/precipitation chamber. The produced droplets are dried, and the resulting particles are collected in a wire mesh placed at the outlet at the bottom of the drying chamber.

Results

Solid-State Control

Powder x-ray diffraction (PXRD) and differential scanning calorimetry (DSC) analysis determined that the processing conditions at DoE point 5 were the only set of conditions which generated the stable (co)crystalline form. This is likely due to larger droplets produced during the atomization step due to the higher solution flow rate, which in turn led to a longer time for complete solvent removal, coupled with a low solution concentration (i.e. low levels of supersaturation) which favored the formation of the stable cocrystalline form. All other sets of process conditions led to the formation of (co)amorphous material due to rapid solvent separation from the solid phase.

The Gordon-Taylor equation (equation 1) was employed to aid in determining the molecular nature of the (co)amorphous samples. This generates a value, T_g12, a theoretical glass transition temperature for a mixture of amorphous celecoxib and amorphous tramadol hydrochloride, assuming weak or no intermolecular interactions. Glass transition temperatures exceeding this theoretical value are considered to display evidence of intermolecular interactions [6]. The theoretical value obtained here is 65.1 °C. (Co)Amorphous samples which displayed increased glass transition values in comparison included samples from DoE point 7 (73.9 °C) and 8 (74.0 °C). Further to this, Fourier transform infrared spectroscopy (FTIR) was employed to further investigate these intermolecular interactions. The FTIR spectra displayed in Figure 2 highlights the peak at 3479.9 cm^-1 which corresponds to the hydrogen bonding network observed in the cocrystal. Unlike other (co)amorphous samples, DoE 8 samples display a peak at this point, indicating hydrogen bonding in the (co)amorphous sample, comparable with the cocrystal.

Particle Size Control

Figure 3 displays SEM images for samples obtained from DoE points 1, 5, 7 and 8. The cocrystalline sample (DoE point 5) displayed a granular morphology and a mean particle size of 1160 ± 220 nm. All (co)amorphous samples were smaller, with a more uniform and spherical morphology. DoE point 4 displayed the smallest particle size of 120 ± 20 nm, due to the low solution flow rate which produces smaller droplets, and the lower solution concentration leading to a lower solid content in the atomized droplets.

Tabletting

Samples from DoE process conditions 1, 5, 7 and 8 were subjected to tabletting (10% w/w sample) and assessed on the basis on tabletability, compressibility and compactability, with fitted tabletting parameters highlighted in Table 1. Samples from DoE 7 displayed the highest tensile strength, with DoE 5 samples displaying the lowest. This may be attributed to the reduced intermolecular interactions in the (co)amorphous samples leading to more efficient packing, however, all samples displayed a tensile strength > 2MPa, which is required for acceptable manufacturability and performance. Formulations including samples from DoE point 5 displayed reduced compactability in comparison to other formulations, as observed in the σ_t0 value of 5.053 MPa, which may be attributed to the samples larger, more irregular shape. The b value observed in Table 1 provides insight into bonding capacity. Samples from DoE 8 has the largest b value due to its uniform, spherical shape (unlike DoE 5), and its increased intermolecular interactions in unlike samples from DoE 1 and 7. Regarding compressibility, large values for the Heckle coefficient, k, indicate good compressibility, with samples from DoE 5 and 8 displaying the best compressibility.

Polymorphic Cocrystals

Due to the production of (co)amorphous (unstable) samples from many of the above tested SASD process conditions, three polymorphic cocrystals were selected to further delve into the solid-state control aspect of this study. These cocrystals are:

- Ethenzamide-Saccharin (EthSacc) (with two polymorphic forms reported) [7]

- Ethenzamide-Gentisic acid (EthGa) (with three polymorphic forms reported) [8]

- Isonicotinamide-Citric acid (IsoCa) (with three polymorphic cocrystal forms and one salt form reported) [9]

A two factorial DoE was employed for the production of these cocrystal systems using the previously described supercritical method, varying solution flow rate and solution concentration. Experimentation has been carried out using SASD, and is ongoing regarding the use of conventional spray drying, and using compressed nitrogen in the SASD apparatus. Briefly, the SASD method led to the production of the stable α and β polymorphs of the IsoCa cocrystal. The SASD method only produced the metastable form II for the EthGa cocrystal, while both the metastable form II, and mixtures of form I and II were produced for the EthSacc cocrystal. Further characterization is ongoing.

Conclusion

The supercritical CO₂-assisted nano-spray drying method described here has been successful in controlling both particle size and solid-state of the most recently FDA approved cocrystal, comprised of celecoxib and tramadol hydrochloride. Variations within acceptable ranges were observed in tabletting in regard to tabletability, compactability and compressibility, demonstrating the methods versatility in manipulating solid-state and particle size to achieve the desired rheological properties. While experimentation and characterization are still ongoing, the supercritical method has shown promise in isolating metastable and stable polymorphs of polymorphic cocrystals, demonstrating the robustness and transferability of this technique to other multicomponent pharmaceutical systems.

References

1. Sathisaran, I. and S.V. Dalvi, Engineering Cocrystals of Poorly Water-Soluble Drugs to Enhance Dissolution in Aqueous Medium. Pharmaceutics, 2018. 10(3): p. 108.

2. Rodriguez-Aller, M., et al., Strategies for formulating and delivering poorly water-soluble drugs. Journal of Drug Delivery Science and Technology, 2015. 30: p. 342-351.

3. Gupta, D., et al., Salts of therapeutic agents: chemical, physicochemical, and biological considerations. Molecules, 2018. 23(7): p. 1719.

4. O'Sullivan, A., et al., Solid-State and Particle Size Control of Pharmaceutical Cocrystals using Atomization-Based Techniques. International Journal of Pharmaceutics, 2022: p. 121798.

5. Long, B., et al., Controlling Polymorphism of Carbamazepine Nanoparticles in a Continuous Supercritical-CO2-Assisted Spray Drying Process. Crystal Growth & Design, 2019. 19(7): p. 3755-3767.

6. An, L., et al., Effects of molecular weight and interaction parameter on the glass transition temperature of polystyrene mixtures and its blends with polystyrene/poly (2, 6-dimethyl-p-phenylene oxide). European polymer journal, 1997. 33(9): p. 1523-1528.

7. Aitipamula, S., P.S. Chow, and R.B. Tan, Dimorphs of a 1: 1 cocrystal of ethenzamide and saccharin: solid-state grinding methods result in metastable polymorph. CrystEngComm, 2009. 11(5): p. 889-895.

8. Aitipamula, S., P.S. Chow, and R.B. Tan, Trimorphs of a pharmaceutical cocrystal involving two active pharmaceutical ingredients: potential relevance to combination drugs. CrystEngComm, 2009. 11(9): p. 1823-1827.

9. Stainton, P., et al., First comparative study of the three polymorphs of bis (isonicotinamide) citric acid cocrystals and the concomitant salt 4-carbamoylpyridinium citrate isonicotinamide. Crystal Growth & Design, 2018. 18(7): p. 4150-4159.