Assessing Process Effects on the Recrystallisation of Co-Amorphous Form Products

Amorphous drug delivery systems can improve the performance of poorly soluble, active pharmaceutical ingredients (APIs) especially for category II and IV APIs of the Biopharmaceutics Class System. However, poor physical stability of an amorphous
form can hamper application to drug delivery. To overcome this challenge, co-amorphous systems (CASs) can be obtained by combining an API with a small excipient molecule. In this work, different co-amorphisation methods were applied to valsartan (VAL), an angiotensin II receptor blocker with poor aqueous solubility (Class II of BCS), and nicotinamide (NIC), an effective hydrotrope to enhance solubility of the VAL. Multiple CASs of varying ratios of VAL to NIC were prepared by melt amorphisation and characterised by X-ray, microscopic, thermal and, spectroscopic analyses. The CASs were X-ray amorphous, dominated by a single phase with multiple VAL-NIC binding sites and show enhanced aqueous solubility of VAL, but hydration of the NIC was shown to lower the glass transition temperature below that of the nominal starting composition, particularly for the NIC-rich CASs. The equimolar system was identified as optimal because it retained the least crystallinity with the best aqueous solubility of VAL and was the most stable against recrystallisation. CASs of equimolar amounts of VAL and NIC were then prepared by different processes to produce X-ray amorphous systems, all of which demonstrated enhanced aqueous solubility and dissolution rates. Inhomogeneous mixing during the different preparation routes was identified by polarised light microscopy, which showed a range of retained crystallinity in all samples. Systems prepared by solvent evaporation techniques demonstrated the best VAL-NIC mixing, aqueous solubility, and dissolution rate, suggesting these approaches are ideal for making a homogenous, single-phase CAS. The as-produced, equimolar CASs were subjected to ageing in different relative humidities, with characterisation every two weeks to identify the early stages of recrystallisation. Slurry mixing was the most effective against recrystallisation in 85% RH, still retaining the CAS’s X-ray amorphous state at 32 weeks and having 20 times better solubility than VAL alone. The main conclusion of this study is that the stability and solubility of each CAS was limited by amorphous phase separation and not recrystallisation (although both are enhanced in the presence of moisture). The next steps would be to understand reasons for inhomogeneous mixing during processing and the resulting effect on phase separation and recrystallisation in CASs will lead to improvement of these drug delivery systems.