Evaluation of Dose-Exposure-Response Relationships of Miltefosine and Discovery of Multicomponent Crystal Forms of Artesunate

Background: Miltefosine is the first and only oral medication to be successfully utilised as an antileishmanial agent. However, the drug is associated with differences in exposure patterns and cure rates among different population groups e.g. ethnicity and age (i.e., children v adults) in clinical trials. Artesunate (ATS), a BCS class II drug widely used for anti-malaria therapy, exhibits not only poor solubility but also poor stability in aqueous solution.

Aim: This study aimed to characterise the solid-state properties of miltefosine, attempt to synthesis its anhydrous form, and screen for potential polymorphism through recrystallisation in various solvents. Additionally, this study sought to develop mechanistic population physiologically-based pharmacokinetic (PBPK) models to investigate the dose-exposure–response relationship of miltefosine in in silico clinical trials, with a particular focus on evaluating differences between children and adults. Furthermore, the study aimed to discover novel multicomponent crystal forms of ATS with enhanced physicochemical properties. The solubility of ATS and the dissolution performance of ATS and its crystal forms in buffer solutions were also evaluated.

Method: The solid-state properties of miltefosine were characterised using Powder X-ray Diffraction(PXRD), Differential Scanning Calorimetry (DSC), Thermogravimetric analysis (TGA), and Fourier Transform Infrared Spectroscopy (FTIR). Dehydration studies were performed in an oven at a specific temperature of 120°C for varying durations (24h to 144h), with samples characterised by PXRD, DSC, TGA, and FTIR. Recrystallisation experiments were conducted in different organic solvents to screen for polymorphs, with the resulting materials characterised by PXRD,DSC, TGA, and FTIR. Finally, the solvates formed were desolvated by heating in an oven at 60°Cfor 24h and characterised using PXRD and DSC. This work employed the Simcyp population-PBPK platform to predict miltefosine plasma concentrations in a virtual population under different dosing regimens. The cure rate of miltefosine treatment is related to systemic drug exposure in plasma and also depends on its concentration in the host cells because leishmania parasites are intracellular pathogens. Therefore, the prediction of the miltefosine distribution in peripheral blood mononuclear cells (PBMCs) in a virtual population was implemented within the developed PBPK model using the PD (pharmacodynamics) model facility in the software. Finally, the dose-exposure–response relationships of miltefosine in adults and children were assessed by the developed PBPK models with a PK target of AUC d0-28>535μg⋅day/mL in plasma. In total, 117 potential coformer candidates were examined, including carboxylic acids, amino acids, nutraceuticals, and drug candidates for treating neglected tropical diseases (such as malaria and leishmania). A mechanochemical solvent drop grinding (SDG) approach using methanol was used in the initial screening experiments. For those which were formed as gel-like resultants, further grinding was carried out with other solvents such as acetonitrile, ethanol, propan-2-ol,tetrahydrofuran, or acetonitrile. The samples were analysed by PXRD, FTIR, and thermal analyses[i.e., DSC, TGA, and hot stage microscopy (HSM)] to confirm new crystal solid formation. Dynamic apparent equilibrium solubility and dissolution experiments were conducted in 0.01M chloride buffer (pH 1.2) and 0.01M PBS (pH 4.5 and 6.8). The residues recovered were characterised by PXRD.

Result: The solid-state characterisation of miltefosine showed that it exists as a hydrate. Dehydration studies resulted in a loss of crystallinity in the miltefosine hydrate. Polymorph screening through recrystallisation led to the formation of miltefosine solvates, which were subsequently desolvated upon heating. It is shown that both adult and paediatric PBPK models of miltefosine can be developed to predict the PK data of the clinical trials accurately. There was no significant difference in the predicted dose-exposure–response of the miltefosine treatment for different simulated ethnicities under the same dose regime and the dose-selection strategies determined the clinical outcome of the miltefosine treatment. A lower cure rate of the miltefosine treatment in paediatrics was predicted because a lower exposure of miltefosine was simulated in virtual paediatric in comparison with adult virtual populations when they received the same dose of the treatment. Experimental screening of ATS with the 117 coformers resulted in discovering five novel multicomponent crystal forms of ATS with 4-aminobenzoic Acid (ABA), 1,4-diazabicyclo[2.2.2]octane (DABCO), 1,2-di(pyridine-4-yl)ethane (DPE), 1,10-phenanthroline (PHEN) and urea(URE). Based on the ΔpKa rule and FTIR results, 1:1 ATS-ABA, 2:1 ATS-DABCO Form 1, 2:1ATS-DPE, 1:1 ATS-PHEN, and 1:1 ATS-URE are cocrystals and 2:1 ATS-DABCO Form 2 is salt. Additionally, ATS-URE cocrystals can be crystallised as different solvate forms, including methanol, ethanol, and acetonitrile. Two polymorphs of 2:1 ATS-DABCO cocrystal/salt have also been discovered. The crystal structures of these multicomponent ATS crystals were determined by SXRD and characterised by PXRD, FTIR and different thermal analytical techniques (i.e., DSC,TGA and HSM). It has been shown that some of the multicomponent ATS crystals can significantly improve the in vitro dissolution performance of ATS and its stability in solution. Unfortunately, the solid-state stability study shows that these multicomponent cocrystals/salts do not exhibit better stability than the raw ATS under the conditions studied. Dynamic apparent equilibrium solubility profiles indicated higher ATS solubility in neutral (pH6.8) and slightly acidic (pH 4.5) conditions compared to acidic (pH 1.2) environments. CocrystalsATS²-DABCO Form 2 and ATS²-DPE enhanced ATS dissolution and stability at pH 4.5 and 6.8.

Conclusion: In conclusion, it was determined that miltefosine contains one hydrate molecule within its crystal lattice, as dehydration studies produced a poorly crystalline material. Recrystallisation of miltefosine in various solvents resulted in solvate successfully desolvated upon heating. Therefore, miltefosine demonstrates stability in its hydrated form. The mechanistic PBPK model suggested that the higher fraction of unbound miltefosine in plasma was responsible for a higher probability of failure in paediatrics because of the difference in the distribution of plasma proteins between adults and paediatrics. The developed PBPK models could be used to determine an optimal miltefosine dose regime in future clinical trials. The discovery of more multicomponent crystals of ATS with improved physicochemical properties(e.g., solubility and/or stability) could help to enhance its therapeutic efficacy and stability. The dynamic apparent equilibrium solubility profiles indicated that ATS exhibits greater solubility in neutral and slightly acidic conditions (pH 4.5 and 6.8) compared to highly acidic conditions (pH1.2). Cocrystals such as ATS²-DABCO Form 2 and ATS²-DPE demonstrated potential in enhancing ATS dissolution and the stability of ATS and DHA under pH conditions of 4.5 and 6.8, suggesting their suitability for formulation optimisation.