Process Optimization and Upscaling of Spray-Dried Drug-Amino acid Co-Amorphous Formulations.

The feasibility of upscaling the formulation of co-amorphous indomethacin-lysine from lab-scale to pilot-scale spray drying was investigated. A 2² full factorial design of experiments (DoE) was employed at lab scale. The atomization gas flow rate (Fatom, from 0.5 to 1.4 kg/h) and outlet temperature (Tout, from 55 to 75 °C) were chosen as the critical process parameters. The obtained amorphization, glass transition temperature, bulk density, yield, and particle size distribution were chosen as the critical quality attributes. In general, the model showed low Fatom and high Tout to be beneficial for the desired product characteristics (a co-amorphous formulation with a low bulk density, high yield, and small particle size). In addition, only a low Fatom and high Tout led to the desired complete co-amorphization, while a minor residual crystallinity was observed with the other combinations of Fatom and Tout. Finally, upscaling to a pilot scale spray dryer was carried out based on the DoE results; however, the drying gas flow rate and the feed flow rate were adjusted to account for the different drying chamber geometries. An increased likelihood to achieve complete amorphization, because of the extended drying chamber, and hence an increased residence time of the droplets in the drying gas, was found in the pilot scale, confirming the feasibility of upscaling spray drying as a production technique for co-amorphous systems.

Co-former selection for co-amorphous drug-amino acid formulations.

We have previously developed a fast screening method on the ability of twenty amino acids (AA) to form co-amorphous formulations with six drugs upon ball milling. In this work, the potential advantages in physical stability and dissolution rate of the 36 successful co-amorphous formulations, compared to the pure amorphous drug, were further investigated. The physical stability of the formulations at dry conditions was assessed by X-ray powder diffraction (XRPD) and their thermal behavior by differential scanning calorimetry (DSC). In addition, the intrinsic dissolution rate (IDR) of all formulations was determined in phosphate buffer (10 mM, pH 6.8). Finally, all the co-amorphous formulations were summarized into different groups, according to the outcome of the co-formability, physical stability and dissolution rate screenings, and guidelines could be drawn for selection of co-formers for a new given drug: (i) For acidic drugs, basic AAs (arginine, histidine, and lysine) are good co-formers with respect to the three critical quality attributes: co-formability, physical stability and dissolution. High glass transition temperatures (Tg), physical stability for 1-2 years, and accelerated IDR were observed. (ii) For basic and neutral drugs, non-polar AAs with aromatic groups such as tryptophan (TRP) and phenylalanine (PHE) should be explored as first choice. These combinations presented high Tgs, which generally translated into good physical stability. The IDR of TRP- and PHE-based formulations were usually superior to the IDR of the pure amorphous drugs; (iii) Non-polar AAs with aliphatic structures such as leucine, isoleucine, methionine and valine did not provide an increase in Tg or IDR compared to the pure amorphous drug, and appear to be less feasible AAs for co-amorphous formulations.

In vitro and in vivo comparison between crystalline and co-amorphous salts of naproxen-arginine.

Liquid-assisted grinding (LAG) and dry ball milling (DBM) have recently been used to obtain different physical forms of drug-amino acid salts with promising dissolution and physical stability properties. In this work, crystalline and co-amorphous naproxen-arginine mixtures were prepared using LAG and DBM, respectively, and compared with regard to their in vitro and in vivo performance. X-ray powder diffraction and Fourier-transformed infrared spectroscopy showed that LAG led to the formation of a crystalline salt, while DBM led to a co-amorphous salt. These results agreed with the differential scanning calorimetry profiles: a melting point of 230 °C was determined for the crystalline salt, while the co-amorphous formulation showed a single glass transition temperature at approx. 92 °C. Both solid state forms of the salt showed increased intrinsic dissolution rates (14.8 and 74.1-fold, respectively) and also higher solubility (25.3 and 29.8-fold, respectively) compared to the pure crystalline drug in vitro. Subsequently, the co-amorphous salt revealed an improved bioavailability in a pharmacokinetic study, showing a 1.5-fold increase in AUC0-t and a 2.15-fold increase in cmax compared to the pure crystalline drug. In contrast, even though showing a better in vitro performance, the crystalline salt interestingly did not show an increase in bioavailability in comparison to pure crystalline naproxen.

The Role of Glass Transition Temperatures in Coamorphous Drug-Amino Acid Formulations.

The improved physical stability associated with coamorphous drug-amino acid (AA) formulations may indicate a decrease in mobility of the amorphous drug molecules, compared to the neat amorphous form of the drug. Since the characteristic glass transition temperatures ( Tgα and Tgß) represent molecular mobility in amorphous systems, we aimed to characterize Tgα and Tgß and to determine their role in physical stability as well as their potential usefulness to determine the presence of an excess component (either drug or AA) in coamorphous systems. Indomethacin (IND)-tryptophan (TRP) and carvedilol (CAR)-TRP were used as model coamorphous systems. The analytical techniques used were X-ray powder diffractometry (XRPD) to determine the solid-state form, dynamic mechanical analysis (DMA) to probe Tgα and Tgß, and differential scanning calorimetry (DSC) to probe thermal behavior of the coamorphous systems. Tgα analysis showed a gradual monotonous increase in Tgα values with increasing AA concentration, and this increase in the Tgα value is not the cause of the improved physical stability. The Tgß analysis for the IND-TRP sample with 10% drug had a Tgß of 226.8 K, and samples with 20-90% drug had similar Tgß values around 212.5 K. For CAR-TRP, samples with 10-40% drug had similar Tgß values around 230.5 K, and samples with 50-90% drug had similar Tgß values around 223.3 K. The similar Tgß values in coamorphous systems at different drug ratios indicate that they in fact are the Tgß of the component that is in excess to an ideal drug-AA coamorphous mixture. DSC and XRPD analysis showed that for IND-TRP, IND is in excess if the drug concentration is 30% or above and will eventually recrystallize. For CAR-TRP, CAR is in excess and recrystallizes when the drug concentration is 50% or above. We have proposed a means of estimating, on the basis of Tgß, which drug to AA ratios will lead to optimally physically stable coamorphous systems that can be considered for further development.

The use of molecular descriptors in the development of co-amorphous formulations.

Co-amorphous systems consisting of a drug and an amino acid have been investigated extensively for the enhancement of drug solubility and amorphous stability. The purpose of this study is to investigate which molecular descriptors are important for predicting the likelihood of a successful co-amorphisation between amino acid and drug. The predictions are thought to be used in an early screening phase to identify potential drug-amino acid combinations for further studies. A large variety of molecular descriptors was calculated for six drugs (carvedilol, mebendazole, carbamazepine, furosemide, indomethacin and simvastatin) and the twenty naturally occurring amino acids. The descriptor differences for all drug-amino acid combinations were calculated and used as input in the X-matrix of a Partial Least Square Discriminant Analysis (PLS-DA). The Y-matrix of the PLS-DA consisted of the X-ray powder diffraction response ("co-amorphous" or "not co-amorphous") obtained by ball milling all combinations for 60 min. The PLS-DA model showed a clear separation of the not co-amorphous and the co-amorphous samples and was successfully predicting the class membership of 19 out of the 20 completely left out drug-amino acid combinations of mebendazole. The approach seems to be promising for predicting the ability of new drug-amino acids combinations to become co-amorphous.

Performance comparison between crystalline and co-amorphous salts of indomethacin-lysine.

The introduction of a highly water soluble amino acid as co-amorphous co-former has previously been shown to significantly improve the dissolution rate of poorly water soluble drugs. In this work, dry ball milling (DBM) and liquid assisted grinding (LAG) were used to prepare different physical forms of salts of indomethacin (IND) with the amino acid lysine (LYS), allowing the direct comparison of their solid-state properties to their in vitro performance. X-ray powder diffraction and Fourier-transformed infrared spectroscopy showed that DBM experiments led to the formation of a fully co-amorphous salt, while LAG resulted in a crystalline salt. Differential scanning calorimetry showed that the samples prepared by DBM had a single glass transition temperature (Tg) of approx. 100°C for the co-amorphous salt, while a new melting point (223°C) was obtained for the crystalline salt prepared by LAG. Intrinsic dissolution and powder dissolution studies demonstrated an increased dissolution rate of the drug in the co-amorphous salt compared to pure amorphous IND and also the crystalline drug-LYS salt. Furthermore, the co-amorphous IND-LYS salt presented long term physical stability in dry conditions at 25°C and 40°C. Overall, it has been shown that the co-amorphous form of a salt has a superior performance in comparison to a crystalline salt.

Influence of drug load and physical form of cinnarizine in new SNEDDS dosing regimens: in vivo and in vitro evaluations.

The aim of this work was to evaluate the influence of drug load and physical form of cinnarizine (CIN) in self-nanoemulsifying drug delivery systems (SNEDDS) on absorption in rats. Further, the predictivity of the dynamic in vitro lipolysis model was evaluated. The following dosing regimens were assessed: (1) CIN dissolved in SNEDDS at 80% of equilibrium solubility (Seq) (SNEDDS 80%); (2) supersaturated SNEDDS with CIN dissolved at 200% Seq (super-SNEDDS solution); (3) SNEDDS suspension with CIN added at 200% Seq (CIN partially dissolved and partially suspended) (super-SNEDDS suspension); (4) drug-free SNEDDS co-dosed with aqueous CIN suspension (Chasing principle), and (5) CIN aqueous suspension. The CIN dose was kept constant for all dosing regimens. Therefore, the super-SNEDDS solution and super-SNEDDS suspension contained 2.5-fold less SNEDDS pre-concentrate than SNEDDS 80% and the Chasing principle. In vivo, a higher AUC after dosing CIN in SNEDDS 80% and the Chasing principle was obtained when compared to the super-SNEDDS solution, super-SNEDDS suspension, and aqueous suspension. In vitro, a higher extent of CIN in the aqueous phase was observed for all SNEDDS-containing dosing regimens, compared to the aqueous suspension. Since the drug level in the aqueous phase is traditionally considered as the fraction available for absorption, a lack of in vitro-in vivo relation was observed. This study revealed that the physical form of CIN in the current SNEDDS does not affect CIN absorption and solubilization, whereas the drug load, or amount of co-dosed lipid, significantly influenced CIN bioavailability.

Development of a screening method for co-amorphous formulations of drugs and amino acids.

Using amino acids (AA) as low molecular weight excipients in the preparation of co-amorphous blends with the aim to stabilize the drug in the amorphous form have been discussed in a range of studies. However, there is currently no theoretical consensus behind which AA would be a suitable co-former for a given drug. In this work, a fast screening process to assess the co-former feasibility in co-amorphous drug-AA blends has been developed on the basis of the amorphization kinetics upon oscillatory ball milling. For this purpose, six model drugs were combined with 20 different AAs and co-milled at an equimolar ratio for different times (1, 5, 15, 30 and 60min). The degree of amorphization was then studied for the different time points by determination of the area under the curve of the diffraction peaks in X-ray powder diffraction measurements. The results of this study suggest that the choice of AA as co-formers for the formation of the co-amorphous blend could be significantly inferred after 15min of milling, since a crystallinity decrease higher than 90% after 15min resulted in successful co-amorphization in approximately 90% of the mixtures after 60min of milling. The results furthermore suggested that non-polar AAs, such as tryptophan, phenylalanine, leucine, isoleucine, methionine, valine and proline, are a good first choice in the selection of a co-former for a given drug in a co-amorphous formulation. Basic AAs appear suitable for amorphous salt formation in the case of acidic drugs. Acidic AAs however, were shown to be generally poor co-formers for co-amorphous systems.

Development of low density azithromycin-loaded polycaprolactone microparticles for pulmonary delivery.

CONTEXT: The development of low-density polymeric microparticles may be a useful approach to deliver antibiotics such as azithromycin into the lung. OBJECTIVE: The aim of this study was to develop azithromycin-loaded low density polycaprolactone microparticles by the double emulsion/solvent evaporation method. MATERIALS AND METHODS: Microparticles were prepared and characterized according to their physicochemical properties, drug loading, and drug release profiles. A full 23 factorial design was used to evaluate the effect of some independent variables on the drug loading and aerodynamic diameter of the particles. An in silico pulmonary deposition model was used to predict the lung deposition profiles for the formulations. RESULTS AND DISCUSSION: The resulting particles presented drug loading up to 23.1% (wt%) and mean geometric diameters varying from 4.0 µm to 15.4 µm. Bulk and tapped densities were low, resulting in good or excellent flow properties. SEM images showed spherical particles with a smooth surface. However, hollow inner structures were observed, which may explain the low values of bulk density. The estimated aerodynamic diameters ranged from 2.3 µm to 8.9 µm. The in silico pulmonary deposition profiles indicated, for some formulations, that a significant fraction of the particles would be deposited in the deeper lung regions. CONCLUSIONS: Statistical analysis demonstrated that not only drug loading but also the aerodynamic diameter of the microparticles is greatly affected by the preparation conditions. Overall, the results indicated that the low-density azithromycin-loaded microparticles with a relatively high respirable fraction may be obtained for the local treatment of lung infections.

Development of low density azithromycin-loaded polycaprolactone microparticles for pulmonary delivery.

CONTEXT: The development of low-density polymeric microparticles may be a useful approach to deliver antibiotics such as azithromycin into the lung. OBJECTIVE: The aim of this study was to develop azithromycin-loaded low density polycaprolactone microparticles by the double emulsion/solvent evaporation method. MATERIALS AND METHODS: Microparticles were prepared and characterized according to their physicochemical properties, drug loading, and drug release profiles. A full 2(3) factorial design was used to evaluate the effect of some independent variables on the drug loading and aerodynamic diameter of the particles. An in silico pulmonary deposition model was used to predict the lung deposition profiles for the formulations. RESULTS AND DISCUSSION: The resulting particles presented drug loading up to 23.1% (wt%) and mean geometric diameters varying from 4.0 µm to 15.4 µm. Bulk and tapped densities were low, resulting in good or excellent flow properties. SEM images showed spherical particles with a smooth surface. However, hollow inner structures were observed, which may explain the low values of bulk density. The estimated aerodynamic diameters ranged from 2.3 µm to 8.9 µm. The in silico pulmonary deposition profiles indicated, for some formulations, that a significant fraction of the particles would be deposited in the deeper lung regions. CONCLUSIONS: Statistical analysis demonstrated that not only drug loading but also the aerodynamic diameter of the microparticles is greatly affected by the preparation conditions. Overall, the results indicated that the low-density azithromycin-loaded microparticles with a relatively high respirable fraction may be obtained for the local treatment of lung infections.