Formulation and optimization of polymeric agglomerates of Bosentan monohydrate by crystallo-co-agglomeration technique

The polymeric spherical agglomerate of Bosentan monohydrate was prepared by crystallo-co-agglomeration technique, for enhancing the micrometric properties and solubility of the drug. The agglomerates were developed using two distinct solvents, DCM as a good solvent and bridging liquid and water as a weak solvent, respectively. Hydrophilic polymer like HPMC K100M is used as a hardening agent which gives mechanical strength to the agglomerates, and PEG6000 is used as a wetting agent. Other excipients like talc which was used as a size enhancer, and PVA was used as an emulsifier agent. The formulation was optimized by Box–Behnken design. The concentration of talc and PEG6000, as well as rotation speed, was considered as independent variables. The particle size, angle of repose, and % drug content were used as dependent variables to investigate the effect of independent variables on dependent variables. The spherical crystal agglomerates were subjected to various physicochemical evaluations. The results revealed that as the concentration of talc and PEG6000 increases, the sphericity and particle size of the agglomerates increase, and at a low agitation, speed agglomerates become more spherical and coarser, which is confirmed by FESEM. The characterization like FTIR confirms no interaction with excipients, while XRPD confirms the polymorphic changes, and gas chromatography (GC) confirms the concentration of residual organic solvents in PDE limits. The optimized formulation of SAs showed a good angle of repose which is 30.33 ± 0.35, and the % cumulative drug release at 20 min was 94.14 ± 0.628%. Finally, the FDTs of the optimized batch were prepared. The comparison of the in vitro release study of pure drugs with agglomerates and fast dispersible tablets of agglomerates confirms the solubility improvement. Finally, it can be concluded that the polymorphic crystal agglomerates enhance the solubility and micrometric properties of Bosentan monohydrate.


Background
The process of obtaining larger particles through agglomeration during crystallization is referred to as spherical crystallization of drugs. Spherical crystallization is a novel agglomeration technique that transforms crystalline drugs into compact spherical crystals. Many pharmaceutical drugs have problems due to inappropriate physical and mechanical properties, as well as low aqueous solubility. The micromeritic properties of drug particles, such as shape and size, are critical for the formulation of solid high-dose units. Spherical agglomeration is a technique used to improve the physical and micromeritic properties of drugs with low bioavailability and poor physical properties such as flowability, compressibility, wettability, solubility, and dissolution rate. The polymorphism of the converted crystal agglomerates may improve drug bioavailability. A drug solution in a good solvent is poured into a poor solvent that is miscible with the good solvent in the spherical crystallization method. Otherwise, crystals would form immediately because the affinity between the solvents is greater than the affinity between the good solvent and the compound. A third solvent, known as the bridging liquid, is also added in a small amount to the spherical crystallization method and acts as an interparticle binder, promoting crystallization rate. The bridging liquid collects the crystals suspended in the system by forming liquid bridges between the crystals due to capillary negative pressure and interfacial tension at the solid-liquid interphase. It should not be miscible with the poor solvent and should wet the precipitated crystals. Surfactants are typically avoided because the strength of liquid bridges is proportional to the interfacial tension between the bridging liquid and solid. As the term crystallo-co-agglomeration (CCA) implies, crystallization occurs in the presence of an external inert substance or diluents. The CCA was effective for low-dose active ingredients when combined with another active ingredient or diluent such as talc, sodium starch glycolate, or starch. Another pharmaceutical entity has been used as a substrate by some researchers in the development of mixed-dose spherical crystals.
Bosentan monohydrate is an endothelin receptor antagonist that is used to treat pulmonary arterial hypertension patients. It belongs to the BCS class-2 of drugs. As a result, it is poorly water-soluble and has a 45% bioavailability. Endothelin-1 (ET-1) is a neurohormone, and the endothelium and vascular smooth muscle contain two types of ETA and ETB receptors. Because pulmonary artery hypertension causes serious liver damage, a liver function test must be carried out during treatment. Bosentan monohydrate is metabolized in the liver. Longterm use of Bosentan monohydrate causes liver damage due to dose accumulation. Formulating spherical agglomerates may increase the drug's solubility, allowing the dose of Bosentan monohydrate to be reduced. Crystal-coagglomeration (CCA) is a straightforward and efficient method for producing polymeric Bosentan monohydrate spherical agglomerates. Formulation of spherical agglomerates using various hydrophilic polymers improves drug solubility, dissolution rate, micromeritic properties, and physical properties for use in direct compression tablets (Feeley et al. 1998;Villiers 1995).

Solubility of the drug in different solvents
Solubility of the drug in different solvents was done by quantitative method. The solubility determination of Bosentan monohydrate in various solvents was performed by adding Bosentan monohydrate in increments of 1 mg until it failed to dissolve further in the fixed 1 ml of solvent. The amount of drug dissolved in solvents was determined. The experiment was conducted in triplicate.

Selection of solvent system and excipients for the formulation of agglomerates by CCA technique
Based on the solubility and miscibility study, dichloromethane (DCM) was selected as a good solvent as well as a bridging liquid. Furthermore, water was selected as a poor solvent for the preparation of spherical agglomerates. The various hydrophilic polymers like PEG6000, PVP K30, Poloxamer 407, Poloxamer 188, and β-cyclodextrin were added in the poor solvents. The effect of different polymers was observed on the formulation of spherical agglomeration (Kumar et al. 2008;Patil et al. 2014).

Method for preparation of agglomerates
The 500 mg of Bosentan monohydrate dissolved in DCM was containing 1% talc and 1% of HPMC K100M. The poor solvent was containing PVA as an emulsifier which was added to the good solvent under continuous stirring at a specific temperature of 10 °C for 45 min until the agglomerates were prepared. Then, crystal agglomerates were filtered and washed through the distilled water and dried at room temperature.

Compatibility of drug and excipients by FTIR
The interaction between drug and excipients can be identified by Fourier-transform infrared (FTIR) spectroscopy. Potassium bromide was mixed with the sample in the ratio of 1:1 and pellets were prepared using KBr pellet press and the spectrum was taken using FTIR. The FTIR spectrum of Bosentan monohydrate was compared with the mixture of the solvents and polymers/excipients. The disappearance of the Bosentan monohydrate peak or shifting of the peak of the functional group was studied.

Optimization of formulation by using Box-Behnken design
To obtain optimized formulation with a minimum number of runs, the Box-Behnken design was used utilized to study the effect of independent variables on the dependent variables in the formulation of spherical agglomerates (SAs). Stat-Ease design expert v7.0.0 software was used for optimization. In the formulation of spherical agglomerates, the amount of HPMC K100M and PVA was fixed. Independent variables like the concentration of talc (A), the concentration of PEG6000 (B), and rotation speed (C) were selected. Dependent variables like particle size (Y1), angle of repose (Y2), and drug content % (Y3) were evaluated for the effect of independent variables. The level and responses of both independent and dependent variables are listed in Table 1. Based on the 3 levels, 13 batches were prepared, and the composition of batches is shown in Table 2.

Evaluation of model/checkpoint analysis
Evaluation of the model was done by checkpoint analysis. To evaluate the dependability of the model and to check the effectiveness of the established contour plot and reduced polynomial equation in the development of spherical agglomerates, two checkpoint batches were prepared to evaluate by comparing the experimental and predicted value of responses. Each formulation was fabricated three times, and the average was calculated.

Preparation of optimized formulation based on the desirability function
Optimization was carried out to ascertain the level of independent variables (X1, X2, and X3) that would provide data of Y1, Y2, and Y3. At the time of developing the formulation, the response has been united to design the product of the required attribute. The main function of the desirability was to join every response in a single experiment and provide the probability of predicting the highest level for independent variables. The last optimized formulation suggested by the software was prepared, and parameters were compared to the expected value given by the software.

Practical yield
The practical yield of the prepared formulation was calculated to determine the loss of product during the process. The % yield of the spherical agglomerates was calculated by using the following equation (Pawar et al. 2004;Amorim et al. 2010;Kim et al. 2000).

Angle of repose
The angle of repose was performed by the Tilting box method. A fixed quantity of prepared SAs was weighed and placed within a box with a transparent side to observe the angle of the slide. Then, the box was slowly tilted at a rate of approximately (3 degrees/second) and tilting was stopped, when the powdered drug begins to slide at a particular angle which was considered the angle of repose (Pawar et al. 2004;Amorim et al. 2010;Kim et al. 2000).

Carr's index
For the determination of Carr's index, a fixed quantity of prepared SAs was weighed and transferred into the 10-ml measuring cylinder and the volume (bulk density) was noted down. The cylinder was tapped 100 times, and then, decrease in volume (tapped density) was noted down. Carr's index was calculated using the following equation (Pawar et al. 2004;Amorim et al. 2010;Kim et al. 2000).

Hausner ratio
The Hausner ratio was calculated by the division of bulk density and tapped density and measured Hausner ratio for SAs compared with the criteria for Hausner's ratio (Pawar et al. 2004;Amorim et al. 2010;Kim et al. 2000).

Drug content
The drug content of spherical agglomerates was carried out by triturating the agglomerates equivalent to 50 mg of Bosentan monohydrate and dissolving into dichloromethane. The solution was diluted 1 ml in 10 ml, and the drug content was determined at 269 nm using a UV spectrophotometer (Li and Ge 2012).

Characterization of optimized spherical agglomerates Optical microscopy
For the determination of the shape of the particle, the spherical agglomerates were observed under optical microscopy (Leonardi et al. 2015).

Field emission scanning electron microscopy (FESEM)
FESEM provides topographical and elemental information at a magnification of 10 × to 300000x, with virtually unlimited depth of field. Compared with convention SEM, FESEM produces clearer, less electrostatically distorted images with spatial resolution down to 1½ nm three to six times better (Leonardi et al. 2015).

Fourier-transform infrared spectroscopy (FTIR)
The FTIR spectral measurements were performed using Shimadzu (model 8033) at ambient temperature for prepared optimized spherical agglomerates. Spherical agglomerates of Bosentan monohydrate were dispersed in KBr, and the mixture was transformed into pellets by compressing them with hydraulic pressure. Then, they Carr's index = Tapped density − Bulk density Bulk density × 100 Hausner ratio = Tapped density Bulk density were placed in the compartment and analyzed in the range of 400-4000 nm (Leonardi et al. 2015;Sirianni et al. 1969).

X-ray powder diffraction (XRPD)
The polymorphic changes in the drug are important as it affects its dissolution rate. For evaluating the changes in the crystalline nature of the pure drug and SAs, the X-ray powder diffraction (XRPD) study was performed. XRPD analysis was done for pure drug and SAs with the help of using an X-ray diffractometer (Leonardi et al. 2015;Sirianni et al. 1969).

Gas chromatography (GC)
The estimation of the amount of the residual solvents is mandatory to validate while using the organic solvents in any formulation because it may produce toxicity in humans if it is not in the permitted daily exposure (PDE) limits described as per ICH. For the estimation of the amount of the residual solvent DCM in prepared spherical agglomerates, Auto system 2 (version 6.3.2.0646) was used. (Kawashima et al. 1981).

Formulation of fast dispersible tablets (FDTs) of polymeric Bosentan monohydrate spherical agglomerates
The prepared polymeric Bosentan monohydrate spherical agglomerates, equivalent to 45 mg of the drug, will be taken and mixed with other pharmaceutical excipients (glidant, diluent, super disintegrant, lubricant, etc.,) and directly compressed into tablets (Patil and Upadhye 2022;Chow and Leung 1996). The composition of fast dispersible tablet with optimized SAs of Bosentan monohydrate is shown in Table 3.

Evaluation of prepared FDTs of SAs Hardness or crushing strength test
This test measures the degree of force in kilogram or pound required to fracture the tablet. Besides the concentration of binder used and the compression force, the hardness of the tablet also depends upon the characteristic of granules to be compressed, the type and concentration of lubricant used, and the space between upper and lower punches at the time of compression. The hardness of the tablet was tested by the Monsanto hardness tester. The tablet was placed in the tester, and the pressure required for the tablet to be broken was noted (Hansen and Kleinebudde 2022).

Thickness
Tablet thickness is measured with a vernier caliper, and the thickness of a tablet should be controlled within ± 5% variation of a standard value depending on the size of the tablet.

Friability
The friability test is carried out by the Roche friability tester to evaluate the tablet's ability to withstand abrasion in handling, packaging, and shipping. Pre-weighed tablet was placed in Roche friabilator and rotated at 25 rpm for 4 min for 100 revolutions. The dropping distance of the tablet was 6 inches. Then, the tablet was dusted and weighed again and the % friability was calculated. It can be determined by following the formula (Patil and Upadhye 2022).
where W 1 = weight of tablet before test and W 2 = weight of tablet after test

Weight variation
The weight variation test of the tablet can be done by weighing 20 individual tablets, and the average weight of the tablet was calculated. Not more than 2 of the individual tablet weight deviate from the average weight by more than the ± 7.5% deviation, as per the I.P. limit (Patil and Upadhye 2022).

Drug content
The drug content of FDTs was carried out by triturating 10 tablets, weighing the powder of 62.5 mg equivalent to Bosentan monohydrate, and dissolving into methanol. The solution was diluted 1 ml in 10 ml, and the drug content was determined at 269 nm using a UV spectrophotometer (Chow and Leung 1996).

Disintegration time
The disintegration time is the time taken by the tablet to disintegrate into the particles. The 6 tablets were placed in the disintegrating machine, and the (0.1 N HCl) media was poured into the jar. The tablet was dropped into the jar, and the time for disintegrating of the tablet was noted %Friability = W 1 − W W 1 × 100 (Patil and Upadhye 2022;Chow and Leung 1996;Hansen and Kleinebudde 2022).

In vitro dissolution study of FDTs
The in vitro dissolution study of FDTs was carried out using the USP dissolution apparatus type II. The study was carried out by using 900 ml of 7.4 pH phosphate buffer with the addition of 1% SLS. At the end of 30 min %, drug release was checked. The temperature of the dissolution medium was maintained at 37 ± 0.5 °C, and the paddle was rotated at 50 rpm. 5 ml of aliquots was withdrawn at regular time intervals of 5 min and analyzed by UV spectrophotometer (Patil and Upadhye 2022;Chow and Leung 1996;Hansen and Kleinebudde 2022).

Accelerated stability study for the FDT of SAs
The frontline stability chamber was used for the study of accelerated stability. The tablet was wrapped in an aluminum foil and placed into the stability chamber at 40 ± 2 °C/75 ± 5% relative humidity. At the predetermined time interval of 0, 15, and 30 days, the tablet was evaluated for the different physicochemical parameters (Patil and Upadhye 2022;Chow and Leung 1996;Hansen and Kleinebudde 2022).

Solubility
The solubility of the drug in the various solvents was done by quantitative method. Solubility data are shown in Table 4. From the data, it can be concluded that the solubility of Bosentan monohydrate is 0.570 ± 1.6 gm in 1 ml of dichloromethane.

Selection of excipients for the formulation of agglomerates by CCA technique
The technique was performed by using various hydrophilic polymers. The effect of various polymers was observed on the formation of agglomerates which is shown in Table 5.

Drug excipient compatibility study
The drug excipient compatibility was studied by comparing FTIR spectra of pure drug with FTIR spectra of a mixture of drug and excipients, which are shown in Figs. 1 and 2.

Optimization of parameters by Box-Behnken design
The effect of formulation and process parameters has its importance in the formulation of SAs. According to the design, 13 formulations were prepared with variations in the concentration and RPM of the independent variable. The data in Table 6 show the effect of independent variables on dependent variables.

Effects of independent variables on dependent variables
The effects of independent variables on dependent variables were explained by response surface plot and contour plot. Figures 3, 4 and 5 show how the concentration of talc, the concentration of PEG6000, and speed affect the particle size, angle of repose, and drug content of spherical agglomerates.

Checkpoint batches
The checkpoint batches were prepared to find the predictability of the model. Both predicted values and actual values of both the responses were compared. Two checkpoint batches were prepared and studied for particle size, angle of repose, and drug content. It was found that both the values are by each other as shown in Table 7.

Formulation of optimized batch
Based on the criteria like high drug content, small particle size, and low angle of repose, 18 solutions were suggested by software with a different desirability. The optimized batch having desirability 0.823 was selected.
The composition and results of the optimized batch are shown in Table 8, in which 500 mg of the drug, 15 mg of HPMC K100M, and 15 mg of PVA were constant.

Evaluation of prepared spherical agglomerates
Physical properties of spherical agglomerates As per the data shown in Table 9, the physical properties of prepared agglomerates were compared with pure drug.

SAs characterization
Optical microscopy For particle size determination, the spherical crystal agglomerates were observed under  Fig. 6.
FESEM The surface morphology result in Fig. 7 shows that the pure Bosentan monohydrate powder was in the form of rod-shaped crystals, which leads to its poor flowability and compression difficulties, whereas the prepared agglomerates of Bosentan monohydrate are having improved surface and spherical shape and they are formed by tiny crystals which are compacted closely in a spherical shape, which enables them to flow easily.
FT-IR IR spectra of an optimized batch of SAs of Bosentan monohydrate are shown in Fig. 8. The wavenumber obtained in the IR spectra of the drug was correlated with the functional group of Bosentan monohydrate.
XRPD Figure 9 represents the XRPD diffractogram of Bosentan monohydrate and SAs of Bosentan. While comparing the X-ray diffractogram of pure drug and SAs of Bosentan, it was observed that diffractogram of the pure drug shows a highly intense peak when the drug is in a crystalline form and X-ray diffractogram of Bosentan  SAs, and the intensity of the peak is reduced because of changes in crystalline behavior to amorphous form which indicates the improvement in the solubility (Dangre et al. 2017;Montenegro et al. 2011).

GC (gas chromatography)
A GC-headspace chromatogram of DCM and Bosentan monohydrate agglomerates was observed. The detector voltage (y-axis) was plotted as a function of time (x-axis). The identity of each peak was determined by injecting samples and noting their retention times.

Comparison of in vitro dissolution study of pure drug, SAs of Bosentan monohydrate, and FDTs
The comparison of in vitro dissolution studies of pure drug, an opti-   mized formulation of spherical agglomerates, and FDTs of spherical agglomerates are shown in Fig. 10.

Stability study
The stability studies for the FDTs of SAs show no remarkable changes in the physicochemical properties like hardness, thickness, diameter, friability, weight variation, drug content, disintegration time, and % drug release after 3 months. The physicochemical evaluation is shown in Table 10.

Discussion
Based on the drug's solubility in various solvents, DCM was chosen as a solvent because it is immiscible with water and can be used as both a solvent and a bridging liquid. The PEG6000 exhibits spherical-shaped agglomerates among all polymers, whereas other polymers exhibit    and C are significant and for Y3, model terms C and C 2 are significant. Adjusted R-squared values of responses Y1, Y2, and Y3 were 0.7983, 0.8145, and 0.8993 which was a reasonable agreement with predicted values. From the surface response plot and contour plot, it was observed that as the concentration of talc and PEG6000 increases, the particle size was increased up to optimum level and  after that, particle size decreases with increasing the concentration of talc and PEG6000. The particle size and angle of repose are proportional to each other; as the particle size increases, so does the angle of repose. The greater the angle of repose, the poorer the flow properties, demonstrating that the angle of repose is inversely proportional to the concentrations of talc and PEG6000. The speed also influences particle size; at low speeds, particle size increases, while at high speeds, particle size decreases, affecting the angle of repose. Because talc is used to harden the particles, the concentration of talc had an effect on the drug content. Agglomerates with a higher concentration of talc have a low drug content, whereas agglomerates with a low concentration of talc have a high drug content. The statistical evaluation of all the obtained data was carried out by analysis of variance (ANOVA) using DoE (design of expert) software. The results of ANOVA (p value) showed the effect of independent variables on dependent variables. The full polynomial model was obtained after regression analysis of all formulations. Checkpoint analysis was used to confirm the model. The physical properties of the optimized batch with desirability 0.823 were investigated further. When compared to pure drug, there were improvements in the angle of repose, compressibility index, and Hausner ratio. The conversion of fine powder into spherical agglomerates is responsible for the improvement in physical properties. The crystalline pure Bosentan monohydrate was clearly shown to be converted into spherical agglomerates in the microscopic image. The FESEM image of SAs and pure drug shows how the crystalline form of the drug transformed into a spherical shape, which was responsible for the improved flow properties. The FTIR spectra of optimized batch confirmed the presence of -OH stretching at 3657 cm −1 , N-H stretching at 1592 cm −1 , and S=O stretching at 1280 cm −1 . There were no discernible differences in the spectra of SAs when compared to the pure drug. The intensity of the peak was found to be high in the XRPD investigation due to the pure drug's crystalline nature. The crystalline behavior of the drug was transformed to an amorphous form after conversion into spherical agglomerates, as evidenced by the low intensity of the detected peak in the XRPD spectra of the agglomerates. The gas chromatography was used to identify the residual solvents. Following the injection of a standard DCM solution, a peak corresponding to DCM content appeared at a retention time of 3.25 min. At the retention time of 3.25, an extremely low intensity residual solvent peak of Bosentan monohydrate agglomerates was observed. It was discovered that the majority of the solvents were evaporated, with only a trace of solvent remaining in the agglomerates. The amount of DCM found was 2.58 ppm. According to ICH guidelines, the PDE amount of DCM was defined as 6 ppm, indicating that the amount of entrapped DCM in agglomerates was minimal and did not cause toxicity in humans. According to the ICH guidelines, the residual solvent concentration was found to be safe. The physicochemical properties of the spherical agglomerates containing fast dispersible tablets were investigated, while they were being prepared. The post-formulation parameters were found to be within acceptable limits. According to the in vitro drug release data, the percent cumulative drug release (CDR) of spherical agglomerates and FDT of SAs is improved at the end of 20 min compared to the pure drug.

Conclusions
In this current study, the polymeric spherical crystal agglomerates of Bosentan monohydrate have been developed successfully using the crystallo-co-agglomeration technique. The study reveals that the speed of agitation and the concentration of polymers and diluent play a remarkable role in the sphericity, micrometric properties, and % drug release of the agglomerates. The results of physical properties and in vitro dissolution study show that the micrometric properties and solubility of the polymeric Bosentan monohydrate spherical crystal agglomerates have been improved as compared to the pure drug because of the sphericity of the agglomerates, smoother surface, flowability, and conversion of crystalline form into an amorphous form. The optical microscopy and SEM results show the conversion of pure Bosentan monohydrate crystal form into spherical form and smoother surface. The XRPD studies show the reduced intensity of the peak which indicates the reduced crystallinity and leads to the amorphous form. The FTIR study shows no remarkable changes shown in the peak of SAs as compared to pure drug. The GC results show that the amount of the residual solvents in the SAs is under the limits of PDE. So, there is no chance of producing toxicity in humans. Finally, the spherical agglomerates have been converted into FDTs by direct compression technique.