*3.1. Materials*

Praziquantel (PZQ) Ph. Eur. grade ((11bRS)-2-(Cyclohexylcarbonyl)-1,2,3,6,7,11b-hexahydro-4-Hpyrazino[2,1-a]isoquinolin-4-one) was kindly donated by FATRO S.p.A., Ozzano Emilia, Bologna, Italy. PZQ impurities: Impurity A (2-Benzoyl-1,2,3,6,7,11b-hexahydro-4-H-pyrazino[2,1-a]isoquinolin-4-one) and impurity B (2-Cyclohexanecarbonyl-2,3,6,7-tetrahydro-pyrazino[2,1-a]isoquinolin-4-one) were of Ph. Eur. grade and were purchased from Endotherm Gmbh (Saarbruecken, Germany). Povidone (Kollidon K30, PVP K30) was supplied by BASF (Ludwigshafen, Germany) while Gelucire 50/13 was kindly supplied by Gattefossè (Milan, Italy).

### *3.2. Preparation of Activated Materials by Neat Grinding*

These experiments were performed in a vibrational mill-Retsch MM400 (Retsch GmbH, Haan, Germany), which was equipped by two screw-type zirconium oxide jars, each with a capacity of 35 mL. A ceramic material like zirconium oxide was selected, due to its high density (5.9 g/cm3). Three zirconium oxide spheres of 15 mm (weighing 10.72 g) were used as the milling media.

In particular, to obtain the PZQ crystalline polymorphic form, Form B, PZQ was ground by itself. The vibrational frequency was set at 20 Hz, for 240 min, without interruption. The amount of powder to be introduced in the milling jar was determined to be 0.800 g per jar, and no cooling was provided to the grinding jar, during room temperature milling. These process parameters were necessary to obtain the new polymorphic form, Form B, which were selected on the basis of our previous work [11]. A 20 Hz vibrational frequency was applied for a duration of 4 h. The experiment was performed, twice, to obtain enough material for both the experimental analysis and the spray congealing process.

For the preparation of the coground system, PZQ and PVP were manually gently mixed in an agate mortar, in a 1:1 drug-to-polymer weight ratio, for the standardized time of 3 min (batch size ranging about 1g). On the basis of previous experiences [11], the amount of powder to be introduced

in the milling jar was determined to be 1.072 g per jar, and a vibrational frequency and a milling time of 60 min were set. Prior to milling, the jars containing the samples were immersed in liquid nitrogen for 1 min; re-cooling of the milling jars with liquid nitrogen for 1 min, was performed, every 15 min of milling. The experiment was repeated four times, to obtain enough material for both the experimental analysis and the spray congealing process.

Post milling, all samples were collected and stored in glass vials in the dark, in a desiccator, over anhydrous calcium chloride, at 25 ◦C, for further characterization and processing.

For comparison purposes, the properties of the raw PZQ and binary physical mixtures (prepared in the same agate mortar, by manually mixing PZQ and PVP), were investigated.

### *3.3. Preparation of Microparticles by Spray Congealing*

Four batches of microparticles (indicated as MPs A–MPs D) using Gelucire® 50/13 as the carrier were produced by the spray-congealing technology, using an external-mix two-fluid atomizer, called Wide Pneumatic Nozzle (WPN) [18]. Gelucire® was heated up to about 10 ◦C, above its melting point. The active material (whose amount was determined in order to have a final percentage of PZQ equal to 15% *w*/*w*) was added to the molten carrier, as a powder, and magnetically stirred. The suspension or solution obtained was then loaded into the feeding tank of the spray congealing apparatus. The temperature of the feeding tank of the nozzle and the inlet air pressure were set at 60 ◦C and 3 bar, respectively. The atomized molten droplets hardened during the fall into a cylindrical cooling chamber, which was held at room temperature. Finally, the microparticles were collected from the bottom of the cooling chamber and stored in polyethylene closed bottles, at 25 ◦C. The batch size was 10 g for each formulation. The composition of the produced microparticles is reported in Table 2.

### *3.4. HPLC Analysis*

HPLC analysis was performed for the quantification of both PZQ and the related impurities (or other detectable related products), in the activated samples, after the milling process and for the determination of the real drug content in the microparticles. HPLC analysis was also used to calculate the solubility and the dissolution properties of the different samples. The method was adapted from literature [26] and have been already validated and employed for PZQ quantification in previous studies [11–13]. Briefly, the HPLC system consisted of two mobile phase delivery pumps (LC-10ADvp, Shimadzu, Japan) and a UV–vis detector (SPD-10Avp, Shimadzu, Japan). An autosampler (SIL-20A, Shimadzu, Japan) was used to inject samples (20 μL) onto a Kinetex 5 μm C18 column (150 mm × 4.60 mm; Phenomenex, Bologna, Italy). The mobile phase was methanol and water at a ratio of 65:35 V/V, the flow rate and the wavelength of the UV detector were set at 1 mL/min and 220 nm, respectively. The linear calibration curve of the PZQ was obtained in the range of 0.4–40 mg/L (r2 = 0.99985). The retention time of PZQ was about 5.5 min and the run time was set at 12 min. The PZQ standard solution was prepared by dissolving 10 mg of PZQ in 20 mL of methanol and diluting 1:200 in the mobile phase, in order to have a final standard PZQ concentration in solution of 2.5 mg/L. The calibration curve of the PZQ specified impurities (impurity A and B according to the PZQ monograph) [27], was obtained in the range of 0.01–1 mg/L (r2 = 0.99927 and 0.99941, for impurity A and B, respectively). The retention time of the impurities were at 3.45 min and 11.2 min.

PZQ content was determined by dissolving a variable quantity of the specific sample (PZQ milled, PZQ:PVP cryo-comilled and loaded-microparticles) accurately weighed in 20 mL of methanol. In the case of the milled samples, the obtained solution was then diluted 1:200, in the mobile phase, corresponding to about 2.5 mg/L of PZQ, while for the microparticles, the solution was stirred for 24 h, protected from light, to assure a complete solubility. Finally, the solution was diluted in the mobile phase 1:100. For all samples prior to injection, the solutions were filtered through a 0.2 μm membrane filter and the drug content was assayed by HPLC. Each sample was analyzed in triplicates and the mean of the sum of the peak responses of PZQ was then calculated and have been reported along with the SD. Moreover, for the activated materials, the PZQ recovery was expressed as the percentage

of PZQ, with respect to the sum of all peaks (PZQ and related impurities or other detectable related products). However, for the microparticles, the encapsulation efficiency (%) was then calculated by dividing the experimental drug content with the theoretical one and then multiplying it by 100.

### *3.5. Solubility and Dissolution Studies*

Solubility and dissolution studies were carried out for both raw and activated materials and for the microparticles. For the solubility test, an excess amount of sample was added to the 10 mL of distilled water. The samples were magnetically stirred for 48 h, at 20 ◦C, and were protected from light by means of an aluminum foil, throughout the experiment. After equilibrium, the samples were centrifuged at 10,000 rpm, for 20 min, and the supernatant was filtered through a 0.20 mm membrane filter. After diluting the samples in a ratio of 1:200 in the mobile phase, they were finally analyzed by HPLC. The measurements were performed in triplicates, for each formulation, and the mean ± SD was reported. The statistical assessment of the obtained values was performed using one-way ANOVA, while comparison between means was performed using the *t*-Test. Differences were considered statistically significant for *p* values < 0.01.

In vitro dissolution studies of all samples were performed in 1000 mL of water, maintained at 37 ± 0.5 ◦C, and stirred at 100 rpm, using a paddle apparatus (Erweka DT800, Heusenstamm, Germany). Sink conditions were ensured by considering the PZQ water solubility at 37 ◦C of 215.0 ± 4.9 mg/L, as found by Trastullo et al. [7], and different amount of samples (corresponding to 16 mg of PZQ) were added to the vessel, according to their composition. Then aliquots of 2 mL were withdrawn at specified times, through a 8 μm filter, in order to only collect the dissolution media and leave the formulation in the vessel. At each sampling time, the PZQ content was assayed by the HPLC. Withdrawn samples were replaced with an equal volume of fresh medium. The dissolution tests were performed, at least in triplicates, and the mean ± SD was reported. Comparison between drug release profiles from the pellets were carried out, using the similarity factor (*f* 2).

$$f\_2 = 50 \ast \log \left\{ 1 + \left[ \frac{1}{n} \ast \sum\_{t=1}^{n} (R\_t - T\_t)^2 \right]^{-0.5} \ast 100 \right\}$$

where *n* is the sampling number, *Rt* and *Tt* are the cumulative percentage drug dissolved of the reference and the test products at each time point *t*. For the *f* 2 values calculation, sampling number obtained within 20 min of the dissolution test were considered. The similarity factor fits the result between 0 and 100. The two drug release profiles were similar if the *f* 2 was greater than or equal to 50.

### *3.6. Wettability Studies*

The measurements of contact angle were carried out according to the sessile drop method on compressed non-disintegrating disks, using deionized water as a wetting liquid, as previously reported for Gelucire-based matrices [28]. Disks were prepared by compressing the mixtures in a manual press Perkin Elmer, imparting a force of 1 tons for 1 min. The flat tablets produced were then analyzed with the Drop Shape Analysis System (Krüss DSA 30, Krüss GmbH, Germany), using a single drop of purified water (25 μL). The contact angle (between the disk and the drop) measurements, performed in triplicates, were taken after 10 s. The pure PZQ and milled PZQ were analyzed and the mean of at least three determinations, was calculated.

### *3.7. Viscosity Measurements*

The viscosity of the molten mixtures, heated to the temperature set for the spray congealing process (60 ◦C), was measured with a Brookflied DVzT viscosimeter (Ametek GmbH, Lorch, Germany) using a spindle number RV04 and a rotating speed of 200 rpm.

### *3.8. Scanning Electron Microscopy (SEM)*

Powder samples (pure PZQ, polymorphic form B and cryo-coground) were metallized with S150A Sputter Coater (Edwards High Vacuum, Crawley, West Sussex, UK) and then observed under a scanning electron microscope Leica Stereoscan 430i (Leica Cambridge Ltd., Cambridge, UK).

### *3.9. Environmental Scanning Electron Microscopy (ESEM)*

The shape and surface characteristics of the microspheres were observed by environmental scanning electron microscopy (ESEM) (Quanta 200 FEI, FEI Company, Czech Republic).

### *3.10. Particle Size Analysis*

Particle size measurements of the starting PZQ, Form B, cryo-coground and corresponding PM, were carried out, using a laser diffractometer (Malvern Mastersizer 2000, Malvern, UK). Before analysis, about 10 mg of each sample were dispersed by sonication in 100 mL of water containing 0.1% of polisorbate 80; sample containing PVP were dispersed in silicon oil (Silico DC 245 DOW Corning, Biesterfeld Spezialchemie GmbH, Hamburg, Germany).

Size distribution of microparticles was evaluated by sieve analysis, using a vibrating shaker (Octagon Digital, Endecotts, London UK), and a set of six sieves, ranging from 75 to 500 μm (Scientific Instrument, Milan, Italy).

### *3.11. Differential Scanning Calorimetry (DSC) Studies*

Raw materials, activated materials, and MPs were analyzed by DSC, using a Perkin Elmer DSC 6 (Perkin Elmer, Beaconsfield, UK), with nitrogen as the purge gas (20 mL/min). The instrument was calibrated with indium and lead, for temperature, and with indium for the measurements of enthalpy. About 8–9 mg of the sample were placed in an aluminum pan and heated from 30 to 180 ◦C, at a scanning rate of 10 ◦C/min. Residual crystallinity (%) was then calculated by the measurements of the enthalpy of fusion ( ΔH) of the PZQ, using the following equation: Residual crystallinity (%) = (Δ Hsample × 100)/ Δ HPZQ.

### *3.12. Hot Stage Microscopy (HSM) Analysis*

HSM was performed on raw PZQ, PZQ:PVP physical mixture, activated materials and MPs, using a hot stage apparatus (Mettler-Toledo S.p.A., Novate Milanese, Italy), under a Nikon Eclipse E400 optical microscope. Solubilization processes, phase transitions, or polymorphic changes occurred during the heating (from 25 to 200 ◦C, scanning rate 10 ◦C/min) were observed and images were captured by means of a Nikon Digital Net Camera DN100. The magnification was set at 10×.

### *3.13. X-Ray Powder Diffraction Studies (PXRD)*

PXRD patterns were recorded using a Bruker AXS D5005 X-ray Diffractometer (Karlsruhe, Germany) with Cu-K α radiation (1.5418 Å), monochromatized by a secondary flat graphite crystal. The analyses were performed in duplicates, using a current of 30 mA and the voltage was set at 40 kV. The powder samples were scanned in the range of 3◦–35◦ of the 2θ angle, steps were of 0.05◦ of 2θ, and the counting time was of 5 sec/step. The samples subjected to the analysis were the following: Gelucire 50/13, raw PZQ, polymorphic form B, MPs B and MPs D, and the corresponding PMs.

### *3.14. Fourier Transform-Infrared Spectra (FT-IR) Analysis*

Studies of infrared spectra of the excipients (PVP and Gelucire 50/13), raw PZQ, PZQ:PVP physical mixture, the activated materials, and the MPs were conducted with an IR spectrophotometer (Jasco FT-IR A-200, Pfungstadt, Germany), using the KBr disc method. The samples were mixed with KBr and compressed into a tablet (10 mm in diameter and 1 mm in thickness), using a hydraulic press

(Perkin Elmer, Beaconsfield, UK), at 3 tons, for 3 min. The scanning range was 650–4000 cm<sup>−</sup><sup>1</sup> and the resolution was 1 cm<sup>−</sup>1.

### *3.15. Physical Stability During Storage*

In order to check possible modifications of the solid state within time, PXRD analyses was done, and dissolution tests of the MPs were carried out after 1 year.

### *3.16. In Vitro Studies on S. Mansoni*

Newly transformed Schistosomula (NTS) were obtained using the mechanical transformation method [29] and maintained at 37 ◦C, 5% CO2, for 24 h before the experiments. One hundred NTS were placed in each well of a 96 well plate, containing Medium 199, supplemented with 5% iFCS and 1% penicillin/streptomycin. A concentration of 100 μg/mL PZQ and Form B and MPs B were used and serially diluted 1:3, up to 1.23 μg/mL. The plate was then incubated at 37 ◦C, 5% CO2, and monitored at 24 h, 48 h, and 72 h. The assay was performed with biological triplicates, for each concentration. At each time point the NTS were assessed, microscopically, using a viability scale (3 = motile, no changes to morphology; 2 = reduced motility or some damage to the tegument noted; 1 = severe reduction to motility or damage to the tegument observed; 0 = dead). NTS incubated in the highest concentration of DMSO served as the negative controls.

Adult *S. mansoni* were collected from the hepatic portal and mesenteric veins of the infected mice. In a 24-well plate, 2–3 worm pairs were placed in the culture medium (RPMI supplemented with 5% iFCS and 1% penicillin/streptomycin) with 0.33, 0.11, and 0.037 μg/mL of the test compounds for 24 h, 48 h, and 72 h, at 37 ◦C, 5% CO2, in biological duplicates. Effects were assessed microscopically, as described above. Adult worms exposed to the maximum concentration of DMSO served as the negative control. IC50 values for both NTS and adult worms were calculated using the CompuSyn software (ComboSyn Inc., Paramus, NJ, USA).
