3.5.6. Moisture content %

Moisture Loss %

Studying the moisture loss % was undertaken to evaluate the integrity and the physical stability of each OFDF [21]. The moisture loss % of all the prepared OFDFs ranged from 0.93 ± 0.02 to 1.24 ± 0.03%, as shown in Table 3. These results were in an acceptable range, indicating little moisture loss and the good physical integrity and stability of the prepared OFDFs [27]

#### Moisture Absorption %

The moisture absorption % of the films is important, because it influences the friability, mechanical strength, adhesive properties, disintegration, and dissolution behaviors of each film [16]. Table 3 shows that the moisture absorption % of all the prepared OFDFs were from 1.06 ± 0.08% to 8.73 ± 0.38%. It was noticed that the moisture absorption % of the OFDFDs was slightly increased when increasing the polymer concentration. This could have been because of the hydrophilic nature of the used polymers [54]. It was also noticed that the OFDFs made of CMC polymer showed a high water sorption; therefore, they seemed to be not suitable for use at high humidity, which makes the films sticky and unsuitable for this application [13].

#### 3.5.7. Mechanical Characteristics of the OFDFs

Tensile Strength

An ideal OFDF should exhibit an adequately high tensile strength value to be able to withstand normal handling. In spite of this, a very high value (very high rigidity) is not desired, because it could retard the drug release from the polymer matrix [29]. The prepared OFDFs had tensile strength values from 1.04 ± 0.11 Mpa to 15.5 ± 0.68 Mpa, as displayed in Table 3. Tensile strength values were analyzed using the following equation:

$$\text{Tensile strength} = 6.39 - 1.72 \times \text{X}\_1 + 1.25 \times \text{X}\_2 - 3.04 \times \text{X}\_3 \,\text{[1]} + 3.73 \times \text{X}\_3 \,\text{[2]} \tag{9}$$

where X<sup>1</sup> is the polymer type, X<sup>2</sup> is the polymer concentration, and X<sup>3</sup> is the plasticizer type. X<sup>3</sup> [1] represents the first plasticizer type (glycerol) and X<sup>3</sup> [2] represents the second plasticizer type (propylene glycol).

Table 3 and Figure 2a show that all the independent variables (X1, X2, and X3) had a significant impact on the tensile strength (rigidity) of all the OFDFs loaded with PX. It was obvious that the tensile strength values were significantly changed when changing the type of polymer used (X1). Where OFDFs prepared using pectin polymer showed significantly higher tensile strength values than those prepared with CMC polymer. This could be accredited to the differences in the nature and the molecular weight between the two polymers. These findings are in good agreement with that detected by Maher et al., who found that the type of polymer used (hydroxypropyl methyl cellulose and carboxymethyl cellulose) had a significant effect on the tensile strength of the prepared films [55].

Moreover, it was noticed that increasing the polymer concentration (X2) significantly increased the tensile strength values of the prepared OFDFs. This could have been because of the formation of a densely packed network of the used polymer chains at higher concentration, leading to formation of a stronger matrix. Bharti et al. stated similar findings, where they found that the tensile strength of the prepared films was increased by increasing the film's former concentration (hydroxypropyl methyl cellulose polymer) [33].

**Figure 2.** Response 3-D plots for the effect of polymer type (X1), polymer concentration (X2), and plasticizer type (X3) on (**a**) tensile strength, (**b**) elongation %, and (**c**) Young's modulus of the OFDFs loaded with PX nanosuspension. **Figure 2.** Response 3-D plots for the effect of polymer type (X<sup>1</sup> ), polymer concentration (X<sup>2</sup> ), and plasticizer type (X<sup>3</sup> ) on (**a**) tensile strength, (**b**) elongation %, and (**c**) Young's modulus of the OFDFs loaded with PX nanosuspension.

Percentage Elongation OFDFs should possess a large elongation percentage, in order to exhibit the desired flexibility and stretchability, which is important for facile handling and application of the film to the buccal cavity [15]. The percentage elongation values for all the prepared OFDFs were found to be from 6.03 ± 0.45% to 53.08 ± 1.28%, as displayed in Table 3. The following equation was used to analyze the percentage elongation values: On the other hand, the tensile strength of PX OFDFs prepared with different plasticizer types (X3) was increased in this order: PG > PEG 400 > glycerol. It can be noticed that OFDFs prepared with glycerol as a plasticizer showed the lowest tensile strength values. This could be because of the high hygroscopicity of glycerol, which causes humidity absorption and consequently gives softness to the prepared films and decreases the tensile strength values [56].

On the other hand, the tensile strength of PX OFDFs prepared with different plasticizer types (X3) was increased in this order: PG > PEG 400 > glycerol. It can be noticed that OFDFs prepared with glycerol as a plasticizer showed the lowest tensile strength values. This could be because of the high hygroscopicity of glycerol, which causes humidity absorption and consequently gives softness to the prepared films and decreases the tensile

Percent Elongation = 26.29 + 1.41 × X1 + 0.44 × X2 + 22.68 × X3 [1] − 12.57 × X3 [2] (10) Percentage Elongation

strength values [56].

where X1 is the polymer type, X2 is the polymer concentration, and X3 is the plasticizer type. X3 [1] represents the first plasticizer type (glycerol) and X3 [2] represents the second plasticizer type (propylene glycol). OFDFs should possess a large elongation percentage, in order to exhibit the desired flexibility and stretchability, which is important for facile handling and application of the film to the buccal cavity [15]. The percentage elongation values for all the prepared OFDFs were found to be from 6.03 ± 0.45% to 53.08 ± 1.28%, as displayed in Table 3. The following equation was used to analyze the percentage elongation values:

Percent Elongation = 26.29 + 1.41 × X<sup>1</sup> + 0.44 × X<sup>2</sup> + 22.68 × X<sup>3</sup> [1] − 12.57 × X<sup>3</sup> [2] (10)

where X<sup>1</sup> is the polymer type, X<sup>2</sup> is the polymer concentration, and X<sup>3</sup> is the plasticizer type. X<sup>3</sup> [1] represents the first plasticizer type (glycerol) and X<sup>3</sup> [2] represents the second plasticizer type (propylene glycol).

Table 3 and Figure 2b show that the percentage elongation of the prepared OFDFs was significantly affected only by the plasticizer type (X3). Where, the percentage elongation of the OFDFs prepared with glycerol had the highest percentage elongation values.

The increase in the OFDFs elongation percentage can be attributed to the fact that glycerol replaces the intermolecular bonds present between polymer matrixes with hydrogen bonds created between polymer and glycerol molecules. This disruption and reconstruction of polymer molecular chains allows greater chain mobility, resulting in decreasing the rigidity and providing flexibility and stretching to the films [57]. These findings are in accordance with those stated by Junmahasathien et al., who reported that glycerol was the best plasticizer for increasing the elongation percentage of the prepared films, in comparison with other plasticizers [58].

#### Young's Modulus

OFDFs should have low Young's modulus values, to exhibit the desired elasticity; whereby, high values of Young's modulus lead to the formation of stiff and brittle films [59]. All the prepared OFDFs showed Young's modulus values from 8.09 ± 0.15 to 383.66 ± 11.06 Mpa, as displayed in Table 3. The following equation was used to analyze the Young's modulus values:

$$\text{Young's modulus} = 234.50 - 3.64 \times \text{X}\_1 - 1.64 \times \text{X}\_2 - 224.60 \times \text{X}\_3 \text{ [1]} + 151.74 \times \text{X}\_3 \text{ [2]} \tag{11}$$

where X<sup>1</sup> is the polymer type, X<sup>2</sup> is the polymer concentration, and X<sup>3</sup> is the plasticizer type. X<sup>3</sup> [1] represents the first plasticizer type (glycerol) and X<sup>3</sup> [2] represents the second plasticizer type (propylene glycol).

Table 3 and Figure 2c show that the plasticizer type (X3) was the only factor significantly impacting the Young's modulus of the prepared OFDFs. Where, the OFDFs prepared using glycerol had the lowest Young's modulus values.

This could be attributed to glycerol being capable of dispersing between the spaces of the polymer chains, reducing their intermolecular attraction and, hence, providing flexibility to the films [60].

#### 3.5.8. In Vitro Disintegration Time

Disintegration time is a very important parameter for OFDFs, to indicate the onset of drug action. A low value of disintegration time, allows a faster release and absorption of the loaded drug through the buccal mucosa. The mean disintegration time of all the prepared OFDFs ranged from 17.09 ± 1.30 to 160.06 ± 4.20 s, as displayed in Table 3. The following equation was used to analyze disintegration time values:

Disintegration time = 75.57 + 45.41 × X<sup>1</sup> + 9.98 × X<sup>2</sup> − 11.92 × X<sup>3</sup> [1] − 3.36 × X<sup>3</sup> [2] (12)

where X<sup>1</sup> is the polymer type, X<sup>2</sup> is the polymer concentration, and X<sup>3</sup> is the plasticizer type. X<sup>3</sup> [1] represents the first plasticizer type (glycerol) and X<sup>3</sup> [2] represents the second plasticizer type (propylene glycol).

Table 3 and Figure 3a show that all the independent variables (X1, X<sup>2</sup> and X3) had a significant effect on the disintegration time of the prepared OFDFs. Disintegration time values were significantly changed when changing the type of polymer used (X1). Where, OFDFs prepared using pectin polymer, showed significantly lower disintegration time values than those prepared with CMC polymer. This could be because of the difference in the nature of the used polymers; wherein, pectin is more hydrophilic than CMC [61]. Thus, OFDFs prepared with pectin had a faster hydration, and hence faster disintegration, than those prepared with CMC.

inhibited its intake and retarded the film disintegration [20].

films loaded with desloratadine [50].

**Figure 3.** Response 3-D plots for the effect of polymer type (X1), polymer concentration (X2), and plasticizer type (X3) on (**a**) disintegration time, (**b**) % PX dissolved after 10 min, and (**c**) desirability of OFDFs loaded with PX nanosuspension. **Figure 3.** Response 3-D plots for the effect of polymer type (X<sup>1</sup> ), polymer concentration (X<sup>2</sup> ), and plasticizer type (X<sup>3</sup> ) on (**a**) disintegration time, (**b**) % PX dissolved after 10 min, and (**c**) desirability of OFDFs loaded with PX nanosuspension.

3.5.9. In Vitro Dissolution Studies For OFDFs, time is an important factor, because the loaded drug should be dissolved within a minute. Figure 4 illustrates the dissolution profiles of all the prepared OFDFs; where, the % PX dissolved after 10 min from all the prepared OFDFs was found to be from 12.14 ± 0.08% to 96.02 ± 3.46%, as displayed in Table 3. The following equation was used to analyze the % PX dissolved after 10 min: % PX dissolved after 10 min = 36.27 − 7.47 × X1 − 15.69 × X2 + 15.53 × X3 [1] − 10.15 × X3 [2] (13) where X1 is the polymer type, X2 is the polymer concentration, and X3 is the plasticizer Additionally, it is worth noting that increasing the polymer concentration (X2) resulted in a significant increase in the disintegration time of the prepared OFDFs. This could have been because increasing polymer concentration leads a need for more fluids to wet the films and increasing the film thickness, which retarded the penetration of water. Moreover, Shen et al., who formulated fast-dissolving films loaded with herpetrione nanoparticles, stated that the disintegration time of the formulated films increased with increasing HPMC concentration. He attributed this to the fact that increasing the HPMC concentration increases water viscosity when the film comes into contact with it, which inhibited its intake and retarded the film disintegration [20].

Moreover, Shen et al., who formulated fast-dissolving films loaded with herpetrione nanoparticles, stated that the disintegration time of the formulated films increased with increasing HPMC concentration. He attributed this to the fact that increasing the HPMC concentration increases water viscosity when the film comes into contact with it, which

On the other hand, changing the plasticizer type (X3) significantly influenced the disintegration time of the prepared OFDFs. Where, OFDFs containing glycerol as a plasticizer were found to have a lower disintegration time than OFDFs prepared with PG and PEG 400. These results were in harmony with those stated by Singh et al., who prepared oral

type. X3 [1] represents the first plasticizer type (glycerol) and X3 [2] represents the second plasticizer type (propylene glycol). On the other hand, changing the plasticizer type (X3) significantly influenced the disintegration time of the prepared OFDFs. Where, OFDFs containing glycerol as a plasticizer were found to have a lower disintegration time than OFDFs prepared with PG and PEG 400. These results were in harmony with those stated by Singh et al., who prepared oral films loaded with desloratadine [50].

3.5.9. In Vitro Dissolution Studies

For OFDFs, time is an important factor, because the loaded drug should be dissolved within a minute. Figure 4 illustrates the dissolution profiles of all the prepared OFDFs; where, the % PX dissolved after 10 min from all the prepared OFDFs was found to be from 12.14 ± 0.08% to 96.02 ± 3.46%, as displayed in Table 3. The following equation was used to analyze the % PX dissolved after 10 min:

$$\% \text{ PX dissolved after 10 min} = 36.27 - 7.47 \times \text{X}\_1 - 15.69 \times \text{X}\_2 + 15.53 \times \text{X}\_3 \text{ [1]} - 10.15 \times \text{X}\_3 \text{ [2]} \tag{13}$$

rate [63].

where X<sup>1</sup> is the polymer type, X<sup>2</sup> is the polymer concentration, and X<sup>3</sup> is the plasticizer type. X<sup>3</sup> [1] represents the first plasticizer type (glycerol) and X<sup>3</sup> [2] represents the second plasticizer type (propylene glycol). after 10 min. Where, OFDFs contained glycerol as a plasticizer had a higher %PX dissolved than those films contained PG or PEG 400. As explained earlier, this could be attributed to the high glycerol hygroscopicity, which led to more humidity absorption. This resulted in increasing the film hydrophilic character and increasing the %PX dissolved [56].

On the other hand, plasticizer type (X3) had a significant impact on %PX dissolved

In addition, the statistical analyses clarified that polymer concentration (X2) had a significant impact on % PX dissolved after 10 min. Where, higher % PX dissolved after 10 min was shown in OFDFs with lower polymer concentration. Shaikh et al. reported similar findings, where they found that using a low polymer concentration led to needing a lower amount of water to dissolve the film and leading to faster drug release [64].

Table 3 and Figure 3b show that all the independent variables (X1, X2, and X3) had a

The polymer type (X1) had a significant influence on the % PX dissolved after 10 min. Where, OFDFs prepared by pectin polymer showed higher % PX dissolved after 10 min upon comparison with CMC-based OFDFs. This could have been because pectin has a more hydrophilic nature than CMC, resulting in faster hydration of pectin OFDFs, as explained previously [61]. Moreover, this could be because of the ionization of pectin at pH 6.8 (pH of the utilized dissolution medium), which is >pKa value of pectin (3.5) [62]. The ionization of pectin resulted in the presence of negative charges on the pectin backbone. Thus, the pectin polymer was uncoiled in the form of an extended structure, because of the negative charge repulsion, and the diffusion of positive charges within the pectin matrix generated an extra difference in the osmotic pressure across the matrix, which caused a higher water uptake. Hence, the pectin polymer swelled, resulting in drug diffusion from the films at a higher

significant influence on the % PX dissolved after 10 min from the prepared OFDFs.

*Pharmaceutics* **2021**, *13*, x FOR PEER REVIEW 17 of 25

**Figure 4.** Dissolution profiles of the prepared (**A**) pectin and (**B**) CMC OFDFs loaded with PX nanosuspension. **Figure 4.** Dissolution profiles of the prepared (**A**) pectin and (**B**) CMC OFDFs loaded with PX nanosuspension.

*3.6. Selection of the Optimized OFDF Loaded with PX Nanosuspension*  To identify the optimal OFDF, it was nearly impossible to fulfill all the desired re-Table 3 and Figure 3b show that all the independent variables (X1, X2, and X3) had a significant influence on the % PX dissolved after 10 min from the prepared OFDFs.

sponses at the same time, because the optimum condition fulfilled for one response could adversely affect other responses [22]. However, the desirability function combined all the desired responses in one variable, in order to determine the optimum levels of the examined factors [31]. Figure 3c shows that the highest desirability value was 0.659 for the optimized PX OFDF (F1) containing pectin polymer with a concentration of 1% *w/v* and glycerol as a plasticizer. This optimized OFDF (F1) collectively showed the maximal tensile strength, elongation %, % PX dissolved after 10 min, and minimal Young's modulus and The polymer type (X1) had a significant influence on the % PX dissolved after 10 min. Where, OFDFs prepared by pectin polymer showed higher % PX dissolved after 10 min upon comparison with CMC-based OFDFs. This could have been because pectin has a more hydrophilic nature than CMC, resulting in faster hydration of pectin OFDFs, as explained previously [61]. Moreover, this could be because of the ionization of pectin at pH 6.8 (pH of the utilized dissolution medium), which is >pKa value of pectin (3.5) [62]. The ionization of pectin resulted in the presence of negative charges on the pectin backbone. Thus, the pectin polymer was uncoiled in the form of an extended structure, because of the negative charge repulsion, and the diffusion of positive charges within the pectin matrix generated an extra difference in the osmotic pressure across the matrix, which caused a higher water uptake. Hence, the pectin polymer swelled, resulting in drug diffusion from the films at a higher rate [63].

In addition, the statistical analyses clarified that polymer concentration (X2) had a significant impact on % PX dissolved after 10 min. Where, higher % PX dissolved after 10 min was shown in OFDFs with lower polymer concentration. Shaikh et al. reported similar findings, where they found that using a low polymer concentration led to needing a lower amount of water to dissolve the film and leading to faster drug release [64].

On the other hand, plasticizer type (X3) had a significant impact on %PX dissolved after 10 min. Where, OFDFs contained glycerol as a plasticizer had a higher %PX dissolved than those films contained PG or PEG 400. As explained earlier, this could be attributed to the high glycerol hygroscopicity, which led to more humidity absorption. This resulted in increasing the film hydrophilic character and increasing the %PX dissolved [56].

#### *3.6. Selection of the Optimized OFDF Loaded with PX Nanosuspension*

To identify the optimal OFDF, it was nearly impossible to fulfill all the desired responses at the same time, because the optimum condition fulfilled for one response could adversely affect other responses [22]. However, the desirability function combined all the desired responses in one variable, in order to determine the optimum levels of the examined factors [31]. Figure 3c shows that the highest desirability value was 0.659 for the optimized PX OFDF (F1) containing pectin polymer with a concentration of 1% *w/v*

market tablet.

and glycerol as a plasticizer. This optimized OFDF (F1) collectively showed the maximal tensile strength, elongation %, % PX dissolved after 10 min, and minimal Young's modulus and disintegration time. Where, OFDF (F1) showed a tensile strength of 3.89 ± 0.19 Mpa, elongation % of 53.08 ± 1.28, Young's modulus of 8.12 ± 0.13 Mpa, disintegration time of 17.09 ± 1.30 s, and 96.02 ± 3.46% PX dissolved after 10 min. Upon comparing the observed and predicted values, they were found to be very similar; as shown in Table 2. Consequently, the optimized OFDF (F1) was chosen for further investigation.

### *3.7. Characterization of the Optimized OFDF Loaded with PX Nanosuspension*

#### 3.7.1. Re-Dispersion of PX Nanoparticles from the Optimized OFDF

There was no significant difference between the PS of PX nanosuspension (217.09 ± 4.18 nm) and PS measures after re-dispersion of the optimized OFDF (231.88 ± 3.50 nm) (*p* < 0.05). This slight increase in PS could be attributed to coating the embedded nanoparticles with the polymeric matrix (pectin) and the plasticizer (glycerol) used in preparing the film. These results indicate the stability of the PX nanoparticles within the polymeric matrix [15].

3.7.2. Comparative Dissolution Study of the Optimized OFDF (F1), Pure Drug, and the Market Tablet

The optimized OFDF (F1) showed a significant increase in PX dissolution rate and extent in comparison with the PX dissolution from pure drug powder and the market tablet, with *f* <sup>2</sup> values of 6 and 11, respectively. Figure 5 illustrates that 96.02% (more than 75%) PX was dissolved within just 10 min from the optimized OFDF (F1), compared with 10.04 and 26.37% from pure PX powder and the market tablet, respectively. The extent of dissolution of the optimized OFDF (F1) after 10 min was increased by more than 9.5 and 3.6 fold compared to the drug released from the pure drug and the market tablet, respectively. Additionally, the release T50% of the optimized OFDF (F1) was 3.96 min, while the release T50% of PX pure powder and the market tablet were 112.24 min and 54.16 min, respectively. *Pharmaceutics* **2021**, *13*, x FOR PEER REVIEW 19 of 25

**Figure 5.** Dissolution profiles of the optimized OFDF loaded with PX nanosuspension (F1), PX pure powder, and the **Figure 5.** Dissolution profiles of the optimized OFDF loaded with PX nanosuspension (F1), PX pure powder, and the market tablet.

3.7.3. Ex Vivo Permeation Studies Many permeability studies have revealed the importance of choosing an appropriate This could be attributed to the addition of tween 80, which enhanced the drug release from the optimized OFDF (F1), due to aiding in the acceleration of the disintegration of the film and releasing the incorporated drug more rapidly [15].

mucosal membrane; where, the buccal mucosa of many experimental animals, such as rabbits and rats, is entirely covered with keratin [66]. On the other hand, chicken buccal Another important reason for enhancement of the PX dissolution from the optimized OFDF was the formulation of the loaded drug as nanosized drug particles, which increased

The PX permeation from the optimized OFDF (F1) and PX pure powder through the freshly excised chicken buccal mucosa (pouch) was examined. HPLC analysis showed a linearity coefficient, LLOQ, and accuracy range of 0.9994, 10 μg/mL, and 100% ± 10, respectively. The optimized OFDF (F1) showed a significant increase in PX permeation rate and extent when compared to PX pure powder, as illustrated in Figure 6. The flux (J) values were detected and there was a significant difference (*p* value < 0.001) between the

). The enhancement ratio (ER) value of the optimized OFDF (F1) was also

estimated and was found to be 3.18, reflecting a more than three-fold increase in PX

These results could be attributed to coating the loaded PX NPs with hydrophilic polymer, which enhanced the drug solubility and increased the surface area in contact with the mucosal membrane surface, resulting in increased permeability [15]. These results could also be attributed to the incorporation of tween 80 into the optimized film. Whereby, this surfactant provided an elastic effect that loosened or fluidized the mucosal membrane lipid bilayer and made the drug capable of squeezing into deeper layers of the biological membrane, leading to the enhancement of drug permeability through the buccal mucosa [31]. Regarding the mechanism of PX transport across the buccal mucosal membrane, the uptake process was concentration-dependent, via simple diffusion [67].

) when compared with PX pure powder (26.66

keratinized and thin oral lining mucosa [67].

optimized OFDF (F1) (85.00 μg/h/cm<sup>2</sup>

permeation through the buccal mucosa.

μg/h/cm<sup>2</sup>

the drug solubility due to embedding the nanosized PX particles in the hydrophilic matrix. Moreover, the presence of the drug at nano-size could also have decreased the diffusion layer thickness, increasing the concentration gradient, which consequently increased the drug dissolution rate from the optimized film [65].

The low PX release from the pure PX powder could be attributed to the poor dissolution rate of paroxetine [6].

#### 3.7.3. Ex Vivo Permeation Studies

Many permeability studies have revealed the importance of choosing an appropriate mucosal membrane; where, the buccal mucosa of many experimental animals, such as rabbits and rats, is entirely covered with keratin [66]. On the other hand, chicken buccal mucosa (pouch) is considered the best alternative, because it resembles the human nonkeratinized and thin oral lining mucosa [67].

The PX permeation from the optimized OFDF (F1) and PX pure powder through the freshly excised chicken buccal mucosa (pouch) was examined. HPLC analysis showed a linearity coefficient, LLOQ, and accuracy range of 0.9994, 10 µg/mL, and 100% ± 10, respectively. The optimized OFDF (F1) showed a significant increase in PX permeation rate and extent when compared to PX pure powder, as illustrated in Figure 6. The flux (J) values were detected and there was a significant difference (*p* value < 0.001) between the optimized OFDF (F1) (85.00 µg/h/cm<sup>2</sup> ) when compared with PX pure powder (26.66 µg/h/cm<sup>2</sup> ). The enhancement ratio (ER) value of the optimized OFDF (F1) was also estimated and was found to be 3.18, reflecting a more than three-fold increase in PX permeation through the buccal mucosa. *Pharmaceutics* **2021**, *13*, x FOR PEER REVIEW 20 of 25

**Figure 6.** Permeation profiles of PX from the optimized OFDF (F1) and the pure drug through the chicken buccal mucosa. **Figure 6.** Permeation profiles of PX from the optimized OFDF (F1) and the pure drug through the chicken buccal mucosa.

3.7.4. Stability Study Table 4 illustrates that there were no significant changes (*p* > 0.05) concerning all the investigated parameters of the optimized OFDF (F1) upon storage under the applied These results could be attributed to coating the loaded PX NPs with hydrophilic polymer, which enhanced the drug solubility and increased the surface area in contact with the mucosal membrane surface, resulting in increased permeability [15]. These

> **Disintegration Time (s)**

**% PX Dissolved after 10 min**

**Modulus (Mpa)**

The recorded results for texture evaluation showed that 87.5% of the enrolled volunteers stated that the optimized OFDF (F1) was flexible, easy to handle, and non-

The enrolled volunteers declared that the mean in situ disintegration time of the optimized OFDF (F1) was less than 1 min (14.84 ± 2.15 s), which is similar to the previously recorded in vitro disintegration time of the optimized OFDF (17.09 ± 1.30 s). This fastdisintegration could have been due to citric acid's incorporation in the OFDF, which stimulated the saliva secretion in the buccal cavity, promoting rapid disintegration of the

Regarding the results of the taste evaluation, 100% of the volunteers stated that PX pure powder was very bitter. While, in the case of the optimized OFDF (F1), 12.5% of the volunteers stated an acceptable taste, 25% stated pleasant taste, and 62.5% stated a very

**Strength (Mpa) % Elongation Young's**

Freshly prepared 96.68 ± 3.62 3.89 ± 0.19 53.08 ±1.28 8.12 ± 0.13 17.09 ± 1.30 96.02 ± 3.46 After 3 months 95.70 ± 3.14 4.02 ± 0.25 48.34 ± 0.03 8.06 ± 0.32 15.24 ± 0.87 96.50 ± 1.78 After 6 months 93.89 ± 4.08 3.93 ± 0.12 48.29 ± 0.16 7.99 ± 0.30 20.33 ± 1.01 95.63 ± 2.44

3.8.1. In Situ Disintegration Time and Palatability Studies

stability conditions.

**Tensile**

*3.8. In Vivo Clinical Studies*

**Content Uniformity (%)**

sticky.

OFDF [51].

**Optimized OFDF (F1)**

results could also be attributed to the incorporation of tween 80 into the optimized film. Whereby, this surfactant provided an elastic effect that loosened or fluidized the mucosal membrane lipid bilayer and made the drug capable of squeezing into deeper layers of the biological membrane, leading to the enhancement of drug permeability through the buccal mucosa [31]. Regarding the mechanism of PX transport across the buccal mucosal membrane, the uptake process was concentration-dependent, via simple diffusion [67].

#### 3.7.4. Stability Study

Table 4 illustrates that there were no significant changes (*p* > 0.05) concerning all the investigated parameters of the optimized OFDF (F1) upon storage under the applied stability conditions.


**Table 4.** Stability test parameters for the optimized OFDF (F1).

#### *3.8. In Vivo Clinical Studies*

3.8.1. In Situ Disintegration Time and Palatability Studies

The recorded results for texture evaluation showed that 87.5% of the enrolled volunteers stated that the optimized OFDF (F1) was flexible, easy to handle, and non-sticky.

The enrolled volunteers declared that the mean in situ disintegration time of the optimized OFDF (F1) was less than 1 min (14.84 ± 2.15 s), which is similar to the previously recorded in vitro disintegration time of the optimized OFDF (17.09 ± 1.30 s). This fastdisintegration could have been due to citric acid's incorporation in the OFDF, which stimulated the saliva secretion in the buccal cavity, promoting rapid disintegration of the OFDF [51].

Regarding the results of the taste evaluation, 100% of the volunteers stated that PX pure powder was very bitter. While, in the case of the optimized OFDF (F1), 12.5% of the volunteers stated an acceptable taste, 25% stated pleasant taste, and 62.5% stated a very pleasant taste. These findings represent a success in masking PX's bitter taste by incorporating sucralose as a sweetening agent [52]. Moreover, approximately 87.5% of the volunteers reported a mouth refreshment feeling. This feeling can be attributed to the incorporation of menthol in the OFDF [53].

All the enrolled volunteers also stated that the aftertaste of the optimized OFDF (F1) was significantly improved when compared to PX pure powder. Overall, these findings indicate that the optimized OFDF has the desired properties to be an easily handled and palatable fast-dissolving film.

3.8.2. Pharmacokinetic Parameters of PX in Healthy Human Volunteer s

LC-MS/MS Method for Detection of Paroxetine in Human Plasma

No significant interference with paroxetine or Paroxetine-6D maleate (IS) was noticed in the chromatographed human plasma utilized in the preparation of quality control samples and calibration standards. The retention times of paroxetine and Paroxetine-6D maleate were 1.6 and 1.7 min, respectively. The linear relationship between PX concentrations and peak area ratio of PX/Paroxetine-6D maleate exhibited a linearity coefficient equal to 0.995, and the LLOQ was 1 ng/mL.

Estimation of Bioequivalence Estimation of Bioequivalence

All the volunteers tolerated the procedures executed in this study and the investigated drug well. Figure 7 illustrates PX mean plasma concentration–time profiles following oral administration of both the treatments. The PX pharmacokinetic parameters determined for the two treatments are shown in Table 5. All the volunteers tolerated the procedures executed in this study and the investigated drug well. Figure 7 illustrates PX mean plasma concentration**–**time profiles following oral administration of both the treatments. The PX pharmacokinetic parameters determined for the two treatments are shown in Table 5.

3.8.2. Pharmacokinetic Parameters of PX in Healthy Human Volunteers

LC-MS/MS Method for Detection of Paroxetine in Human Plasma

coefficient equal to 0.995, and the LLOQ was 1 ng/mL.

*Pharmaceutics* **2021**, *13*, x FOR PEER REVIEW 21 of 25

incorporation of menthol in the OFDF [53].

palatable fast-dissolving film.

pleasant taste. These findings represent a success in masking PX's bitter taste by incorporating sucralose as a sweetening agent [52]. Moreover, approximately 87.5% of the volunteers reported a mouth refreshment feeling. This feeling can be attributed to the

All the enrolled volunteers also stated that the aftertaste of the optimized OFDF (F1) was significantly improved when compared to PX pure powder. Overall, these findings indicate that the optimized OFDF has the desired properties to be an easily handled and

No significant interference with paroxetine or Paroxetine-6D maleate (IS) was noticed in the chromatographed human plasma utilized in the preparation of quality control samples and calibration standards. The retention times of paroxetine and Paroxetine-6D maleate were 1.6 and 1.7 min, respectively. The linear relationship between PX concentrations and peak area ratio of PX/Paroxetine-6D maleate exhibited a linearity

**Figure 7.** Mean paroxetine plasma concentration time curve after administration of the optimized OFDF (F1) and the market tablet to six healthy human volunteers. **Figure 7.** Mean paroxetine plasma concentration time curve after administration of the optimized OFDF (F1) and the market tablet to six healthy human volunteers.

Statistical Analysis of Paroxetine Pharmacokinetic Parameters It was noticed that the optimized OFDF (F1) had a significant increase in Cmax (1.74 **Table 5.** Drug pharmacokinetic parameters after the administration of the optimized OFDF (F1) compared to the market tablet.


<sup>a</sup> Data are the mean values (*<sup>n</sup>* = 6) <sup>±</sup> SD.

Statistical Analysis of Paroxetine Pharmacokinetic Parameters

It was noticed that the optimized OFDF (F1) had a significant increase in Cmax (1.74 folds), AUC0–48 (1.56 folds), and AUC0-<sup>∞</sup> (1.78 folds), when compared with the market tablet (*p*-value < 0.05). These results clarified that the extent of PX absorption from the optimized OFDF (F1) was significantly higher than the absorption from the market tablet. The optimized OFDF (F1) had a significantly lower Tmax value when compared with the market tablet (*p*-value less than 0.05). This is might have been because of the rapid disintegration of the optimized OFDF and rapid dissolution of the PX in saliva, leading to fast absorption of PX through the buccal mucosa, and reaching higher plasma concentrations more rapidly [55]. The relative bioavailability of the optimized OFDF (F1) upon comparison with the market tablet was 178.43%. These findings fulfill the goal of this study, for enhancement of paroxetine bioavailability.

This increase in PX bioavailability can be credited to many reasons. First, the presence of the drug at nano-size within the OFDF, which led to increased drug solubility and dissolution rate, in comparison with the market product [16]. Second, the rapid disintegration of the optimized OFDF resulted in rapid absorption through the buccal mucosa and the prevention of large amounts of PX being metabolized in liver, achieving a higher bioavailability [53]. Third, the incorporation of tween 80 in the film formulation could have led to increased PX particle permeability at the absorption sites, which boosted the absorbed fraction of PX [17,21,27].

#### **4. Conclusions**

A PX nanosuspension, prepared using a solvent–antisolvent precipitation method, was successfully loaded into OFDFs prepared using a solvent casting method. The OFDFs loaded with PX nanosuspension represent a palatable and stable dosage method, which can be easily taken by pediatric, geriatric, and psychiatric patients. More than 90% of PX was dissolved within 10 min from the optimized OFDF, compared with 10.04 and 26.37% from the pure drug and the market tablet, respectively. A permeation study utilizing chicken buccal pouch revealed increasing drug permeation with the optimized OFDF, with a more than three-fold increase in permeation over the pure drug. Moreover, an in vivo bioavailability estimation in healthy human volunteers clarified that the optimized OFDF (F1) increased the PX bioavailability significantly more than the market tablet. Hence, the prepared OFDF can be considered a promising, convenient, and economical approach to boosting paroxetine bioavailability.

**Author Contributions:** Conceptualization, R.M.E.-D.; Methodology, A.H.E. and R.M.E.-D.; Software, R.M.E.-D.; Formal Analysis, A.H.E. and R.M.E.-D.; Investigation, R.M.E.-D.; Resources, R.M.E.-D.; Data Curation, A.H.E.; Writing-Original Draft Preparation, R.M.E.-D.; Writing-Review & Editing, A.H.E.; Visualization, R.M.E.-D.; Supervision, A.H.E. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Genuine Research Center (GRC/1/21/R4). The date of approval was (14/03/2021).

**Informed Consent Statement:** Written informed consent has been obtained from the patients to publish this paper.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **In Vitro-In Silico Tools for Streamlined Development of Acalabrutinib Amorphous Solid Dispersion Tablets**

**Deanna M. Mudie \* , Aaron M. Stewart , Jesus A. Rosales † , Molly S. Adam, Michael M. Morgen and David T. Vodak**

> Global Research & Development, Lonza, Bend, OR 97703, USA; aaron.stewart@lonza.com (A.M.S.); rosaleja@uw.edu (J.A.R.); molly.adam@lonza.com (M.S.A.); michael.morgen@lonza.com (M.M.M.); david.vodak@lonza.com (D.T.V.)

**\*** Correspondence: deanna.mudie@lonza.com

† Current address: Pharmaceutics Department, University of Washington, Seattle, WA 98195, USA.

**Abstract:** Amorphous solid dispersion (ASD) dosage forms can improve the oral bioavailability of poorly water-soluble drugs, enabling the commercialization of new chemical entities and improving the efficacy and patient compliance of existing drugs. However, the development of robust, highperforming ASD dosage forms can be challenging, often requiring multiple formulation iterations, long timelines, and high cost. In a previous study, acalabrutinib/hydroxypropyl methylcellulose acetate succinate (HPMCAS)-H grade ASD tablets were shown to overcome the pH effect of commercially marketed Calquence in beagle dogs. This study describes the streamlined in vitro and in silico approach used to develop those ASD tablets. HPMCAS-H and -M grade polymers provided the longest acalabrutinib supersaturation sustainment in an initial screening study, and HPMCAS-H grade ASDs provided the highest in vitro area under the curve (AUC) in gastric to intestinal transfer dissolution tests at elevated gastric pH. In silico simulations of the HPMCAS-H ASD tablet and Calquence capsule provided good in vivo study prediction accuracy using a bottom–up approach (absolute average fold error of AUC0-inf < 2). This streamlined approach combined an understanding of key drug, polymer, and gastrointestinal properties with in vitro and in silico tools to overcome the acalabrutinib pH effect without the need for reformulation or multiple studies, showing promise for reducing time and costs to develop ASD drug products.

**Keywords:** acalabrutinib; amorphous solid dispersion; bioavailability enhancement; acid reducing agent; proton pump inhibitor; kinase inhibitor; in silico prediction; absorption modeling; spray drying; GastroPlus

#### **1. Introduction**

Oncology is the top therapeutic area in pharmaceutical drug development, accounting for 25% of drugs approved by the FDA over the last decade [1]. However, many oncology drugs are poorly water soluble across at least part of the gastrointestinal (GI) pH range, often manifesting as drug–drug or food–drug interactions that can limit oral bioavailability [2]. Calquence® (crystalline acalabrutinib) is an example of an oral oncology drug that exhibits pH-dependent absorption, whereby the area under the plasma drug concentration– time curve (AUC) is reduced by 43% when taken with a proton pump inhibitor (PPI), which elevates gastric pH [3,4]. This drug–drug interaction (DDI) can result in decreased patient compliance and efficacy, since many cancer patients are prescribed PPIs and other gastric acid reducing agents (ARAs) [5].

The mechanism of reduced absorption of Calquence when taken with ARAs is decreased solubility at elevated pH, which is a common mechanism for Biopharmaceutics Classification System (BCS) 2 weak base drugs such as acalabrutinib [6,7]. A previous publication from our laboratory demonstrated that amorphous solid dispersion (ASD) tablets overcame the ARA effect at the human-prescribed 100 mg dose in a fasted beagle

**Citation:** Mudie, D.M.; Stewart, A.M.; Rosales, J.A.; Adam, M.S.; Morgen, M.M.; Vodak, D.T. In Vitro-In Silico Tools for Streamlined Development of Acalabrutinib Amorphous Solid Dispersion Tablets. *Pharmaceutics* **2021**, *13*, 1257. https:// doi.org/10.3390/pharmaceutics13081257

Academic Editors: Vitaliy Khutoryanskiy and Hisham Al-Obaidi

Received: 18 June 2021 Accepted: 9 August 2021 Published: 13 August 2021 Corrected: 2 December 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

dog model [8]. Results demonstrated that ASD tablets achieved similar AUC values at low and high gastric pH conditions and outperformed Calquence capsules 2.4-fold at high gastric pH.

The improved performance of the ASD tablet compared to Calquence was driven by the increased aqueous solubility of the amorphous compared to the crystalline form, which is enough to provide rapid and high extent of GI dissolution at high and low gastric pH values (i.e., in the presence and absence of ARAs). In addition, ASD tablets were 60% smaller than Calquence capsules and showed good physical stability, chemical stability when stored desiccated or refrigerated, and manufacturability as described by Mudie et al. [8]. This outcome highlighted the utility of ASD dosage forms for improving the performance of the numerous weak base and oral oncology drugs that show decreased absorption when taking ARAs [3,9,10].

While the previous publication focused on the in vivo study outcome, the current publication describes a streamlined approach to develop ASD tablets, including formulation selection, in vitro dissolution testing, and in silico simulations used to build confidence in the ASD tablet, overcoming the ARA effect observed with Calquence. A priori in silico plasma concentration–time profiles are compared with in vivo plasma profiles, and in silico prediction accuracy is calculated. This paper describes:


#### **2. Materials and Methods**

#### *2.1. Material Sourcing*

Acalabrutinib (CAS 1420477-60-6, >98% purity) was purchased from LC Laboratories (Woburn, MA, USA). Form I was prepared by recrystallizing the purchased acalabrutinib according to WO 2017/002095 Al, Example 1 [11]. See Appendix A.1 for Form I verification. Hydroxypropyl methylcellulose acetate succinate (HPMCAS) (Aqoat, HF grade, MF grade and LF grade) was purchased from Shin-Etsu Chemical Co., Ltd. (Tokyo, Japan). Poly(methyl methacrylate-co-methacrylic acid (Eudragit L100®) was purchased from Evonik (Evonik Industries AG, Essen, Germany). Vinylpyrrolidone-vinyl acetate copolymer (PVPVA) (Kollidon® VA64) was purchased from BASF (Ludwigshafen, Germany). Polyvinylpyrrolidone (PVP) (Kollidon® 30) was purchased from BASF. Hydroxypropyl methylcellulose (HPMC) (MethocelTM E3 LV) was purchased from DuPont de Nemours, Inc. (Wilmington, DE, USA). MethocelTM A4M was purchased from ThermoFisher Scientific (Waltham, MA, USA). Sodium acetate, sodium phosphate, potassium phosphate, hydrochloric acid (HCl), and sodium chloride (NaCl) were purchased from Sigma Aldrich Chemical Company (St. Louis, MO, USA). Fasted-state simulated intestinal fluid (FaSSIF) powder was purchased from Biorelevant.com Ltd. (London, UK). Methanol (HPLC grade) was purchased from Honeywell (Morris Plains, NJ, USA). Tetrahydrofuran (THF) (Optima grade) was purchased from Thermo Fisher Scientific (Waltham, MA, USA). Calquence capsules were purchased from Drug World (Cold Spring, NY, USA). Avicel PH-101 (microcrystalline cellulose) was purchased from FMC Corporation (Philadelphia, PA, USA). Pearlitol 25 (mannitol) was purchased from Roquette America (Geneva, IL, USA). Ac-Di-Sol (croscarmellose sodium) was purchased from Dupont (Wilmington, DE, USA). Cab-O-Sil M5P (fumed silica) was purchased from Cabot Corporation (Alpharetta, GA, USA). Magnesium stearate was purchased from Macron Fine Chemicals/Avantor (Radnor, PA, USA).

#### *2.2. Dispersion Polymer Screening*

A polymer screening test of seven different polymers was performed using a solvent shift method to assess the polymers' abilities to sustain acalabrutinib supersaturated drug concentrations [12,13]. The polymers tested included HPMCAS-H, HPMCAS-M, HPMCAS-L, Eudragit L100, HPMC E3, PVP K30, and PVP-VA64. A 20 mg/mL acalabrutinib stock solution (95/5 (*w/w*) tetrahydrofuran (THF)/water) was delivered into an aqueous buffer containing 200 µg/mL of a dissolved polymer using a calibrated pipette. The aqueous buffer consisted of 67 mM phosphate containing 0.5% (*w/w*) (6.7 mM) FaSSIF powder and 82 mM NaCl at a pH of 6.5. A total of 200 µL stock solution was added to achieve a target final concentration of 400 µg/mL acalabrutinib.

During addition of the stock solution, light scattering was monitored by Pion Rainbow™ (Pion Inc., Billerica, MA, USA) ultraviolet (UV) probes (2 mm path length) at 500 to 600 nm (outside the UV absorbance range of acalabrutinib) to observe light scattering with increasing concentration. The scatter signal was corroborated by monitoring direct UV absorbance from 370 to 374 nm to verify that the light-scattering signal could be attributed to crystallization (loss of UV absorbance).

#### *2.3. ASD and ASD Tablet Manufacturing and Characterization*

Six different acalabrutinib ASDs were prepared using three different grades of HPM-CAS (HPMCAS-H, HPMCAS-M, and HPMCAS-L) and two different drug loadings (25 and 50 wt %). HPMCAS-H and HPMCAS-M were chosen since they performed best in the polymer screening study. HPMCAS-L was also selected to determine the impact of HPM-CAS grade on relative dissolution and precipitation rates of acalabrutinib. Specifically, in progressing from -L, to -M, to -H grades, HPMCAS becomes more hydrophobic, and the minimum pH above which it dissolves increases [14]. According to Friesen et al., HPMCAS grades are sparingly soluble above pH values of 4.8 (-L), 5.2 (-M), and 5.7 (-H), dispersing to form colloidal solutions [14]. When partially ionized, hydrophobic regions of HPMCAS can interact with hydrophobic drugs, and carboxylate groups can interact with the aqueous phase at the drug–water interface to inhibit crystal nucleation and growth [14–17].

Compositions, spray solvents, and total solids loadings are shown in Table 1. Solutions were spray dried with an outlet temperature of 45–50 ◦C and inlet temperature of 142–150 ◦C on a custom laboratory scale spray dryer with a 35 kg/h drying gas capacity and a 0.3 m chamber diameter. A Schlick 1.5 pressure-swirl nozzle was used for the 25% drug loading formulations, and a Schlick 2.0 pressure-swirl nozzle was used for the 50% formulations (model 121, 150 um and 200 um orifice, Schlick Americas, Bluffton, SC, USA). After material was collected in a cyclone, residual solvent was removed by secondary drying in a vacuum dryer (Model TVO-2, Cascade TEK, Cornelius, OR, USA) for >16 h at 40 ◦C with a nitrogen sweep gas (−60 cm Hg, 3 standard liters per minute). Solvent removal was below the International Council for Harmonization (ICH) thresholds for methanol (<3000 ppm) as confirmed using a gas chromatograph with a headspace sampler (GC).


**Table 1.** Compositions of ASDs and spray-drying solutions.

Powder X-ray diffraction (PXRD) was used to confirm that ASDs were amorphous, and modulated differential scanning calorimetry (mDSC) was used to ensure a single T<sup>g</sup> as an indication of drug–polymer homogeneity. Scanning electron microscopy (SEM) was

ASD immediate release (IR) tablets were made using the 50/50 (*w/w*) acalabrutinib/ HPMCAS-H ASD. This ASD was chosen as the lead after dissolution testing, since it maximized both in vitro performance (i.e., AUC) and drug loading. The 50 wt % drug loading ASD resulted in ASD tablets that were 60% smaller than Calquence capsules (by volume) at an equivalent 100 mg unit dosage strength. ASD tablets had a 400 mg total tablet mass and a drug loading of 25 wt %. Refer to Mudie et al. [8] for tablet formulation, manufacturing methods, and characterization.

#### *2.4. In Vitro Dissolution Testing of Intermediates*

ASDs were evaluated for dissolution performance in a gastric to intestinal transfer dissolution test using a Pion µDissTM Profiler with Rainbow™ fiber optic UV probe detection. Tests using all six ASDs were first conducted using simulated gastric media with elevated pH representative of dogs (or humans) taking an ARA. While gastric pH can vary from ≈4 to 7 with ARAs, gastric pH values of 5 and 6 were chosen, since the solubilities of different HPMCAS polymer grades (-L, -M, and -H) are most sensitive in this range [18–22].

Relative ASD performance in the elevated gastric pH tests was used to select lead ASDs (25 and 50 wt % drug loading HPMCAS-H ASDs and 50 wt % drug loading HPMCAS-M ASD). Then, these ASDs were tested at a gastric pH representative of dogs pre-treated with pentagastrin (i.e., pH 2). Simulated intestinal medium was representative of fasted dogs, with a pH of 6.5 and 0.5 wt % (6.7 mM) FaSSIF powder (see Appendix A.3 for detailed medium compositions) [23–25].

Tests were conducted at dose concentrations of 2 (gastric) and 1 (intestinal) mg/mL, which approximated dose/volume in the stomach of fasted beagle dogs taking a 100 mg dose of acalabrutinib and resulted in 'non-sink' conditions (dose/volume/solubility > 1) with respect to both the apparent amorphous and crystalline solubilites in intestinal medium [26]. Refer to Appendix A.3 for dose number calculations.

For the tests conducted using pH 5 and pH 6 simulated gastric media, crystalline acalabrutinib and ASDs were prepared as a suspension in 0.5% Methocel A4M at a concentration of 25 mg/mL acalabrutinib. To begin the experiment, 0.8 mL of suspension was added to 9.2 mL of gastric fluid to achieve a dose concentration of 2 mg/mL acalabrutinib. For the tests conducted using pH 2 simulated gastric medium, neat crystalline or ASD powder was added directly to the dissolution vessel to begin the experiment. Samples were stirred at 100 rpm and held at 37 ± 2 ◦C by circulating water through a heating block mounted to the Pion µDiss™ profiler. After 30 min, gastric medium was diluted 1:1 with a concentrated intestinal medium to a final volume of 20 mL and a concentration of 1 mg/mL acalabrutinib.

Data were collected for 30 min in gastric medium and for approximately 160 min in intestinal medium with Pion Rainbow™ UV probes. See Appendix A.3 for UV analysis parameters. The apparent concentrations measured consisted of (1) drug dissolved in aqueous medium and (2) drug partitioned into bile salt micelles (present in intestinal medium) as a micelle-bound drug. Each sample was measured in duplicate.

AUC in intestinal medium for each test was calculated in Microsoft Excel (Microsoft Corporation, Seattle, WA, USA) using the rectangular rule using 0.25 min time increments between times 30 and 176 min of the dissolution test (i.e., for a 146 min duration in intestinal medium). AUC enhancement was determined for each ASD sample by dividing AUC in intestinal medium for the ASD by AUC in intestinal medium for crystalline acalabrutinib tested using the same dissolution conditions.

#### *2.5. In Vitro Dissolution Testing of Dosage Forms*

ASD tablets and commercially available Calquence capsules were evaluated for dissolution performance in a gastric to intestinal transfer dissolution test using a Vankel VK7000 (now Agilent, Palo Alto, CA, USA) United States Pharmacopeia (USP) 2 disso-

lution apparatus equipped with 500 mL vessels. Tests were conducted using simulated gastric media (HCl) representative of dogs taking ARAs (pH 6) or dogs pretreated with pentagastrin (pH 2). Simulated intestinal medium was representative of fasted dogs, with a pH of 6.5 and 0.5 wt % (6.7 mM) FaSSIF powder (see Appendix A.3 for detailed medium compositions).

Tests were conducted at dose concentrations of 0.4 (gastric) and 0.2 (intestinal) mg/mL to allow enough of the dose to dissolve in pH 2 and pH 6 gastric media, and in intestinal medium for both the ASD tablet and Calquence capsule to facilitate the determination of zfactors (i.e., dissolution rates) for in silico predictions [27]. This dose concentration allowed for ≈20–100% dose dissolved for the Calquence capsule and ≈40–100% dose dissolved for the ASD tablet across the three media. Refer to Appendix A.3 for dose number calculations. In vito testing at a higher, non-sink dose concentration was conducted in a controlled transfer dissolution (CTD) test as described by Mudie et al. [8].

To begin the test, 250 mL of gastric medium was added to the dissolution vessel, which was followed by a single ASD tablet, or a Calquence capsule (100 mg dose) contained in a capsule sinker. After the dosage form was added, the paddles were started. Samples were stirred at 75 rpm and held at 37 ± 2 ◦C by circulating water through a heater attached to the USP 2 dissolution apparatus. After 30 min, gastric medium was diluted 1:1 with a concentrated intestinal medium to a final volume of 500 mL and a concentration of 0.2 mg/mL acalabrutinib.

Dissolution performance was monitored for 30 min in gastric medium and for 150 min in intestinal medium with Pion Rainbow™ UV probes. See Appendix A.3 for UV analysis parameters. The apparent concentrations measured consisted of (1) drug dissolved in aqueous medium and (2) drug partitioned into bile salt micelles (present in intestinal medium) as a micelle-bound drug. All samples were analyzed in duplicate.

AUC in intestinal medium for each test was calculated in Microsoft Excel using the rectangular rule using 0.25 min time increments between times 30 and 150 min of the dissolution test. AUC enhancement was determined for the ASD tablet by dividing AUC in intestinal medium for the ASD tablet by AUC in intestinal medium for the Calquence capsule tested using the same dissolution conditions.
