Development and Pharmacokinetic Evaluation of Two Parenteral Formulations of Albendazole Using Prodrug and Cosolvent Approaches

Albendazole is a broad-spectrum anthelmintic drug used for parasitic infections. In addition, due to its mechanism of action, it has been studied as an anticancer agent. However, poor and highly variable bioavailability are limiting factors for its use in systemic illnesses. The present study aimed to develop two parenteral formulations of albendazole and to compare its pharmacokinetic profile with the conventional oral administration. Parenteral formulations were developed using two different approaches: a phosphonooxymethylated prodrug and cosolvents. For the albendazole prodrug, once synthetized, its solubility and hydrolysis with alkaline phosphatase were evaluated. A factorial design of experiments was used for the cosolvent formulation. Stability and hemolytic activity were assessed. A pharmacokinetic study was performed on New Zealand rabbits. Both formulations were administered intravenously, and the prodrug was also administered intramuscularly. Results were compared with those obtained after the oral administration of albendazole. A 20,000-fold and 6000-fold increase in albendazole solubility was found with the prodrug and cosolvent formulations, respectively. Both parenteral formulations displayed higher albendazole plasma concentrations for the first 2 h compared with oral administration, even when the oral dose was doubled. The absolute bioavailability of oral albendazole was 15.5% while for the intramuscular administration of the prodrug was 102.6%. Both parenteral formulations showed a significant decrease in the formation of albendazole sulfoxide (ANOVA p<0.05) and allowed greater exposure to albendazole. Albendazole cosolvent parenteral formulation could be a promising option in systemic illnesses considering its ease of preparation and superb pharmacokinetic performance.


Introduction
Albendazole (ABZ), a benzimidazole derivative approved in 1982 as an anthelmintic product, is the drug of choice for cystic hydatid disease as well as for the treatment of parenchymal neurocysticercosis [1]. Its mechanism of action is related to its inhibitory effect on tubulin polymerization, which results in the selective degeneration of cytoplasmic microtubules and the death of the parasite [2].
Due to this effect, ABZ has also been evaluated for its potential against different types of cancer cells including liver [3,4], stomach [5], colon [6], pancreas [7], ovarian [8], and leukemia [9]. A limiting factor in albendazole dosage regimes is its low solubility which leads to a great interindividual variability in plasma concentrations [2].
Although the oral administration is the most common for the convenience of the patients, the parenteral route represents a good option to attain the desired plasma concentrations. Few studies have described the development of ABZ parenteral formulations. In 2015, Noorani et al. assembled ABZ with bovine serum albumin into nanoparticles (nab) and reported a significant improvement in solubility and better efficacy than free ABZ in an ovarian cancer xenograft model [10]. Pillai K. et al. reported a new parenteral formulation, using sulfobutylether-β-cyclodextrin. Results showed that the solubility of ABZ in the complexed product was 8 mg/mL. Also, a potent cytotoxic in vitro and in vivo effect was found in an ovarian cancer cell line and in a BALB/c mice murine model with an ovarian tumor [11]. Recently, Viranov et al. evaluated the effect of different surfactants on ABZ solubility and found that the most promising candidate for parenteral application was a mixture of 60% of sodium dioctyl sulfosuccinate and 40% of the phospholipid sodium dipalmitoyl-phosphatidylglycerol. In this delivery system, a solubility of 4.4 mg/mL of ABZ was found [12].
Although the abovementioned strategies have improved ABZ bioavailability, most of them have some limitations. For instance, the size of the nab nanoparticles may limit the access of the drug into the tumor [10]. In the case of using surfactants, the need to use acidic conditions (pH=3) is a limitation because it might cause local irritation in parenteral administration [12].
Other alternatives to increase the drug solubility are the formation of prodrugs and the use of cosolvents. For example, Chassaing et al. developed phosphonooxymethylated prodrugs of benzimidazoles. This study showed a 195,000-fold increase in the solubility of the fenbendazole prodrug. Also, higher plasma concentrations were obtained after the oral administration of the prodrug compared with the administration of an oral suspension [13].
In the case of cosolvents, when drugs are lipophilic or nonpolar and have low water solubility, they can be solubilized using aqueous solutions and water-miscible organic solvents (cosolvents). If this strategy does not allow to reach the adequate concentration of the drug, it is possible to use nonaqueous solutions composed of cosolvents [14]. The advantages of liquid cosolvent formulations include the following: no need for expensive technology for producing the dosage form, easy sterilization process, and facility in the administration process [15].
In the current study, we developed and evaluated two parenteral formulations of ABZ to improve its solubility and bioavailability. Techniques employed for the formulations were a phosphonooxymethylated prodrug approach and the use of cosolvents.
Distilled or deionized water was obtained from the Elix 3 and Milli-Q water purification system (Millipore, Milford, MA, USA). Solvents used for chromatographic analysis were HPLC grade. All other reagents and solvents were of analytical grade.

Preparation of Phosphonooxymethylated ABZ Prodrug
The ABZ prodrug was prepared according to the synthetic sequence shown in Fig. 1. The synthesis was based on the methodologies previously reported by Chassaing et al. [13] and Flores et al. [16].

Preparation of Sodium (2-((methoxycarbonyl) amino)-6-(propylthio)-1H-benzimidazol-1-yl)methyl phosphate (6)
The crude compound (5) was dissolved in MeOH (7 mL). Then, a solution of NaOH in MeOH:water (6:3) was added dropwise to reach pH 7 while stirring. Volatile solvents were removed on a rotary evaporator, and the solid residue was suspended in anhydrous EtOH with stirring for 3 h. The solid (6) was filtered, re-suspended, and maintained at stirring for 24 h in DCM to remove residual ABZ and other nonpolar compounds. The solid obtained was placed in MeOH under continuous stirring for 24 h to dissolve the prodrug. The solvent was evaporated under a vacuum, and the solid recovered was dried in a vacuum for 72 h. The compound was characterized by 1 H NMR, 13 C NMR, 31 P NMR, and MS data. Th residual content of solvents (DCM and dioxane) was determined by quantitative 1 H NMR using chloroacetic acid as an internal standard.

Solubility of the Prodrug
Solubility was evaluated in buffer solutions at pH 6.0, 7.4, and 8.0 using the shake flask method. Briefly, an excess amount of the compound was added to each solution within a water bath maintained at 25 +1°C. The tubes were kept on an oscillatory shaker at 150 rpm for 48 h. Then, they were filtered and diluted with water. Concentrations were determined by HPLC (Shimadzu LC-10ATvp) using a Spherisorb ODS1 (5μm, 250 × 4.6 mm) column. The mobile phase was 1% NH 4 H 2 PO 4 :methanol:acetonitrile (30:50:20 v/v) at a flow rate of 1 mL/min and a UV detector at 245 nm. Samples were prepared in triplicates.

Hydrolysis of the Prodrug
One milligram of prodrug was dissolved in 1.2 mL of Tris/ HCl buffer pH 8.5 and placed in a water bath at 37°C. After 10 min, 50 μL (88 units) of alkaline phosphatase was added. Samples were withdrawn at 10, 20, and 30 s and 1, 3, 5, 10, and 90 min and mixed with 900 μL of acetonitrile to stop the enzymatic activity. Samples were centrifuged for 10 min at 14,000 rpm. The supernatant was analyzed for ABZ formation by HPLC.

Development of a Formulation Using Cosolvent Approach
Solubility of ABZ in isopropyl myristate, glycerin, castor oil, cremophor EL, PG, ethanol, PEG 600, PEG 400, DMSO, and DMA was evaluated using the shake flask method. For the assay of ABZ, samples were diluted with the mobile phase, and 30 μL was injected into the chromatographic system (Shimadzu LC-10ATvp), using an Hypersil ODS column (5μm, 150 × 4.6 mm), a mobile phase consisting of 1.25% NH 4 H 2 PO 4 :methanol:acetonitrile (30:50:20 v/v), a flow rate of 1.2 mL/min, and a UV detector at 295 nm. The method was linear in the range of 3.0-25.0 μg/mL.
According to ABZ solubility, the solvents selected were DMA, PEG 400, and EtOH. To choose the proportion of each solvent, a 3 2 factorial design of experiments (DOE) was used. DOE was carried out using Statgraphics Centurion XVI version 16.1.18. The response variable was ABZ solubility. The formulation with the highest ABZ concentration was named F1.
To reduce hemolytic potential, different proportions of EtOH (60, 70, and 80% v/v) and 6% sodium benzoate were used. The formulation with the highest ABZ concentration was named F2.

Physical and Chemical Stability
Formulations F1 and F2 were prepared at an ABZ concentration of 5 mg/mL. Formulations were stored at room temperature and refrigerated at 4°C. Samples were analyzed after they were prepared, and at 6 and 12 months for drug content, pH, and physical appearance.

Hemolysis Assay
Hemolysis of both formulations was determined in fresh rabbit blood. Negative and positive controls were 0.9% saline solution and 1% Triton X100, respectively. Briefly, blood was centrifuged at 3500 rpm for 10 min. Blood cells were separated and washed with 0.9% saline solution. Cells were suspended in saline solution, an aliquot of 0.9 mL was placed in a water bath at 37°C, and 0.1 mL of ABZ formulations or control solutions was added, vortexed for 5 s, and maintained at 37°C for 2 min. Samples were centrifuged at 3500 rpm for 10 min. A 0.3 mL aliquot of the supernatant was mixed with 5.7 mL of saline solution, and absorbance was measured at 540 nm. The percentage of hemolysis was calculated as:

Pharmacokinetic Study
Animal experiments were approved by the Institutional Animal Care and Use Committee (CICUAL) of the Chemistry Faculty, UNAM. Protocol was approved on August 16, 2021, with the official number FQ/CICUAL/435/21. All experiments were conducted according to Mexican Guidance (NOM-062-ZOO-1999).
A pharmacokinetic study was performed in twelve male white New Zealand rabbits (2.5-3.5 kg). The animals were individually housed with access to food and water ad libitum. Rabbits were randomly assigned to two different groups. Group A received an oral dose of 0.8 mL of Zentel® suspension (40 mg ABZ/mL) while group B received 1 mL of the aqueous solution of the prodrug intramuscularly at a concentration equivalent to 15.88 mg/mL of ABZ. After a washout period of 15 days, group A received 3 mL of formulation F2 intravenously (5 mg/mL) in the marginal %hemolysis = Abs test − Abs negative control Abs positive control − Abs negative control x100 ear vein, and group B received 1 mL of the prodrug intravenously at a concentration equivalent to 15.88 mg/mL of ABZ. Immediately prior to administration, both parenteral formulations were filtered through 0.22-μm nylon filters. Blood samples were taken from the marginal ear vein prior to dosing (0 h) and at 20 and 40 min and 1, 1.5, 2, 3.5, 5, 8, 12, and 24 h post administration. In the parenteral administration, a sampling time of 10 min was added. Samples were collected on EDTA K 2 tubes and centrifuged. Plasma samples were stored at −20°C until analysis.

Albendazole and Albendazole Sulfoxide Assay
For the assay of ABZ and ABZSO, a liquid chromatographic method coupled with tandem mass spectrometry (LC-MS/ MS) was developed and validated using a Shimadzu Prominence chromatograph (Shimadzu, Japan) coupled to a turbo ion spray ionization-triple quadrupole mass spectrometer, API 4500 (ABSciex, Darmstadt, Germany), with positive ion electrospray ionization using multiple reaction monitoring (MRM) mode to detect and quantify ABZ and ABZSO with mass-to-charge (m/z) transitions at 266.3 → 234.0 and 282.0 → 240.0, respectively.

Pharmacokinetic Analysis
Pharmacokinetic parameters of ABZ and ABZSO were estimated using Phoenix® WinNonlin® Version 8.3.3.33 (Pharsight, Mountain View, CA, USA). To compare the relative formation of ABZSO, the individual ABZSO/ ABZ AUC 0-∞ ratio for each administration was calculated, and the mean ratios were compared. Also, the absolute bioavailability was determined for both extravascular administrations.

Statistical Analysis
ANOVA statistical comparisons were performed using Statgraphics Centurion XVI version 16.1.18 followed by a Tukey test.

Preparation of Phosphonooxymethylated Prodrug of ABZ
The prodrug was obtained through an alkylation of ABZ employing di-tert-butyl(chloromethyl) phosphate in DMF. A strong base was used to create the nucleophile (2) (Fig. 1). In the first instance, we proposed sodium hydride as the strong base; however, it promoted the formation of a methylenebis-albendazole (4), which was the most abundant secondary product [13]. This compound (4) was isolated and characterized by 1 H NMR. A characteristic signal was obtained around 6.5 ppm, which was attributed to methylene binding albendazole molecules in the dimer.
To avoid the formation of compound (4), we decided to look for a weaker inorganic base in the same solvent. Cesium carbonate was selected because it has been used in similar reactions with an adequate performance [17]. When this reactive was used, the quantity of dimer was decreased; nevertheless, ABZ maintained a low reactivity. To achieve better yields, several reaction conditions including temperature, the quantity of electrophilic reagent, and the use of a Finkelstein activator as KI were tested [17]. The best results were obtained with the conditions mentioned in the methodology section.
A column purification of compound (3) was attempted, but ABZ appeared in all fractions. The only way to separate it was to use an acid-mobile phase; however, this led to the formation of compound (5), which was strongly retained on the column. Therefore, we decided to continue with the other reactions without purification.
Once compound (3) was obtained; an acid-promoted deprotection allowed us to obtain prodrug (5). Finally, the conversion of the free acid to the corresponding disodium salt was achieved with NaOH in methanol. These last two reactions were quick and did not require further optimization. However, the purification of the final product was complicated because most of the impurities were soluble in the same solvents as the compound (6). It is important to note that the prodrug was obtained as a mixture of their N1 and N3 regioisomers; however, this fact is irrelevant since both are enzymatically hydrolyzed to give ABZ. The global yield was 11%. Spectroscopical characterization of ABZ prodrug is included in the Supplementary material. Mass ( Figure S1) and NMR spectra (Figures S2-S4) showed results for compound (5), since compound (6) when dissolved in water produces compound (5). Analysis of residual solvents using 1 H NMR spectra showed that dichloromethane was not detected. In the case of dioxane, the concentration was 224 ppm, which is lower than the ICH guidance specification (< 380 ppm) [18].

Prodrug Solubility
As shown in Table I, the prodrug solubility was independent of pH in the range evaluated. A 20,000-fold increase in solubility was found, when compared to the reported ABZ solubility in water of 0.8 mg/L (3×10 −3 mM) [19].

Hydrolysis of the Prodrug
ABZ was rapidly hydrolyzed in an alkaline phosphatase solution with a first-order half-life of 8.72 ± 0.23 min. Since the hydrolysis of both regioisomers resulted in the same active compound, the pharmacokinetic study was performed with a mixture of them. Table II shows the ABZ solubility in the different cosolvents evaluated. Based on these results and considering the toxicity and hemolytic potential reported, the cosolvents selected were DMA, PEG 400, and EtOH. Table III shows the characteristics of the DOE. Different DMA and PEG 400 proportions were evaluated while the EtOH proportion was fixed at 25% v/v. Figure 2 shows the response surface plot of DOE. It was found that the highest ABZ solubility  [20].

Development of a Formulation Using Cosolvent Approach
When the different proportions of EtOH and sodium benzoate were evaluated, it was found that the highest solubility was achieved using 30% v/v DMA, 50% w/v PEG 400, and 25% v/v of sodium benzoate 6% in ethanol 70% v/v, and therefore, this formulation, named F2, was selected. Formulations, F1 and F2, were prepared at an ABZ concentration of 5 mg/mL for further studies.

Physical and Chemical Stability
Both formulations were colorless and transparent at the time of preparation. At room temperature, both formulations maintained their appearance, and no precipitation was observed even after 12 months of storage. Besides, drug content remained between 90 and 110%, which is the limit to be considered stable [21]. The pH was maintained throughout the time interval at a value of approximately 5. On the other hand, under refrigeration conditions, a white precipitate was formed at the end of the sixth month in both formulations, and the evaluation was not continued. Figure 3 shows the mean percentage of hemolysis for formulations F1 and F2. The hemolysis was lower than 12%, and therefore, formulations were considered non-hemolytic [22]. Considering that formulation F2 presented the lowest percentage of hemolysis, it was selected for the in vivo study.

Pharmacokinetic Evaluation
Analytical Assay for ABZ and ABZSO The method was linear using a weight factor of 1/x. The calibration curves were linear in the range of 0.5-1000.0 ng/mL for ABZ and from 10.0 to 1500.0 ng/mL for ABZSO. The method was precise with intraday coefficients of variation in the range of 2.4-8.0% for ABZ and 1.4-5.3% for ABZSO, while the inter-day assay precision was between 11.5 and 14.6% for ABZ and 2.6% and 4.7% for ABZSO. Accuracy was also demonstrated with deviation from the nominal concentration ranging from 0.3 to 6.2% for ABZ and 1.7 to 9.9% for ABZSO. Precision and accuracy were maintained after sample dilution ratios of 1:2 and 1:5.

Pharmacokinetic Study
The mean pharmacokinetic profiles of ABZ and ABZSO after different routes of administration are shown in Fig. 4. ABZ concentrations were quantifiable up to 12 h after the dose, while ABZSO concentrations were quantifiable up to 24 h. It is important to consider that the  The pharmacokinetic parameters of ABZ and ABZSO are presented in Tables IV and V, respectively. There were no statistically significant differences in t 1/2 of elimination (p > 0.05) of both analytes regardless of the route of administration. Similarly, no significant differences were found in Vz and Cl of ABZ after the intravenous administration of cosolvent and prodrug formulations.
After C max and AUC last dose normalization, it was found that there was a significant difference (p < 0.01) between the parenteral and oral routes (Table IV), being greater the exposure of ABZ in the parenteral routes.
For oral administration, the absolute bioavailability of ABZ was 15.5%. For the prodrug, the absolute bioavailability after intramuscular administration was 102.6% with respect to intravenous administration of the same prodrug.

Discussion
Different parenteral albendazole formulations were developed using two different approaches to improve its solubility. The first technique consisted of the synthesis of a phosphonooxymethylated prodrug of ABZ. This methodology has already been used for other benzimidazoles [13,23], and even fosphenytoin, a prodrug of phenytoin, has been approved by FDA and is available for parenteral administration [24].
Despite the low yield obtained during the synthesis, the prodrug possesses characteristics that make it a good candidate for a parenteral formulation. First of all, we found an improvement of 20,000-fold in ABZ solubility. In addition, the prodrug was able to rapidly release ABZ using alkaline phosphatase.
The other technique that was selected to improve ABZ solubility was the use of cosolvents. Due to the low solubility of ABZ, a combination of non-aqueous solvents was  selected for the cosolvent formulation. This strategy has been used for other low-solubility drugs [14]. Formulations F1 and F2 resulted in a significant 6000fold increase in ABZ solubility. In addition, both formulations were stable for 12 months at room temperature, and the hemolytic potential was lower than the proposed maximum value of 20% [22].
Pharmacokinetic parameters obtained by the oral administration, which is the only route of administration currently available, showed that t 1/2 and t max of ABZSO were similar to those reported in other studies in rabbits [25,26]. Different studies have reported the difficulty to quantify ABZ concentrations due to its high and rapid metabolism after the oral route [27][28][29]. To our knowledge, this is the first study that reports ABZ pharmacokinetic parameters, including absolute bioavailability, in a rabbit model after oral administration. Results showed that the half-life of ABZ is relatively shorter than that of ABZSO. In addition, a higher variability in plasma concentrations of ABZ and ABZSO was found after oral administration than in parenteral routes. Furthermore, the C max after the oral route was found to be much lower than concentrations after parenteral administrations despite the fact that double the dose was administered, which is consistent with the expected low bioavailability.
By comparing the pharmacokinetics of the intravenous cosolvent formulation vs. the oral suspension, a similar elimination half-life in both routes was found. The pharmacokinetic profile of ABZ (Fig. 4) showed that the intravenous administration maintained a much higher concentration of ABZ than the oral route at least for the first 2 h even when the dose was half the dose of the oral administration.
When the formulations (prodrug and cosolvent) administered by intravenous route were compared, there were no significant differences in any of the pharmacokinetic parameters neither for ABZ nor for ABZSO, demonstrating that in vivo, ABZ was rapidly released from the phosphonooxymethylated prodrug corroborating the in vitro results.
With respect to the prodrug administration by intravenous and intramuscular routes, no changes in pharmacokinetic performance were found. This indicates a very rapid disposition of ABZ from the prodrug. Furthermore, a complete bioavailability of ABZ was obtained (F=102.6%). Due to these observations, the intramuscular route could be considered the route of choice for ABZ prodrug administration considering that it is safer than intravenous administration.
ABZ has been proposed for cancer treatment, and some studies have shown that it has a greater activity than its metabolites [30,31]. Considering this information, the relative formation of metabolite after each administration route was compared using the ABZSO/ABZ AUC 0-inf ratio. The results in Table V indicate that the mean ABZSO/ABZ AUC 0-∞ ratio was similar for parenteral administrations. However, a statistical difference was found with the oral route, with higher metabolite formation in the latter.
The lower formation of ABZSO in the parenteral route results in a higher proportion of ABZ, which added to its more lipophilic character [30], and the fact that it is not a substrate of Bcrp1 and P-gp [32] allows the drug to reach its site of action and to penetrate the target cells more easily. The results are promising for the further evaluation of its efficacy in in vivo cancer models.

Conclusion
In the present study, two parenteral formulations of ABZ were developed using prodrug and cosolvent approaches. Both formulations showed a significant improvement in ABZ plasma exposure compared to oral administration. In addition, the parenteral route exhibited a significant decrease in the formation of the ABZSO.
Considering that there were no significant differences in the pharmacokinetic parameters of the parenteral formulations, either of them would be a good choice for the treatment of systemic diseases. However, the cosolvent formulation has the advantage that it does not need expensive technology, Values are mean ± SD, except for T max where median (range) is reported. **p < 0.05 in Tukey vs all other treatments. K e , elimination rate constant; t 1/2 , terminal elimination half-life; C max , maximum plasma concentration; T max , time to C max ; AUC last , area under the curve from 0 h to the time of the last quantifiable plasma concentration; AUC0-∞, AUC from 0 to infinity time