Validated Capillary Zone Electrophoretic Determination of Avanafil and Dapoxetine Hydrochloride in their Pure form and Pharmaceutical Preparation

Aims: In this study, a simple, green, and rapid capillary zone electrophoresis (CZE) method coupled with a diode array detector (DAD) was applied for the analysis of avanafil (AVA) and dapoxetine hydrochloride (DAP) as a binary mixture using vardenafil (VAR) as an internal standard (IS) in pure form and pharmaceutical formulation. Methodology: The separation was done using fused silica capillary (58.5 cm total length, 50 cm effective length, and 50 μm internal diameter) and the running background electrolyte (BGE) was 100 mM acetate buffer at pH 3.6. During the separation process, the applied voltage was 30 KV, while the temperature was 25 °C. The sample injection was applied at a pressure of 50 mbar for 10 s, and detection was carried out at 210 nm for DAP and 248 nm for AVA and VAR. Original Research Article Ali et al.; JPRI, 33(46B): 446-459, 2021; Article no.JPRI.75615 447 Results: Analysis of the tested drugs and the internal standard was carried out in less than 6.5 min, where the migration times were 4.29, 4.90, and 6.02 min for IS, DAP and AVA respectively. The proposed method showed linearity in the concentration range 5-80 and 5-70 μg/mL with correlation coefficients 0.9996 and 0.9999 for AVA and DAP respectively. The limit of detection (LOD) was 0.523 and 0.531 for AVA and DAP respectively, while the limit of quantification (LOQ) was 1.585 and 1.608 in respective order. The Peak purity and identity in the proposed method were validated by DAD. Conclusion: The proposed CZE method was validated according to ICH guidelines and applied successfully for the estimation of AVA and DAP in their combined pharmaceutical preparation.


INTRODUCTION
AVA is a class of drugs called phosphodiesterase (PDE) inhibitors, and it works by improving blood flow to the penis during sexual stimulation leading to an erection [1]. AVA is absorbed quickly and has a faster onset of action than other PDE inhibitors [2]. It is chemically identified as: (S)-4-((3-chloro-4methoxybenzyl) amino)-2-(2 (hydroxymethyl) pyrrolidin-1-yl)-N-(pyrimidin-2-ylmethyl) pyrimidine -5-carboxamide, as shown in Fig. 1. Dapoxetine hydrochloride is a drug that belongs to the selective serotonin reuptake inhibitors (SSRIs) family and it was employed for the treatment of premature ejaculation [3].After oral administration, dapoxetine is absorbed and eliminated rapidly [4]. It is Chemically recognized as: (S)-N, N-Dimethyl-3-(naphthalen-1-yloxy)-1phenylpropan-1-amine hydrochloride, as shown in Fig. 1. Recently, AVA was co-formulated with DAP to treat impotence and premature ejaculation. Several techniques were reported for the estimation of AVA in drug products including spectrophotometry [5,6], HPLC [7-9], LC-MS/MS [10] and voltammetry [11]. Different studies were developed for the determination of DAP alone or in a combination with other drugs including UV-Vis spectrophotometry [12][13][14][15], HPLC [16][17][18][19][20][21][22], capillary electrophoresis [23,24], screen printed electrode [25] and potentiometric ion-selective electrodes [26].The analytical techniques for determination of AVA and DAP in binary mixtures were reported including spectrophotometry [27,28], HPLC [29,30], HPTLC [31] and fluorimetry [32]. Capillary zone electrophoresis (CZE) has several advantages in drug analysis compared to other alternative techniques such as high separation efficiency, few waste products, short analysis time, ease of automation, and consuming small nanoliters injection volumes [33,34]. CZE methods are applied mostly in aqueous mediums and the concept of green chemistry can apply easily in opposed to the liquid chromatographic methods that need large amounts of organic solvents [35]. There is no reported electro-chromatographic method for the estimation of AVA and DAP in a binary mixture. Our objective in this study was to develop a CZE method for efficient separation, rapid, and selective analysis of AVA and DAP in a pure form and pharmaceutical preparation.

CE System
The experimental study was carried out using Agilent 7100 CE device (Agilent Technologies Deutschland, GmbH, Hewlett-Packard-Str.8 Waldbronn, Germany) with Diode Array Detector (DAD) and Agilent ChemStation software for data analysis. In the same run, DAD can develop electropherograms at different wavelengths. As a result, each analyte can be detected easily at its maximum absorption wavelength that enhancing the sensitivity of the applied technique. DAP and AVA were detected using DAD at 210 nm and 248 nm respectively. Deactivated fused silica capillary (Agilent Technologies, Waldbronn, Germany) is the capillary that was used in the experiment with dimensions of: 50 cm effective length,58.5 cm total length, and 50μm internal diameter.

Materials and Chemicals
AVA was supplied by Andalous pharma, Cairo, Egypt. DAP and VAR were supplied by EVA Pharma, Cairo, Egypt. Methanol (HPLC grade) was purchased from Sigma-Aldrich Chemie GmbH, Switzerland. Analytical grade reagents including glacial acetic acid, sodium acetate, and sodium hydroxide were obtained from Oxford Company for Laboratory Reagents (Mumbai, India). commercial preparation containing AVA and DAP is super-avana® tablets claimed to contain 100 mg AVA and 60 mg DAP (Sunrise Remedies Pvt. Ltd., India). Acetate buffer (100 mM at pH 3.6) was selected as the best buffer for the proposed experiment. The running buffer was prepared by mixing sodium acetate (0.572 g) and glacial acetic acid (0.55 ml), and the mixture was dissolved in 100 mL of distilled water. The pH of the final mixture was approximately3.6.After that, the final pH value was adjusted to be3.6 with 0.5 M sodium hydroxide or glacial acetic acid.

Capillary conditioning
The capillary was flushed using 0.5 M sodium hydroxide for 15 min at the beginning of each working day, and then with distilled water for 15 min. After that, the capillary was washed with 0.1 M sodium hydroxide for 5 min, waiting for 2.5 min to confirm that, the internal wall of the fused silica capillary was activated completely. The capillary was rinsed with the distilled water for 5 min then, the capillary was washed with the selected background electrolyte for 10 min. Between every two successive injections, The capillary was washed with the BGE for 2 min. The applied voltage was 30 KV, where the sample injection was applied for 10 s at a pressure of 50 mbar.

Preparation of standard solutions and construction of calibration graphs
AVA, DAP, and VAR stock solutions were prepared by dissolving 100 mg of each drug in 100 ml HPLC-grade methanol to produce 1000 μg/mL stock solutions and stored in the refrigerator. Working standard solution of each drug was prepared by accurately transferring aliquots of each drug into a series of volumetric flasks (10 mL), then diluted with distilled water to achieve the concentrations of the linearity range (5-80 and 5-70 μg/ mL for AVA and DAP respectively), while the concentration of the working solution of the internal standard was 30 μg/mL. Furthermore, each concentration was examined three times. Finally, the calibration curve for each drug was produced by plotting the peak area ratios against the respective concentrations.

Assay of tablets dosage form
Ten super-avana® tablets were accurately weighed, crushed and mixed. A suitable quantity of the powder equivalent to 1 tablet (100 mg AVA, and 60 mg DAP) was put into a 100 ml volumetric flask and dissolved in 60 mL of HPLCgrade methanol. The resulted solution was sonicated for 10 min, centrifuged at 5000 rpm for 10 min, and then filtered into a 100 mL volumetric flask. Methanol was added to reach the filtrate up to the final volume. The working solution was prepared by transferring a suitable portion of the filtrate solution into a 10 ml volumetric flask and the solution was spiked with 300 μL VAR stock solution, then diluted to 10 ml with distilled water to achieve the desired concentrations inside the linearity ranges. All procedures in the method were completed as described above.

Buffer type and Ph
The influence of the type of a buffer on the experiment was investigated by using borate buffer of different pH values (8, 8.5 and 9), phosphate buffer of different pH values (6, 7 and 7.5). It was found that borate buffer produced a distorted peak shape of DAP while phosphate buffer produced bad resolution and peak broadening. 100 mM Acetate buffer of different pH values (3.6, 4.2, 4.8 and 5.4) was examined. It was found that sharper peaks were obtained at a reasonable time by using the acetate buffer of pH 3.6.

Buffer concentration
As buffer concentration increased the resolution between AVA and DAP increased. A small alteration occurred in the migration times, but a significant improvement of peak symmetry and the resolution occurred by increasing buffer concentration. Different concentrations of BGE were examined (from 20 to 100 mM), 100 mM acetate buffer exhibit the best peak symmetry ( Fig. 2) and the highest resolution ( from 2.55 in 20 mM buffer to 7.35 in 100 mM buffer) with reasonable migration time.

Applied voltage
The influence of applied voltage on the experiment was studied. Different values of the applied voltage were examined (15, 20, 25 and 30 kV) using the optimized BGE. The results showed that the migration times increased as the applied voltage decreased (Fig. 3) because of decreasing in the electro-osmotic flow (EOF). The resolution was unaffected by changing voltage and its value remained around 7.3 in spite of changing the migration times. As applied voltage decreased the peak broadening increased. As a result, 30 kV ( maximum applied voltage that can be reached by the CE device) was chosen to achieve effective separation with less migration time.

Sample injection and use of internal standard
Sample injection times affect the peak height and width so different injection times (4.0-18.0 s) were investigated at a pressure of 50 mbar. Generally, the peak area (signal response) increased by increasing the injection time. On the other hand, a further increase in the injection time resulted in a loss of linear response and peak tailing. The optimum results showed at 10 s injection time. Vardenafil is a drug that belongs to the phosphodiesterase (PDE) inhibitors class like AVA. VAR is used as an internal standard because of applying an internal standard in the proposed method is used to improve the quantitative estimation of the drugs and to decrease the possible errors caused by fluctuations in migration times and injection volumes [36].

Selection of detection wavelength
AVA and DAP were measured at their maximum absorption wavelength as DAD can measure samples at several wavelengths and you can choose the best one which results in greater sensitivity. DAP was measured at 210 nm, while AVA and internal standard (VAR) were measured at 248 nm. The proposed method exhibit effective separation between AVA, DAP and the IS (VAR) with a retention time less than 6.5 min. Fig. 4 showed electropherograms for VAR, DAP and AVA respectively at the two selected wavelengths. The proposed method showed well-defined peaks of VAR, DAP, and AVA with migration times 4.29, 4.90, and 6.02 min respectively. Table 1 describes system suitability parameters for the analyzed drugs such as selectivity (α), retention factors (k), number of theoretical plates (N) and resolution [37].

Method Validation
The applied technique was validated following the International Conference on Harmonization (ICH) guidelines. [38].

Linearity and ranges
From the proposed method described previously, the recorded responses (ratios of the peak area of AVA or DAP against the IS) were observed to be proportional to concentrations of the analyzed drugs where the linearity was found to be 5-80 and 5-70 μg/mL for AVA and DAP respectively. The regression equations were produced using the least squares treatment of the calibration data. Table 2 describes the linearity data as well as other statistical parameters such as concentration ranges, regression equations, correlation coefficients, slope (Sb), standard deviations of the intercept (Sa), and standard deviations of residuals (Sy/x). Statistical parameters showed good linearity where the correlation coefficient values were 0.9996 and 0.9995 for AVA and DAP respectively. Moreover, the linearity was checked by measuring the RSD % value of the slope (Sb%) for each drug, which was less than 1%. High F values and low significance F in the analysis of variance (ANOVA) test indicate an increase in the mean of squares because of regression and a decrease in the mean of squares because of residuals.

Limit of detection and quantification
The LOD and LOQ values for AVA and DAP were determined following the ICH recommendations, where LOD and LOQare defined as the sample concentration with a signal-to-noise ratio of 3:1 and 10:1 respectively. Table 2 represents the values of LOD and LOQ that indicate the sensitivity of the applied method.

Accuracy and precision
The accuracy and intra-day (within-day) precision of the applied method were investigated by testing each compound three times at three concentration levels within one day, while the inter-day (between-day) precision was studied by testing each compound three times at three concentrations levels repeated on three different days. Statistical results of precision data of AVA and DAP (Table 3) exhibited low RSD% values, which were less than 2%, indicating the method's precision. The Er% values of AVA and DAP did not exceed 2%, indicating the accuracy of the applied method.

Selectivity
The selectivity of the applied method was investigated by analyzing several laboratoryprepared mixtures of the AVA and DAP at various concentrations ratios within their cited linear ranges. The mixtures of the analyzed drugs were prepared in different concentrations ratios lower or higher than those found in the commercial tablet. Table 4 demonstrates the ability of the proposed method in the separation and quantification of AVA and DAP in different concentrations ratios, where RSD% and Er% values proved the selectivity of the applied method. DAD can verify the purity of the resulted peaks, therefore, the proposed method when applied in commercial tablets didn't show coeluted peaks from the inactive ingredients that indicate the selectivity of the proposed method. furthermore, the superimpositions of a set of spectra at five different time intervals across the peak confirm the purity of the peak and the selectivity of the method (Fig. 5 (A1, B1, C1)). Fig. 5 (A2, B2, C2) confirmed the high similarity of all peaks because the evaluation included all the spectra obtained throughout the migration or elution of the peak rather than only three or four spectra. The similarity curves did not exceed the noise threshold value that appeared as a redcolored area which indicates the reliability and the sensitivity of the peak purity evaluation.

Robustness
The robustness of the applied technique was investigated by monitoring the migration times and peak area ratios after small alterations in the method conditions. The studied conditions include buffer pH 3.6 ± 0.2, acetate buffer concentration 100 ± 2 mM, and wavelength 210 ± 2 nm and 248 ± 2 nm. Analysis was carried out by injection of 20 μg/mL mixture of AVA, DAP and 30 μg/ mL of VAR three times with only one alteration occurred at a time. The investigated parameters had no noticeable impact on migration times or peak area ratios. The RSD% values in Table 5 reflect the robustness of the applied method.

Stability of solutions
The stability of working sample solutions was examined using a 30 μg/mL mixture of AVA, DAP and IS within 24 hr at room temperature. The examined drugs showed good stability, and no noticeable change in peaks areas or migration times of the tested drugs as the %RSD values were not more than 1.9. Furthermore, no  chromatographic alterations were observed. The absorption spectra of the analyzed drugs were also investigated and showed no significant change. The stability of stock solutions was examined, and they were observed to remain stable for at least three weeks when kept in a refrigerator at 4 °C. The concentrations of the previously prepared solution (from two weeks) and the newly prepared one were calculated using the applied method, and the RSD% values were observed to be less than 2.0%.

Application of Dosage Forms
The ability of the applied method for the quantitative determination of AVA and DAP as a binary mixture in a tablet dosage form was examined. There were no interfering peaks from the inactive components and no change in migration time for the analyzed drugs (Fig. 6).
Analysis results showed good Recoveries and low RSD% values as showed in (Table 6). DAD can check the peak purity, where the examined drugs exhibited homogenous and pure peaks. As a result, the applied method is successfully applied for the estimation of AVA and DAP in combined dosage form with good recovery and selectivity.

CONCLUSION
The validated CZE technique was applied and assessed for the quantitative estimation of AVA and DAP in a pure form and pharmaceutical preparation. The proposed method is fast, simple, robust and showed acceptable performance concerning all validation parameters for the analyzed drugs in pure form and pharmaceutical preparation. Furthermore, the use of DAD provides a significant benefit for the verification of peak purity. The proposed method is the first electro-chromatographic method applied for the binary mixture of AVA and DAP, and this method can be used as an alternative to HPLC reported methods because it is also effective for routine analysis with a lower cost and the proposed method applied the concept of green chemistry.

DISCLAIMER
The products used for this research are commonly and predominantly use products in our area of research and country. There is absolutely no conflict of interest between the authors and producers of the products because we do not intend to use these products as an avenue for any litigation but for the advancement of knowledge. Also, the research was not funded by the producing company rather it was funded by personal efforts of the authors.

CONSENT
It is not applicable.

ETHICAL APPROVAL
It is not applicable.