Design of Experiment Assisted UV-Visible Spectrophotometric and RP-HPLC Method Development for Ambrisentan Estimation in Bulk and Formulations

Veiw figure View Figure

2.4.1. Experimental Design for Optimization

A two-level full factorial design was employed to evaluate the effect of four independent factors, viz. MBTH reagent concentration (X1), volume of MBTH reagent (X2), FeCl₃ concentration (X3) and Volume of FeCl₃ solution (X4) on absorbance Y. The table matrix of the two-level full factorial design and experimental runs is provided in Table 1. From the resultant design matrix it was observed that maximum absorbance (Y) was shown with 2% MBTH and 2% FeCl₃ each of 2 mL and 1.5−2 mL respectively. Thus in the optimized method, 10 mg of ABN was dissolved in 10 mL methanol to get a concentration of 1000 µg/mL (standard stock solution) and further diluted with the same solvent to get a concentration of 10 µg/mL. This solution was scanned in the range of 400−800 nm against blank and 625 nm was selected as the λ_max for analysis of ABN by MBTH. To each flask, 1.5−2 mL of aqueous FeCl₃ solution (2% in distilled water) and 2 mL of MBTH reagent (2% in distilled water) were added. Volumes were adjusted in each tube with distilled water. The absorbance of bluish green colored chromogen was measured at 625 nm against the blank. Different aliquots of ABN were prepared in 10 mL volumetric flasks in the concentrations of 10−30 µg/mL. The amount of ABN present in the sample solution was computed from its calibration curve ^[14].

2.4.2. Validation of the Method

The above method was validated as per ICH guidelines in terms of linearity, accuracy, precision, robustness, LOD and LOQ ^[22]. The linearity range for the estimation of ABN by MBTH was determined by preparing aliquots in the concentration range of 10−30 µg/mL and absorbances measured at 625 nm. Calibration curves (concentrations vs absorbance) were plotted and R²value not less than 0.99 was regarded as acceptance criterion.

Accuracy of the proposed method was ascertained by recovery studies using analyte recovery method and percent recovery for ABN in the range of 98.8−99.6% were set as the acceptance criterion. The precision was studied by intra-day variations in the test method of ABN and expressed as percent relative standard deviation (RSD) where these values should not be greater than 2%. The LOD and LOQ parameters were determined from the calibration curves basing on the formulae: LOD = 3.3 σ / S and LOQ = 10 σ / S, where S is slope of calibration curve and σ is standard deviation of y-intercepts of calibration graph.

The robustness of an analytical procedure is a measure of its capacity to remain unaffected by small but deliberate variations in the process parameters. As a measure of robustness of the method, intentional alterations were made in the absorbance maximum to evaluate the impact of the method. A 20 μg/mL of ABN was prepared by above procedure and the absorbance was observed at 627 and 623 nm.

Accurately weighed 10 tablets (10 mg) were pooled and powdered of which 10 mg equivalent of ABN was transferred into 10 mL volumetric flask, dissolved in methanol, sonicated, filtered through 0.45 µm filter (Millipore^®), and properly diluted to get a solution concentration of 100 μg/mL. The absorbance of ABN sample solution at 20 μg/mL concentration was measured at 625 nm against blank and the percentage assay was calculated using the formula:

where, AT is absorbance of test, WS= wt. of standard sample, WT= wt. of test sample, AS is absorbance of standard, DS is dilution factor of standard, DT is dilution factor of test and P is potency of standard.

2.5. Estimation of ABN by UV Spectroscopy

The 0.1 M pH 7.4 buffer solution was prepared by dissolving 2.62 g of monobasic sodium phosphate and 11.5 g of anhydrous dibasic sodium phosphate in 1000 mL of water. Stock solution was prepared by dissolving 10 mg of ABN in pH 7.4 phosphate buffer and diluted to get a concentration of 10 µg/mL. This solution was scanned in 200−400 nm range against blank and 262.5 nm was selected as λ_max for analysis of ABN. The standard calibration curve was constructed with the standard stock solution in the concentration range of 10−50 µg/mL and absorbances were measured at 262.5 nm. The UV method developed was validated in terms of linearity, accuracy, precision, LOD, LOQ and robustness as discussed above.

For preparation of ABN sample solution, accurately weighed 10 tablets of ABN were powdered of which 10 mg equivalent of the drug was transferred into 10 mL volumetric flask, dissolved in 7 mL of phosphate buffer, sonicated well, filtered through 0.45 µm filter (Millipore^®) and diluted to get a concentration of 100 μg/mL. The absorbance of ABN sample (20 μg/mL) was measured at 262.5 nm against blank and the percentage assay was calculated using the above formula.

2.6. Estimation of ABN by RP−HPLC
2.6.1. Experimental Design and Optimization of Mobile Phase

Before proceeding to the optimized RP−HPLC chromatographic conditions for ABN estimation, numerous trails were conducted with different mobile phase compositions. Amongst fifteen preliminary trials, injection volume (10 µL) and flow rate (1 mL/min) were kept constant. The absorbances were measured between 260−264 nm wavelength and trial runs were performed using different mobile phase compositions. Using triethylamine (TEA) as mobile phase with pH adjusted by 0.05M orthophosphoric acid (OPA), a very broad peak was obtained. The same result was reported when methanol was used as diluents, when water: methanol (80:20) was used as diluents the peak retention time became unduly delayed. On changing the mobile phase composition to TEA: OPA: acetonitrile (ACN) with diluents of varying compositions like water: acetonitrile (50:50) and methanol: water (80:20), the peaks were not within symmetry limits with a noisy baseline indicating instability of the drug in the mobile phase. A two-level full factorial design ^{[13, 14, 15, 16]} was employed to evaluate the effect of three independent factors, viz. wavelength (X1), flow rate (X2) and OPA percentage in mobile phase (X3) on retention time (Y1) and USP tailing factor (Y2). The table matrix of the two-level full factorial design and experimental runs is provided in Table 2.

The optimized chromatographic conditions for ABN estimation were obtained with isocratic separation mode in a C₁₈ inertsil column (250 mm × 4.6 mm, 5 μm) using a degassed mixture of orthophosphoric acid (OPA): acetonitrile (40:60) as mobile phase with water: acetonitrile (30:70) as diluent, injection volume (20 µL), flow rate (1.2 mL/min) and run time (10 min) with PDA (Photo Diode Array) detector set at 264 nm.

The working standard and sample solutions of ABN were prepared by dissolving 65 mg ABN in diluent (water: acetonitrile, 30:70), sonicated and volume adjusted with diluent. Further dilutions from 5−50 µg/mL concentration range were prepared from the standard and sample solutions, and the volume was adjusted in each case with mobile phase.

2.6.2. System Suitability

The ABN standard solution was injected into the HPLC system six times to evaluate the system suitability parameters from the standard chromatograms obtained by calculating the% RSD of retention times, tailing factor, theoretical plates and peak areas from six replicate injections. The limit of RSD and tailing factor were set below 2.0%. The theoretical plate count was considered if less than 2000.

2.6.3. Validation

The proposed RP−HPLC method was validated as per ICH guidelines ^[22]. Accuracy of the proposed method was ascertained by recovery studies using analyte recovery method, where the drug substance was spiked with placebo in concentrations of 50, 100 and 150% and the acceptable criteria of percent recovery for ABN were set as 98−102%. Precision was studied in terms of repeatability (system precision), where 50 µg/mL of sample was injected for six times into the HPLC system as per test procedure. For method precision, from sample and stock solution, six replicates of standard and sample of 50 µg/mL were prepared and injected into the HPLC system and% RSD was calculated. Intermediate precision study or ruggedness of experimentation was carried out by different analyst, on different instrument and on different days. From the sample and stock solutions, six replicates of 50 µg/mL were prepared and injected into the HPLC system and% RSD was calculated. The LOD and LOQ parameters were determined from the calibration curves basing on the formulae above.

As a measure of robustness of the method, deliberate alterations in the absorbance maximum, flow rate and mobile phase composition were made to evaluate the impact of the method. The λ_max was changed (± 5 nm) from 264 nm to 269 and 259 nm. The flow rate was increased from original value (1.2 mL/min) to 1.4 mL/min and decreased to 0.8 mL/min. The ratio of mobile phase composition was also altered (± 10%) from 40:60 to 30:70 and 50:50. The tailing factor of ABN standard, RSD of asymmetry and retention time of ABN standard should not be more than 2% due to the intentional alterations of the said parameters.

2.6.4. Interference Studies

For studying blank and placebo interference, diluents and sample of placebo were injected in to the system with the objective that they should not show any peak at the retention time of analyte peak.

2.6.5. Degradation Studies

Chemical degradation studies such as; acidic, alkaline, oxidative, reductive, thermal, photolytic and humidity were conducted under stress conditions. From the sample stock solution, aliquots of 5 mL was transferred into 20 mL volumetric flask. To this, a 2 mL of 5% either HCl or NaOH was added for acid or alkali degradation studies respectively. In oxidation study, 2 mL of H₂O₂ was added, while in reduction study a 2 mL of sodium bisulphite was added. A small volume of diluent was added in all solutions, heated at 80C on water bath for 20 min and cooled. The solutions were neutralized if necessary and the volume was adjusted with diluent. In case of thermal degradation studies the diluent was added to adjust the volume in the flask and the solution was heated at 105C for 24 h, whereas in photolytic degradation, the solution was kept under sunlight for 24 h. During humidity stress investigation, the solution was subjected to 90% RH at 25C for 24 h. The above solutions were filtered through 0.45 µm nylon filter (Millipore^®) by discarding first 4 mL of solution. The blank solutions were prepared in the same manner without using samples.

2.6.6. Bench Top Solution Stability of Standard and Sample

To evaluate the solution stability, standard solution was prepared in single and samples were prepared in duplicate and kept on bench top. Solutions were analyzed at 0, 4, 8, 12 and 24 h. The differences of ABN content (%) for bench top samples were calculated from the results as follows: Difference of assay (%) = % of assay at each time interval (h) − % assay at initial time (0 h).

2.6.7. Assay of ABN in Bulk and Dosage Forms

Prior to the injection of drug (ABN) solutions, the column was equilibrated for at least 30 min with the mobile phase with a flow rate of 1.2 mL/min. Then 20 µL of standard and sample solutions were injected for six and two times respectively and the drug content was determined. For the estimation of ABN in dosage forms, accurately weighed 10 tablets of ABN were powdered. A 10 mg equivalent of the drug was transferred into 10 mL volumetric flask, the diluent was added, sonicated well, filtered through 0.45 µm filter (Millipore^®) and suitably diluted and injected into HPLC system and the drug content was determined using standard curve.

Table 1. Two Level Full Factorial Design and Response Values for Visible Spectroscopy

Download as

Veiw figure View Table

3. Results

3.1. Visible and UV Sspectroscopy

The matrix of the two-level full factorial design and experimental runs for visible spectroscopy is provided in Table 1. The contour plots of absorbance vs. FeCl₃ and MBTH concentration and volume are presented in Figure 2. Data of the calibration curve of ABN by visible and UV spectroscopy are summarized in Table 2. The percent assay of ABN by visible spectroscopy was 100.58−101.37% with 0.38% RSD and that detected by UV spectroscopy was found 99.1−99.8% with 0.5% RSD. Results of visible and UV spectroscopic method validations in terms of accuracy (recovery studies), precision, specificity, LOD and LOQ and robustness are provided in Table 3.

Figure 2. Contour plots of (A) absorbance vs FeCl₃ concentration and volume, (B) absorbance vs. FeCl₃ and MBTH volume, (C) absorbance vs. MBTH and FeCl₃ concentration, and (D) absorbance vs. MBTH concentration and volume

Download as

Veiw figure View Figure

Table 2. Data of the Calibration Curve of Ambrisentan by Visible and UV Spectroscopy

Download as

Veiw figure View Table

Table 3. Validation Studies of Ambrisentan by Visible and UV Spectroscopy (n=6)

Download as

Veiw figure View Table

3.2. RP−HPLC Method

The response values of experimental runs of two-level full factorial design on HPLC method are given in Table 4. The different contour plots of the RP-HPLC experimental design are shown in Figure 3. As regards method validation, the linearity of the method was demonstrated over the concentration range of 10−300 µg/mL with R² value not less than 0.99 and RSD of peak areas of the solution not more than 2.0%. From the assay calculations, the assay of ABN tablets was found 100.9% and 0.35% RSD. There was no interference due to blank and placebo at the retention time of the analyte (Figure 4). The recovery of the spiked drug was found 100.7% with 0.23% RSD which falls within the acceptance criterion. Regarding precision, the system precision or repeatability of the method showed 0.632% RSD for the peak areas. The label claim for method precision was found 100.6% with 0.23% RSD. For intermediate precision or ruggedness of the method, 0.23% RSD was found for the peak area. Thus all parameters of precision studies were less than 2.0. LOD and LOQ values were calculated as 1.967 and 5.961 respectively.

Table 4. Two Level Full Factorial Design and Response Values for RP-HPLC

Download as

Veiw figure View Table

Figure 3. Contour plots of (A) retention time vs. OPA concentration and flow rate, (B) retention time vs. wavelength and OPA concentration, (C) USP tailing factor vs. OPA concentration and flow rate, and (D) USP tailing factor vs. OPA concentration and wavelength

Download as

Veiw figure View Figure

Figure 4. Typical chromatograms of (A) blank and (B) placebo

Download as

Veiw figure View Figure

Results of robustness of the method as determined by deliberate alterations in the ratios of mobile phase composition, flow rate and wavelength range are presented in Table 5. The results of forced degradation studies are presented in Table 6. Results of bench top solution stability studies are provided in Table 7 and Figure 5.

Table 5. Robustness of the Developed RP−HPLC method (n=6)

Download as

Veiw figure View Table

Table 6. Forced Degradation Studies of Ambrisentan by Developed RP-HPLC Method (n=6)^a

Download as

Veiw figure View Table

Table 7. Bench Top Solution Stability Studies of Ambrisentan by Developed RP-HPLC Method (n=6)

Download as

Veiw figure View Table

4. Discussion

A simple, rapid and properly validated UV spectroscopic method was developed for estimation of ABN in bulk and tablet dosage forms. Earlier work reported a single UV method ^[10] for ABN quantifications using acetonitrile as the diluent, whereas the current method employed pH 7.4 phosphate buffer as the solvent. Two methods reported ABN estimation in visible range by forming colored ion-pair complexes of ABN with methylene blue and safranine ^[8] and by charge transfer reaction of ABN with dichloro benzoquinone derivatives ^[9]. But the present work applied principle of oxidation and coupling reaction of ABN with MBTH reagent in presence of FeCl₃ to form bluish-green chromogen which is detectable in the visible range. The two reported RP-HPLC methods ^{[11, 12]} used mobile phase in gradient mode but we developed an optimized method using isocratic separation mode.

Figure 5. Chromatograms of solution stability at (A) initial time, (B) 6 h, (C) 12 h and (D) 24 h

Download as

Veiw figure View Figure