Stability-Indicating High-Performance Liquid Chromatographic Determination of Apixaban in the Presence of Degradation Products

Abstract A simple, robust, and stability-indicating reversed-phase high-performance liquid chromatographic (HPLC) method for the analysis of apixaban and its related substances has been successfully developed. Chromatography was performed on a 250 mm × 4.6 mm, 5 μm C18 column with a gradient mixture of a phosphate buffer–methanol 60:40 (v/v) at 1.0 mL min-1. Ultraviolet detection of apixaban was at 220 nm. The method was validated for linearity, precision, repeatability, sensitivity, and selectivity. Selectivity was validated by subjecting apixaban solution to photolytic, acidic, basic, oxidative, and thermal degradation. The peaks from the degradation products did not interfere with that from apixaban. The method was used to quantify the related substances in apixaban in the bulk drug and can be used for routine quality control purposes.

The structure of apixaban The FDA approved apixaban in December 2012 with an indication of reducing the risk of stroke and dangerous blood clots (systemic embolism) in patients with atrial fibrillation that is not caused by a heart valve problem [2,3].
Apixaban appears as a white-to-pale yellow, non-hygroscopic crystalline powder, with an aqueous solubility of 0.028 mg/mL at 24°C. Apixaban is a non-ionisable compound and its partition coefficient at 24°C is 44.7 (log Po/w = 1.65) at pH 7.4 (n-octanol / aqueous buffer). The molecule has no chiral centres, therefore, no stereoisomers exist [4].
There are some methods of estimation of apixaban from human plasma by LC-MS [5,6], but there is no related substance method for apixaban by HPLC using a UV detector.

Fig. 2.
Chemical names and structures of apixaban's process impurities

Results and Discussion
The present study was aimed at developing a liquid chromatographic method for the separation and quantitative determination of apixaban and its related impurities.
The structures of apixaban and its known impurities (RS1, RS2) are depicted in Fig. 2. The compounds RS1 and RS2 are intermediates.

Optimization of HPLC Conditions
The chromatographic conditions were optimized with respect to specificity, resolution, and time of analysis. The specificity of the method was established through the study of the resolution factor of the apixaban peak from the nearest resolving peak. Peaks were identified using retention times compared with those of standards and the characteristic spectra were confirmed by photodiode array detection (range 200-400 nm) and spectroscopic techniques (NMR and mass).
Several binaries were tested using different proportions of solvents, such as acetonitrile, methanol, and water. A flow rate of 1.0 mL/min was selected after preliminary tests.
Effects of pH (2.5-6) were investigated using phosphate and acetate buffers. Initial attempts were made by trying different proportions (or compositions) of acetonitrile with a buffer, but there was an interference of the blank with the RS2 peak. Columns of different compositions viz. Cosmosil, Qualisil were tried, but there was a problem with peak tailing and peak dissymmetry. Methanol-potassium dihydrogen phosphate with 0.1% TEA (and the pH of the buffer adjusted to 3.0 with diluted orthophosphoric acid (pH 4.5)) in gradient mode was found to achieve the complete separation of apixaban and known impurities within 60.0 min. In the optimized conditions, apixaban was well-separated from RS1 and RS2 as shown in Fig. 3.

Structural Elucidation of Related Impurities and the Drug Substance
The identification of apixaban and its known impurities (RS1) and (RS2) was confirmed by ESI-MS, 1 H NMR, and FT-IR. The spectral data on the 1 H NMR chemical shift, FT-IR, and m/z of apixaban and the impurities are presented in tables 1, 2, and 3.

Results of Forced Degradation
When apixaban was subjected to solution state forced degradation (acid, alkaline, and peroxide), the drug molecule degraded up to 12% under strong acidic conditions (heating at 60°C for 3.5 hours), but in alkaline conditions and peroxide it was stable (Figures 4-6). It was observed that apixaban did not degrade under UV light (photolytic) and heating at 105°C (thermal conditions) ( Table 4). The unknown degradants produced in the forced degradation were well-separated (Rs > 2.0) from apixaban and its known impurities. The peak homogeneity of apixaban and its impurities was verified using a photodiode array detector by using Empower software which indicated that the peak purities of all components were passing.

Tab. 4.
Results from stress testing of apixaban

Linear Range
Calibration plots were linear over the range of 0.01-0.22 μg mL −1 . Linear regression analysis of the concentration against the peak-area ratio resulted in an average coefficient of determination (R) greater than 0.99. The linear regression results (Table 5) indicated that detector responses at 220 nm were linear over the concentration range studied [6].

Limits of Quantitation and Detection
The LOD and LOQ were determined for apixaban and each of the impurities based on the % relative standard deviation (SD) of the response.

System Suitability Test
The solution of apixaban 0.1% w/w with 0.15% w/w of known impurities was analyzed during the validation studies. It may be noted that the resolution value of more than 2.0 was achieved for all the compounds under study with no significant variation noticed in the RRTs and tailing factors of the compounds. The percentage relative standard deviation of the peak of the six replicate injections of the standard solution was observed to be no more than 5.

Accuracy
The recovery of each impurity was observed in the range of 93.4-100.5% (Table 7).

Robustness
The robustness of the method was checked by doing deliberate changes in the method parameters viz. flow rate (± 20%), column temperature (± 2°C), and change in pH (± 0.2).
It was observed that the resolution between apixaban and RS1 was more than 1.5 and the %RSD of the six replicate standard injections of the impurity solutions was no more than 5. It is shown in Table 8.

Tab. 8.
Results from the evaluation of the robustness parameter for apixaban

Stability of the Analytical Solution
The % difference in the peak area of all the known and unknown impurities from the initial to relevant time interval was calculated. Based on the data, the sample solution was declared to be stable for at least 24 h at room temperature.

Application of the method
Three batches of APX drug substance were analysed using the proposed method. The levels of impurities relative to APX were in the range of 0.07-0.08% and total impurities were in the range of 0.39-0.43%. It is shown in Table 9.

Tab. 9.
Results of analysis of apixaban for impurities

Analytical Procedure
Apixaban reference standard at 1.0 mg/mL and subsequent concentrations were prepared in diluent (water: acetonitrile), respectively. Solutions of all the impurities (0.15% w/w) were prepared by dissolving known amounts of the compounds initially in diluent and diluted up to the mark with the same diluent. These solutions were prepared freshly and diluted further quantitatively to study the validation attributes. The specification limit considered for the validation studies was 0.15% for the known impurity. Known and unknown impurities were determined against the mean areas obtained from six replicate injections of the standard solution and relevant impurity standard solution.

Method Validation
Validation of the chromatographic method was carried out with regards to linearity, sensitivity (limit of detection and limit of quantification), precision, accuracy, and robustness.