Development and Validation of Bioanalytical Method for Determination of Nebivolol and Valsartan in Human Plasma by Using RP-HPLC
Article Category: Original Paper
Published Online: Mar 12, 2022
Page range: 49 - 58
Received: Apr 14, 2018
Accepted: Jun 01, 2018
DOI: https://doi.org/10.2478/afpuc-2021-0019
Keywords
© 2021 R. N. Kachave et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.
Nebivolol (NL) is chemically described as 1-(6-fluoro-3, 4-dihydro-2H-1-benzopyran-2-yl)-2-{[2-(6-fluoro-3, 4-dihydro-2H-1-benzopyran-2-yl)-2-ydroxyethyl]amino} ethan-1-ol (Figure 1a). Its molecular formula is C22H25F2NO4 and its molecular weight is 405.435 g/mol. Valsartan (VL) is chemically described as (2S)-3-methyl-2-[N-({4-[2-(2H-1, 2, 3, 4-tetrazol-5-yl) phenyl] phenyl}methyl) pentanamido] butanoic acid (Figure 1b). Its empirical formula is C24H29N5O3 and its molecular weight is 435.52 g/mol [1,2,3,4,5].
Figure 1a
Structure of Nebivolol.
Figure 1b
Structure of Valsartan.
VL belongs to the angiotensin II receptor blocker (ARB) family of drugs, which also includes telmisartan, candesartan, losartan, olmesartan and irbesartan. ARBs selectively bind to angiotensin receptor-1 (AT1) and prevent the protein angiotensin II from binding and exerting its hypertensive effects, which include vasoconstriction, stimulation and synthesis of aldosterone and ADH, cardiac stimulation and renal reabsorption of sodium, among others. Overall, the physiological effects of VL lead to reduced blood pressure, lower aldosterone levels, reduced cardiac activity and increased excretion of sodium. NL is a racemic mixture of two enantiomers, wherein one is a β-adrenergic antagonist and the other acts as a cardiac stimulant without β-adrenergic activity. Treatment with NL leads to a greater decrease in systolic and diastolic blood pressure than atenolol, propranolol or pindolol. NL and other β-blockers are generally not first-line therapies as many patients are first treated with thiazide diuretics. They are indicated for hypertension to lower the blood pressure and reduce the risk of fatal and non-fatal cardiovascular events, primarily strokes and myocardial infarction. NL is a competitive and selective β1-receptor antagonist, has little or no effect on β2 receptors at doses <10 mg, lacks intrinsic sympathomimetic and membrane-stabilising activity at therapeutically relevant concentrations and reduces systemic vascular resistance. VL blocks the binding of angiotensin II to type 1 angiotensin receptors, causing a reduction in blood pressure, and blocks vasoconstrictor and aldosterone-secreting effects of angiotensin II. Byvalson (VL and NL tablet) 80 and 5 mg is a fixed-dose combination for the treatment of hypertension [1,2,3,4,5,6,7,8].
The literature survey presents different methods available for the detection of NL and VL separately and also in a combined form. Ultra-performance liquid chromatography (UPLC) method for detection of both drugs [9], an isocratic (high-performance liquid chromatography [HPLC]) method [10], an ion-pair HPLC method [11], RP-HPLC method for the estimation of both drugs [12], ultraviolet (UV) simultaneous estimation of NL and VL [13] – these all are the methods available for the detection of NL and VL.
This study aims to develop a rapid analytical technique for the estimation of NL and VL in human plasma. The proposed method is a rapid, precise, selective and sensitive RP-HPLC method that has short and simple extraction procedures that consume small amounts of solvent and biological fluid for extraction and time. This method is used in the bioavailability/bioequivalence (BA/BE) study for analysis of NL and VL in biological samples like blood, plasma and urine.
NL and VL active pharmaceutical ingredients (APIs) were received as a gift sample from Spectrum Pharma Research solutions. HPLC-grade acetonitrile and methanol were procured from Rankem, Avantor Performance Material India Limited. Plasma used for the study was received from Deccan pathological lab, Hyderabad. Analytical grade phosphate buffer, potassium dihydrogen phosphate and orthophosphoric acid were also procured from Rankem, Avantor Performance Material India Limited.
Chromatographic separation was performed at 30°C of column temperature with the mobile phase consisting of 0.01 N potassium dihydrogen phosphate pH 3.0:acetonitrile in the ratio of 60:40. Separation of NL and VL was achieved on Symmetry C18 (150 × 4.6 mm, 5 μm) at a flow rate of 1.0 mL/min; the injection volume was 50 μL and detector wavelength was 280 nm.
Based upon the solubility of the drugs, diluent was selected; 0.01 N potassium dihydrogen phosphate and acetonitrile were taken in the ratio of 50:50.
Take 5 mg of NL in a 100-mL volumetric flask and add diluent to give a volume of 50,000 ng/mL. Take 1 mL of the prepared solution and dilute to 10 mL with diluent to produce 5000 ng/mL.
From the above-mentioned NL stock solution, 0.01, 0.02, 0.03, 0.08, 0.10, 0.12, 0.16 and 0.20 mL was pipetted and transferred to eight individual 10-mL volumetric flasks and the volume was made up to the mark with diluent to produce 5, 10, 15, 40, 50, 60, 80 and 100 ng/mL. Calibration standards and quality control (QC) samples were prepared by spiking blank plasma with working stock dilutions of analytes to produce 0.5, 1, 1.5, 4, 5, 6, 8 and 10 ng/mL.
Take 40 mg of VL in a 100-mL volumetric flask and add diluent to give a volume of 400,000 ng/mL.
From the above-mentioned VL stock solution, 0.1, 0.2, 0.3, 0.8, 1.0, 1.2, 1.6 and 2.0 mL was pipetted and transferred to eight individual 10-mL volumetric flasks, and the volume was made up to the mark with diluent to produce 4.0, 8.0, 12, 32, 40, 48, 64 and 80 μg/mL. Calibration standards and QC samples were prepared by spiking blank plasma with working stock dilutions of analytes to produce 400, 800, 1200, 3200, 4000, 4800, 6400 and 8000 ng/mL.
Fifty milligrams of atorvastatin was weighed, dissolved and diluted to 100 mL by diluent (500 μg/mL). From this stock solution, 2 mL was pipetted, diluted to 100 mL with diluent and used as a spiking solution (10 μg/mL).
Take 750 μL of plasma and add 500 μL of IS, 250 μL of NL and 250 μL of VL from the spiking solutions into a centrifuging tube; add 1 mL of acetonitrile. Mix by cyclomixer for 15 sec. Then vertex for 2 min and finally centrifuge for 5 min at 3200 rpm. After centrifugation, collect the supernatant, filter it and directly inject 50 μL into HPLC.
Take 1250 μL of the patient's sample and add 500 μL of IS into a centrifuging tube; add 1 mL of acetonitrile. Mix by cyclomixer for 15 sec. Then vertex for 2 min and finally centrifuge for 5 min at 3200 rpm. After centrifugation, collect the supernatant, filter it and directly inject 50 μL into HPLC.
Method validation for an analytical method is a process used to verify/confirm whether the newly developed method is suitable for its intended purpose as per the ICH guidelines [14,15,16,17,18,19]
Parameters involved in bioanalytical validation are as follows:
System suitability test Selectivity Matrix factor evaluation Precision and accuracy Linearity Recovery Stability
System suitability test was performed before each batch sample analysis to ensure the reproducibility of the chromatographic system. This test was performed by running six injections of the diluted drug and IS in the linear region of the calibration curve and measuring the percentage of relative standard deviation (% RSD).
To establish the selectivity of the method, possible interferences at the retention time (RT) of NL, VL and IS due to endogenous plasma components were checked during validation. Selectivity was performed by testing six lots of K2EDTA blank plasma, and the analyte detection of extracted plasma has good selectivity of both drugs and IS.
The combined effect of all components of the sample other than the analyte on the measurement of quantity. Three blank specimens from each of not less than six batches of the matrix under screening were extracted. For the matrix effect, LQC and HQC spiking dilutions were spiked, and IS dilutions were also spiked in the previously extracted blank specimens.
The intra-day and inter-day accuracy and precision were assessed by analysing six replicates at four different QC levels like LLOQ, LQC, MQC and HQC. Accuracy and precision method performance was evaluated by six replicate analyses for NL at four concentration levels, that is, 0.5 ng/mL (LLOQ), 1.5 ng/mL (LQC), 5 ng/mL (MQC) and 8 ng/mL (HQC), and for VL at 400 ng/mL (LLOQ), 1200 ng/mL (LQC), 4000 ng/mL (MQC) and 6400 ng/mL (HQC). The intra-day and inter-day accuracies of plasma samples were assessed and excellent mean % accuracies were calculated. The precision of the method was expressed in terms of % RSD, and accuracy was expressed as a percentage of theoretical concentration (observed concentration × 100/theoretical concentration).
The calibration curve in a bioanalytical method is a linear relationship between concentration (independent variable) and response (dependent variable) using a least-squares method. This relationship is built to predict the unknown concentrations of the analyte in a complicated matrix. The linearity sample should consist of a blank sample, a zero sample and six to eight non-zero samples covering the expected range, including LLOQ.
Recovery was determined by comparing the peak areas obtained from prepared plasma samples with those extracted from blank plasma spiked with the same amount of NL and VL standards at three QC concentration levels.
Long-term stock solution stabilities for NL and VL were determined at LQC and HQC concentration levels after a storage period of 37 days at −20°C and −70°C in the deep freezer. The % means stabilities of NL and VL were calculated.
Chromatographic separation of NL and VL by RP-HPLC method was performed by various mobile phase trials. The ideal separation of NL and VL was obtained by using a mobile phase consisting of 0.01 N potassium dihydrogen phosphate (pH 3.0):acetonitrile in the ratio of 60:40. A representative chromatogram of the optimised chromatographic method is shown in Figure 2.
Figure 2
Chromatogram of Optimized Chromatographic Method.
All the system suitability parameters were within the range and satisfactory as per the ICH guidelines. The % coefficient of variation (CV) for system suitability test was in the range of 1.0 ± 0.0316 for RT of NL, 1.18 ± 0.0512 for RT of VL and 0.23% for the area ratio (analyte area/IS area) of atorvastatin. Full results of the system suitability test are shown in Table 1.
Observation of optimised method chromatogram.
| RT | 2.597 | 3.189 | 4.374 |
| Area | 97,579 | 5671 | 11,302 |
| USP plate count | 9792.8 | 9135.0 | 8011.3 |
| USP tailing | 1.4 | 1.4 | 1.1 |
| USP resolution | 4.6 | 6.9 | |
| Area ratio | 0.05686 | 0.3545 | |
| SD | 0.0316 | 1.00 | |
| % CV | 0.0512 | 1.18 |
Evaluation of selectivity was performed by testing six batches of K2EDTA blank plasma. The analysis of extracted blank plasma showed that the developed method has good selectivity for both drugs and IS. Representative chromatograms are shown in Figure 3 of standard blank and blank with IS sample using pooled plasma.
Figure 3
Extracted standard blank sample.
The matrix effect plays a vital role in the assessment of pharmacokinetic studies. It was expressed as an IS-normalised matrix factor and it varied from 0.90 to 0.99, which was close to 1, which indicates that there was no analytical signal suppression or enhancement in plasma samples. Results are shown in Table 2.
Matrix factor evaluation of nebivolol and valsartan.
| n | 18 | 18 | 18 | 18 |
| Mean | 7.9592 | 1.4937 | 6400.5556 | 1190.4444 |
| SD | 0.07785 | 0.01768 | 44.02035 | 30.91016 |
| % CV | 0.98 | 1.18 | 0.69 | 2.60 |
| % Mean accuracy | 99.49 | 99.58 | 100.01 | 99.20 |
| No. of QC failed | 0 | 0 | 0 | 0 |
The intra-day and inter-day accuracies of plasma samples were assessed and excellent mean % accuracies for NL and VL were obtained with range varying from 99.32% to 99.70% and from 99.13% to 100.44% for intra-day accuracy and from 99.31% to 99.61% and from 99.84% to 99.98% for inter-day accuracy, respectively. The precision (% CV) of the analytes in the plasma samples was calculated and found to be 0.66%–2.65% and 0.17%–2.68% for intra-day accuracy and 0.94%–2.95% and 0.19%–2.00% for inter-day accuracy, respectively. Results are shown in Tables 3–5.
Intra-day precision and accuracy of nebivolol and valsartan.
| n | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 |
| Mean | 7.9513 | 4.9662 | 1.4943 | 0.4985 | 6403.1667 | 4017.5000 | 1196.000 | 396.5000 |
| SD | 0.05211 | 0.10071 | 0.01707 | 0.01320 | 10.60974 | 107.61924 | 11.6666 | 8.06846 |
| % CV | 0.66 | 2.03 | 1.14 | 2.65 | 0.17 | 2.68 | 0.97 | 2.03 |
| % Mean accuracy | 99.39 | 99.32 | 99.62 | 99.70 | 100.05 | 100.44 | 100.44 | 99.13 |
Inter-day precision and accuracy of nebivolol and valsartan.
| n | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 |
| Mean | 7.9238 | 4.9362 | 1.4907 | 0.4977 | 6394.3333 | 3999.0000 | 1194.8333 | 393.5000 |
| SD | 0.08962 | 0.08039 | 0.01770 | 0.01799 | 15.83246 | 70.11419 | 11.72035 | 6.05805 |
| % CV | 1.13 | 1.63 | 1.19 | 3.61 | 0.25 | 1.75 | 0.98 | 1.54 |
| % Mean accuracy | 99.05 | 98.72 | 99.38 | 99.53 | 99.91 | 99.98 | 99.57 | 98.38 |
Batch-to-batch precision and accuracy of nebivolol and valsartan.
| n | 18 | 18 | 18 | 18 | 18 | 18 | 18 | 18 |
| Mean | 7.9577 | 4.9657 | 1.4917 | 0.4981 | 6398.5000 | 3993.7222 | 1194.17 | 396.1667 |
| SD | 0.07478 | 0.08347 | 0.01596 | 0.01470 | 11.92254 | 79.93071 | 11.1596 | 7.10634 |
| % CV | 0.94 | 1.68 | 1.07 | 2.95 | 0.19 | 2.00 | 0.93 | 1.79 |
| % Mean accuracy | 99.47 | 99.31 | 99.44 | 99.61 | 99.98 | 99.84 | 99.44 | 99.04 |
Calibration was found to be linear over the concentration range of 0.5–10 ng/mL for NL and 400–8000 ng/mL for VL. The coefficient of determination (
Figure 4a
Calibration Curve of Nebivolol.
Figure 4b
Calibration Curve of Valsartan.
Observation table for linearity of nebivolol and valsartan.
| 0.5 | 570 | 0.006 | 400 | 3415 | 0.0350 |
| 1 | 1235 | 0.013 | 800 | 6724 | 0.0689 |
| 1.5 | 1785 | 0.018 | 1200 | 10,160 | 0.1041 |
| 4 | 4643 | 0.048 | 3200 | 26,678 | 0.2731 |
| 5 | 5880 | 0.060 | 4000 | 34,216 | 0.3506 |
| 6 | 6902 | 0.071 | 4800 | 41,162 | 0.4217 |
| 8 | 9176 | 0.094 | 6400 | 52,728 | 0.5402 |
| 10 | 11,175 | 0.115 | 8000 | 67,420 | 0.6910 |
| Correlation coefficient | 0.998 | 0.9994 | |||
| Regression equation | y = 1140× | 8.4073× | |||
| Linearity range | 0.5–10 ng/mL | 400–8000 ng/mL | |||
The overall % mean recoveries for NL and VL were found to be 97.96% and 98.11%, respectively. The recoveries were obtained for NL and VL at three QC concentration levels. The overall % mean recovery for atorvastatin (IS) was found to be 99.00%. The results are shown in Table 7.
Recovery study of nebivolol and valsartan.
| n | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 |
| Mean | 9445 | 9253 | 5965 | 5782 | 1886 | 1857 | 54,658 | 53,310 | 35,376 | 34,761 | 10,665 | 10,507 |
| SD | 59.17 | 70.20 | 53.57 | 66.05 | 12.32 | 31.02 | 657.59 | 400.50 | 416.36 | 210.54 | 143.61 | 70.53 |
| % CV | 0.62 | 0.75 | 0.89 | 1.14 | 0.65 | 1.67 | 1.20 | 0.75 | 1.17 | 0.88 | 1.34 | 0.67 |
| % Mean recovery | 97.96 | 96.93 | 98.46 | 97.54 | 98.26 | 98.52 | ||||||
| Overall % mean recovery | 97.784 | 98.112 | ||||||||||
The % mean stability of six replicates of LQC and HQC samples (1.5 and 8 ng/mL) was analysed after 9 h of storage at room temperature on the laboratory bench. These were calculated and found to be 99.51% for LQC and 99.92% for HQC. The result is shown in Table 8.
Long-term stock solution stability of nebivolol and valsartan.
| n | 6 | 6 | 6 | 6 |
| Mean | 7.9938 | 1.4927 | 6391.6667 | 1193.5000 |
| SD | 0.01950 | 0.02082 | 21.42584 | 20.08731 |
| % CV | 0.24 | 1.39 | 0.34 | 1.68 |
| % Mean accuracy | 99.92 | 99.51 | 99.87 | 99.46 |
The % mean stability of six replicates of LQC and HQC samples (1200 and 6400 ng/mL) was analysed after 9 h of storage at room temperature on the laboratory bench. These were calculated and found to be 99.51% for LQC and 99.92% for HQC. The result is shown in Table 8.
The % mean stability of NL was found to be 100.18% (LQC) and 98.68% (HQC) at 20°C ± 5°C and 99.65% (LQC) and 100.33% (HQC) at 70°C ± 5°C. The % mean stability of VL was found to be 99.61% (LQC) and 99.95% (HQC) at 20°C ± 5°C and 99.51% (LQC) and 100.23% (HQC) at 70°C ± 5°C. The results are shown in Tables 9a and b.
Matrix samples’ stability of nebivolol and valsartan at −20°C ± 5°C for 37 days.
| n | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 |
| Mean | 7.9930 | 7.8872 | 1.4872 | 1.4898 | 6394.5000 | 6391.1667 | 1193.1667 | 1188.5000 |
| SD | 0.02481 | 0.15394 | 0.02729 | 0.01393 | 23.57753 | 18.50856 | 20.48821 | 18.98157 |
| % CV | 0.31 | 1.95 | 1.83 | 0.94 | 0.37 | 0.29 | 1.72 | 1.60 |
| % Mean accuracy | 99.91 | 98.59 | 99.14 | 99.32 | 99.91 | 99.86 | 99.43 | 99.04 |
| % Mean stability | 98.68 | 100.18 | 99.95 | 99.61 | ||||
Matrix samples’ stability of nebivolol and valsartan at −70°C ± 5°C for 37 days.
| n | 6 | 6 | 6 | 6 | 6 | 6 | 6 | 6 |
| Mean | 7.9880 | 8.0145 | 1.4953 | 1.4902 | 6386.5000 | 6401.5000 | 1194.5000 | 1188.6667 |
| SD | 0.02458 | 0.05858 | 0.02273 | 0.01646 | 22.29574 | 21.37054 | 17.14351 | 14.17980 |
| % CV | 0.31 | 0.73 | 1.52 | 1.10 | 0.35 | 0.33 | 1.44 | 1.19 |
| % Mean accuracy | 99.85 | 100.18 | 99.69 | 99.34 | 99.79 | 100.02 | 99.54 | 99.06 |
| % Mean stability | 100.33 | 99.65 | 100.23 | 99.51 | ||||
After studying literature about NL and VL, the method development was started according to hydrophobic and hydrophilic interaction of the drugs. The absorbance spectra of both drugs were measured, overlayed and the wavelength of optimal sensitivity of determination was chosen (280 nm). The IS was selected based on the presence of chemical properties similar to the compounds of interest. Atorvastatin was chosen as the IS. After selecting the wavelength, the next step was the selection of the extraction procedure by observing the % recovery. The protein precipitation method was selected as the extraction procedure for better recovery. After selection of the extraction method, the method development process started with a flow rate of 1.0 mL/min at a column temperature 30°C, mobile phase 0.1% orthophosphoric acid (pH 2.2):acetonitrile (50:50), column Agilent C18 (150 mm × 4.6 mm, 5 μm) and injection volume 50 μL. In these parameters, VL was not eluted. So, further trial was carried out. In the second trial, the mobile phase and its ratio were changed to 0.01 N potassium dihydrogen phosphate (pH 3.0):acetonitrile (55:45) and other chromatographic conditions were the same as first trial. In the second trial, the resolution between atorvastatin and NL was unacceptable. So, third trial was carried out using 0.1% orthophosphoric acid (pH 2.2):acetonitrile (60:40) mobile phase and Symmetry C18 (150 mm × 4.6 mm, 5 μm) column. VL was not eluted in this trial. In the fourth trial, mobile phase 0.01 N potassium dihydrogen phosphate (pH 3.0):acetonitrile (70:30) and column Agilent C18 (150 mm × 4.6 mm, 5 μm) were used. In this trial, NL United States Pharmacopeia (USP) plate count was low and RTs were unnecessarily high. So, in the next trial, 0.01 N potassium dihydrogen phosphate (pH 3.0):acetonitrile (60:40) was used as the mobile phase and Symmetry C18 (150 × 4.6 mm, 5 μm) was used as the column. All peaks eluted with good peak shape and RT. The tailing tests were also passed. So, this method is used as an optimised method for the separation of NL and VL from human plasma. After optimisation was completed, the method was evaluated for various validation parameters as per the ICH M10 guideline.
A simple, accurate and precise method was developed for the estimation of NL and VL in human plasma using atorvastatin as the IS. RTs of NL and VL were found to be 3.146 and 4.346 min, respectively. The % CV of NL and VL was found to be 1.0% and 0.58%, respectively. Per cent recoveries were obtained as 98.28% and 98.11%, respectively. The linearity concentrations were in the range of 0.5–10 ng/mL for NL and 400–8000 ng/mL for VL. The lower limits of quantification were 0.5 ng/mL for NL and 400 ng/mL for VL, which reached the levels of both drugs possibly found in human plasma. Further, the reported method was validated as per the ICH guidelines and found to be well within the acceptable range. The proposed method is simple, rapid, accurate, precise and appropriate for pharmacokinetic studies and therapeutic drug monitoring in clinical laboratories.