Two stability-indicating methods were developed for the determination of atomoxetine hydrochloride (ATM) and validated in the presence of its degradation products. Method I is based on (UPLC) separation of ATM from its alkaline, oxidative, and acidic degradation products on Zorbax SB C18 column using acetonitrile -aqueous 0.01M triethylamine, pH 4.2 (50:50, v/v) mobile phase. Photodiode array detection at 205 nm was used for quantitation of ATM over the range of 0.1-35 μg/ml. The run time was 2.5 min within which ATM and its degradation products were well separated. The method was also applied to the determination of ATM in spiked human plasma over the range of 0.1-4 μg/ml. Moreover, the produced acidic degradation products were isolated, and structural elucidation of the degradates was done by LC/MS spectrometry studies. A proposal of the acid hydrolysis pathway was presented. Method IIA describes direct measurement of the intrinsic fluorescence intensity of both ATM and its known acid degradates using sodium dodecyl sulfate as fluorescence enhancer in aqueous solutions. This method was extended to (Method IIB) to apply first derivative synchronous fluorescence spectroscopy for the simultaneous analysis of ATM and its acidic depredates. The proposed methods were successfully applied to quantify ATM in commercial capsules and the results were in good agreement with those obtained using a reference method.
Keywords: Atomoxetine HCL; UPLC; Stability indicating studies; Spiked human plasma; First derivative; Synchronous fluorescence spectroscopy
Atomoxetine HCl (ATM), (R) n-methyl-3-(2-methylphenoxy)-3- phenyl propylamine hydrochloride, has a formula of C17H21NO. HCl and MW of 291.8 (Figure 1). It is a norepinephrine selective reuptake inhibitor that is used in the treatment of ADHD (attention-deficit hyperactivity disorder). It is thought to improve ADHD symptoms by blocking norepinephrine reuptake and thereby increasing norepinephrine levels in the prefrontal cortex. ATM has several advantages over the amphetamines, including a lower abuse/addiction potential and a longer plasma half-life that allows for once daily dosing. ATM increases peripheral as well as central norepinephrine levels, and thus increases heart rate and blood pressure [1,2].
The assay of ATM is not officially recorded up till now in any pharmacopoeia, owing to the therapeutic importance of ATM, Various analytical procedures have been established for their quantitative determination in drug substance, drug products, and/or biological fluids, either singly or in combination. These procedures include spectrophotometry and/or fluorimetry [3-6], high performance thin layer chromatography , voltammetry , HPLC [9-11], and LC/MS [12,13]. Forced degradation study was not reported in these articles [4-13]; in addition, a paper on the stability indicating HPLC assay of ATM in presence of unknown degradation products was developed . However, the paper did not deals neither with the determination of ATM in spiked plasma nor with analysis of the drug in presence of known degradates using (DSF) derivative synchronous fluorometry.
Ultra performance liquid chromatography (UPLC) is a recent technique in LC, which enables significant reduction in separation time and solvent consumption. Literature survey reveals that UPLC system allows about nine fold decreases in analysis time as compared to the conventional HPLC system using 5 μm particle size analytical columns [15,16]. The proposed UPLC method for analysis of ATM in authentic samples, spiked human plasma and capsules has yet been developed before. Identification and structural elucidation of the acid degradate were studied. The three major degradation products were confirmed as; N-methyl-3-hydroxy-3-phenylpropylamine; N-methyl-3-phenyl-2, 3-propenylamine; and O-Cresol).
Synchronous fluorescence spectroscopy (SFS) has been found to have several advantages , such as simple spectra, high selectivity, low interference, etc. It has attracted the attention of many researchers and developed rapidly since it was firstly proposed by LIoyd .
The combination of SFS and derivative synchronous fluorometry (DSF) is more advantageous than differentiation of the conventional direct spectrofluorometry in terms of sensitivity, because the amplitude of the derivative signal is inversely proportional to the bandwidth of the original spectrum .
The normal synchronous fluorescence spectra of ATM and its acidic degradation products were greatly overlapped. This observation led us to utilize first derivative synchronous fluorescence spectroscopy (FDSFS) to separate the drug and its acidic degradation without any interference. The peaks intensities of ATM and the acid degradates were measured at 276 and 265 nm, respectively.
Up till now neither direct nor (DSFS) method has been reported for the simultaneous analysis of ATM and its acid degradates. The proposed method able to prevent cross-interferences arising from absorption and/or emission by the acid degradate; therefore our target was to develop simple sensitive and selective stability-indicating UPLC and DSF methods for analysis of the drug. The parent drug stability guidelines issued by the international conference of harmonization (ICH) .
1. UPLC was performed with Agilent 1200 SL RRLC auto sampler instrument containing a Bin pump SL, model G131213, an Autosampler injector ALS SL, model G132913, and a photo diode array detector (PDAD) SL detector, model G1315C. Compounds were separated on Zorbax SB C18 column (100 mm×2.1 mm i.d, 1.8 μm particle size). A Radwag analytical balance, model 60/220/x and an ultrasonic sonicator 1505JAC were used. The solvents were filtered through a 0.45 μm membrane filter (Millipore, Milford, MA) before use.
2. The fluorescence spectra and measurements were recorded using Hitachi F-7000 FL Spectrofluorometer; model 5J1-0004 connected with FL Solutions 2.1 Software program. It is equipped with a 150 watt Xenon arc lamp. Slit widths for both monochromators were set at 5 nm. A 1 cm quartz cell was used. Derivative spectra can be evaluated using FL Solutions 2.1 Software.
3. Degradation products were identified by use of liquid chromatography/mass spectrometry, performed in the Center of Applied Research and Advanced Studies, Faculty of pharmacy, Cairo University. Mass spectrometry was performed with LC/MS Waters Acquity quadrupole spectrometer using TQ detector.
Atomoxetine HCl (ATM) was kindly supplied by Lilly Company, Cairo, Egypt, certified to contain 99.72 ± 1.32%, according to the reported HPLC method.
Strattera® capsules (BN# A944003), Lilly Company, Spain; each capsule Contains 40 mg atomoxetine base equivalent to 45.71 mg atomoxetine HCl/capsule
Chemicals and solvents
• Acetonitrile and methanol (HPLC grade Lab-Scan, Poland), Triethylamine and hydrochloric acid (Fischer Chemical, UK), orthophosphoric acid and hydrogen peroxide (Adwia, Egypt), sodium hydroxide (BDH,UK).
• Human serum plasma, (Cealb Co., Amsterdam, B.N. 07126H122A, Biotest Pharma GmbH 63303 Dreieich, Germany, Lot. N. A137054 was kindly supplied form Vacsera, Egypt, and four samples were obtained from National Blood Bank, Egypt, and frozen until use after gentle thawing. Distilled water and the mobile phase used for UPLC were prepared by double glass distillation and filtration through a 0.47 μm membrane filter (Alltech Associates, USA).
UPLC method (Method I)
Zorbax SB C18 column (100 mm×2.1 mm i.d, 1.8 μm particle size) was used as the stationary phase at ambient temperature. The mobile phase was Acetonitrile-aqueous 0.01M triethylamine pH 4.2 50:50 (v/v) at a flow rate of 0.2 ml/min, Elution was monitored with a Photodiode array detection at 205 nm, and the injection volume was 0.8 μl. Triethylamine solution was adjusted to pH 4.2 with orthophosphoric acid. All measurements were performed at ambient temperature.
Spectrofluorometric method (Method II)
Instrumental parameters: Measurement type: 2-D; wavelength scan, Scan mode: Emission; Excitation; Synchronous, Data mode: Fluorescence, EX sampling interval: 20 nm, EM sampling interval: 10 nm, EX and EM Slit: 10 nm, Scan speed: 12000 nm/min, PMT Voltage 400 V.
Processing performed derivative: Derivative order: 1, Smoothing order: 4, Number of points: 99
Peak integration: Integration method: Rectangular, Sensitivity: 1, Threshold: 1
Stock standard solution of ATM for (Method I and Method II): A stock solution was prepared by dissolving 100 mg of ATM in 100 ml distilled water (1 mg/mL).
A standard solution of fluoxetine HCl (FLX) for (Method I): (0.3 mg ml-1) was prepared in mobile phase. Further dilutions with the same solvent in a 10 ml volumetric flask containing (30 μg ml-1) of FLX was used as internal standard I.S. The standard solutions were stable for 7 days when stored at 4°C.
Stock solutions of alkaline, acidic or oxidative degradation products: 100 mg ATM powder were accurately weighed and transferred into three separate 250-mL round flask, each was dissolved in 20 ml water then refluxed with 100 ml of each of 2M NaOH, 5MHCL or H2O2 3% for 1, 10, or 1 h, respectively, at thermostatically controlled water bath at 80°C. The solutions were left at room temperature. After neutralization of alkaline and acidic solutions, the three solutions were evaporated to dryness under vacuum (in a rotavapour device). The residues were then extracted with methanol (4×20 mL), to exclude NaCl resulted from the neutralization process, filtered into 100 mL volumetric flask and completed to volume with methanol to get (1000 μg/mL) stock solution for each degradation product.
Working solution of ATM for (Method I and Method II)
Ten milliliters of ATM stock solution were transferred into two separate 100-mL volumetric flask and completed to the volume with distilled water to get (0.1 mg/mL) working solution for each. Then further dilutions were made in the mobile phase or distilled water for (Method I and Method II), respectively. All solutions were stored at 4°C
Working solutions of alkaline, acidic or oxidative degradation products
Ten milliliters of each stock degradation product (1000 μg/mL) were transferred into three separate 100 mL volumetric flask, and completed to the volume with mobile phase or distilled water for UPLC and fluorometric methods, respectively, to get (100 μg/mL) working solution for each degradation product.
Aliquot volumes of each of (alkaline, acidic or oxidative degradation product) solutions were transferred to 10 ml volumetric flask, and then (Method I) was performed as described under Construction of Calibration Curve.
Acid degradation products were identified by use of LC/MS. Then aliquot volumes of (acidic degradation product) solution were transferred to 10 ml volumetric flask and then (Method II) was performed as described under Construction of Calibration Curve.
UPLC method (Method I)
Linearity and construction of calibration curve: Portions (0.01- 3.5 ml) of ATM standard solution (0.1 mg ml-1) were transferred into a series of 10 ml measuring flasks, completed to volume with the mobile phase to get the working concentrations 0.1-35 μg ml-1 of ATM. The procedure was completed as under experimental conditions. The calibration graph was plotted representing the relationship between the recorded area under the peak ratio and concentration, and then the corresponding regression equation was computed.
Analysis of laboratory-prepared mixtures: The peak areas of different laboratory prepared mixtures containing different ratios of ATM and the acidic degradation products (5-90%) were measured and the concentration of the drug in each mixture was obtained by applying in the corresponding regression equation.
Spectrofluorometric method (Method II)
Linearity and construction of calibration curve: Aliquots of ATM standard working solution were transferred into a series of 10 mL volumetric flasks to get final concentration range (1-12 µg ml-1) of ATM followed by adding 2 mL of 3.5 mM SDS and completed to the volume with distilled water.
Aliquots of acidic degradation working solution were transferred into a series of 10 mL volumetric flasks to get final concentration range (1-7 µg ml-1) of the degradate, follwed by adding 2 mL of 3.5 mM SDS and completed to the volume with distilled water.
Synchronous fluorescence spectra of the solutions were recorded by scanning both monochromators at a constant wavelength difference at Δλ=50 nm as under the instrumental parameters. The fluorescence intensities of the normal synchronous spectra for ATM and the acid degradates were estimated at 225 nm and 275 nm, respectively.
Linearity and construction of calibration curve: After following the procedures mentioned in method IIA, the first derivative fluorescence spectra of ATM and its acid-induced degradation products was derived from the normal synchronous spectra using FL Solutions software. The fluorescence intensity of the first derivative synchronous spectra was estimated at 276 and 265 nm for ATM and its acid degradates, respectively.
A blank experiment was performed simultaneously. The fluorescence intensity of the first derivative technique was plotted versus the final concentration of the drug (μg/mL) to get the calibration graph. Alternatively, the corresponding regression equations were computed.
Analysis of laboratory-prepared mixtures: The fluorescence intensities of different laboratory prepared mixtures containing different ratios of ATM and its acid degradates in the ranges of 5-90% were measured as under the instrumental parameters, and percentage recoveries of the drug were calculated.
Application to commercial capsules
The content of ten capsules of Strattera® was emptied and mixed well. An amount of the powder equivalent to 100 mg of ATM hydrochloride was weighed and transferred into a 100 ml volumetric flask, about 80 ml of water was added and the flask was sonicated for 15 min, mixed well and filtered, then completed to the volume to get 1000 μg/ml stock solution. Then as mentioned in details under each method appropriate working solutions of ATM were prepared and analyzed as described under Construction of Calibration Curve.
The drug concentrations were calculated from the corresponding regression equations. The validity of each method was assessed by applying the standard addition technique by mixing different concentrations of the standard drug to a fixed amount of its formulation. The concentrations of standard added were calculated from the corresponding regression equations
Application to spiked human plasma
Control sample of plasma (0.5 ml) was spiked with increased quantities of ATM, 1 ml of I.S solution (30 ug/ml) was also added and transferred into a series of 10 ml polypropylene tubes to get a final concentration range (0.1-4 µg/ml) of ATM. The samples were stored in a freezer at -20°C until analysis, and then allowed to thaw at 25°C before processing. The plasma samples were centrifuged at 4000 rpm for 10 min, for each concentration, 2 ml of acetonitrile was added to an aliquot containing (plasma, drug and IS). The mixture was vortex mixed briefly, and after standing for 5 min at room temperature, the mixture was centrifuged at 4000 rpm for 20 min. The supernatant was carefully transferred into vial and injected into UPLC system, then procedure was performed as described under UPLC method (Construction of Calibration Curve). The nominal content of the drug was determined using the corresponding regression equation.
Quality control (QC) samples
QC samples were prepared separately and pooled at three different concentration levels (0.5-1.5-2 μg/ml). The samples were stored in a freezer at -20°C until analysis. A calibration curve was constructed from a blank sample (plasma sample processed without I.S), a zero sample (plasma processed with I.S) and seven non-zero samples covering the total range (0.1-4 μg/ml).
For UPLC method
The aim of the present study was to develop a stability-indicating RP-UPLC assay method for separation of ATM and its degradation products in authentic sample, pharmaceutical and spiked human plasma. The experimental conditions mobile phase, several columns, different PH and wavelengths were optimized. Acceptable resolution with reasonable peak shapes and peak purity was achieved by using Acetonitrile-0.01M triethylamine pH 4.2 (50:50) (v/v), pH adjusted with orthophosphoric acid, as the mobile phase.
Also dissolving solvent has a great effect on separation, when mobile phase was used as a dissolving solvent; the peak of ATM eluted very quickly and interfered with endogenous biological substances. So water was the best choice for good separation and resolution.
Fluoxetine HCl was chosen as I.S because it was eluted with reasonable resolution from ATM at its absorbance characters that show high absorbance at the chosen wavelength 205 nm, that increases the sensitivity of the method.
Stability indication of the method
Forced degradation studies: In order to establish whether the analytical method was stability-indicating or not, ATM was exposed to different forced degradation studies. The stability of ATM has been studied and the degradation behavior of the drug under individual stress conditions is described below (Table 1).
|Stress conditions and time studies||Degradation % of the drug|
|Oxidizing medium 3 % H2O2, 80°C, 1 h||25%|
|Basic medium 2N NaOH, 80 °C,1 h||30 %|
|Acidic medium 5N HCL, 80 °C 10 hrs||100 %|
Table 1: Forced degradation of ATM.
Oxidative degradation: Degradation of 25% was observed when solution of ATM in distilled water containing 3% H2O2 was refluxed for 1 h. The peroxide peak appears and the peak of the drug decreases in area which indicates that the drug is degraded, as indicated by a comparison with the chromatogram of the standard solution of the drug (Figures 2a, 2b).
Alkaline degradation product: It was observed that around 30% of the drug was degraded when it was refluxed in 2M NaOH for 1 h in distilled water during UPLC analyses of alkali-degraded samples, Figure 2c).
Acidic degradation products: It was observed that the drug was completely degraded when it was refluxed in 5 M HCl for10 h, and there was a corresponding formation of degradation products, as indicated by a comparison with the chromatogram of the standard solution of the drug (Figure 2a).
Three new peaks appeared in the chromatograms of acid-degraded samples of ATM at 1.22, 1.42, and 2.08 min. The drug concentration gradually decreased with time when the sample was refluxed in 5 M HCl, and the degradation of the drug was completed in 10 h; (Figure 2d). Three acidic degradation products have relative retention times of 0.49, 0.58, and 0.84 min were well separated (Figure 2e) and subjected to LC/MS analysis, which confirm the suggested structures. Mild degradation was seen in 1 h by using 2M NaOH and 3% H2O2. On the other hand, the drug was completely degraded in 10 h by 5M HC.
The mass spectrum showed a mass ion peak at mlz 108,147,165 corresponding to acid degradation products which gave the following suggestion for the molecular formula: C7H8O (mlz 108), C10H15NO (mlz 165) and C10H13N (mlz 147) (Figure 3). The proposal pathway of hydrolysis of ATM with 5 M HCl for 10 h was presented in (Scheme 1).
Stability of ATM in human plasma: The stability of the studied drug in spiked human plasma was assessed at varying stability conditions (short-term stability, long-term stability and freeze thaw stability), using QC samples at (0.5, 1.5, and 2 μg/ml). The samples were analyzed and the results were compared with that obtained with the corresponding QC sample freshly prepared and processed immediately. It was observed that the studied drug showed stability in spiked human plasma when stored at ambient temperature at least 24 hrs, also when stored at -20°C for one month as long term stability, and over three freeze-thaw cycles  (Table 2).
|Parameters||Conc μg /ml||Accuracy %||RSD %||RE %|
At 25°C, 24 hr
At -20°C, 30 days
|Three freeze-thaw cycles
Thawed at 25°C for 2 h, refrozen 24h
Table 2: Summary of stability of ATM in human plasma at varying condition by the proposed UPLC method.
The high sensitivity of the proposed UPLC method allowed the determination of ATM with its acid-degradates in spiked human plasma. From the results, it is clear that there is no interference from the plasma matrix demonstrating the efficiency of the proposed UPLC method for analysis of the drug in human plasma (Figure 4). ATM is well absorbed in gastrointestinal tract and predominantly. ATM is given orally in a dose of 1.0 mg/kg twice-daily; this leads to estimate a maximal plasma concentration level of 1.073 μg/ml , which can be determined by the proposed method. The results are shown in Table 3. The extraction procedure described by Jordana SB. et al  was adopted here, since acetonitrile was found to be the best organic plasma protein precipitant.
|Parameter||UPLC method Authentic sample Spiked plasma||Derivative synchronous spectroscopy FDSFS|
|Linearity range μg /ml||0.1-35||0.1- 4||1-12|
|Slope ( b)||115.12||2.01||0.298|
|SD of slope||0.65||0.014||0.71|
|Intercept ( a )||-12.328||0.0158||0.3122|
|SD of intercept||0.92||0.02||0.03|
|Correlation coefficient r||0.9998||0.9998||0.9978|
|Standard error SE %||0.35||2.23||0.011|
|Recovery ± SD %
|98.55 ± 0.78
99.75 ± 0.93
99.98 ± 0.83
|95.6 0 ± 5.70
94.33 ± 7.40
|98.32 ± 0.66
98.66 ± 0.86
99.11 ± 0.71
|Precision ± RSD %
Intra – day
Inter – day
|98.80 ± 0.56
100.1 ± 0.39
|100.60 ± 3.90
103.2 ± 5.30
|98.75 ± 0.75
98.82 ± 0.65
|LOD μg /ml||0.03||0.03||0.33|
|LOQ μg /ml||0.10||0.10||1|
Table 3: Validation of the proposed UPLC method and derivative synchronous fluorescence spectroscopy for the determination of ATM.
Figure 4: Chromatogram of (a) a blank sample (plasma without drug or I.S),(b) a zero sample ( plasma spiked with 3 μg/ml IS,(c) non-zero sample (plasma spiked with 1 μg/ml drug and 3 μg/ml I.S), (d) plasma spiked with 1 μg/ml drug after acid degradation, μg/ml drug and 3 μg/ml I.S Where (A) is atomoxetine HCl, (F) is fluoxetine HCl.
Both ATM and its acid degradates are non fluorescent, after addition of SDS they exhibit fluorescence with λ maximum of 320 nm and 310 nm, after excitation at 270 and 280 nm for ATM and the acid degradate respectively (Figures 5a, 5b). The excitation and emission spectra of ATM and its acid degradation product are greatly overlapped (Figure 6). This fact hindered the use of this method for determination of ATM as a stability indicating one. The two-dimensional plot 2D showed the total fluorescence spectra of 2 μg/mL ATM and its acid degradation, respectively, after addition of 2 mL of 3.5 mM SDS in aqueous solution (Figures 7a, 7b). It was necessary to first record, the normal synchronous spectra for ATM and its acid-induced degradation product in order to derive the first derivative synchronous spectra. Figure 8a represents the synchronous spectra of different concentrations of ATM (1-12 μg/ mL) at 225 nm in presence of constant concentration of its degradation products (5 μg/mL) (Figure 8b). illustrates the SF spectra of different concentrations of ATM acid-induced degradation product (1-7 μg/ mL) at 275 nm in presence of constant concentration of ATM (2 μg/ mL).
Figure 8: Normak synchronous fluorescence spectra of (a) ATM (1,2,4,6,8,10,12) μg/mL at 225 nm in presence of 5 μg/mL of its acid induced degradation products; after addition of 2 mL 3.5 mM SDS in aqueous solution. (b) Acid- induced degradation products (1,3,5,6,7) μg/mL at 275 nm in presence of 2 μg/mL of ATM; after addition of 2 mL 3.5 mM SDS in aqueous solution.
Therefore we performed first derivative synchronous fluorescence spectroscopy technique for simultaneous determination of each of ATM and acid degradates. Spectra of ATM and its degradations were well separated using FDSFS with zero-crossing technique of measurement (Figures 9a, 9b). Under the experimental conditions the two peaks was measured at 276 & 265 nm for ATM and its acid-induced degradation product, respectively.
Figure 9: (a) First derivative synchronous fluorescence spectra of ATM (1-12) μg/mL at 276 nm in presence of 10 μg/mL of its acid degradation products; after addition of 2 mL 3.5 M SDS in aqueous solution. (b) First derivative synchronous fluorescence spectra of different concentrations of ATM degradation products (1,3,5,6,7) μg/mL at 265 nm in presence of 2 μg/mL ATM; after addition of 2 mL 3.5 mM SDS in aqueous solution.
Optimization of spectrofluorometric method
Different experimental parameters affecting the performance of the proposed method were carefully studied and optimized. Such factors were changed individually while others were kept constant. These factors included Δλ selection, effect of micellar medium, concentration of surfactant, volume of surfactant, diluting solvents.
Effect of diluting solvent
Dilution with different solvents including water, methanol, ethanol, acetonitrile and acetone, was attempted. In these compounds, the emission wavelengths present a bathochromic shift as solvent polarity increases, the larger red shift observed in water. It was found that water was the best solvent for dilution as it gave the highest fluorescence intensities, and the lowest blank reading, moreover its choice adds to advantages of the proposed method. Distinct decrease in the fluorescence intensities was attained upon ethanol, acetonitrile and acetone, as shown in Figure 10.
Effect of different surfactants
Micellar systems of an anionic surfactant (SDS), a cationic surfactant (Cetrimide) and non ionic surfactant (Tween 80) were investigated by measuring the fluorescence intensity of the analytesurfactant system. The influence of the micellar medium on the fluorescence intensity of both drug and acid degradates is studied. It proved that (SDS) enhanced the FI of ATM much more than other surfactants. The fluorescence intensity was increased 10 fold compared to that in aqueous solution. Apparently, the enhancement ability sequence was SDS>Tween 80> Cetrimide>aqueous solution. The enhancement occurs for the following reasons: ATM is a nitrogenring compound whose alkalinity is stronger when excited than in the ground state. So Molecules of ATM easily formed hydrogen bonds with water molecules in aqueous solution. With the surfactant added, the hydrogen bonds were disrupted and ATM was enriched in the hydrophobic microenvironment produced by the micellar medium. Also the fluorescence quenching caused by dissolved oxygen was prevented and therefore the fluorescence intensity of ATM increased .
Moreover, as the anionic surfactant SDS has four oxygen atoms; it can easily form hydrogen bonds with ATM. These hydrogen bonds not only prevent the formation of hydrogen bonds between the analyte and water, but also produce electron donating effect, which can strengthen the delocalization effect of electrons in conjugated system and therefore the fluorescence increased.
Effect of time
The effect of time on the stability of the synchronous fluorescence intensity of the drugs was also studied. It was observed that the (FI) developed instantaneously and remained stable for more than 2 hours.
The validity of the proposed methods was tested regarding linearity, accuracy specificity, repeatability and precision according to ICH recommendations [20,21] and stability according to the currently accepted U.S food drug administration (FDA) bio-analytical method validation guidance .
Linearity of the proposed UPLC and fluorometric methods was evaluated by analyzing a series of different concentrations of the drug. Each concentration was repeated three times. The assay was performed according to experimental conditions previously established. Plot of peak area Method I and fluorescence intensity Method II versus the concentration of the drug within the respective ranges are stated in Table 3.
Accuracy studies of the analytical procedures were done on triplicate solutions of ATM prepared at each recovery level and analyzed versus standard solution. The means ranged from 98.32% to 98.55%, indicating accurate results (Table 3). The results obtained were satisfactory compared with a reported HPL C method , and no significant difference was found are Moreover, the proposed methods are more sensitive.
Also the accuracy of the proposed methods for analysis of the drug in capsules was evaluated by applying the standard addition technique. The results obtained indicate good recovery (Table 3).
The intra and inter day precisions of the proposed methods were estimated by triplicate analysis of three different concentrations (0.5, 15, 30 and 0.5, 1.5, 20 μg/ml) and (1, 6, 12 μg/ml) for UPLC and FDSFS, respectively) over a period of three days. The small value of % RE and % RSD for the proposed procedures indicates high accuracy and high precision of the proposed methods (Table 4).
|Drug substance||The proposed UPLC Authentic sample||Derivative synchronous spectroscopy FDSFS||Reference methodb|
|Meana ±a SD||98.55 ± 0.78||98.32 ± 0.66||99.72 ± 1.32|
Table 4: Statistical comparison between the proposed and the reference method for the determination of atomoxetine HCl in drug substance.
Detection and quantification limits
The limits of detection (LOD) and the limits of quantification (LOQ) were calculated using the following equations:
Where SD is the standard deviation of response and S is slope of graph Table 3.
The selectivity of the proposed methods was checked by analyzing different synthetic mixture of ATM and its acid degradation products at various concentration ranges using the optimized procedures, and satisfactory results were obtained (Table 5).
|% degradation||Drug recovery %|
Table 5: The determination of atomoxetine HCl in a mixture of its acid degradation products in laboratory prepared mixture by the proposed methods.
The proposed UPLC method was able to determine ATM in the presence of up to 90% of its acidic degradation products without interference.
The proposed DSFS method was applied to the simultaneous determination of ATM & its degradants in synthetic mixture containing different concentrations of both in the ratio of (10:3 μg/mL ) (Figure 11). The relative fluorescence intensities of first derivative technique were measured. The first derivative signal of ATM was measured at 276 nm which is considered as zero-crossing point for the degradants and the first derivative signal of the degradants was measured at 265 nm which is considered as zero-crossing point for ATM. The concentrations of of ATM & its degradants in the synthetic mixture were calculated according to the linear regression equation of the calibration graphs. The results indicate high accuracy of the proposed method as shown in Table 5. The proposed DSFS method was able to determine ATM in the in the presence of up to 100% of its degradants without interference.
The selectivity of the two methods was investigated by observing any interference encountered from the common capsule excepients, such as starch and dimethicone. These excepients did not interfere with the proposed methods.
The developed methods are applicable to analysis of the drug in its commercial capsules (Table 6). Results obtained were in good agreement with a reference HPLC method for assay of capsules . Statistical comparison of the results were performed with regard to accuracy and precision using Student’s t-test and variance F-ratio -test at the 95%confedence level, and there were no significant differences between the proposed methods and the reference one .
|Commercial product||UPLC method||Derivative SFS FDSFS||Reference methodb|
|Meana ± SD||99.75 ± 0.93||98.66 ± 0.86||99.29 ± 1.54|
Table 6: Statistical comparison between the proposed UPLC and the reference method for the determination of atomoxetine HCl in Strattera® capsules.
The UPLC system suitability parameters including capacity factor (k′), selectivity (α), resolution (Rs), tailing factor (T), and theoretical plate (N) listed in Table 4. All parameters were satisfactory with good specificity for the stability assessment of ATM. The results are given in Table 7.
|Compounds||Rt||k'||α||Rs||( N )||( T )|
Table 7: system suitability parameters by the proposed UPLC method.
The robustness of the developed methods was demonstrated by the constancy of area under the peak value or the constancy of the fluorescence intensity with the deliberated minor changes in the experimental parameters such as, change in pH 4.2 ± 0.1, the change in the detection wavelength 205 ± 3 nm and change in volume of SDS 2 mL ± 0.1. These minor changes that may take place during the experimental operation didn’t affect the peak area value or FI of the analyte.
The effect of pH variation of the mobile phase on the resolution and retention time were studied and the results are provided in Table 8.
Table 8: Effect of changing pH on resolution and retention time of ATM peak.
The developed UPLC and FDSFS methods are simple, reproducible, selective and time saving methods. They allow analysis of ATM in raw material, in the presence of its degradation products as well as in capsules without interference from excipients. LC/MS detection was used to find the mass values of the degradation products. Three acidic degradation products were isolated and confirmed as (N-methyl-3-hydroxy-3-phenylpropylamine), (N-methyl-3-phenyl-2, 3-propenylamine) and (O-Cresol).