Simultaneous Spectrophotometric Determination of Sacubitril and Valsartan in their Recently Approved Pharmaceutical Preparation

Objectives: Simultaneous estimation of sacubitril and valsartan in their new pharmaceutical tablets by bivariate method and two multivariate methods namely; classical least square and principal component regression. Methods: (A) Bivariate method, where absorbance values were measured at two optimum wavelengths (220 nm and 250 nm). (B) Multivariate methods, where partial factorial design was constructed within the wavelength range from 210-290nm at 1 nm interval. Results: The bivariate method is based on simple mathematic algorithm using two optimum wavelengths for the analysis of the mixture. On the other hand, two multivariate methods considering all the variables at the same time have been used to improve the quality of spectral analysis of the mixture. Conclusion: The proposed methods have been successfully applied for the determination of the two drugs in Entresto tablets. Statistical comparisons between the obtained results and those obtained by the reported method have been performed showing no significant differences by applying t-test and F-test.


INTRODUCTION
Entresto ® (sacubitril/valsartan) is a new FDA approved mixture for the treatment of heart failure.It contains sacubitril(SAC) Figure 1, a prodrug that results in neprilysin inhibition and valsartan (VAL) which is angiotensin II Type-1 receptor blocker, Figure 2 1,2 .
The resolution of multi-component preparations is often the main goal of analytical chemistry.The spectrophotometric techniques have advantages of simplicity, solvent and time saving.The only drawback is the presence of severely overlapped spectra in multi-component preparations 3,4 .Hence, several spectrophotometric methods have been developed for simultaneous estimation of these components [5][6][7][8][9][10][11] .

Standard solutions
Standard stock solutions of SAC and VAL (1000 µg/mL) were prepared separately by dissolving 100 mg of each drug powder in methanol.Working solutions (100 µg/mL) were prepared before use by dilution from the stock solutions with methanol.

Procedure For BM
Different aliquots equivalent to 10-150 µg of SAC and VAL were accurately transferred from their working standard solutions (100 µg/mL) into two separate series of 10-mL volumetric flasks and completed to volume with methanol.The absorption spectra (from 200 to 400 nm) of these solutions were recorded using methanol as a blank.The absorbance was measured at 220 and 250 nm from which the calibration graphs were constructed and then the corresponding regression equations were computed at the selected wavelengths.The obtained slope and intercept values were used for construction of two mathematical equations used for calculating the concentrations of both drugs in their binary mixture.

Assay of laboratory prepared mixtures
Laboratory prepared mixtures containing SAC and VAL in different ratios was prepared by transferring the aliquots of two drugs into 10-mL volumetric flasks and complete to volume with methanol.The absorbance values of the mixtures were recorded at 220 and 250nm and substituted in the following two equations to obtain the concentrations of both drugs: CSAC = (AM1 -eSAC+VAL1-mVAL1CVAL) / mSAC1 CVAL = [mSAC2 (AM1 -eSAC+VAL1) + mSAC1 (eSAC+VAL2 -AM2)] / mSAC2mVAL1 -mSAC1mVAL2 Where: -CSAC, CVAL are the concentration of component SAC, component VAL.
-AM1, AM2 are the absorbance values of the binary mixture at λ1, λ2.
For CLS and PCR Partial factorial design based on five levels and two factors was constructed 24 .The design results in 25 mixtures of different ratios of SAC and VAL, Table 1.Thirteen samples were used as a training set and twelve were used as a validation set.The chosen concentrations for each compound are based on its linearity.The absorption spectra of the samples were scanned from 200 -400 nm at 1 nm interval against methanol as a blank.The noisy region from 200-210 nm and the zero absorbance after 290 nm were rejected.CLS and PCR models were constructed by transferring the spectral data to MATLAB for subsequent calculations.The 2D plot of the experimental space showing the positioning of the training set and the validation set samples, Figure 3.

Application to pharmaceutical preparation
Ten Entresto ® tablets (each tablet labeled to contain 97 mg SAC and 103 mg VAL) were weighed and finely powdered.A portion of powder equivalent to one tablet was weighed, transferred into conical flask and dissolved in 75 mL of methanol.The solution was  shaken vigorously for 15 min then sonicated for about 30 min and filtered into 100-mL volumetric flask.The volume was completed to 100-mL with methanol to get a stock solution containing 970 µg/mL of SAC and 1030 µg/mL of VAL.The solution was suitably diluted with methanol to obtain sample solutions containing SAC and VAL in the concentrations ratio of 1:1.06 µg/mL, respectively, as in the tablet formulation.Then the procedure was completed as described under the procedure of each method.

RESULTS AND DISCUSSION
The zero order absorption spectra of SAC and VAL show sever overlap which doesn't permit direct determination of both drugs in their mixture, Figure 4.In order to resolve these overlap, bivariate and multivariate methods have been developed.

Bivariate method (BM)
Kaiser method 19 has been used for the selection of two optimum wavelengths for the analysis of two drugs.The absorbance of the two components individually at six different selected wavelengths was recorded in the region of overlapping; 220, 230, 240, 250, 260 and 270 nm.The slope values of the linear regression equations were estimated for both SAC and VAL at the selected wavelengths.A series of sensitivity matrices (K) was created for each binary mixture and for every pair of the preselected wavelengths: The shaded rows represent the calibration set.
Where, mSAC1,2 and mVAL1,2 are the slopes of the components SAC and VAL at the two selected wavelengths [16][17][18] .The determinants of these matrices were calculated and the wavelength set selected for which the highest matrix determinant value was obtained, Table 2.It was found that; the slopes at 220 and 250 nm gave the maximum value of K and thus chosen for the analysis.

Method validation
The proposed method was validated according to ICH recommendations 25 .

• Linearity
The linearity of the developed method was evaluated by analyzing different concentrations of SAC and VAL in triplicates.Beer-Lambert concentration ranges were found to be 1-15 µg/mL for both drugs.The values of coefficient of determination were close to unity indicating good linearity.The regression parameters were summarized in Table 3.
• Limit of detection (LOD) and limit of quantitation (LOQ) LOD and LOQ were calculated from the following equations: LOD = 3.3 Sa / slope LOQ = 10 Sa / slope Where Sa is the residual standard deviation of a regression line.Low values of LOD and LOQ values of SAC and VAL indicated the sensitivity of the proposed methods, Table 3.

• Accuracy
The accuracy of the results was checked by applying the proposed method for triplicate determination of three concentration levels covering the specified range for each drug (5, 9, 13 μg/mL).The concentrations were obtained from the constructed mathematical equations.The mean percent recovery of the concentrations was calculated.Good recovery percent of the studied concentration with acceptable value was obtained, Table 3.Moreover, accuracy of the method was further assessed by applying the standard addition technique to prove that the proposed methods can selectively analyze each drug without any interference from the other drug or the excipients, Table 4.

• Precision
Three concentrations of (5, 9, 13 μg/mL) for either SAC or VAL were analyzed three times intraday for repeatability and on three successive days for intermediate precision using the proposed method.The small calculated relative standard deviations indicated the high precision of the proposed methods, Table 3.

• Specificity
The specificity of the method was confirmed by the analysis of different laboratory prepared mixtures of SAC and VAL within the linearity range.Perfect determination of each drug in the presence of the other indicated by acceptable calculated recovery percent and confirmed the high specificity of the proposed methods, Table 5.

• Stability of solutions
Working standard solutions of SAC and Val showed no spectrophotometric changes up to 7 days when stored at 4 °C.

Multivariate methods
Multivariate models are a useful tool for data analyses.CLS involved the application of multiple linear regressions to the classical expression of the Beer-Lambert law of spectroscopy:  Where A is absorbance matrix, absorptivity coefficient and path length are combined in single constant K matrix, C is constituent concentrations matrix and E is matrix of absorbance error (the residual errors between the least squares fit line and the actual absorbance).
The calibration absorbance matrix (13×81) and their corresponding concentration matrix (13×2) were used to find the (k) matrix which was further used for predicting the concentrations of SAC and VAL in the validation set and in pharmaceutical preparation.
However, CLS method is the most simplest of the multivariate methods, but it is a rigid model that require the whole information of all the components in the mixture and their concentrations Unlike CLS, PCR method is more flexible and can be applied even in the presence of interfering substances.PCR is a factors analysis method which involves decomposition of spectral matrix into a set of eigenvectors and scores followed by regression of the scores against the constituent concentrations.
A cross validation method leaving out one sample at a time was used to select the optimum number of latent variables (factors) in PCR 26 .If the number of factors was more than required, more noise would be added to data and if the number of factors was too small, meaningful data that could be necessary for the calibration might be discarded 11 .The method developed by Haaland and Thomas was used for selecting the optimum number of factors 27 .It was found that two latent variables are sufficient for PCR method, Figure 5.
Table 6 shows percent recoveries, mean, standard deviation (SD) and root mean square error of calibration (RMSEC) of the calibration set in both models.While, Table 7 shows percent recoveries, mean, standard deviation (SD) and root mean square error of prediction (RMSEP) of the validation set in     7 shows percent recoveries, mean, standard deviation (SD) and root mean square error of prediction (RMSEP) of the validation set in both models.

Statistical analysis
The developed methods have been applied for determination of SAC and VAL in Entresto ® tablets and the results obtained were acceptable with small %RSD values.Results obtained by the proposed methods were statistically compared to those obtained by the reported method 14 and no significant difference was observed, Table 8.Moreover, a one-way analysis of variance (ANOVA) test was carried out to compare the proposed methods with each other, and no significant difference was found, Table 9.

CONCLUSION
In this work, bivariate and multivariate spectrophotometric methods have been successfully applied for simultaneous determination of a recently FDA approved binary mixture of sacubitril and valsartan.The developed methods have advantages over chromatographic methods that required sophisticated procedures and time consuming.

Figure 3 .
Figure 3.The 2D plot of the experimental space showing the positioning of training set (∆) and the validation set (•) samples for SAC and VAL.

Figure 4 .
Figure 4. Absorption spectra of SAC and VAL.

Figure 5 .
Figure 5. RMSECV plot of the cross validation results of the calibration set as a function of the number of latent variables (LVs) used to construct the PCR model.

Table 3 . Regression and validation data for determination of SAC and VAL by the BM
* Average of three determinations for three concentrations repeated three times.

Table 4 . Determination of SAC and VAL in Entresto ® tablets by BM and application of standard addition technique
*Average of five determinations.**Average of three determinations.

Table 8 . Statistical comparison of the results obtained by applying the proposed methods and the reported method 14
The mean of percent recovery of pharmaceutical preparation.**** The values in parenthesis are tabulated values of "t "and "F" at (P = 0.05).