Quality by design (QbD) based development and validation of RP-HPLC method for buserelin acetate in polymeric nanoparticles: Release study.

Rashmi S. Tambare: Writing – original draft, Investigation, Conceptualization. Sadhana R. Shahi: Supervision. Vishal C. Gurumukhi: Writing – review & editing, Software, Formal analysis, Data curation. Suhas M. Kakade: Validation, Resources, Methodology. Ganesh G. Tapadiya: Supervision, Resources.

Approval of an ethics committee as well as informed consent was not required for this study because this study did not involve any kind of animal study or living beings.

This research received no external funding.

For the development of RP-HPLC method for BA, the statistical optimization and analysis is a significant part. We assessed the effect of experimental variables and their interaction that have an impact on the CAAs reflecting better performance of the established method. Here, the Rt of the developed method can be a result of the ratio and composition of the mobile phase and was found in the range of 6.56 min–6.98 min ( Table 3 ). The Rt depends on the flow rate of the mobile phase which was in the range of 0.72–1.28 mL/min as per experimental design. The significant effect of the composition of the mobile phase and flow rate was observed to achieve the peak resolution. The results of ANOVA study show a linear model for the retention time. The respective responses such as contour plot and 3D response surface plot for the Rt are shown in Fig. 3 A. Furthermore, performing residual analysis using the normal probability plot ( Fig. 4 A) and found to be normally distributed which looks similar to a straight line with no outlier points met. Moreover, a good agreement between actual and predicted responses was established on the plot of predicted value versus actual value ( Fig. 4 B) that gives the indication of actual and predicted response of Rt. Fig. 4 The predicated Vs actual plot. Fig. 4 The predicated Vs actual plot. The polynomial equation representing the influence of flow rate (X1), and pH of buffer (X2) was found to have a level of significance (P < 0.0017) statistically. Furthermore, the adjusted R 2 and PRESS values were found to be 0.6659 and 0.087 respectively. (1) Y1 (retention time) = 6.82–0.13×X1-0.037×-0.037×X2 From the above equation (1) , the negative impact of the flow rate (X1) and pH of buffer (X2) was observed. It indicates that an increase in flow rate, and pH of buffer decreases the Rt. The Pa was the result of the analyte molecule passed through the detector. The Pa of BA was found in the range of 881232–927834 ( Table 3 ). The Pa could be obtained that has a positive effect on the flow rate (X1) and pH of the buffer (X2). The Pa and its resolution can be significantly achieved by the appropriate tunning of the flow rate (X1) and adjusting the pH of the buffer (X2). The ANOVA study shows a linear model for the Pa and the respective response contour plot and 3D response surface plot are shown in Fig. 3 B. Furthermore, in the case of Pa, by performing the residual analysis, the normal probability plot ( Fig. 4 C) was found and normally distributed on a straight line with no deviation points met. Moreover, after careful observation, it was found that there was a good agreement between actual and predicted responses on the plots of predicted value versus actual value ( Fig. 4 D). These values depict the evidence of actual and predicted response of peak area. The polynomial equation representing the influence of flow rate (X1), and pH of buffer (X2) on the Pa was found to have a probability value (P < 0.0479) enabled significant level statistically. Furthermore, the values of adjusted R 2 and PRESS were found to be 0.3465 and 0.054 respectively. (2) Y2 (Area of peak) = 9.085 + 0.05+9264.94× X1+493.48 ×X2 From the above equation, it was observed that positive impact on the flow rate (X1) and pH of buffer (X2). It means that an increase in flow rate (X1), and pH of buffer (X2) increases the Pa simultaneously. The optimization is a crucial part of the estimation of the method based on the AQbD approach employing the desirability approach that can generate the design space [ 40 ]. The design space shows the target value in terms of the upper and lower levels to use for an optimum method for BA [ 41 , 42 ]. The design space in optimization was constructed by overlay plot ( Fig. 5 ). Herein, optimization method was performed numerically and graphically in which optimum value for CMPs such as flow rate (X1) (in the range of 0.8–1.2 min/mL) and pH of buffer (X2) (3.1–3.5 pH) was generated as predictor to come out the desired CAAs values. Practically CMP value should be the desirability value closest to 1 that becomes optimization. The identified desirability points for flow rate (X1) 1.6 min/mL and pH of buffer (X2) 3.19 for the mobile phase for the experiment of the optimized method. The predicted values were verified experimentally and found to be identical to Fig. 2 (representative chromatogram). The Rt and Pa were found to be 6.84 min and 911234 respectively. Fig. 5 Overlay plot for optimization of the method. Fig. 5 Overlay plot for optimization of the method. The sample was analyzed after 1st, 4th , and 8th days using HPLC. The obtained peak and resolution of the chromatogram were compared with the initial chromatogram. The % recovery of the analyte concentration for the samples was found. That indicates the standard stock solution was stable for longer duration up to 8th days. It was verified using statistical method and found to be within limit. The developed method for BA using an AQbD-based approach was validated using validation parameters according to the ICH guidelines Q2 R (1) [ 43 ]. A high degree of accuracy in the method development study was confirmed by the system suitability test. At the beginning, six injections of the same concentrations were given and RSD value was found to be less than 2 % for Tp and Tf ( Table 5 ). Table 5 System suitability of developed method. Table 5 Parameter Value (Mean ± SD) n = 6 Acceptable limits Tailing factor (Tf) 1.09 ± 0.36 Tf ≤ 2 Theoretical plate (Tp) 7589 ± 134 Tp ≥ 2000 Retention time (Rt) 6.78 ± 0.05 Rt ≤ 1 % System suitability of developed method. For linearity, the peak area versus six different concentrations BA solution (10–60 μg/mL) was plotted. Linearity shows that all the results were within acceptable limits that were achieved by regression analysis and the least square method. The observed calibration curve was linear and the regression coefficient was found to be R 2  = 0.9991. The linearity of the BA was validated by the analysis of variance and found to be the significant linearity deviations (P < 0.05) indicating validation of linearity. The equation of the linearity was found to be Y = 29482X+34221 ( Fig. S3 ). The specificity study has shown that no interference between the polymer and the excipients with the peak of BA ( Fig. 2 ). Moreover, no peak was eluted at the retention time of blank BA. Thus, the developed method was selective for the determination of BA in nanoparticle formulations. Therefore, the HPLC method is useful for quantifying BA in the developed formulations. The LOD and LOQ of the BA were determined by the signal-to-noise ratio (S/N). The acceptable value of signal-to-noise ratio is 3 for LOD and 10 for LOQ. The result indicates that the method has high sensitivity. The LOD and LOQ were observed to be 0.051 μg/mL and 0.254 μg/mL, respectively. The accuracy means the closeness between the obtained value to the true value and expressed in terms of % recoveries. It was performed using percent recovery ranging from 100.55 ± 0.93 to 103.45 ± 0.32 using triplicate injection ( Table S2 provided as supplementary). The mean percent recovery at each concentration of BA was found to be within the acceptance limit indicating that the method was accurate. Whereas precision measured the effect of intraday and interday results (% RSD). The % RSD for the BA peak assay from six replicate injections of the standard solution was less than 1.0 %. The intraday and interday precision results for the quantification of BA in nanoparticle formulations are presented in Table S3 . This indicates that the present developed method for the quantification of BA was precise. All the observed values were within the range of recommended guidelines. The robustness of the estimated method was evaluated by intentionally changing the various chromatographic conditions. The results did not change significantly and found were within the range [ 44 ]. Thus, the developed method was robust ( Table 6 ) according to the recommended guidelines. Table 6 Robustness at different flow rate, volume of mobile phase, and wavelength. Table 6 a Rt a Tp a Tf a Pa 1. Flow rate (1.0  ±  0.2 mL/min) 6.80 7434.00 1.48 907976.00 6.83 7532.00 1.49 907634.00 6.87 7453.00 1.53 903427.00 Mean 6.83 7473.00 1.50 906345.67 SD 0.04 51.97 0.03 2533.42 % RSD 0.51 0.70 1.76 0.28 2. Change in mobile phase volume (79:21, 81:19) 6.76 7487.00 1.43 907463.00 6.81 7357.00 1.49 907323.00 6.93 7696.00 1.54 903487.00 Mean 6.83 7513.33 1.49 906091.00 SD 0.09 171.03 0.06 2256.22 % RSD 1.28 2.28 3.70 0.25 3. Change in wavelength (218, 222 nm) 6.73 7534.00 1.43 907463.00 6.93 7622.00 1.46 904322.00 6.89 7485.00 1.43 913837.00 Mean a 6.85 7547.00 1.44 908540.67 SD 0.11 69.42 0.02 4848.18 % RSD 1.54 0.92 1.20 0.53 a indicates the value of three determination (n = 3) and Rt=Retention time, Tp = Theoretical plate, Tf = Tailing factor, Pa=Peak area. Robustness at different flow rate, volume of mobile phase, and wavelength. indicates the value of three determination (n = 3) and Rt=Retention time, Tp = Theoretical plate, Tf = Tailing factor, Pa=Peak area. Assay values of BA in nanoparticle formulation ranged from 99.85 ± 0.85 % to 101.06 ± 0.57 %. The assay values for the formulations were the same as mentioned in the labeled claim. The results of the assay demonstrate that the method for the BA is selective without interference from the polymer and excipients used in the dosage form ( Fig. 6 ). The estimated drug content with low standard deviation values justified the accuracy of the proposed method. Since all validation parameters examined met the acceptance criteria, the proposed method was considered validated according to the ICH Q2 (R1) guidelines. Fig. 6 Comparison of blank versus sample chromatogram. Fig. 6 Comparison of blank versus sample chromatogram. The established method of BA was successfully applied to the BA encapsulated in polymeric nanoparticles ( Fig. 6 ) which were found identical to the chromatogram of BA ( Fig. 2 ). In this study, blank versus sample chromatogram was compared, and observed that no interference of any excipient was found ( Fig. 6 ). Furthermore, the method can be applied for the estimation of % drug release from the polymeric nanoparticle using dialysis bag method. In addition, the mechanism of drug release can be predicted using mathematical model of kinetics. In vitro drug dissolution study was performed for BA-loaded polymeric nanoparticles and BA-suspension using the dialysis bag method ( Fig. 7 ). As can be seen, the controlled release of the drug was observed from the polymeric nanoparticles. The controlled release of drug was due to the presence of the polymer PLGA- Resomer RG505 and Resomer RG750 (50:50) within 48 h whereas the BA suspension drug release maximized up to 44.65 %. The obtained data from the dissolution study was fitted to the different kinetic model equations and showed the Korsmeyer-Peppas model (regression coefficient, r 2  = 0.9832) with a Fickian diffusion release pattern (n < 0.432). Thus, the controlled release of drug was due to the release of drug from the pores of the encapsulated nanoparticles matrix over the period of time. Fig. 7 In vitro drug dissolution study. Fig. 7 In vitro drug dissolution study.

Buserelin acetate (Purity >98.0 %) was purchased from Henan Ting Xuan New Material Technology Co., Ltd. (China). Poly D, L-lactic-co-glycolic (PLGA- Resomer RG505 and Resomer RG750) (Purity 100 %) was purchased from Evonik Roehm (GmbH, Weiterstadt, Germany). Poloxamer 188 (Pluronic F 68) (95–99 % pure) was gifted by (BASF, Mumbai, India). HPLC grade ethyl acetate (99.7 % pure), acetonitrile (99.9 % pure), orthophosphoric acid (99.7 % pure), and polyvinyl alcohol (PVA) (99.6 % pure) were purchased from Merck Ltd. (Mumbai, India). All other reagents used were of analytical grade and were used for this method. RP-HPLC method was developed using an Agilent HPLC system (LC-1220 series, USA) equipped with an intelligent autosampler, a quaternary pump, and a solvent degasser. The chromatogram was detected and recorded with LC solution software. The digital pH meter (Thermo Scientific, Sr. No. V 04082, Mumbai, India) and an ultrasonic water bath (Aczet CUB 10) were used for the measurement of the pH and degassing of the mobile phase respectively. The mobile phase Water: acetonitrile (80: 20 v/v) and orthophosphoric acid was used using (pH modifier) and used as shown in experimental design. The mobile phase, thereafter, was filtered using 0.45 μm membrane before degassing ultrasonically. The analytical method development was performed using Zorbax Eclipse plus C18 column (4.6 mm × 150 mm × 5 μm) maintaining the column temperature at 30 °C. An automated sample injector was used to inject 20 μL sample from the sample container to the column with varying flow rates as per the experimental design. The chromatogram was observed at 220 nm using a photodiode array detector (Waters 2996). The optimized chromatographic condition is displayed in Table 1 . Table 1 An optimized chromatographic condition. Table 1 Parameters Conditions Mobile phase Water: Acetonitrile: Orthophosphoric acid (80:20:0.5) Stationary phase Zorbax Eclipse plus C18 column (4.6 mm × 150 mm × 5 μm) Flow rate 1.0 mL/min Column temperature 30 °C Volume of injection loop 20 μL Detection wavelength 220 nm Retention time 6.78 min An optimized chromatographic condition. Accurately weighed BA (10 mg) was placed into a 10 mL volumetric flask containing a mixture of water and acetonitrile (80:20, v/v) to obtain a primary stock solution of concentration 1000 μg/mL. Further, a standard stock solution of 100 μg/mL concentration was made by diluting the primary stock solution. Thereafter, different dilutions were prepared for the construction of the calibration curve. Nanoparticles loaded with BA were prepared by double emulsion techniques using a reported method with some modifications [ 22 ]. In brief, PLGA polymer (50:50) was dissolved in ethyl acetate to obtain an organic phase. BA and Poloxamer 188 were dissolved in water to obtain an aqueous phase. This aqueous phase was emulsified with the organic phase using a high-speed homogenizer to form a primary emulsion. Then the primary emulsion was injected into the continuous phase containing polyvinyl alcohol (PVA) and ethyl acetate followed by homogenized to get a double emulsion. The developed emulsion was washed using water for injection and subsequently lyophilized at −76 °C temperature, 35 mT vacuum pressure for 52 h using Virtis-Bench Top Lyophilizer (Spinco Biotech Pvt. Ltd.) to obtain dry nanoparticles. To determine the BA content in the nanoparticle, an accurately weighed amount of nanoparticles was lysed completely by using a mixture of water and ethyl acetate followed by centrifugation. The mixture was then kept constant for 12 h to separate the residue. The dry obtained residue was then dissolved in acetonitrile and appropriately diluted with mobile phase (water: acetonitrile 80:20 v/v). Samples were filtered through a 0.45 μm membrane filter and sample (20 μL) was injected into HPLC system. The wavelength of drug sample was determined using a UV–visible spectrophotometer. For this purpose, the prepared standard stock solution was diluted with a mixture of water and acetonitrile to get 10 μg/mL of solution. This solution was examined in the spectrum of UV range against blank to obtain λ max of the sample. In order to develop the RP-HPLC method, the elements of the AQbD approach were implemented and described below. ATP is the primary stage and objective of the analytical method development. It is the collection of the characteristics of the target analytical method that can be achieved by the appropriate selection of the CMPs and CAAs [ 23 ]. Herein, flow rate (X1) and pH of buffer (X2) were identified as CMPs whereas Rt and Pa were identified as CAAs The variations in CMPs have an impact on CAAs that can be controlled to ensure the desired quality of the developed method [ 24 ]. The RP-HPLC method also should allow the separation of BA from the polymeric nanoparticles without interference with the sample matrix and degradation products. Thus, the main objective was to optimize the chromatographic condition to enhance the quality of the chromatogram in terms of resolution, retention time (Rt), peak area (Pa), and tailing factor (Tf) [ 9 ]. Therefore, to achieve the ATP, the CAAs such as Rt and Pa were chosen for the attainment of the ATP. The ATP and CAAs with their justifications are described in Table S1 . In HPLC, several factors can affect the quality of resolution and separation such as mobile phase, pH, column temperature, sample composition, and injection volume. Thus, to identify the CMP, risk assessment was performed for the identification of high-risk, low-risk, and medium-risk factors of the chromatographic method development [ 25 ]. Herein, the risk assessment was a crucial step of AQbD and was used to identify the possible risk factors that can affect the quality of the method and critical quality attributes (CAA) [ 26 ]. It was performed based on sound science and a review of literature. To achieve ATP, the fishbone diagram ( Fig. 1 ) was created in order to find the relationship between the CMP and CAA of the developed method [ 27 , 28 ]. Fig. 1 Fishbone diagram to indicate the relationship between the CMP and CAA. Fig. 1 Fishbone diagram to indicate the relationship between the CMP and CAA. Factor screening of each analytical step was performed based on the risk assessment matrix (RAM) method ( Table 2 ) for identifying method parameters (CMP) that affect the selected CAAs, i.e. Rt and Pa [ 29 ]. Risk assessment for the independent variables such as mobile phase, flow rate, solvent, pH of buffer, stationary phase, and detector were performed by the RAM method [ 30 ]. The flow rate and pH of buffer got high scores and were selected as CMPs in proposed AQbD method development. Furthermore, selected parameters were highly influenced on the selected CAAs that could be used in the design of the experiment (DoE) [ 31 ]. Table 2 Risk assessment matrix (RAM) method for identifying method parameters. Table 2 Red area indicates- High (H) risk factors, Yellow area indicates- Medium (M) risk factor, Green area indicate – Low (L) risk factors. Risk assessment matrix (RAM) method for identifying method parameters. Red area indicates- High (H) risk factors, Yellow area indicates- Medium (M) risk factor, Green area indicate – Low (L) risk factors. After the risk assessment analysis, preliminary trials were performed to identify the method parameters. Here, the flow rate (X1) and pH of buffer (X2) were selected based on preliminary trials, sound science, and understanding. In DoE, efficient and comprehensive central composite design (CCD) was selected for method development at different levels i.e. low (−1), medium (0), and high (+1) with 13 experimental runs including 5 center point replicates. The center point replicates indicate reproducibility and enhance the validity of the experimental design [ 9 ]. In CCD the extra edge value (−alfa, + alfa) was seen that had additional exposure to the response surface model generating few experimental runs resulting in better establishment of a method [ 32 , 33 ]. All the remaining parameters were kept constant during the experimental method runs. Thus, AQBD-based method for BA was developed and subsequently validated using validation parameters. Here, the benefit for the AQBD-based method development was to reduce the cost and the experimental run to optimize the optimal factor for optimization of the method. The AQbD-based method development is considered as a paradigm shift that is accepted by various industries [ 24 ]. In the present studies, we identified the absorption wavelength ( λ max  = 220 nm) from the spectrum of the UV–visible spectrophotometer ( Fig. S2 ). Therefore, for the detection wavelength, 220 nm was set for the analysis in the detector (PDA detector). The mobile phase, Water: Acetonitrile: Orthophosphoric acid (80:20:0.5) was optimized based on the solubility, polarity and the pH of the buffer was adjusted using orthophosphoric acid in the range of 3.02–3.58 respectively to improve peak resolution. To remove the traces of the component and smooth functioning of column, the mobile phase was run with blank and with sample. The representative chromatogram was obtained ( Fig. 2 B). The chromatographic conditions ( Table 1 ) were used as the basis for the method and ranges of the CMPs were determined which was further applied in the design expert software using CCD. Fig. 2 The Blank chromatogram (A) and the representative chromatogram of BA (B). Fig. 2 The Blank chromatogram (A) and the representative chromatogram of BA (B). AQbD-based systematic analysis of the experimental variables was performed using CCD. The DoE was employed in order to study the effect of independent variables i.e. flow rate (X1) and pH of buffer (X2) were assessed on the responses i.e. retention time (Rt) (Y1) and peak area (Pa) (Y2). All 13 experiments were run using CCD as shown in Table 3 and a chromatogram was obtained individually. The statistical data such as model determination, predicted vs actual plot, ANOVA, lack of fit test, and response surface plots were obtained for individual retention time (Rt) and peak area (Pa) [ 9 ]. Table 3 Actual and coded experimental values and its responses obtained from different run using CCD. Table 3 Independent factors (Coded levels) Dependent variables (Responses) Experiments Flow rate (X 1 ) (mL/min) pH of buffer (X 2 ) Retention time (Y 1 ) (min) Area of peak (Y 2 ) 1 1.00 (0) 3.58 (+α) 6.80 907975 2 1.00 (0) 3.30 (0) 6.76 927834 3 1.20 (+1) 3.50 (+1) 6.56 909884 4 1.00 (0) 3.02 (-α) 6.89 907973 5 1.00 (0) 3.30 (0) 6.83 907966 6 0.80 (−1) 3.50 (+1) 6.9 905387 7 0.80 (−1) 3.10 (−1) 6.96 903472 8 1.00 (0) 3.30 (0) 6.87 907973 9 0.72 (-α) 3.30 (0) 6.94 881232 10 1.00 (0) 3.30 (0) 6.9 907993 11 1.00 (0) 3.30 (0) 6.98 907966 12 1.28 (+α) 3.30 (0) 6.65 927364 13 1.20 (+1) 3.10 (−1) 6.67 907854 All values in parentheses indicate coded level. Actual and coded experimental values and its responses obtained from different run using CCD. All values in parentheses indicate coded level. The selected responses of each experiment were further analyzed and best-fitted model was obtained. The generated data was applied in the ANOVA calculations ( Table 4 ), and the statistically significant (P < 0.05) values were observed for the selected CAAs. Thus, the relationship between CMPs such as flow rate (X1) and pH of buffer (X2) on CAAs such as retention time (Rt) and peak area (Pa) was governed by generating the contour and 3D response surface plots ( Fig. 3 ) [ 34 ]. The lack of fit test should be insignificant for the fitted model to obtain a significant level (p < 0.05) [ 25 ]. The polynomial equation was developed in coded equation that indicates the main and interactive effects of the critical method parameters (independent variables). The generated response shows the effect of the CMP on the CAA i.e. main and interaction effects. Furthermore, the developed response-factor relationship was presented graphically via 2D contour plot and 3D response surface plot. Further, the statistical data obtained from the DoE were analyzed by the ANOVA. Thus, the significant impacts were critically observed on the selected CAAs from the CMPs. Table 4 ANOVA for responses surface of linear model. Table 4 Sum of Squares df F Value Prob > F P-value Y 1 Y 2 Y 1 Y 2 Y 1 Y 2 Y 1 Y 2 Linear Model 0.15 6.887E+008 2 2 12.96 4.18 0.0017 (Significant) 0.0479 (Significant) A 0.14 6.867E+008 1 1 23.96 8.34 0.0006 0.0162 B 0.011 1.948E+006 1 1 1.96 0.024 0.1920 0.8808 Residual 0.56 8.235E+008 10 10 Lack of fit 0.030 5.080E+008 6 6 0.74 1.07 0.6456 (Non-significant) 0.4955 (Non-significant) Pure error 0.27 3.155E+008 4 4 Cor Total 0.20 1.512E+009 12 12 A-Flow rate, B-pH of buffer, Y 1 - Retention time; Y 2 -Area of peak; df-Degree of freedom; F value- Fischer's ratio; p-value- Probability value. Fig. 3 The contour and 3D response surface plots. Fig. 3 ANOVA for responses surface of linear model. A-Flow rate, B-pH of buffer, Y 1 - Retention time; Y 2 -Area of peak; df-Degree of freedom; F value- Fischer's ratio; p-value- Probability value. The contour and 3D response surface plots. The optimization of the method and model validation was performed on the design expert software 11. The numerical and graphical approaches were employed for the optimization of the method. The upper and lower levels were set for the desired response for the optimization process to outline the design space [ 19 ]. A stability study of prepared standard stock solution was performed to determine the stability for a longer duration of time. This solution can be used for analytical method development. Herein, the standard solution of BA (10 mg/mL) was stored at 4–6 °C for further analysis. These solution samples were filtered and degassed. The samples were injected in HPLC column for their stability parameters such as reproducibility, peak area, sharpness, and quality of chromatogram. The analysis was performed thrice the quality and resolution of the peak area were compared over the period of time [ 9 ]. The method validation was performed in terms of various validation parameters such as linearity, accuracy, precision, robustness, LOD, LOQ, etc according to ICH Q2 (R1) guidelines [ 35 , 36 ]. System suitability is a principal test of the establishment of the method and it was performed in accordance with USP 40/NF 35 to confirm the performance of resolution and reproducibility of a chromatographic system for suitable and reproducible analysis [ 25 ]. It was performed by injecting the BA-prepared solution (100 μg/mL) six times for assessment of column efficiency, retention time (Rt), tailing factors (Tf) theoretical plate (Tp), and reproducibility respectively. The relative standard deviation (RSD) measurement was used for determining Rt, Tf, and Tp and was limited to RSD ≤2 %. The linearity is test results that can be obtained using different concentrations of analyte. The prepared different concentrations in a range 10–60 μg/mL were injected using an autosampler six times per concentration (n = 6). The linearity was constructed by plotting the prepared analyte concentrations on the x-axis and peak area on the y-axis ( Fig. S3 ). The specificity of the method provides the information about suitability of the method for the analysis and measuring of the drug concentration without any interference of formulation excipients in BA. The placebo solutions with formulation excipients were spiked with drug sample concentration for specificity. The resulting chromatograms were compared to verify excipient-induced interference in drug quantification [ 37 ]. The limit of detection LOD ( k  = 3.3) and limit of quantitation LOQ ( k  = 10) of the analyte was determined by equation ( A = kσ / S), w here A is LOD or LOQ, and σ is the standard deviation of the response and S is the slope of the calibration curve [ 21 , 38 ]. The accuracy of the method is a closeness of the observed concentration value of the method and the injected value determined by the recovery test. These tests were carried out at three different levels such as 80, 100, and 120 μg/mL of BA. It was determined by the standard addition of known concentrations of BA to predetermined samples and subsequently analyzed [ 39 ]. It was evaluated as a percentage error for BA according to the following equation: % Accuracy = Observed concentration Injected concentration × 100 Precision is expressed in terms of repeatability. The intraday and interday precision was calculated by injecting six different preparations (30 μg/mL) of the pure drug on the same day and six different days in a week. The RSD <2 % is considered as a precise method. The results were depicted as mean ± SD and % RSD. Robustness is referred as the ability of reliable and remain unaffected by small variations in method parameters during normal use. It was evaluated by employing intentional changes in the chromatographic conditions. Thus, the results of an intentional change of parameters on the analyte were evaluated by Rt, Tp, Tf, and Pa as per the ICH guidelines. The in vitro drug release profile was estimated for the prepared BA-loaded polymeric nanoparticles. For this purpose, the dialysis bag was used. It was then filled with the developed formulation and BA-suspension. The release media, phosphate buffer, pH 7.4, was used in which the prefilled formulation and BA-suspension were hung separately. The 1.0 mL of sample was removed at the set time such as 0.5, 1, 2, 3, 4, 8, 12, 24 h, and replenished the freshly prepared same media in order to maintain the sink condition. The samples were then diluted suitably and analyzed using the developed HPLC method. The drug release mechanism kinetics was studied using different kinetic models such as Zero order, First order, Higuchi and Korsmeyer-peppas, and Hixon-Crowel. Based on the highest correlation coefficient value, the best-fitted model was selected for the drug release. The results from the dissolution test were depicted as mean ± standard deviation (n = 3).

A simple, precise, sensitive, and reproducible RP-HPLC method for BA and its nanoparticle formulation was developed using a QbD based approach. With the help of response surface methodology and CCD the estimated RP-HPLC method was optimized [ 48 ]. The optimized method showed good separation and better resolution for BA using green and white analytical concept i.e. environment friendly, cost-effective and eco-friendly idea of white analytical chemistry. Thereafter, the polymeric nanoparticles were prepared and the developed method was utilized for the subsequent evaluation of the nanoparticles. The developed RP-HPLC method was rapid, reliable, and versatile which can be applicable to the various formulations without the interference of excipients. The validation reveals that the developed method was linear and specific. Moreover, it has accuracy, robustness, and reproducibility as per the validation test limits. According to the literature, a developed method for BA employing QbD-based approach for the estimation of BA in polymeric nanoparticles. Thus, this method will be helpful for various research groups to analyze the various biological matrix and various products to predict the percent drug release along with the mechanism of kinetics.

Breast cancer is a common disease among women in which uncontrollable growth of breast tissues occurs [ 45 ]. Various drugs are ineffective for the management of hormone-responsive breast cancers due to the unpredictable pharmacokinetic profiles. BA is often recommended for the treatment of hormone-responsive breast cancer that inhibits the production of hormones such as estrogen which is responsible for breast cancer. A robust and reliable analytical RP-HPLC method has been developed for BA using a AQbD-based approach. The variability in method performance was identified and controlled which leads to more consistent and reproducible results. Nevertheless, implementation of AQbD requires a thorough understanding of the principles because it is a more complex, and systematic approach as compared to traditional method development [ 18 ]. In the present study, the RP-HPLC method has been developed successfully using various AQbD elements including the desired characteristic such as ATP and risk assessment. The ATP of the developed method was achieved and validated by focusing CMAs and CAAs. The risk assessment using RAM method was performed to assess the failure mode of the experimental factors. The high-risk parameters such as pH of buffer and flow rate were identified and further validated according to ICH Q2(R2) guidelines. The literature survey depicted no reports on the application of AQbD for BA and in vitro assessment. In our study, 13 experiments were performed using CCD based on design expert software (version 11). The selected responses (CAAs) such as Rt and Pa were analyzed and interactions of experimental variables were studied. Here, the negative impact of CMPs (pH of Buffer and flow rate) were found during the analysis of Rt. This effect was due to the interactions between the analyte and a stationary phase. Importantly, when a larger time was spent by the analyte in the column, a higher Rt was observed. In addition, the reduced flow rate of the mobile phase could be the result of higher retention time [ 46 ]. Moreover, the change in pH of mobile phase could be the reason for higher Rt because pH of the mobile phase was adjusted by the orthophosphoric acid. Hence, the negative effect of the pH of buffer on the retention time was revealed [ 46 ]. The adjusted pH of buffer plays a crucial role in determining the Rt that results in the developed method being robust. Similarly, optimum Pa could be the contribution of the optimum flow rate (X1) and appropriate pH of buffer (X2). Equation (2) indicates that flow rate (X1) and pH of buffer (X2) have a prominent effect on Pa of chromatogram i.e. an increase in flow rate (X1) and pH of buffer (X2), the broadening of peak (higher Pa) occurs [ 47 ]. The developed RP-HPLC method was optimized and validated as per ICH Q2(R2) guidelines in order to generate a reliable, robust, and accurate method. Nevertheless, it is a requirement of various regulatory agencies such as the United States Food and Drug Administration (USFDA) and European Medicines Agency (EMA) that the method to be validated to meet specific standards. Therefore, the parameters such as system suitability, linearity, specificity, LOD and LOQ, accuracy, precision, and robustness were validated. These parameters signify the standard requirement and authenticity of the chromatographic method. Further employing validation parameters, the prepared polymeric nanoparticles containing BA were evaluated for in vitro assessment. The controlled release of drug from the polymeric nanoparticles was observed over the period of time. It was due to the release of drug from the pores of the encapsulated polymeric nanoparticles.

Breast cancer is a cancer that forms in the cells of breast tissue. About 8 out of 10 breast cancers are hormone-positive caused due to the rise in the level of estrogen [ [1] , [2] , [3] ]. To reduce the level of estrogen for the treatment of premenopausal breast cancer, buserelin acetate (BA) is recommended clinically [ 2 ]. BA ( Fig. S1 ) is represented by 5-oxo-L-prolyl-L-histidyl-L-tryptophyl-L-seryl-L-tyrosyl-O-tert-butyl-D-seryl-L-leucyl-L-arginyl-N-ethyl-L-prolinamide-acetate [ 4 ]. It is a synthetic gonadotropin-releasing hormone and active synthetic analog with higher activity than endogenous GnRH [ 5 ]. The pKa of BA is 11.85 due to its strong basic nature. BA is not recommended for oral administration because it becomes extensively harmful if swallowed and may damage the fertility of the unborn child. Secondly, for the treatment of advanced prostate cancer a marketed BA implant is recommended [ 6 ]. It reduces not only the intensity of prostate cancer by lowering level of testosterone in the blood but reduces the painful condition of endometriosis caused by the growth of extra tissue inside or outside the uterus. BA reduces the levels of estradiol by suppressing the aromatase enzyme which is responsible for increasing the level of circulating estrogen in ovaries and muscle tissues. Till date, a few analytical methods have been estimated and validated for BA. Likewise, a limited analytical quantification method of BA in a pharmaceutical product was estimated [ 2 ]. A novel simultaneous HPLC-ESI-MS/MS quantitation method for gonadotropin-releasing hormone (GnRH) was developed and validated [ 7 ]. A novel NMR spectroscopic method was established for originality and reliability of BA using 13 C NMR spectra and [5-D-tyrosine] BA in different solvents [ 8 ]. Thus, in above research work, several experimental runs have been repeated in order to optimize the method with variation in experimental method factors achieving consistency [ 9 ]. Moreover, a lot of time is required to optimize the HPLC method. Moreover, developing a method can not be determined by the joint influence of experimental factors. A traditional approach to method development is more expensive and time-consuming. Thus, such limitations can be overcome by applying the analytical quality-by-design (AQbD) based approach for estimation of BA using RP-HPLC which is a newer and first-ever estimated work. For the development of RP-HPLC method using hazardous organic solvent such as acetonitrile, toluene, methanol is often used in RP-HPLC method [ 10 , 11 ]. These solvents are harmful to the environment, and aquatic and terrestrial life. In comparison to this, water is sustainable, non-toxic and non-hazardous. It can be used in combination to other organic solvents to reduce toxicity. Using water in combination with other solvent such as acetonitrile becomes cost effective and environment friendly [ 12 ]. Thus, the water is considered as a green solvent and pure solvent to be used in RP-HPLC method development. Thus, the greenness and whiteness approach can be achieved for this method development [ 13 , 14 ]. AQbD can be applied to overcome several issues during the development and optimization of the analytical method [ 15 ]. It is popular and consistent analytical method with understanding and minimal failure. Based on the AQbD-based frame work, the development of a robust analytical method with high level of confidence can be possible [ 16 ]. Analytical target profile (ATP) can be achieved by a thoughtful and systematic understanding of independent and dependent variables. Independent variables are selected with risk identification followed by risk assessment. The risk identification and risk assessment have been highlighted in ICH Q8, Q9, and Q10 guidelines respectively [ 17 ]. These guidelines contain a systematic approach for development of an analytical method using AQbD that involves the contribution of various stages like risk assessment, Ishikawa fishbone diagram, and failure mode effect analysis (FMEA). It was a challenge to estimate a robust analytical method for BA. The aim of present analytical study was to achieve enough retention time and sharp peak area with higher resolution. The organic phase, buffer concentration, flow rate, and PH of buffer were the critical factors that influence the retention time and resolution of the analyte. Desirable results can be achieved by controlling all the variables by employing systematic steps based on AQbD approach [ 18 ]. In this framework, ATP can be achieved with an appropriate understanding of the method, previous experience, and scientific knowledge [ 19 ]. ATP was the outcome of the AQbD-based method development which was linked with the critical method parameters (CMPs) and critical analytical attributes (CAAs). The choice of the CMP depends on various factors such as risk identification by the Ishikawa diagram, risk assessment, control process, and continuous improvements [ 20 ]. The overall development of the method was highlighted by the International Conference on Harmonization (ICH) Q8 (R2) which can achieve quality products with minimum experiments and minimum time. However, the new ICH Q14 guideline elicited the method development using AQbD [ 9 , 21 ]. To date, there is no study for the estimation of BA in polymeric nanoparticle formulations or in any other dosage form that has been reported as per the literature. Hence, there was an urgent need to develop a method for the estimation of BA. The present study reports RP-HPLC method development using a QbD-based approach and its validation using reliable and robust method development. Further, the developed nanoparticles were assessed by dissolution study, and the release mechanism of kinetic of BA-loaded nanoparticles was highlighted by mathematical modeling. Besides, the developed analytical method will also help to quantify the BA in human plasma samples.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data will be made available on request.