Development and Validation of a Novel RP-HPLC Analytical Method for Sitagliptin Determination in Human Plasma

Background: Different bio-analytic methods have been developed for determining drug concentration in plasma, but methods for sitagliptin determination are still very rare. In this study, RP-HPLC based method has been developed for assessing sitagliptin concentration in plasma. Aim: To develop and validate RP-HPLC based analytical method for estimating sitagliptin in human plasma for pharmacokinetic applications. Methods: In the present study, the mobile phase composed of acetonitrile: 0.5% triethanolamine (20:80) with pH 6.5 has been utilized. Samples of plasma containing sitagliptin and internal standard (IS)-rosiglitazone were extracted with dichloromethane:diethyl ether (4:6; v/v) at pH 7.4. The rate of flow was 1 ml/min. The retention time was about 5,232 and 6,903 minutes respectively for sitagliptin and rosiglitazone. Results: At concentrations of 100-3200 ng/ml in plasma, calibration curves of sitagliptin were linear. The interand intra-day precision and accuracy ranged in between 93.56-98.56% and 1.09Original Research Article Pamu et al.; JPRI, 33(42B): 92-101, 2021; Article no.JPRI.72886 93 4.55% respectively. For specificity, the study findings showed no co-eluting peaks occurring with IS drug (rosiglitazone) and confirmed that no percentage of interferences at analyte (sitagliptin) retention in presence of rosiglitazone. The sitagliptin recovery was 96.442%. Chromatographic separations were performed on HI Qsil C-18 HS column (250mm x 4.6mm x 5μm). The stability of stock solutions of sitagliptin and IS at room temperature was 98.06% and 100.79% respectively while under refrigerated conditions stability was 98.19% and 96.59% respectively. Freeze-thaw stability for sitagliptin was performed with low & high QC and shown as 98.26% for and 97.45% respectively. The limits of detection and quantification were 8.592 ng/ml and 28.641 ng/ml respectively. Conclusion: A simple, sensitive and accurate method was developed for bio analytical estimation of sitagliptin in human plasma using liquid-liquid extraction technique. The validation results of linearity, accuracy, precision, stability, selectivity, ruggedness, LOD and LOQ were good under acceptable limit and can be applied for pharmacokinetic studies.

Different bio-analytic methods have been developed for determining drug concentration in plasma, but methods for sitagliptin determination are still very rare. There are few studies which have developed analytical methods based on RP-HPLC [6,7] and LC-MS/MS principles [8][9][10][11][12] for the determination of sitagliptin in biological fluids. But LC-MS/MS methodological approaches are very expensive and require the availability of expensive equipments. So through this study, RP-HPLC based approach was explored as it is cost effective. As per literature review, two earlier studies have developed an analytical method for sitagliptin determination in human plasma. One study was conducted as per United States-Food and Drug Administration [7] guidelines and the other one has not specified any guidelines. The present study was performed as per International Conference on Harmonisation (ICH) guidelines with newly developed mobile phase by using an advanced RP-HPLC equipment and detector. Earlier studies were validated only with the parameters of selectivity, specificity, recovery, linearity, accuracy, precision, limit of detection (LOD), limit of quantification (LOQ) and stability (freeze thaw, short/long term) but the present study in addition to these conducted extensive validation parameters like calibration curves, ruggedness, dilution integrity etc. Thus the current work developed and validated a simple, sensitive, accurate, reproducible and cost effective RP-HPLC technique for estimating sitagliptin in human plasma even within a very low detection range [13].

Apparatus and Chromatographic Conditions
The symmetry shield RP18 HPLC system (Millford, MA, USA), a separation module of waters 2695 comprised of a quaternary pump, an auto-sampler, a column heater, a degasser and a 2998 PDA (Photo Diode Areditector) detector. Empower 3 software was used for data acquisition and computer processing. Chromatographic separations were performed on HI Qsil C-18 HS column (250mm x 4.6mm x 5μm) (Kromatek, Great Dunmow, England) attached to a guard column Hypersil BDS C-18 (20mm x 4 mm) (Thermo Fisher Scientific India Pvt Ltd, Hyderabad, India). The mobile phase composed of acetonitrile and 0.05% TEA at pH 6.5 (20:80, v/v). Freshly prepared buffer (acetonitrile) solution was utilized for chromatograohic purpose. Whatman membrane filters of size 0.45 µm Wipro GE Healthcare, Bangalore, India) were used to filter all solutions. The volume of injection was 8.0 µl maintained with the flow rate 1ml/min at an ambient temperature of the column. For detection purpose, diode array detector at 281 nm was utilized. Before injecting drug standards, equilibration of the column was made for at least 20 minutes with the mobile phase flowing through the system.

Preparation of Stock and Working Solutions
Stock solutions of sitagliptin and rosiglitazone were prepared by dissolving 100mg of each drug in 10 mL of acetonitrile and volume was made up to 100mL by adding mobile phase to produce a concentration of 1000 µg/mL for each drug. 10 mL of the prepared solution of each drug was withdrawn and volumes made up to 100 mL to achieve 100 µg/mL of sitagliptin and rosiglitazone.

Preparation of Sample Solutions for Calibration Curve (CC) and Quality Control (QC)
Blank human plasma (0.95 mL) aliquots were spiked with the 50µL (0.05 mL) aliquots of prepared working solutions (2,4,8,16, 32 and 64 µg/mL) to prepare different calibration standard concentrations like 100, 200, 400, 800, 1600 and 3200 ng/mL (Table  1). Similarly,different quality control samples were also prepared from the working solutions (Table  2). By further diluting stock solution in acetonitrile, fresh working solutions were prepared. Working solutions of rosiglitazone were prepared by diluting the stock solution with acetonitrile at a concentration of 500 ng/ml. Fifty microliters of rosiglitazone were used for every analysis.

Extraction Procedure
The liquid-liquid extraction method was used to prepare the sample. An aliquot of spiked human plasma (1 mL) was placed in stoppered test tube, then 50 mL of IS working solution (10 mg/mL) was added and vortex-mixed for 1 minute. 5 mL dichloromethane:diethyl ether (4:6, v/v) was added into this solution, and the contents of the tubes were mixed in a reciprocating shaker at 100 strokes per minute for 30 mins and centrifuged at 3000 rpm for 10 minutes to separate the phases. The supernatant organic layer (4 mL) was transferred to a new tube and allowed to evaporate under nitrogen stream. The residue was then reconstituted in 250 mL of mobile phase and analyzed chromatographically.

Method Validation
Plasma calibration assays were done in 6 runs to test linearity, accuracy, precision, selectivity, dilution integrity, ruggedness, stability studies, recovery study, the limit of detection LOD and LOQ.

Optimization of Chromatographic Conditions
Several combinations of buffer and organic phase were tested for mobile phase. It was found that the combination of acetonitrile-0.05% and TEA (pH 6.5) (ratio 20:80, v/v) provided good peaks for sitagliptin. Fig. 2 shows the representative chromatograms of blank human plasma and plasma samples spiked with sitagliptin or rosiglitazone.
The organic buffer, i.e., acetonitrile, was identified as the most suitable buffer that doesn't absorb at low wavelength [14]. Acetonitrile was selected as it was adequate to avoid band tailing. At low pH (pH < 2) silica-based particles were unstable [15] while at pH 6.5, best area counts with the least band tailing were observed, and so pH was fixed as 6.5. The run time was short and required just 10 min (retention times for sitagliptin and rosiglitazone were approximately 5.2 min and 6.9 min, respectively). This short analytical time was considered good for plasma samples.

Liquid-Liquid Extraction
In this experiment, the liquid-liquid extraction method was chosen. To develop a single step liquid-liquid extraction procedure with good recovery, a large range of extraction solvents such as dichoromethane, diethyl ether, and a mobile phase (acetonitrile: 0.05% of TEA) were investigated. As per literature [16], extractions of medications from biological fluids are typically the most difficult step in any analysis because of the interference's presence. The absolute recoveries of sitagliptin after single extraction from plasma using these organic solvents were less than 95% unless diethyl ether used where the absolute recovery was high (over 95% for both sitagliptin and rosiglitazone). Diethyl ether was considered better extracting solvent, most likely because of its moderate polarity, low cost, and volatility [17]. Diethyl ether was consequently selected to further explore the impact of pH on extraction efficiencies. It was found that the plasma pH of 7.4 gave the most noteworthy and best percentage recovery for sitagliptin (96%) and rosiglitazone (97%). This pH was chosen for liquid-liquid extraction method.

Recovery
Recovery was quantified by finding the ratio of the slopes of the calibration curves for extracted to non-extracted samples. Recovery for sitagliptin in plasma was found to be 96.42%.

Calibration Curves, Linearity, Accuracy and Precision
The calibration curve for sitagliptin was linear in 100-3200 ng/ml concentration range in human plasma, and linearity is shown in Fig. 3. The inter-and intra-day calibration curves showed consistent linearity, as seen in the consistency of intercept, slope and coefficient of correlation. A typical concentration curve for sitagliptin had a slope of 0.0013, an intercept of 0.0088, and r2 = 0.9983. An accuracy and precision assay determined from the low (250 ng/ml), medium (1000 ng/ml) and high (3000 ng/ml) QC plasma samples. The inter-day assay estimated analysis of 6 QC samples and investigated on four different days. Intra-day estimated for each 6 QC sample on a single day. The study of inter-and intra-day precision was showed as percent of the coefficient of variation (CV) and accuracy showed as the mean percentage of the analyte [18]. According to the ICH [13], the estimated precision should not overtake 20% of the CV at each concentration. The accuracy and precision results of the intra-and inter-days are expressed in Table 3.

Specificity and Selectivity
In this HPLC study, selectivity was evaluated by analyzing the six different plasma lots and no coeluting peaks occurred with IS and rosiglitazone. Fig. 4 shows the chromatogram of one of the tested drugs (rosiglitazone) with sitagliptin and the peaks of interests. The results showed no percentage of interferences at analyte retention in presence of rosiglitazone. Therefore, it was concluded that this method is selective and suitable for sitagliptin quantification in human plasma samples.

Stability
As per ICH guidelines, stability studies were conducted [13]. Three concentrations of sitagliptin in plasma were prepared in six QC samples at low QC (250ng/ml), medium QC [1000ng/ml], and high QC (3000ng/ml). Three different stability studies were carried out, and results are shown in Table 4.   In these stability studies, sitagliptin and rosiglitazone did not show any significant degradation (Table 4). When plasma was stored at -70 o C, sitagliptin and rosiglitazone peaks didn't have any interference during storage with the matrix components. These results suggest that under conditions of storage and during injections, sitagliptin was stable in plasma.

Ruggedness
The ruggedness of the developed method, accuracy and precision were examined with the same make of the different column and with different reagent lots. The ruggedness of analyzed sitagliptin QC samples were accurate and precise as given in Table 5.

Dilution Integrity
The developed method's dilution integrity was tested by using six samples diluted 2 times and another six samples diluted 4 times. The precision and accuracy for sitagliptin at 2-fold dilution were found to be 3.179 and 98.04%, and at 4-fold dilution were 0.84 and 96.08%, respectively.

LOD and LOQ of Sitagliptin
LOD and LOQ are the lowest analyte concentration at which detection is feasible. LOD and LOQ of sitagliptin were found to be 8.592 ng/ml and 28.641 ng/ml by using signal-to-noise method.

CONCLUSION
The newly developed method for sitagliptin was validated using RP-HPLC. Sample preparation and quantification of sitagliptin is simple and convenient for human plasma samples. The validated method has good linearity, precision, accuracy, stability, selectivity, ruggedness, LOD and LOQ over the relevant therapeutic concentration range. Considering these points, the developed method is preferable for clinical monitoring of sitagliptin levels in human plasma and can be applied for pharmacokinetic studies.

DISCLAIMER
The products used for this research are commonly and predominantly used products in the area of research and country. There is absolutely no conflict of interest between the authors and producers of the products because we do not intend to use these products as an avenue for any litigation but for the advancement of knowledge. Also, the research was not funded by the producing company rather it was funded by personal efforts of the authors.