Application of drug–metal ion interaction principle in conductometric determination of imatinib, sorafenib, gefitinib and bosutinib

Abstract An analytical method for the quantification of anticancer agents such as imatinib, sorafenib, gefitinib and bosutinib using conductometry was developed. Each drug solution was mixed with measured concentration of metal ion (Cu2+) solution resulting in drug–metal ion complexation in the titration cell. Conductance was progressively decreased on addition of the analyte solution up to a point of maximum reduction, that is, the end point. Corrected conductance values were calculated from the observed conductance and used to plot a graph against the volume of drug solution added. No interferences were observed from blank and placebo as they gave no clear inflection in the conductivity during titration. The precision and the accuracy of the developed method was established by the analysis of quality control samples; %RSD of corrected conductance values <2% and recovery results within 100 ± 2% were achieved. The calibration graphs obtained were linear over the concentrations 1.0–1.4 mM for all the drugs (R 2 > 0.99). The drugs were successfully analyzed in their respective dosage forms prepared in-house. The method has offered easier, faster and cost-effective analysis of the selected drugs and can be used for routine determinations in the quality control laboratories. More importantly, it is an environmental friendly procedure, as no organic solvent was used throughout the analysis.


Introduction
Tyrosine kinase (TK) is one of the more than 500 members of protein kinase enzyme family, which play a pivotal role in various cellular processes such as proliferation, apoptosis, immune response and cell development. The dysregulation of TK may result in pathological conditions including cancer, and in the last few decades, TK inhibitors have emerged as one of the most studied drugs for cancer therapy [1][2][3]. Imatinib is the first clinically successful TK inhibitor mainly used in the treatment of chronic myeloid leukemia (CML). It is believed to act through inhibition of BCR-ABL protein kinase that is responsible for the regular proliferation of the myeloid cells [4]. Following the development of imatinib, more than 30 protein kinase inhibitors have been approved for clinical use in the treatment of different types of cancers including lung, kidney, blood, gastrointestinal tract, breast and skin. Other important TK inhibitors such as sorafenib, gefitinib, sunitinib and bosutinib are mainly used for the treatment of clear cell renal carcinoma, advanced non-small cell lung cancer, metastatic renal cell cancer and chronic myelogenous leukemia, respectively [5][6][7].
The literature survey revealed several analytical procedures that have been developed for estimation of TK inhibitors either as alone or in combinations. These methods were mainly based on the chromatographic techniques such as HPLC and LC-MS/MS [8][9][10][11][12]. These techniques are expensive and involve complex and timeconsuming experimental procedures, in addition to the use of large amounts of hazardous solvents and high equipment cost. Therefore, the analytical methods based on these techniques are less preferred for the routine analysis in the quality control laboratories. Nevertheless, the use of such techniques for the estimation of drugs in biological samples is recommended. Other relatively economical and faster analytical techniques such as UV spectrophotometry [13][14][15] and potentiometry [16] have also been used for the determination of these drugs. The routine quality control analysis of pharmaceuticals demands for simple, economical and fast analytical techniques. The conductometric analysis might prove to be more suitable for the purpose as it is one of the most cost-effective techniques for drug analysis and offers simple, sensitive and rapid analytical procedures. The conductometric titration involves the measurement of conductance of solution, which varies due to interaction between ionic species before and after the end point.
Previously, we successfully utilized conductometric titration methods to perform the quantitative estimation of diphenhydramine hydrochloride [17] and pioglitazone hydrochloride [18] in a pure bulk form and pharmaceutical formulations. Moreover, this technique was also used by several other researchers for the determination of various pharmaceutical substances. However, to the best of our apprehension, no conductometric method was developed for quantification of tested drugs. Therefore, the present investigation was undertaken to develop a simple, economical and fast analytical method to quantitatively analyze four TK inhibitors imatinib, sorafenib, gefitinib and bosutinib in bulk drug and dosage forms using conductometry.
The functional moieties present in the selected drugs were methylpiperazine, pyridine and pyrimidine in imatinib; pyridinecarboxamide in sorafenib; morpholine and quinazoline in gefitinib and methylpiperazine, quinoline and carbonitrile in bosutinib (Figure 1), which get ionized in suitable solvents resulting in generation of negatively charged ions (bases) that are capable of binding to the Lewis acids (metal ions). Therefore, in the method development, principle of drug-metal ion interaction was applied, and in the titration vessel, the drug species was made to bind to metal cations, leading to alteration in the conductivity of the solution. Owing to the speed, simplicity and cost-effectiveness of the experimental procedure, the present method can be used for routine analysis of the selected drugs in quality control laboratories. Moreover, the procedure was ecofriendly as no organic solvents were used throughout the experiment.
2 Experimental procedure

Chemicals and reagents
The samples of imatinib, sorafenib, gefitinib and bosutinib were purchased from Sigma Aldrich (Steinheim, Germany). Copper sulphate (CuSO 4 ), ferrous sulphate (FeSO 4 ), ferric chloride (FeCl 3 ), silver chloride (AgCl) and sodium chloride (NaCl) were purchased from Merck, Germany. Starch, anhydrous lactose and magnesium stearate were acquired from Loba Chemie (India), and carboxymethyl cellulose CMC) was obtained from Sigma Aldrich (Steinheim, Germany). The chemicals used in this experiment were of analytical grade, and doubledistilled grade water was used to prepare the analytical solutions.

Instrumentation and conductometric titration
Conductometric titration was performed using portable conductivity/TDS meter (Jenway 470, Staffordshire, United Kingdom). Double-distilled grade water was produced in-house using Milli-Q (Millipore, Molsheim, France) water purifier system. An aliquot (10 mL) of metal ion solution was transferred to the titration vessel maintained at room temperature (25 ± 2℃), and the probe was immersed into the titration cell. It was followed by quantitative addition of sample solution containing one of the analyte drugs. After each 5 μL addition, the solution was stirred for 30 seconds, and the resulting conductance was recorded. Each observed conductance was used to calculate the corrected conductance values by following the reported procedure [19,20]. Considering conductivity as a linear function of dilution, Equation (1) is utilized for the calculation of correction of conductance [21].
where Ω −1 obs and Ω −1 correct are the observed and corrected electrolytic conductivity values, respectively, and v1 and v2 represent the initial and final volumes of each added titrant, respectively. A plot of corrected conductivities versus volumes of added drug solution (titrant) was prepared to determine the end point.

Preparation of 10 mM ammonium formate buffer (pH 3.5)
Accurately weighed quantity (63.06 mg) of ammonium formate was dissolved in 100 mL double-distilled water followed by adjustment of pH to 3.5 by adding dilute formic acid. The solution obtained was then filtered using 0.45 µm filter.

Preparation of standard drug and metal ion solutions
Standard stock solution of imatinib (2 mM) was prepared by dissolving 59.0 mg of imatinib in 5 mL water. However, sorafenib, gefitinib and bosutinib stock solutions were prepared separately by dissolving 63.7 mg, 44.7 mg and 54.8 mg, respectively, into 5 mL freshly prepared ammonium formate buffer solution (pH 3.5) to achieve a concentration of 2 mM for each drug. Similarly, the standard stock solutions for metal ions were prepared by dissolving 11.6 mg, 32.4 mg, 30.4 mg, 49.8 mg and 28.6 mg of NaCl, FeCl 3 , FeSO 4 , CuSo 4 and AgCl, respectively, in 100 mL of double-distilled water to obtain a concentration of 2 mM for each solution. The stock solutions of drugs and metal ions were diluted appropriately with their respective diluents to achieve working standards of required concentrations. All the prepared solutions were filtered using 0.45 µm filter.

Preparation of pharmaceutical formulations
The pharmaceutical formulations for imatinib, sorafenib, gefitinib and bosutinib used in this study were prepared in-house and consisted of synthetic mixtures of each drug and placebo. To prepare the mixtures, 50 mg of each drug was added and mixed in calculated quantities of starch, anhydrous lactose, magnesium stearate and carboxymethyl cellulose.

Preparation of solutions from drug formulations
In-house mixtures equivalent to 59.0 mg, 63.7 mg, 44.7 mg and 54.8 mg of imatinib, sorafenib, gefitinib and bosutinib, respectively, were accurately weighed and transferred to separate 5 mL volumetric flasks. The mixture containing imatinib was dissolved in doubledistilled water, while sorafenib, gefitinib and bosutinib were dissolved in ammonium acetate buffer (pH 3.5) to obtain stock solutions of 2 mM concentration for each drug. The stock solutions were diluted appropriately with their respective diluents to attain working standards of required concentrations. All the prepared solutions were filtered using 0.45 µm filter.

Preparation of placebo solutions
The placebo was prepared by properly mixing the same amount of ingredients of pharmaceutical formulations without drugs. Equivalent amounts of placebo powder containing anhydrous lactose, starch, carboxymethyl cellulose and magnesium stearate were weighed and transferred to separate 5 mL volumetric flasks. The placebo mixture for imatinib was dissolved in doubledistilled water, while sorafenib, gefitinib and bosutinib were dissolved in ammonium acetate buffer (pH 3.5). Further dilutions were made by following the procedure as described in the preparation of sample solutions. The final solutions were filtered through 0.45 μm Millipore filter.

Method validation
The final optimized conductometric method was validated according to the ICH and USP guidelines [22,23], and different validation parameters including specificity, sensitivity, linearity, precision, accuracy and solution stability were evaluated. This method was checked for specificity by conductometrically analyzing the placebo as well as blank solutions and comparing their conductivities with the ones observed from the analysis of sample solutions. The sensitivity of the method was examined through the calculation of limit of detection (LOD) and limit of quantification (LOQ) of all the analytes. Linearity of the developed conductometric method was evaluated at five concentration points including 1.0, 1.1, 1.2. 1.3 and 1.4 mM for each analyte separately. The calibration curve for each analyte was prepared by plotting corrected conductivities against the concentrations of the analyte. The regression analysis was performed by least square method, and y-intercept, slope and correlation co-efficient were calculated. For each concentration level, the six replicates were performed and conductance was measured.
For the determination of accuracy and precision of the present method, quality control solutions possessing 1.1, 1.2, and 1.4 mM concentrations at low (LQC), medium (MQC) and high levels (HQC) of analytes, respectively, were selected. These solutions were titrated on the same day at different times to evaluate intraday, whereas they were analyzed for three consecutive days to find out the interday precision and the accuracy of the proposed conductometric method. The precision was represented by calculating the %RSD of the conductance values, while the accuracy was determined by recovery studies at aforementioned three concentration levels. A %RSD of <2% and recovery in the range of 98-102% were considered as the acceptance criteria for method precision and accuracy, respectively. Each quality control solution was analyzed in six replicates. Recoveries of drug analytes in pharmaceutical formulations were tested using MQC (1.2 mM) sample at 80%, 100% and 120% concentrations. The standard addition method was employed for the preparation of recovery samples, which were analyzed in triplicates. The standard and test solutions were evaluated for stability after storing at laboratory temperature (25 ± 2℃) for 7 days. At the end of the storage time, the solutions were analyzed by using the current conductometric method in triplicate, and recoveries of all the analytes were calculated.
Ethical approval: The conducted research is not related to either human or animals use.

Method optimization
According to Pearson's Hard-soft acid base theory (HSAB), depending on the mass, charge and polarizability, metals can behave as acids. The metals exhibiting low polarizability index are considered to be hard acids, whereas the ones with high polarizability indices are classified as soft acids, and the metals exhibiting intermediate properties are known to be borderline acids or bases. As per the HSAB theory, usually hard acids favorably coordinate with hard bases and the soft acids coordinate with soft bases [24]. Among the metal ions used in the present investigation, Na + and Fe 3+ were hard acids, Ag + was soft acid and Cu 2+ and Fe 2+ were considered to be borderline acids. The drugs used in the present experiment possess ligands, which can complex with metal ions. These ligands are methylpiperazine, pyridine and pyrimidine in imatinib; pyridinecarboxamide in sorafenib; morpholine and quinazoline in gefitinib and methylpiperazine, quinoline and carbonitrile in bosutinib [25,26].
The present investigation was a development and validation of a conductometric method for the quantitative estimation of four important anticancer agents, imatinib, sorafenib, gefitinib and bosutinib. The method was based on the principle of drug-metal ion interaction, where the ionizable moieties of the drug molecules complexed with certain metal ions. As a result of their complexation, the solution conductance was changed. The equivalence or end point of the titration was determined by the highest alteration in conductance values. The addition of analytical solution (titrant) to the metal ion solution present in the titration vessel led to the continuous decrease in the molar conductance of the solution until the end point. However, further addition of the drug solution resulted in a gradual increase in the conductance values. The reduction in molar conductance could be due to a decrease in metal ion concentration and the formation of lower mobility drug-metal ion complex, whereas the increase in molar conductance was due to an increase in free drug concentration in the titration cell. The curve between corrected conductance values and volume of drug solution (titrant) is shown in Figure 2.
During the optimization process, complexation behavior of certain metal ions including Ag + , Na + , Cu 2+ , Fe 3+ and Fe 2+ with selected drugs was investigated. After remarkable efforts, Cu 2+ was selected for the current analysis because Cu 2+ exhibited superior binding interaction with all the selected drug molecules, and sharp end points were observed. Moreover, among the tested metal ions, better analytical results were obtained when CuSO 4 solution was used as a titrant. The medium to dissolve analytes in conductometric titration may affect the end point as well as the analytical results. In this experiment, to find out the suitable medium, water was tried first, because water is considered to be the most appropriate and economical medium for conductometric determinations. However, only imatinib showed optimum solubility in water and displayed sharp end point with acceptable analytical performance, whereas for sorafenib, gefitinib and bosutinib, water could not be used as medium because of poor solubility, resulting in undistinguishable end point and poor analytical results. Consequently, different acidic and basic buffers were examined, and finally, the most suitable medium to dissolve sorafenib, gefitinib and bosutinib was identified to be ammonium formate buffer (pH 3.5) as all the analytes exhibited optimum solubilities and satisfactory analytical results were achieved in this buffer. Subsequently, metal ions were dissolved in water in the conductometric analysis of imatinib and buffer for the determination of sorafenib, gefitinib and bosutinib.
It is well known that temperature has a marked influence on the analytical results, and conductance of the solutions can be changed with temperature in conductometry. Therefore, after finding out the suitable solution medium, the influence of temperature on the conductivities of the tested drugs was also studied. A temperature range between 25 and 45℃ was tested. However, during trials, no significant alterations in the solution conductance values were observed with an increase in the temperature. Consequently, laboratory temperature (25 ± 2℃) was finalized for the current analysis. To determine the optimum concentrations of drug analytes and metal ion, which can offer stable conductivities and sharp equivalence point, several concentrations were tried during optimization. The concentrations of drugs as well as metal ion less than 0.5 mM were found to be not optimum for the current conductometric determination as poor inflection at equivalence point and unstable conductance values were observed at these concentrations. After several trials, it was found that the concentration between 1.0 and 1.5 mM for analyte drugs as well as metal ions displayed acceptable analytical results and sharp end points. Furthermore, in this investigation, best results were obtained when equimolar concentrations for drug and metal ion solutions were used.

Method validation
The optimized method for estimating imatinib, sorafenib, gefitinib and bosutinib by the conductometric analysis was evaluated for various validation parameters in accordance with the ICH/USP guidelines. To establish the linearity of the optimized method, analytical solutions possessing a range of concentrations of imatinib, sorafenib, gefitinib and bosutinib were conductometrically analyzed. A calibration curve was constructed for each drug by plotting corrected conductance values versus concentrations of analyte ( Figure 3). Consequently, linear plots were obtained for all four analytes over a concentration range of 1.0-1.4 mM, and R 2 > 0.99 was achieved, indicating good linearity of the proposed method. The linearity data have been summarized in Table 1.
The limit of detection (LOD) values for imatinib, sorafenib, gefitinib and bosutinib were calculated to be 0.00021, 0.00037, 0.00037 and 0.0028 mM, whereas the limit of quantification (LOQ) values were found to be 0.00071, 0.00122, 0.00122 and 0.0093 mM, respectively. Remarkably lower LODs and LOQs for all the drugs indicated that the developed conductometric method was highly sensitive. The method specificity was demonstrated via comparison of the conductometric plot drawn from all the drug solutions, with those obtained from analyzing of blank and respective placebo. After careful observation, no sharp change was seen in the conductance during blank and placebo titrations, and there was no significant inflection at the end points. This has indicated that there was no interference from the solution medium, and the components of the placebo in the conductometric analysis using the proposed method and hence, the method was considered to be specific for the selected analytes.
The precision and the accuracy of the proposed conductometric method were evaluated by analyzing the quality control samples for all the drug analytes at low, intermediate and high concentrations in triplicate. The % RSD of the corrected conductance recorded for each drug in both intraday and interday analyses was less than 2%, which was considered to be in the acceptable limit. The accuracy of the method was determined using the same solutions by calculating the recovery of respective drugs at all concentration levels. The recovery values of all the drugs in intraday and interday analysis were within the 100 ± 2% range. The obtained results suggested that the current method was precise and accurate. The precision and the accuracy data are presented in Table 2. After performing the precision and the accuracy experiment, the medium quality control (MQC) solutions was kept aside for 7 days at laboratory temperature (25 ± 2℃) to investigate the stability of imatinib, sorafenib, gefitinib and bosutinib in their respective solutions. The recovery of all the analytes after end of the storage time was calculated and found to be within the 100 ± 2% range. The results showed no significant loss of analytes in any of the four drug solutions, and all the analytes were stable in their analytical solutions, which can be used for at least 7 days at laboratory temperature without degradation. The results of the solution stability are presented in Table 3.

Application to determine imatinib, sorafenib, gefitinib and bosutinib in their respective dosage forms
After validation, the present conductometric method was evaluated for its applicability for the estimation of imatinib, sorafenib, gefitinib and bosutinib in their respective dosage forms, which were prepared in-house. The tested drugs were mixed separately with measured quantities of anhydrous lactose, carboxymethyl cellulose, magnesium stearate and starch. These dosage forms were then quantitatively analyzed using the developed method. First, the conductometric titration curve for each drug solution was carefully examined and compared with those obtained from titration of placebo solution. The method showed good specificity towards the analyses of all drug formulations because no interferences in the conductance values were recorded from placebo solution, and no sharp inflection of conductance values near the end point was observed. Furthermore, the accuracy of the method was also established by performing the recovery experiment at 80%, 100% and 120% of the target MQC concentration (1.2 mM) using drug formulations, and the results for all tested analytes were within the acceptable limit (100 ± 2%). The acceptable recovery of analytes from their respective dosage forms suggested suitability of the current method for quantitative analyses of the tested drugs in their respective dosage forms. The observed recovery of imatinib, sorafenib, gefitinib and bosutinib in their formulations is presented in Table 4.

Conclusion
In this study, complex forming behavior of important anticancer agents imatinib, sorafenib, gefitinib and bosutinib with metal ions was employed to develop a new conductometric method for quantitation in bulk and dosage forms. The analytical procedure involved gradual addition of drug solution to the Cu 2+ solution in the conductometric titration vessel, resulting in a sharp change in the conductance values of the solution owing to the interaction between the test drugs with Cu 2+ . The point of maximum conductance variation was considered as end point. The method validation results indicated that the proposed method was accurate,  precise, sensitive and linear. The proposed method was applied for the assay of imatinib, sorafenib, gefitinib and bosutinib in their respective pharmaceutical dosage forms, which was formulated in-house by mixing each drug with placebo components. No interferences were observed from blank and placebo solutions, indicating good specificity of the method for all drugs in the formulation analysis. The fast analysis and cost-effectiveness are considered to be the major advantages of the method in addition to having acceptable analytical performance and being environmental friendly. As a result, this method can be used for the routine analysis of these drugs in quality control laboratories.