METHOD DEVELOPMENT ON ANALYSIS CYCLOPHOSPHAMIDE AND 4- HYDROXYCYCLOPHOPHAMIDE IN DRIED BLOOD SPOT AND ITS APPLICATION IN BREAST CANCER PATIENTS

Objective: To develop the method for the simultaneous analysis of cyclophosphamide and 4-hydroxycyclophosphamide (4-OHCP) in Dried Blood Spot (DBS) using Ultra-High-Performance Liquid Chromatography-Tandem Mass Spectrometry (UPLC-MS/MS) and its application in breast cancer patients for therapeutic drug monitoring. Methods: Sample preparation used protein precipitation with methanol and acetonitrile (2:1 v/v). The separation was conducted using 1.7μm (2.1 x 100 mm) Waters AcquityTM Results: The method was linear in the range of 50–30.000 ng/ml for cyclophosphamide and 10–1000 ng/ml for 4-OHCP. Lower Limit of Quantification (LLOQ) concentration of cyclophosphamide was 50 ng/ml and 4-OHCP was 10 ng/ml. Accuracy and precision within-run and between-run met the requirements with % diff and CV, not exceeding ±15% and not more than ±20% for LLOQ concentration. The results from DBS samples of cancer patients showed that the level of cyclophosphamide was in the range of 6045.980 ng/ml to 37024.403 ng/ml and 4-OHCP was in the range 33.155 ng/ml to 246.362 ng/ml. UPLC C18 column; mobile phase consists of 0.01% formic acid and methanol (50:50 v/v) with isocratic elution, column temperature 30 °C, flow rate 0.3 ml/min and hexamethylphosphoramide (HMP) used as an internal standard. Analysis was performed by a triple quadrupole mass spectrometry with a positive ion mode of Electrospray Ionization. Cyclophosphamide was detected at m/z 260.968>139.978, 4OHCP at m/z 338.011>224.979, and HMP at m/z 180.17>92.08. The method was applied to quantify cyclophosphamide and 4-OHCP in DBS of breast cancer patients. Blood samples were collected at 2 and 4 h after cyclophosphamide administration for therapeutic drug monitoring. Conclusion: The developed method met the requirements of all validation parameters under the Guideline on Bioanalytical Method Validation by the European Medicines Agency in 2011. Method can be applied on DBS of cancer patients and the results showed that cyclophosphamide and 4-OHCP was detected on 17 samples of breast cancer patients. This can be one of the parameters for therapeutic drug monitoring.


INTRODUCTION
Cyclophosphamide is an anticancer drug that acts as an alkylating agent and is used singly, as well as being a regimen in multiple types of cancer therapies such as non-Hodgkin's lymphoma, sarcoma, and breast cancer [1]. Cyclophosphamide is a prodrug that requires activation by the cytochrome P450 enzyme to convert into its active metabolite, 4-OHCP. This metabolite is in balance with its tautomeric form, aldophosphamide, which will become its active form, phosphoramide mustard and acrolein [2]. The number of enzymes that play a role in the metabolism of cyclophosphamide can cause various expressions and activities among patients, thus contributing to the metabolism, efficacy and toxicity of this drug [3]. Therefore, a method that can monitor the pharmacokinetic parameters of cyclophosphamide is needed to determine the relationship between drug dosage and its efficacy.
Quantitative analytical methods of cyclophosphamide and 4-OHCP have been described previously using High-Performance Liquid Chromatography (HPLC) [4] and Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS) [5][6][7]. However, quantitative analytical methods of cyclophosphamide and 4-OHCP in DBS simultaneously using UPLC-MS/MS have never been done before. UPLC is a technical advancement of liquid chromatography where there are innovations in instrumentation and particle size to improve the resolution, speed, and sensitivity of the method [8]. 4-OHCP metabolites are unstable in biological fluids such as in plasma, which has a half-life of 6 min [7]; this can be overcome by performing derivatization procedures prior to analysis [7,9,10]. In this study, an optimization and validation method for the analysis of cyclophosphamide and 4-OHCP in DBS was carried out using UPLC-MS/MS. Validation parameters are tested according to the requirements of the Guidelines on Bioanalytical Method Validation by the European Medicines Agency in 2011.
Cancer patients who receive cytostatic therapy are the priority of therapeutic drug monitoring to ensure that patients receive safe, effective and rational drugs, so it will minimize drug-related problems and optimize therapeutic effects [11]. After obtaining an optimal and valid method, the method was applied by analyzing dried blood spot samples of breast cancer patients in "Dharmais" Cancer Hospital as one of the effort for therapeutic drug monitoring. DBS samples were analyzed at Cmax

MATERIALS AND METHODS
and elimination phase. a. Patients is not willing to take part in the research and not signing the informed consent.

Stock solution
A 10 mg cyclophosphamide was dissolved in methanol until a concentration of 1 mg/ml cyclophosphamide was obtained. A 2.5 mg 4-OHCP was dissolved with methanol until a concentration of 0.25 mg/ml 4-OHCP was obtained. A 10 mg HMP as internal standard was dissolved with methanol until a concentration of 1 mg/ml HMP was obtained.

Derivatization solution
Semicarbazide hydrochloride standard was dissolved with 10 ml of 50 mmol potassium phosphate (pH 7.4) buffer until 2 M concentration was acquired.

Optimization of derivatization procedure
The derivatization procedure is done by applying 2 M semicarbazide hydrochloride on DBS paper, then dried for 2 h prior to blood spotting. Furthermore, the blood containing the analyte was applied on paper and sample preparation was carried out. The volume of derivative solution tested was 10 µl, 20 µl, 30 µl, 40 µl, and 50 µl.

Optimization of sample preparation
Optimization of sample preparation began with the selection of extraction methods between protein precipitation with methanol, liquid-liquid extraction with ethyl acetate-acetonitrile, and protein precipitation followed by liquid-liquid extraction. The next step is optimization of selected extraction methods. First, blood spotting volume was optimized for 10 µl; 20 µl; and 30 µl, then the DBS paper was left to dry at room temperature and was optimized for 1; 2; and 3 h. Blood spots were cut and put into 1.5 ml microtube, then 20 µl HMP 1 µg/ml solution as the internal standard was added to the microtube and vortex mixed for 20 s. After that, the mixture was extracted with methanol and acetonitrile with the following combination: 1:0 v/v; 1:1 v/v; 1:2 v/v; 2:1 v/v; and 0:1 v/v and the volume was optimized for 200 µl; 500 µl; 800 µl; and 1000 µl. The mixture was vortex mixed, and was optimized for 4; 5; and 6 min, sonication was optimized for 5; 10; and 15 min, and centrifugation optimization was performed for 10 min at 5223.296 g; 8161.4 g; and 11752.416 g. A total of 900 µl of supernatants were evaporated at 60 °C for 20 min under N2 Method validation gas flow. The residue was reconstituted with 100 μl of mobile phase (0,01% formic acid in water-methanol = 50:50 v/v), vortex mixed for 20 s, and centrifuged for 5 min at 1147.068 g. Finally 10 μl of aliquot was injected into the chromatography system.
The full validation of cyclophosphamide and 4-OHCP analysis method in DBS with parameters such as selectivity, carry over, LLOQ, linearity of calibration curve, accuracy and precision, recovery, dilution integrity, matrix effect, and stability [12].
Application of the method-sampling procedure "Dharmais" Cancer Hospital patients was selected as the subject of the study according to the inclusion and exclusion criteria. Patients were given explanations related to the research and if the patients willing to be the subject, patients would sign an informed consent. Blood samples were taken by finger prick method at 2 and 4 h after cyclophosphamide regiment chemotherapy was given to the patients.

Optimization of derivatization procedure
The optimum derivatization procedure was done by applying 20 µl of semicarbazide hydrochloride 2 M on DBS paper, then dried for 2 h prior to blood spotting. Furthermore, the blood containing the analyte was applied on paper and sample preparation was carried out.

Optimization of sample preparation
The optimum sample preparation was obtained by protein precipitation. Blood containing analytes was spotted 30 µl on DBS paper and left to dry at room temperature for 3 h. Blood spots were cut and put into 1.5 ml microtube, then 20 µl HMP 1 µg/ml solution as the internal standard was added to the microtube and vortex mixed for 20 s. After that, the mixture was extracted with 1000 µl methanol-acetonitrile = 2:1 v/v. The mixture was vortex mixed for 4 min, sonication for 15 min, and centrifugation was performed for 10 min at 11752.416 g. A total of 900 µl of supernatants were evaporated at 60 °C for 20 min under N2 Method validation gas flow. The residue was reconstituted with 100 μl of the mobile phase, vortex mixed for 20 s, and centrifuged for 5 min at 1147.068g. Afterward, 10 μl of aliquot was injected into the chromatography system.

a. LLOQ and calibration curve
The LLOQ of cyclophosphamide was 50ng/ml with CV of 4.42% and %diff between 7.16% to 19.62% and 4-OHCP was 10ng/ml with CV of 6.43% and %diff between-18.27% to-6.76%. The calibration curve was linear in the concentration range of 50-30000ng/ml for cyclophosphamide with correlation coefficient (r)>0.9972 and 10-1000ng/ml for 4-OHCP with correlation coefficient (r)>0.9983. Calibration curve was calculated with a weighted linear curve fit equation 1/x [13].

b. Selectivity
The selectivity was performed on LLOQ concentrations and blanks using six different blood sources. The results show that the interference or impurity on the retention time of analytes and internal standard is still meet the criteria<20% of LLOQ concentration and<5% for the internal standard according to European Medicines Agency guideline of bioanalytical method validation. The chromatogram of the blank sample can be seen in fig.  1. The chromatogram of cyclophosphamide, 4-OHCP-SCZ, and HMP with four-level concentrations which was LLOQ, Quality Control Low (QCL), Quality Control Medium (QCM), Quality Control High (QCH) can be seen in fig. 2.

c. Carry over
The result shows no carry over in blank sample after injection of the highest concentration (ULOQ) sample. The carry over percentage still meets the requirements for analytes<20% and<5% for the internal standard. The result can be seen in table 1.

d. Accuracy, precision, and recovery
This research performs accuracy and precision in within-run and between-run. The test was performed using five replicates on each of four concentrations which contain both cyclophosphamide and 4-OHCP, such as LLOQ (50 and 10 ng/ml); QCL (200 and 40 ng/ml), QCM (15000 and 500 ng/ml) and QCH (24000 and 800 ng/ml) for each time of analysis (within-run) and performed as many as three times the analysis of at least 2 different d (between-run). Accuracy is a parameter to see the proximity of the measured concentration results in the analysis with the actual concentration and represented by the %diff (bias) value, whereas the precision is performed to see the proximity between the repetitions of the measurement analytes represented by the CV value. The method was accurate and precise because all values met the requirements that bias (%diff) and CV for LLOQ<20% while for other concentration levels<15%. The result can be seen in table 2 and 3.  Besides accuracy and precision, recovery was conducted to know the extraction efficiency. The recovery was calculated by comparing the area of analyte obtained after extraction with the area of standard analyteat the same concentration. Recovery of cyclophosphamide at three levels of concentration QCL, QCM, and QCL was 28.95-34.01% and 4-OHCP was 25.88-33.84%.

e. Dilution integrity
Dilution integrity was performed to determine whether dilutions carried out during the analysis affects accuracy and precision. The test was performed with a concentration above ULOQ or two times the QCH concentrations of both cyclophosphamide and 4-OHCP 48000 ng/ml and 1600 ng/ml, diluted to half (24000and 800 ng/ml) and one quarter (12000and 400 ng/ml) using a whole blood. The analysis was conducted in five replicates on each dilution and observed the accuracy and precision parameters. The result showed that until one quarter dilution, the CV and %diff<15%.

f. Matrix effect
Matrix effect was conducted because the analysis of cyclophosphamide and 4-OHCP was using mass spectrometry. The matrix effect describes the presence of other compounds outside the analytes and internal standard in the ionization of the analytes and the internal standard in the ionization process. This effect was represented by the matrix factor by calculating the percentage between the area of analytes added after the blank extraction process with the standard solution area. The result can be seen in table 4.   The BEH C18 column is nonpolar with the mobile phase of 0.01% formic acid-methanol (50:50) is polar, so this analysis used reversephase chromatography system. Acquity BEH precolom was used for analysis. The mobile phase used was 0.01% formic acid-methanol (50:50) with isocratic elution for 4 min, a flow rate of 0.3 ml/min, and a column temperature of 30 °C [5].
UPLC-MS/MS instrument and condition in this study was developed for the first time by Harahap, Samuel, Andalusia, and Syafhan by the year of 2016. In this study, the chromatographic condition was applied to develop a new bio sampling method, called DBS.4-OHCP as cyclophosphamide's metabolite was very unstable against the presence of O2 DBS preparation was conducted using protein precipitation. This method was chosen because it cleans the analytes of many impurities in the sample and resulting the highest area of the analyte. The optimization was performed by methanol and acetonitrile as a combination of extracting solution because they are polar so the analytes will be extracted into the solvent. Besides, that solvent also have been used on the extraction of ifosfamid and cyclophosphamide in DBS [16]. The optimum sample preparation were obtained by protein precipitation using 1000 µl methanolacetonitrile = 2:1 v/v. The mixture was vortex mixed for 4 min, sonication for 15 min, and centrifugation was performed for 10 min at 11752.416g. The supernatants were evaporated and the residue was reconstituted with 100 μl of the mobile phase, vortex mixed for 20 s, and centrifuged for 5 min at 1147.068g. Afterward, 10 μl of aliquot was injected into the chromatography system. The optimization results were selected based on the area of analytes and the internal standard also chromatogram form of each compound. [14]. Therefore, the derivatization procedure is carried out prior to analysis. Derivatization in this study was carried out using semicarbazide hydrochloride to form semicarbazone derivatives [5,14]. The procedure of derivatization was performed before the blood containing the analytes was applied on the DBS paper. The procedure of derivatization was developed first by Mess, Taillon, and Côté by the year of 2012 [15] to derivate racecadotril's metabolite, thiorphan with 2-bromo-3'-metoxyacetophenon because it was unstable of oxidation in biology matrix.
After acquiring the optimum condition, the analysis method validation was conducted. Based on all value results of validation that met the requirements of Guideline on Bioanalytical Method Validation [12], it can be concluded that the developed method was valid. This research showed LLOQ was 50 ng/ml for cyclophosphamide and 10 ng/ml for 4-OHCP. This value fulfilled the European Medicines Agency requirement that was 5% of Cmax The average levels of cyclophosphamide and 4-OHCP at the 2 . Therefore this new method of bio samping using DBS can be applied in pharmacokinetic studies or bioequivalence studies. nd h was higher than the 4 th h. This is in accordance with previous research that at the 2 h after administration is Cmax From the results,method that had been developed before could be applied on the analysis of cyclophosphamide and 4-OHCPfor the therapeutic drug monitoring. Cyclophosphamide metabolite, 4-OHCP, will be converted to phosphoramide mustard which can alkylate DNA so it can provide a cytotoxic effect on cancer cells. time and 4 h after administration is an elimination time of both compounds [16,17]. Drug and their metabolites level in each patient can be different. This can be due to the expression level of the CYP enzyme, especially CYB2B6 in the liver, and also the individual capacity to the enzyme. In addition, the level of expression and activity of ALDH isoenzymes can also vary between individuals. Beside polymorphism factors, other factors such as disease status, weight, age, food, dose, stress, dysfunction of other organs such as the liver and kidneys, etc can also affect both compouds level. Graphic of both compounds level can be seen at fig. 3.
In this research there were 17 subjects from Dharmais Cancer Hospital to verify the applicability of this method. In order to establish a reliable pharmacokinetics data, more samples are needed to be taken from a bigger pool of subject.

Fig. 3: Bar chart of cyclophosphamide and 4-OHC plevel CONCLUSION
In conclusion, this method for quantifying cyclophosphamide and 4-OHCP in DBS was successfully developed, validated, and can be applied on breast cancer patients for therapeutic drug monitoring. The LLOQ obtained in this study was 50 ng/ml for cyclophosphamide and 10 ng/ml for 4-OHCP. This method is potential to be used on bigger scale of samples.