Application of HPLC Analysis of Medroxyprogesterone Acetate in Human Plasma

Generally, progestogens, i.e. progesterone in combination with oestradiol esters, are used primarily in the treatment of menstrual irregularities to maintain endangered pregnancies. Their derivatives are orally active medroxyprogesterone acetate including hydroxyprogesterone caproate, chlormadinone acetate, megestrol acetate. Naturally occurring glucocorticoids are often used in medicine such as cortisone, hydrocortisone, prednisone and prednisolone (Fig. 1). Depot-medroxyprogesterone acetate (Depot-MPA) is commercially available and commonly known as a hormonal contraception used by women. It is a synthetic steroid hormone, which mimics natural progesterone, but its contraceptive activity has been shown to be about 30 times higher. Depot-MPA is widely used by intramuscular administration for long-term contraception [1-3].


Steroid hormones and hormonal contraceptive drugs
Generally, progestogens, i.e. progesterone in combination with oestradiol esters, are used primarily in the treatment of menstrual irregularities to maintain endangered pregnancies. Their derivatives are orally active medroxyprogesterone acetate including hydroxyprogesterone caproate, chlormadinone acetate, megestrol acetate. Naturally occurring glucocorticoids are often used in medicine such as cortisone, hydrocortisone, prednisone and prednisolone (Fig. 1). Depot-medroxyprogesterone acetate (Depot-MPA) is commercially available and commonly known as a hormonal contraception used by women. It is a synthetic steroid hormone, which mimics natural progesterone, but its contraceptive activity has been shown to be about 30 times higher. Depot-MPA is widely used by intramuscular administration for long-term contraception [1][2][3].
Research concerning its pharmacokinetics has been performed in order to improve the hormonal activity of the drug formulation [2,4]. This will help to standardize the dosage of the drug.

Determination of MPA in hormonal contraceptive drugs
Analysis of MPA in plasma or serum has been performed to study its pharmacokinetics and to monitor its residual levels in cancer patients after oral or intramuscular administration [4,5]. Numerous recent researches for this analysis have been reported concerning the method sensitivity and selectivity in association with various instruments including GC, HPLC and others [6][7][8][9][10][11][12][13][14][15]. However, each of which had their advantages and disadvantages. Therefore, development and validation of the method for MPA analysis in spiked blood plasma using internal standard by HPLC method with specific derivatizing agents and clean-up by solid-phase extraction have been carried out. The method sensitivity obtained is suitable for pharmacokinetics study of MPA. Normally, the levels of MPA found in plasma range from 1.75 to 9.0 ng/mL [5]. Formerly, the measurement of MPA in a plasma sample was mostly carried out by radioimmunoassay (RIA) [6]. Trace analysis of MPA in dog plasma was achieved using cyclohexane extraction, followed by heptafluorobutylic anhydride derivatization and gas chromatography with electron capture detector (GC-ECD). However, this procedure could not be applied for MPA analysis in human plasma due to its high matrix interference. Attempts have been made to optimize the methods for MPA analysis using small amounts of plasma samples [7,8]. Quantitative analysis of MPA in plasma by HPLC with ultraviolet (UV) detection was also carried out, but did not give satisfactory detection limit [9,10]. The high sensitivity of the GC-MS has also been used to monitor the MPA level in human serum [11]. This procedure used trifluoroacetic anhydride derivatization of the extracted portion after SPE and gave the detection limit down to 0.5 ng/mL. Although, HPLC-UV analysis of MPA in plasma has been performed, but its sensitivity was about 5 ng/mL [12]. In order to enhance both selectivity and sensitivity, recent method development and validation of MPA analysis has been focused on HPLC separation with various detection systems. HPLC with chemiluminescence detection was used to trace MPA in serum via 4-(N,N-dimethylaminosulphonyl) -7-hydrazino-2,1,3-benzoxadiazole as a fluorogenic agent [13]. Recently, MPA analysis in plasma sample has been conducted by liquid chromatography-electrospray ion trap mass spectrometry (LC-MS/MS) after liquid phase extraction and this gave 10 times higher sensitivity than GC-MS [11,14].
From these reviews, both GC-ECD and HPLC-UV did not give sensitivity high enough for MPA analysis in pharmacokinetic studies. In clinical aspects, the RIA method has a rather high sensitivity but sometimes gives positive result due to metabolite interference. Both GC-MS and LC-MS/MS are methods of choice with high selectivity and sensitivity, but may be considerably inconvenient for routine analysis due to the high instrument cost. The HPLC with amperometric detector (AD) would be a highly sensitive tool. From the molecular structure, MPA is a ketone steroid in which the ketone group can undergo reduction. Since the reduction is normally easily interfered by oxygen molecule dissolved in mobile phase, this may be a problem [15]. Hydrazines including 2,4-dinitrophenyl hydrazine (DNPH) have been introduced as an electroactive labeling reagent for carbonyl compounds [16]. Until now there is no report of the derivatization of MPA with DNPH (Fig. 2). From the chemical structures condensation reaction can occur between the keto group of MPA and DNPH to give the MPA-hydrazone which can be detected by HPLC-AD. The optimum conditions of DNPH derivatization with MPA were studied. Therefore, development and validation of the method for MPA analysis in spiked blood plasma using prednisolone (P) as internal standard (I.S.) by HPLC-AD after derivatization with DNPH and clean-up by solidphase extraction have been carried out. The method sensitivity obtained is suitable for pharmacokinetic study of MPA.
Standard stock solution (500 mg/mL) of MPA was prepared in methanol. Dilute standard solutions of MPA were also prepared in methanol. The 2,4-dinitrophenylhydrazine stock solution used as a labeling reagent (1,000 mg/mL) was prepared by dissolving the hydrazine in 1.2 mL concentrated HCl prior to adjusting the final volume to 25 mL with methanol.
Stock phosphate solution (2 M) was prepared from sodium dihydrogen phosphate monohydrate (NaH2PO4.H2O) in deionized water. A buffer solution of 30 mM potassium dihydrogen phosphate (KH2PO4), pH 3.0 was made by dissolving KH2PO4 in deionized water and adjusting to pH 3.0 with 6 M orthophosphoric acid (H3PO4). Mobile phase was prepared and used to optimize the separation of the mixture of MPA-DNPH and P-DNPH derivatives by amperometric detection (HPLC-AD). The suitable solvent system was composed of ACN : MeOH : 30 mM KH2PO4 (pH 3.0) in the ratio of 39 : 39 : 22 by volume according to Snyder et al. [17]. The mobile phase was filtered through nylon membrane (0.45 µm, 47 mm) and degassed for 15 min in ultrasonic bath before use. Sep-pak C18 cartridge, 100 mg (Waters Associates, U.S.A.) was used.
Human blood plasma samples were kindly obtained from the Blood Bank, Srinagarin Hospital, Khon Kaen University, Khon Kaen, Thailand.

Method development and validation
i. Derivatization of MPA with DNPH Study on the hydrazine derivatization of MPA was carried out using 2,4dinitrophenylhydrazine (DNPH). The standard solution of MPA (10 mg/mL) was reacted with an excess amount of DNPH for an hour. The MPA-DNPH derivative was then injected into HPLC with detection at 375 nm [18]. Prednisolone (P) was also tested for derivatization with DNPH and was used as I.S. There should be a single peak and no overlap with any interfering peaks in the sample. The P-DNPH derivative was obtained from the reaction of 0.10 g DNPH in 10 mL MeOH and 0.48 mL concentrate HCl with 0.05 g prednisolone [19]. The reaction mixture was placed on a heated water bath at 60 o C for 30 min, allowed to cool down to room temperature and deionized water (60 mL) was added until the derivative product had completely precipitated. The derivative product was filtered through a glass filter and washed with 5 mL 2 M HCl twice, followed by 50 mL deionized water. It was then recrystallized in a mixture of chloroform and methanol prior to use.

Application of HPLC Analysis of Medroxyprogesterone Acetate in Human Plasma 221
ii. Solid-phase extraction of MPA-DNPH derivative The clean up method using solid-phase extraction (SPE) was found to be a critical step for plasma sample both before and after derivatization with DNPH. Spiked standard plasma (MPA 6 ng/mL) was extracted by Sep-pak C18 cartridge using a mixture of MeOH : H2O as extraction solvent. The C18 cartridge was first washed with 2 mL MeOH and followed by 2 mL water. The spiked plasma (2 mL) was added into the cartridge and then washed with 2 mL water, followed by 250 mL 50%(v/v) MeOH (3x), and the final elution was made with 1 mL MeOH. The MPA extract portion was kept for derivatizing with DNPH. The solution of DNPH containing 0.1 mg was added into the MPA extract portion and allowed to stand for 30 min, followed by 0.9 mL water and 0.2 mL solution of P-DNPH (10 ng/mL). The sample of the derivative mixture was then introduced into the cartridge followed by 3 times rinsing with 1 mL 50%(v/v) MeOH. The cartridge containing MPA-and P-DNPH products was washed by 250 mL 50%(v/v) MeOH (3x) and again allowed to stand for 15 min before elution by 1 mL MeOH. The elute was then dried over a stream of N2, redissolved in 60 mL MeOH, then 40 mL MeOH added and 60 mL was injected into the HPLC system. Various concentrations of MPA spiked in a plasma sample with 10 ng/mL P-DNPH as I.S. were extracted by Sep-pak C18 cartridge, and run by HPLC-AD (0.85 V) [20,21]. Limit of detection (LOD) is the concentration of MPA giving a peak height 3 times the baseline noise (3SD) and the limit of quantitation (LOQ) is defined as 10SD. The standard curve was obtained from the concentrations of 0.5, 1, 2, 4, 6, 8 and 10 ng/mL. The accuracy and precision including recovery were determined at three concentration levels (1, 4 and 8 ng/mL) of the spiked plasma standard.

Results and discussion
For method development, studies on derivatization of some hydrazines with MPA had been carried out in acidic solution [15]. It was found that DNPH is a suitable reagent. Many internal standards having similar core structure which is derivatized with DNPH were also investigated by the same manner [19]. Prednisolone was found to be the most suitable ketosteroid and its derivatized solid product was prepared. It was eluted along with the MPA-DNPH and was used as I.S. throughout the experiment. Effects of derivatization time and concentration of DNPH for MPA analysis were studied. The DNPH concentrations used were 12.5, 25, 75 and 100 mg/mL. MPA (100 ng/mL) was used to react with various concentrations of DNPH in the ratio of 1 : 1 at room temperature and their reaction times were also done up to 180 min (Fig. 3). It was found that 75 mg/mL of DNPH gave the highest peak area within 120 min derivatization time. In this study, the derivatization time of 30 min wasenough, since the signal output was not significantly different from that of 120 min derivatization. The effect of temperature for 30 min derivatization was also investigated using 100 ng/mL MPA and 75 mg/mL DNPH in acidic solution. There seemed to be no effect of temperature, except a linear increase of peak area when the reaction was performed at room temperature (Fig. 4). Thus, the reaction was rapidly completed at room temperature.
Concerning on RP-HPLC-AD analysis, the optimization conditions of the HPLC-AD for MPA-DNPH analysis were then investigated. Effects of pH (3)(4)(5)(6) and buffer concentration (10-40 mM) on mobile phase (ACN : phosphate buffer) were carried out. It was found that the optimum buffering system was at pH 3.0 and 30 mM phosphate buffer (data not shown). In addition, the ratios of organic solvents were investigated and a good separation was obtained with the mobile phase of ACN : MeOH : 30 mM phosphate buffer (pH 3.0) at the ratio of 39 : 39 : 22 by volume. Figure 5 shows the hydrodynamic voltammogram (HDV) of MPA-and P-DNPH derivatives. The optimum potential that gave the highest peak area was around 0.8-0.9 V for both derivatized products, and 0.85 V was used in this study.
For method validation, the method limits for detection and quantitation were found to be 0.2 and 1.0 ng/mL with lower 15% precision and 80 -120% accuracy, respectively. In Table 1, the results for intra-day and inter-day precision were 5.05 and 11.5 %RSD, respectively. The intra-day accuracy ranged from 102.2 -109.5%. The average recovery was found to be 102.9 + 4.4%. It was found that this electrochemical detection was considerably selective and gave a working range of 1.0 -10.0 ng/mL with linear regression: y = (3.17 + 0.12) x -(0.67 + 0.03), r 2 = 0.9985 for n = 5 with %RSD = 2.34 -7.96.   Since the method sensitivity of MPA analysis was the main objective of this study, it has been shown that the clean-up step, using the SPE method for MPA spiked in human plasma sample, was necessary probably due to both the reagent used and the sample interference matrices (Fig. 6, a-d). The solution of MPA-DNPH derivative products with excess DNPH was not clean enough, so it must be added through the Sep-pak cartridge before being separated on the HPLC-AD system. Without the clean-up step, there might be trouble for the HPLC column used, which may affect the analytical sensitivity as well. This is the first report of the use of DNPH as an electroactive labeling reagent for MPA analysis. Development and validation of the HPLC-AD of the derivatized product were then carried out. The MPA-DNPH was completely separated from the I.S. within a suitable analysis time (Fig. 6, a-d).  The plasma sample size used in this study was still rather high (2 mL) for clinical aspects when compared with other reports [11,13,14], giving LOD and LOQ of 0.2 and 1.0 ng/mL, respectively. Thus, the developed method with electrochemical detection has sufficiently high sensitivity when compared with other methods, except for the tandem MS, as shown in