Abstract
A rapid and sensitive liquid chromatography–tandem mass spectrometry method was developed and validated for the quantification of adefovir (PMEA,9-(2-phosphonylmethoxyethyl) adenine) concentration in human serum and urine. The analysis was performed on a negative ionization electrospray mass spectrometer via multiple reaction monitoring. The monitored transitions were set at m/z 272.0 → 134.0 and m/z 276.0 → 149.8 for PMEA and internal standard, respectively. After protein precipitation, samples were separated by high-performance liquid chromatography on a reversed-phase Dikma Diamonsil C18 (250 × 4.6 mm; 5 µm) column with a mobile phase of 0.1 mM ammonium formate buffer–methanol. The calibration curves were linear over the serum concentration range 0.5–1,000 ng/mL and urine concentration range 2.0–1,000 ng/mL. The intra- and interday precision values of PMEA in both serum and urine were lower than 18.16% for low quality control and 13.70% for medium and high quality control. The accuracy, recovery, matrix factor and stability were also within the acceptable limits. The developed method was successfully applied to the pharmacokinetic study of following oral administration of single dose of pradefovir mesylate (10, 30, 60, 90 and 120 mg) and adefovir dipivoxil (10 mg) to healthy Chinese volunteers.
Introduction
Pradefovir mesylate is designed as an oral target prodrug of 9-(2-phosphonylmethoxyethyl) adenine nucleotide analog adefovir (or adefovir, PMEA), containing a cyclic phosphonate diester linkage (see Figure 1). Pradefovir employed the HepDirect™ technology and could be activated to PMEA by CYP3A4, which exists primarily in liver cells (1–3). However, another prodrug of adefovir, the broad-spectrum antiviral anti-Hepatitis B virus (HBV) agent, adefovir dipivoxil, converted to adefovir by widespread cellular esterases (see Figure 1). It has been manifested that pradefovir has a better liver targeting property than adefovir dipivoxil in rats and cynomolgus monkeys (4). Pradefovir mesylate yielded 15 times higher concentrations of radioactivity in the liver than adefovir dipivoxil, but only one-third of the concentrations in the kidney after oral administration to rats. In addition, after oral dosing the same radiolabeled agents in cynomolgus monkeys, pradefovir mesylate yielded 60 times higher levels of total radioactivity in the liver, but only two-thirds of total radioactivity levels in the kidney than adefovir dipivoxil (4). Since the liver is the target organ for HBV infection and the kidney is the site of toxicity, pradefovir is expected to have better efficacy and lower toxicity compared with adefovir dipivoxil.
Figure 1.
Chemical structures of pradefovir mesylate (A), adefovir dipivoxil (B) and PMEA (C).
Pradefovir and PMEA are the main circulatory substrate after oral dosing of pradefovir mesylate. So far, only one study reported the method of simultaneous determination of pradefovir and PMEA after solid-phase extraction (SPE) from human serum and urine by liquid chromatography–tandem mass spectrometry (LC–MS-MS) under positive and negative ions with switch scanning technology (5). But the authors have not published their method validation data. We tried to develop a method for the determination of pradefovir and PMEA simultaneously in positive ionization mode. However, PMEA produced a much lower intensity in positive ionization mode than pradefovir that could not reach the enough lower limit of quantification (LLOQ) for clinical pharmacokinetic trial, while in negative ionization mode, PMEA produced a higher intensity. Hence, PMEA was analyted in negative ionization mode and was detected separately with pradefovir.
The determination of PMEA using high-performance liquid chromatography (HPLC) with ultraviolet detection and HPLC with fluorescence detection has been proved to be too complicated and/or not sensitive enough for the clinical pharmacokinetic study (6–9). So far, several methods have been reported for the assay of PMEA using LC–MS-MS (10–15). Liu et al. (10) reported a LC–MS-MS method that gave a satisfactory sensitivity with LLOQ of 0.10 ng/mL, whereas the serum samples were treated by a complex SPE procedure. Xie et al. (14) developed a LC–MS-MS method with the long analysis time of 14 min. Furthermore, peaks of analytes in the above-mentioned methods were not perfect enough due to their relatively broad shapes. The purpose of our research was to develop and validate a quantitative LC–MS-MS method for the analysis of PMEA with simple pretreatment process and perfect peak shape in human serum and urine for clinical pharmacokinetic studies of pradefovir mesylate.
Experimental
Chemicals and reagents
Pradefovir mesylate (purity: 99.54%) and PMEA (purity: 94.2%) were obtained from Xi'an New Drug Research Co., Ltd (Xi'an, China). Entecavir (purity: 99.9%) as internal standard (IS) was obtained from Wuhan Belka Biological Pharmaceutical Co., Ltd (Wuhan, China). HPLC-grade methanol and acetonitrile were purchased from Merck (Darmstadt, Germany). Formic acid was purchased from ROE (Newark, New Castle, USA). Ultrapure water was obtained from a UPH Ultrapure Water System (UPH-II-5T, Chengdu, China). Ammonium formate, trichloroacetic acid and other reagents were of analytical grade.
Instruments
Chromatographic analysis was performed on a Shimadzu LC-20A series chromatographic system (Shimadzu Corporation UFLC XR, Kyoto, Japan) with two LC-20AD binary pumps, a DGU-20A3 degasser, a SIL20AC autosampler and a CTO-20A column oven. The separation was performed on a Dikma Diamonsil C18 (250 × 4.6 mm; 5 µm) column, at a temperature of 30°C. The mass spectrometer was operated on an API 4000 triple-quadrupole mass spectrometer equipped with an electrospray ionization (ESI) source (AB Sciex, Foster City, CA, USA).
LC–MS-MS analytical conditions
The mobile phase was composed of 0.1 mM ammonium formate buffer in water (pH 2.5 adjusted with formic acid, mobile phase A) and methanol (mobile phase B) at a flow rate of 1.0 mL/min. The gradient elution program was as follows: 0–2.0 min, 5% B, 2.0–2.5 min, 5–30% B, 2.5–6.0 min, 30% B, 6.0–6.1 min, 30–5% B, 6.1–8.0 min, 5% B. The eluate from the column was diverted into the MS from 5.0 to 7.5 min. The total analytical run time per sample, including equilibration time, was 8.0 min. The autosampler was conditioned at 4°C and the injection volume was 20 µL for analysis.
The mass spectrometer was operated in negative ESI mode with multiple reaction monitoring (MRM), monitoring the transition m/z 272.0 to m/z 134.0 for PMEA and m/z 276.0 to m/z 149.8 for IS. The MS conditions were as follows: curtain gas, 20 psi; source temperature, 550°C; dissociation gas, 6 psi; ion source gas 1, 60 psi; ion source gas 2, 50 psi; ion spray voltage, −4,200 V; declustering potential, −73 V; collision energy, −26 V; entrance potential, −10 V and collision cell exit potential, −10 V. All data were acquired and processed using Analyst 1.5.2 version software (AB Sciex).
Preparation of the stock and working solutions
Primary stock solutions of PMEA were prepared in purified water at a concentration of 1.0 mg/mL. A series of working standard solutions (5, 10, 20, 50, 100, 200, 500, 1,000, 2,000, 5,000 and 10,000 ng/mL) and quality control (QC) samples (10, 1,000 and 8,000 ng/mL for serum samples, and 50, 500 and 8,000 ng/mL for urine samples) were obtained by further diluting the stock solutions in methanol/water (1 : 1, v/v). Stock solutions of IS at 1.0 mg/mL were prepared in methanol/water (1 : 1, v/v). The IS working solution (10 µg/mL) was obtained by diluting the stock solution in methanol/water (1 : 1, v/v). All the solutions were stored at 4°C and were brought to room temperature before use.
Preparation of calibration samples and QC samples
The calibration serum samples were prepared by spiking 30 µL of the standard solutions into 270 µL of blank human serum to the final concentrations of 0.5, 1.0, 2.0, 5.0, 20, 50, 200, 500 and 1,000 ng/mL. QC samples (low, medium and high concentration) at 1, 100 and 800 ng/mL were prepared in the same way as the calibration samples and stored at −20°C.
The calibration urine samples were prepared by spiking 10 µL of the standard solutions into 90 µL of blank urine (100 µL blank urine and 900 µL purified water were mixed and vortexed for 3 min to achieve 10-fold dilute urine) to the final concentrations of 2.0, 10, 20, 100, 200 and 1,000 ng/mL. QC samples were prepared in a similar manner at the concentrations of 5, 80 and 800 ng/mL.
Sample preparation
For serum samples, 10 µL IS (10 µg/mL) working solutions were added into a 2.0-mL centrifuge tube. After evaporation to dryness, an aliquot of 300 µL human serum was mixed and vortexed for 30 s. Then, 900 µL methanol was added and the mixture was vortexed for 3 min. These samples were centrifuged at 12,000 r.p.m. for 6 min. The supernatant was transferred, evaporated to dryness under a gentle stream of nitrogen gas. The residue was reconstituted with 100 µL 5% (w/v) trichloroacetic acid by vortexing for 2 min and then transferred to another clean 1.5 mL centrifuge tube. After further centrifugation at 12,000 r.p.m. for 6 min, 20 µL supernatant was subjected to LC–MS-MS for analysis.
For urine samples, 10 µL IS (10 µg/mL) working solutions were added into a 2.0-mL centrifuge tube. After evaporation to dryness, an aliquot of 100 µL dilute human urine (100 µL urine and 900 µL purified water were mixed and vortexed for 3 min to achieve 10-fold dilute urine) was mixed and vortexed for 30 s. Then, 100 µL trichloroacetic acid 10% (w/v) was added and the mixture was vortexed for 2 min. These samples were centrifuged at 12,000 r.p.m. for 6 min. An aliquot of 20 µL of the supernatant was injected into the LC–MS-MS system for analysis.
Method validation
The method validation assays were performed according to the United States Food and Drug Administration guidelines (16).
Selectivity
Selectivity was investigated by comparing the chromatograms of six different batches of blank human serum or urine with the corresponding samples spiked with PMEA, as well as samples collected from subjects to exclude interference of endogenous substances and metabolites.
Linearity and LLOQ
The calibration curves for serum were performed with nine concentrations (0.5, 1.0, 2.0, 5.0, 20, 50, 200, 500 and 1,000 ng/mL). The calibration curves for urine were performed with six concentrations (2, 10, 20, 100, 200 and 1,000 ng/mL). The linearity of each calibration curve was determined by plotting the peak area ratio (y) of analytes to IS versus the nominal concentration (x) of analytes with weighted (1/x2) least square linear regression. The LLOQ was determined as the lowest concentration on the calibration curve at which accuracy (percentage relative error, RE %) within ±20% and a precision (percentage coefficient of variation, CV %) below 20%, and the signal-to-noise ratio was at least 10.
Precision and accuracy
The precision and accuracy of the method was evaluated by analyzing serum and urine QC samples, prepared separately from calibration standards at serum concentrations of 1, 100 and 800 ng/mL and urine concentrations of 5, 80 and 800 ng/mL, respectively. The intraday precision and accuracy of the method were assessed by determining the QC samples five times on a single day, and the interday precision and accuracy were estimated by determining the QC samples over three consecutive days. CV and RE were used to express the precision and accuracy, respectively.
Recovery and matrix factor
Extraction recovery was determined by comparing the peak areas obtained from serum or urine samples with the analytes spiked before and after extraction. The matrix factor was evaluated by dividing the peak area ratio of serum or urine samples spiked after extraction by an equal peak area ratio corresponding to the concentration in neat solution.
Stability
Stability was determined by analyzing triplicate of spiked samples at each QC level under different conditions. Short-term stability study was performed by analyzing QC samples at the room temperature for 4 h. For freeze and thaw stability, QC samples were subjected to three freeze–thaw cycles (−80°C to room temperature). Long-term stability was assessed by analyzing QC samples stored at −80°C for 153 days. Postpreparative storage stability was assessed by analyzing QC samples left in autosampler vials at 4°C for 24 h. The stability of stock solutions was confirmed by comparing the peak area from stock solutions stored at 4°C for 15 days and at the room temperature (24°C) for 3 h with freshly prepared stock solution.
Carry over
Carry over was tested by injecting two processed blank matrix samples sequentially after injecting an upper limit of quantification (ULOQ) sample.
Dilution integrity
Dilution integrity experiment was performed for study sample concentrations crossing the ULOQ (the highest standard of the calibration curve). Urine samples at the concentration of 80 µg/mL were diluted 10-fold with human blank urine and purified water to obtain the final test concentrations of 8 µg/mL (n = 6), respectively, and then analyzed by LC–MS-MS. Accuracy and precision should be within ±15%.
Application to pharmacokinetic study
The validated LC−MS-MS method was applied to investigate the clinical pharmacokinetic profiles of PMEA after oral administration single dose of pradefovir mesylate (10, 30, 60, 90 and 120 mg) and adefovir dipivoxil (ADV) (10 mg) to healthy Chinese volunteers. Fifty healthy volunteers were enrolled in this study, and randomly divided into five groups. Subjects in each group balanced by gender and body mass index, of which six received pradefovir mesylate, two received 10 mg adefovir dipivoxil and two received placebo. Whole blood samples (6–12 mL) were collected by intravenous catheter at ∼0 h (predosing) and 15 min, 30 min, 45 min, 1 h, 1.5, 2, 3, 4, 6, 8, 12, 16, 24, 36, 48, 72, 96 and 120 h postdosing. The samples were clot at the room temperature for 30 min and centrifuged for 15 min at 3,500 r.p.m. Then upper layer was transferred into a labeled tube. Serum samples were stored at −70°C until analyzed. Urine samples were collected at −12 to 0 h predose, and 0–6, 6–24, 24–48, 48–72, 72–96 and 96–120 h postdosing. The urine samples were transferred into labeled tubes and stored at −70°C until analyzed. The total volume of urine in each interval was recorded. The study protocol was approved by the First Hospital of Jilin University Ethics Committee (reference number: 140109-007), and conducted in accordance with the Declaration of Helsinki and the principles of Good Clinical Practice. Written informed consents were obtained from all subjects before the study.
Results
IS selection
The rationale for selecting IS is that basic structure of IS is similar to our target compound. We attempted tenofovir as IS at first, while the chromatographic peak was so bad and hardly separated from PMEA under the premise of optimizing the chromatographic condition. Then entecavir showed not only the perfect chromatographic peak but also good retention. In addition, entecavir could be separated from PMEA in samples. Based on above reasons, we selected entecavir as IS.
Method development
The optimum mobile phase was found to be 0.1 mM ammonium formate buffer in water (pH 2.5 adjusted with formic acid, mobile phase A) and methanol (mobile phase B) at a flow rate of 1.0 mL/min. The retention times of PMEA and IS in serum samples were 5.83 and 6.77 min, respectively. The retention times of PMEA and IS in urine samples were 5.77 and 6.89 min, respectively. The total analytical run time was 8.0 min.
ESI provided the optimum sensitivity for PMEA and IS in negative-ion mode. The negative product ion scan spectra of [M−H]− for PMEA and IS are shown in Figure 2. PMEA and IS gave protonated parent ion [M−H]− at m/z 272.0 and m/z 276.0, and the fragment ions of the most significant intensity were observed at m/z 134.0 for PMEA and m/z 149.8 for IS, respectively. So the mass transitions chosen for quantitation were m/z 272.0 → 134.0 for PMEA and m/z 276.0 → 149.8 for IS.
Figure 2.
Product ion mass spectra of [M–H]– ion from PMEA and IS.
Method validation
Selectivity
It was proved that under current conditions, no endogenous interferences were observed at the retention times of PMEA and IS. Typical chromatograms of blank serum, serum spiked with PMEA and IS (LLOQ) and subjects after oral administration of pardefovir mesylate are shown in Figure 3, and those of urine samples are shown in Figure 4.
Figure 3.
Representative MRM chromatograms for PMEA in human serum: (A) a blank serum sample; (B) a blank serum sample spiked with PMEA at the 20 ng/mL and (C) serum sample from a healthy volunteer after an oral administration of 10 mg pradefovir mesylate. Peak I: PMEA; Peak II: IS.
Figure 4.
Representative MRM chromatograms for PMEA in human urine: (A) a blank serum sample; (B) a blank urine sample spiked with PMEA at the 20 ng/mL and (C) urine sample from a healthy volunteer after an oral administration of 10 mg pradefovir mesylate. Peak I: PMEA; Peak II: IS.
Linearity and LLOQ
Calibration curves for serum were linear over the concentration range of 0.5–1,000 ng/mL for PMEA with a typical calibration curve equation of y = 0.0554x + 0.00413 (R2 = 0.9992), where y represents the peak area ratios of PMEA to the IS and x represents the serum concentrations of PMEA. Calibration curves for urine were linear over the concentration range of 2.0–1,000 ng/mL with a typical calibration curve equation of y = 0.0178x − 0.000475 (R2 = 0.9994). The LLOQ of PMEA was 0.5 ng/mL in human serum samples and 2 ng/mL in human urine samples, respectively, with acceptable accuracy and precision.
Precision and accuracy
The precision and accuracy data for serum and urine samples are presented in Table I. The intra- and interday precision (CV) value for serum samples was <16.88% for low QC samples and 13.70% for medium and high QC samples, and the accuracy (RE) ranged from −6.88 to 1.16%. The intra- and interday precision (CV) value for urine samples was <18.16% for low QC and 9.14% for medium and high QC samples, and the accuracy (RE) ranged from −1.23 to 5.0%. The data indicated that the accuracy and precision of the method were satisfactory.
Table I.
Intra-Batch and Inter-Batch Precision and Accuracy for PMEA in Serum and Urine
| Matrix | Spiked concentration (ng/mL) | Mean concentration found (ng/mL) | Accuracy (%) | CV (%) |
|
|---|---|---|---|---|---|
| Intraday | Interday | ||||
| Serum | 1 | 0.96 | 96.00 | 16.88 | 6.21 |
| 100 | 101.16 | 101.16 | 13.7 | 4.91 | |
| 800 | 745.00 | 93.13 | 12.42 | 5.01 | |
| Urine | 5 | 5.07 | 101.40 | 18.16 | 4.63 |
| 80 | 80.42 | 100.53 | 9.14 | 5.14 | |
| 800 | 790.13 | 98.77 | 8.14 | 3.36 | |
CV, coefficient of variation.
Recovery and matrix factor
The results of extraction recovery and matrix factor are shown in Table II. The extraction recovery of PMEA and IS in serum ranged from 63.09 to 76.31% and 92.07 to 94.75%. The extraction recovery of PMEA and IS in urine ranged from 72.41 to 85.66% and 96.90 to 102.11%. The extraction recovery CV (%) in both serum and urine was <9.6%, which showed that the extraction procedure for PMEA and IS was consistent and reproducible. The matrix factor of PMEA and IS in serum ranged from 0.53 to 0.61 and 0.89 to 0.93. The matrix factor of PMEA and IS in urine ranged from 0.83 to 0.88 and 0.97 to 0.99, respectively. Both of the matrix factor CV (%) in serum and urine was <7.40%. Although there was suppression of ionization observed for samples, this suppression was constant for all QC samples. Therefore, the matrix effect was not a concern for the robustness of the assay.
Table II.
Recovery and Matrix Factor of PMEA and IS in Serum and Urine
| Matrix | Spiked concentration (ng/mL) | PMEA |
IS |
||
|---|---|---|---|---|---|
| Recovery (%) | Matrix factor | Recovery (%) | Matrix factor | ||
| Serum | 1 | 63.09 | 0.61 | 92.07 | 0.89 |
| 100 | 76.31 | 0.53 | 94.75 | 0.93 | |
| 800 | 69.34 | 0.55 | 93.50 | 0.90 | |
| Urine | 5 | 85.1 | 0.88 | 97.07 | 0.99 |
| 80 | 72.41 | 0.83 | 102.11 | 0.97 | |
| 800 | 85.66 | 0.83 | 96.90 | 0.99 | |
Stability
Results of short-term stability, freeze and thaw stability, long-term stability and postpreparative stability are summarized in Table III. It was demonstrated that the stability offered by this method was satisfactory with the accuracy and RE for all samples, suggesting that this analytical method was applicable for routine analysis. PMEA in the stock solution was also stable stored at 4°C for 15 days and at room temperature (24°C) for 3 h.
Table III.
Stability Data for PMEA in Human Serum and Urine Under Various Storage Conditions (n = 5)
| Matrix | Spiked concentration (ng/mL) | Short-term stability |
Three freeze–thaw stability |
Long-term stability |
Postpreparative stability |
||||
|---|---|---|---|---|---|---|---|---|---|
| (4 h, room temperature) |
(−80°C, to room temperature) |
(153 days, −80°C) |
(21 h, 4°C) |
||||||
| Accuracy (%) | RE (%) | Accuracy (%) | RE (%) | Accuracy (%) | RE (%) | Accuracy (%) | RE (%) | ||
| Serum | 1 | 103.10 | 3.1 | 96.23 | −3.77 | 116.00 | 16 | 107.67 | 7.67 |
| 100 | 106.67 | 6.67 | 95.00 | −5 | 111.67 | 11.67 | 101.33 | 1.33 | |
| 800 | 96.63 | −3.38 | 96.13 | −3.88 | 103.13 | 3.21 | 96.29 | −3.71 | |
| Urine | 5 | 87.00 | −12.93 | 114.4 | 14.47 | 91.00 | −8.93 | 91.60 | −8.47 |
| 80 | 95.75 | −4.25 | 99.63 | −0.13 | 104.5 | 4.5 | 100.25 | 0.25 | |
| 800 | 93.67 | −6.33 | 97.21 | −2.79 | 99.04 | −0.96 | 99.13 | −0.88 | |
RE, relative error.
Carry over
No peak was observed at the retention times of PMEA or IS in the chromatogram of a blank sample analyzed after the injection of ULOQ sample, indicating the absence of carry over.
Dilution integrity
The results of the dilution integrity showed that the accuracy and precision for 10 times diluted test samples were within the acceptance criteria of ±15% and both of the CV (%) were <2.06. In addition, there was no significant difference between the two dilute matrix.
Application to pharmacokinetic study
The validated LC–MS-MS method was successfully applied to investigate the clinical pharmacokinetic profiles of PMEA after oral administration of single dose of pradefovir mesylate (10, 30, 60, 90 and 120 mg) and adefovir dipivoxil (10 mg) to healthy Chinese volunteers. Calibration levels were sufficient to quantitate serum and urine samples obtained in the pharmacokinetic study. The mean serum concentration–time curves and cumulative urinary excretion–time curves of PMEA are shown in Figures 5 and 6, respectively. The pharmacokinetic parameters are presented in Table IV.
Figure 5.
Mean serum concentration–time profile of PMEA following oral single dose of 10, 30, 60, 90 and 120 mg pradefovir mesylate and 10 mg adefovir dipivoxil.
Figure 6.
Mean urine cumulative excretion–time curves of PMEA following oral single dose of 10, 30, 60, 90 and 120 mg pradefovir mesylate and 10 mg adefovir dipivoxil.
Table IV.
Pharmacokinetic Parameters for PMEA in Healthy Subjects After Oral Dosing 10, 30, 60, 90 and 120 mg Pradefovir and 10 mg ADV
| Parameter | ADV-derived PMEA | Pradefovir-derived PMEA |
|||||
|---|---|---|---|---|---|---|---|
| 10 mg (n = 10) | 10 mg (n = 6) | 30 mg (n = 6) | 60 mg (n = 6) | 90 mg (n = 6) | 120 mg (n = 6) | ||
| AUC48 h (ng h/mL) | Mean (CV %) | 252.34 (19.45) | 72.65 (38.88) | 172.95 (34.83) | 372.27 (41.32) | 666.63 (13.98) | 1095.48 (22.68) |
| AUC∞ (ng h/mL) | Mean (CV %) | 255.86 (21.07) | 77.65 (51.04) | 191.41 (35.90) | 444.35 (45.01) | 788.06 (14.43) | 1304.89 (24.79) |
| Cmax (ng/mL) | Mean (CV %) | 24.30 (32.50) | 18.10 (27.39) | 41.80 (33.37) | 106.83 (49.54) | 193.50 (42.26) | 312.33 (36.56) |
| Tmax (h) | Geometric mean (min, max) | 0.99 (0.5, 3) | 0.87 (0.75, 1) | 0.87 (0.5, 2) | 0.82 (0.75, 1) | 0.87 (0.5, 1.5) | 0.83 (0.75, 1) |
| T1/2 (h) | Mean (CV %) | 9.36 (33.21) | 13.86 (49.26) | 19.31 (46.00) | 24.77 (18.82) | 30.15 (12.30) | 33.93 (16.29) |
| Ae120 h (mg) | Mean (CV%) | 1.78 (43.82) | 0.26 (39.42) | 0.70 (52.70) | 2.10 (41.65) | 2.86 (26.47) | 8.38 (56.17) |
| CLR (L/h) | Mean (CV %) | 7.36 (53.48) | 3.81 (47.47) | 4.14 (63.61) | 4.99 (35.28) | 3.64 (24.83) | 6.34 (39.29) |
Cmax, maximum serum concentration; Tmax, time of Cmax; AUC48 h, area under the concentration–time curve from 0 to 48 h after dosing; AUC∞, area under the plasma concentration versus time curve from time 0 to infinity; T1/2, half-life; Ae120 h, amount excreted in urine between 0 and 120 h; CLR, renal clearance; CV %, percentage coefficient of variation.
Discussion
LC–MS-MS method development
As PMEA has amino and phosphonic acid groups in its structure, it could produce good mass spectrometric responses in both positive and negative ionization modes (10–15). At first thought, we tried to detect pradefovir and PMEA simultaneously in positive ionization mode through the same sample preparation. Although the concentration of pradefovir and PMEA was comparative after oral administration of pradefovir mesylate to human, PMEA produced a much lower intensity in positive ionization mode compared with pradefovir, whereas PMEA produced high intensity that could meet the LLOQ for clinical pharmacokinetic trial in negative ionization mode. Given these reasons, PMEA was detected in negative ionization mode and analyted separately with pradefovir in our study.
For optimizing the LC system, different C18 and HILIC columns were attempted using methanol/water/formic acid (20 : 80 : 0.30, v/v/v) as the mobile phase. Only Diamnosil C18 columns (250 × 4.6 mm; 5 µm) showed satisfying retention capability for the chromatographic separation. We tried methanol–water as the basic mobile phase component. The concentration and pH of the ammonium formate buffer were also investigated in the present study. The results showed that addition of 0.3% formic acid improved the response and 0.1 mM ammonium formate improved peak shape of PMEA. Probably acidic modifier (formic acid) and ammonium formate in the mobile phase enhanced ionization efficiency and avoided the peak tailing. Thus, a mixture of methanol–0.3% formic acid and 0.1 mM ammonium formate (pH 2.5) was finally adopted as the mobile phase.
As a prerequisite to better balance the peak shape, time-consuming analysis and the robustness of the method, reducing matrix effect become a challenge in developing a precise LC–MS-MS assay. For strong polar compounds, how to solve the separation between analyte and endogenous analogs is pivotal. To the best of our knowledge, many researchers adopted the mobile phase with large proportion of water under isocratic elution processes in low flow rate to minimize the matrix effects, but the results were with long running time (14–15), weak retention (11–13) and wide peak shape (11–15). Compared with the method (15) in above references, we adopted the same mode and similar pretreatment procedure, while the special gradient elute program in our method showed the shorter analysis time and good peak shape with higher column efficiency. Liu et al. (10) reported a gradient isocratic elution process with the serum samples treated by the SPE procedure that reduced the matrix effect significantly, but the process was too complicated. Li et al. (17) reported a pulse gradient method that started with 100% water in order to enrich the analytes in the column, then the elution was directly changed to 100% CH3CN. Consequently, analytes and a part of biological matrix were eluted from the column. Although this method was able to reduce the interference caused by the non-analog components, the matrix effect caused by endogenous analogs was not decreased completely. Based on these situations, exploring an appropriate mobile phase with the ability of weak elution for analyte and thus making endogenous analogs enriched in the column in order to achieve sufficient endogenous analogs–analyte separation is important. So we explored a gradient elution program. In the first stage, 5% organic phase (start proportion) was selected allowing a part of endogenous analogs be enriched in the column, while analyte and another part of endogenous analogs were eluted slowly. In the second stage, the elution was increased to 30% organic phase (elute proportion) allowing the analyte and another part of endogenous analogs to be eluted and separated. In the last stage, organic phase was decreased to the original proportion and the column was equilibrated to the starting condition. Although matrix effect in our method was not eliminated totally, it kept stable with CV values (<7.70%) in the acceptable limits. Furthermore, the total running time and endogenous analogs were better balanced with perfect capacity factor and peak shape, which were the prominent advantages in our study compared with other methods (11–15).
Extraction procedure optimization
For the highly polar property of PMEA, it is difficult to be extracted from serum with organic solvents. Circumvent ion suppression is also a major challenge when quantitating the concentration of PMEA in serum and urine using LC–MS-MS methods. SPE was applied for the sample extraction and preparation because this method could purify and concentrate the samples (10). However, the pretreatment procedures of these methods are time- and labor-consuming. In the dilution integrity, we diluted samples with blank urine and purified water, respectively. However, results showed that there was no significant difference between the two dilute matrix. In addition, considering the source of the purified water was more easily to be obtained than blank urine, we decided to use purified water instead of blank urine diluting samples as a part of pretreatment. A two-step protein precipitation with methanol and trichloroacetic acid in serum and one-step protein precipitation with trichloroacetic acid in urine were adopted in this study. Methanol was selected as protein precipitant due to its less ion suppression than acetonitrile. Compared with formic acid and methanol, trichloroacetic acid played a key role in improving the peak shape for PMEA and advanced precipitation efficiency.
Conclusion
In conclusion, a sensitive and selective LC–MS-MS method was developed and validated for the determination of PMEA in human serum and urine. The validation data demonstrated that the developed methods were sensitive, efficient and robust with high selectivity and acceptable precision and accuracy. The method provided good linearity with a low LLOQ (0.5 ng/mL in serum samples and 2 ng/mL in urine samples) and was successfully applied to clinical pharmacokinetic studies of PMEA after oral administration of single dose of pradefovir mesylate to healthy Chinese volunteers. And it could be easily extended to further clinical pharmacokinetic studies on PMEA.
References
- 1.Erion M.D., Reddy K.R., Boyer S.H., Matelich M.C., Gomez-Galeno J., Lemus R.H. et al. Design, synthesis, and characterization of a series of cytochrome P(450) 3A-activated prodrugs (HepDirect prodrugs) useful for targeting phosph(on)ate-based drugs to the liver; Journal of the American Chemical Society, (2004); 126: 5154–5163. [DOI] [PubMed] [Google Scholar]
- 2.Erion M.D., van Poelje P.D., Mackenna D.A., Colby T.J., Montag A.C., Fujitaki J.M. et al. Liver-targeted drug delivery using HepDirect prodrugs; Journal of Pharmacology and Experimental Therapeutics, (2005); 312: 554–560. [DOI] [PubMed] [Google Scholar]
- 3.Lin C.C., Fang C., Benetton S., Xu G.F., Yeh L.T.; Metabolic activation of pradefovir by CYP3A4 and its potential as an inhibitor or inducer; Antimicrobial Agents and Chemotherapy, (2006); 50: 2926–2931. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lin C.C., Yeh L.T., Vitarella D., Hong Z., Erion M.D.; Remofovir mesylate: a prodrug of PMEA with improved liver-targeting and safety in rats and monkeys; Antiviral Chemistry & Chemotherapy, (2004); 15: 307–317. [DOI] [PubMed] [Google Scholar]
- 5.Lin C.C., Xu C., Teng A., Yeh L.T., Peterson J.; Pharmacokinetics of pradefovir and PMEA in healthy volunteers after oral dosing of pradefovir; Journal of Clinical Pharmacology, (2005); 45: 1250–1258. [DOI] [PubMed] [Google Scholar]
- 6.Vavrova K., Lorencova K., Klimentova J., Novotny J., Hrabalek A.; HPLC method for determination of in vitro delivery through and into porcine skin of adefovir (PMEA); Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, (2007); 853: 198–203. [DOI] [PubMed] [Google Scholar]
- 7.Hughes W.T., Shenep J.L., Rodman J.H., Fridland A., Willoughby R., Blanchard S. et al. Single-dose pharmacokinetics and safety of the oral antiviral compound adefovir dipivoxil in children infected with human immunodeficiency virus type 1. The Pediatrics AIDS Clinical Trials Group; Antimicrobial Agents and Chemotherapy, (2000); 44: 1041–1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Shaw J.P., Louie M.S., Krishnamurthy V.V., Arimilli M.N., Jones R.J., Bidgood A.M. et al. Pharmacokinetics and metabolism of selected prodrugs of PMEA in rats; Drug Metabolism and Disposition, (1997); 25: 362–366. [PubMed] [Google Scholar]
- 9.Sparidans R.W., Veldkamp A., Hoetelmans R.M., Beijnen J.H.; Improved and simplified liquid chromatographic assay for adefovir, a novel antiviral drug, in human plasma using derivatization with chloroacetaldehyde; Journal of Chromatography B: Biomedical Sciences and Applications, (1999); 736: 115–121. [DOI] [PubMed] [Google Scholar]
- 10.Liu Y., Xu G., Xu C., Garcia L., Lin C.C., Yeh L.T.; Ultra sensitive method for the determination of 9-(2-phosphonylmethoxyethyl)adenine in human serum by liquid chromatography-tandem mass spectrometry; Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, (2004); 803: 293–298. [DOI] [PubMed] [Google Scholar]
- 11.Chen X., Liu D., Zhu L., Zhong D.; Development and validation of a liquid chromatography/tandem mass spectrometry procedure for the quantification of adefovir in human plasma; Rapid Communications in Mass Spectrometry, (2005); 19: 1893–1898. [DOI] [PubMed] [Google Scholar]
- 12.Bi H.C., Zhong G.P., Zhou S., Chen X., Huang M.; Determination of adefovir in human plasma by liquid chromatography/tandem mass spectrometry: application to a pharmacokinetic study; Rapid Communications in Mass Spectrometry, (2005); 19: 2911–2917. [DOI] [PubMed] [Google Scholar]
- 13.Sun D., Wang H., Wang B., Guo R.; Development and validation of a sensitive LC-MS/MS method for the determination of adefovir in human serum and urine; Journal of Pharmaceutical and Biomedical Analysis, (2006); 42: 372–378. [DOI] [PubMed] [Google Scholar]
- 14.Xie H.-T., Wang G.-J., Xu M.-J., Jia Y.-W., Li H., Sun J.-G. et al. A new LC–MS–MS method for quantitative analysis of adefovir, and its use for pharmacokinetic studies in healthy Chinese volunteers; Chromatographia, (2010); 71: 587–593. [Google Scholar]
- 15.Zou J., Di B., Zhang J., Dai L., Ding L., Zhu Y. et al. Determination of adefovir by LC-ESI-MS-MS and its application to a pharmacokinetic study in healthy Chinese volunteers; Journal of Chromatographic Science, (2009); 47: 889–894. [DOI] [PubMed] [Google Scholar]
- 16.U.S. Food And Drug Administration, http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm368107.pdf (accessed September 2013).
- 17.Li Y., Sun Y., Du F., Yuan K., Li C.; Pulse gradient, large-volume injection, high-throughput ultra-performance liquid chromatographic/tandem mass spectrometry bioanalysis for measurement of plasma amrubicin and its metabolite amrubicinol; Journal of Chromatography A, (2008); 1193: 109–116. [DOI] [PubMed] [Google Scholar]