- Email: [email protected]
Hydrophilic interaction chromatography combined with tandem mass spectrometry method for the quantification of tobramycin in human plasma and its application in a pharmacokinetic study
Journal of Chromatography B, 973 (2014) 39–44
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Hydrophilic interaction chromatography combined with tandem mass spectrometry method for the quantification of tobramycin in human plasma and its application in a pharmacokinetic study Lingyun Chen a , Huashan Chen b , Mei Shen a,∗ a b
Hygiene Detection Center, School of Public Health and Tropical Medicine, Southern Medical University, Guangzhou 510515, China State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Science, Guangzhou 510640, China
a r t i c l e
i n f o
Article history: Received 30 June 2014 Accepted 5 October 2014 Available online 12 October 2014 Keywords: Tobramycin HILIC-MS/MS Human plasma Tandem mass spectrometry Pharmacokinetics
a b s t r a c t A fast, sensitive and specific hydrophilic interaction chromatography combined with tandem mass spectrometry (HILIC-MS/MS) method was developed for the determination of tobramycin in human plasma. With sisomycin as internal standard, the analysis was carried out on a hilic column (150 mm × 2.1 mm, 3.5 m) using a mobile phase consisting of acetonitrile: 5 mM ammonium acetate and 0.1% formic acid (60:40, v/v). The detection was performed by tandem spectrometry via electrospray ionization (ESI). Linear calibration curves were obtained in the concentration range of 10.51–1051 ng/mL for tobramycin, with a lower limit of quantification of 10.51 ng/mL. The intra- and inter-day precision (RSD) values were below 15% and accuracy (RE) was 1.3–5.7% at all QC levels. The method was applicable to the clinical study of the pharmacokinetics of tobramycin in healthy volunteers. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Tobramycin is a broad spectrum aminoglycoside antibiotic and widely used for the treatment of infections caused by different bacteria. It is particularly useful for the treatment of P. aeruginosa in patients with cystic fibrosis. It has a narrow therapeutic range and monitoring of the drug is required to reduce serious side effects such as nephro and ototoxicity [1–4]. Dosage alterations based on the results of drug monitoring have been found to improve efficacy and minimize toxicity [5]. Analysis of tobramycin is quite challenging due to its physicochemical properties. The lack of a UV chromophore makes direct UV detection unfeasible. Therefore, traditionally, analysis of aminoglycosides has been performed using derivatization with ultraviolet, fluorescent or electrochemical detection [6–14]. However, derivatization methods require complicated sample preparation procedures. Moreover, the sensitivity of these techniques is relatively low. These may not meet the requirement of desired throughput speed and sensitivity in biosample analysis. An alternative for the detection of aminoglycosides is mass
∗ Corresponding author. Tel.: +86 20 6164 8564; fax: +86 20 6164 8966. E-mail address: [email protected] (L. Chen). http://dx.doi.org/10.1016/j.jchromb.2014.10.007 1570-0232/© 2014 Elsevier B.V. All rights reserved.
spectrometry with high selectivity and specificity. Thus, LC–MS [15] and LC–MS/MS [16,17] were used in the determination of tobramycin in biosamples with lower limit of quantification (LLOQ) of 200 ng/mL [15], 50 ng/mL [16], 150 ng/mL [17] and 100 ng/mL [18]. However, aminoglycosides are extremely hydrophilic compounds. Due to their highly polar characteristics, the use of simple chromatographic methods for separation are not applicable: aminoglycosides are positively charged at the pH range employed in reversed-phase HPLC and are not retained on conventional C18 bonded silica columns without an ion pairing reagent [16,18]. Ion-pair chromatography [17] has been described to prolong the retention of aminoglycosides. But electrospray ionization (ESI) MS detection of ion pairs is not ideal because the sensitivity of mass spectrometry will be reduced (due to suppression of ionization). An alternative for separation of hydrophilic compounds is hydrophilic interaction chromatography (HILIC). HILIC is a type of liquid chromatography that allows high-resolution separation of highly polar compounds. Only one HILIC-MS/MS method was used in the determination of tobramycin in serum [19]. However, this method needs a time-consuming and expensive sample – extraction. Moreover, this method has a long analysis time (10 min) and a high LLOQ (100 ng/mL). This paper describes a fast, selective and highly sensitive approach, which enables the determination of tobramycin at 10 ng/mL in plasma with good accuracy using hydrophilic
40
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
interaction chromatography combined with tandem mass spectrometry (HILIC-MS/MS) method. The total analysis time was 3.5 min. 2. Experimental 2.1. Reagents and chemicals Tobramycin reference standard (99.2% of purity) and sisomycin (99.3% of purity) (Fig. 1) were obtained from the National Institute for Control of Pharmaceutical and Biological Products (Beijing, PR China). Acetonitrile, ammonium acetate (HPLC grade) and formic acid (HPLC grade) were purchased from Dikma (11 Orchard Road, Suite 106, Lake Forest, CA 92630, USA). All other chemicals were of analytical grade and purchased from Guangzhou chemical reagent factory (Guangzhou, China). Water was purified by redistillation and filtered through 0.22 m membrane filter before use (Grade 1 water). 2.2. Apparatus and operation conditions 2.2.1. Liquid chromatography The chromatography was performed on Agilent 1200 system with autosampler and column oven enabling temperature control of analytical column. The Inertsil HILIC column (150 mm × 2.1 mm, 3.5 m) was employed. The column temperature was maintained at room temperature. The mobile phase consisted of acetonitrile: 5 mM ammonium acetate and 0.1% formic acid (60:40, v/v) at an isocratic flow rate of 0.30 mL/min. The injection volume was 5 L. 2.2.2. Mass spectrometry Detection was performed on a Sciex API 4000 Qtrap MS system equipped with a Turbo Ionspray interface. Mass spectral setting to operate in positive-ion mode (ESI+) were: ion source voltage: 5000 V; ion source temperature: 400 ◦ C; collision gas (N2 ): medium; curtain gas: 20 psi; nebulizer gas: 40 psi; auxiliary gas: 60 psi. Quantification was performed using multiple reaction monitoring (MRM) of the transitions of m/z 468.2 → m/z 163.0 for tobramycin and m/z 448.2 → m/z 160.0 for sisomycin, respectively. The product ion spectra of tobramycin and sisomycin are shown in Fig. 2. Data acquisition and processing were performed with the Analyst software 1.4.2.
Fig. 2. Product ion spectra of tobramycin (1) and sisomycin (2).
2.3. Preparation of standards and quality control samples Standard stock solutions of tobramycin and sisomycin were both prepared in water at the concentration of 105.1 g/mL and 112.1 g/mL. Then the stock solutions were serially diluted with methanol: water (1:1) to provide working standard solutions of desired concentrations. All the solutions were stored at 4 ◦ C.
Fig. 1. Chemical structures of tobramycin (1) and sisomycin (I.S.) (2).
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
Calibration standards were prepared by spiking 0.2 mL of blank human plasma with 400 L working standard solutions of tobramycin. The effective concentrations in standard plasma samples were 10.51, 21.02, 52.55, 105.1, 210.2, 525.5, 1051 ng/mL for tobramycin. One calibration curve was constructed on each analysis day using freshly prepared calibration standards. The quality control samples (QCs) were prepared with blank plasma at LLOQ, low, middle and high concentrations of 10.51, 26.28, 105.1, 840.8 ng/mL for tobramycin. The standards and quality controls were extracted on each analysis day with the same procedure for plasma samples as described below. 2.4. Plasma sample preparation To a 0.2 mL aliquot of plasma sample in 1.5 mL centrifuge tube, 50 L of internal standard (1121 ng/mL) and 400 L methanol:water (1:1) were added. The mixture was vortex-mixed thoroughly for 1 min and then centrifuged at 13,000 r.p.m. for 10 min. The supernatant was directly injected into the HILICMS/MS system. 2.5. Method validation Validation runs were conducted on three consecutive days. Each validation run consisted of a minimum of one set of calibration standards and six replicates of LLOQ and QC plasma samples at three concentrations. The results from LLOQ and QC plasma samples in three runs were used to evaluate the precision and accuracy of the method developed. 2.5.1. Selectivity Selectivity was studied by comparing chromatograms of six different batches of blank plasma obtained from six subjects with those of corresponding standard plasma samples spiked with tobramycin and sisomycin (1121 ng/mL). 2.5.2. Linearity and lower limit of quantification (LLOQ) Calibration curves were prepared by assaying standard plasma samples at seven concentrations of tobramycin ranging 10.51–1051 ng/mL. The linearity of each calibration curve was determined by plotting the peak area ratio (y) of tobramycin to sisomycin versus the nominal concentration (x) of tobramycin. The calibration curves were constructed by weighted (1/x2 ) least square linear regression. The lower limit of quantification, defined as the lowest concentration on the calibration curve, was validated using an LLOQ sample for which an acceptable accuracy (relative error, RE) within ±20% and a precision (relative standard deviation, RSD) below 20% were obtained. 2.5.3. Precision and accuracy For determination of the intra-day accuracy and precision, a replicate analysis of QC plasma samples of tobramycin was performed on the same day. The run consisted of a calibration curve and six replicates of each LLOQ, low, mid and high concentration quality control samples. The inter-day accuracy and precision were assessed by analysis of three batches on different days. The precision was expressed as the relative standard deviation (RSD) and the accuracy as the relative error (RE). 2.5.4. Extraction recovery and matrix effect The recovery was calculated by comparing the peak area of the tobramycin added into blank plasma and extracted using the protein precipitation procedure with those obtained from the compound spiked into post-extraction supernatant at three QC concentration levels. The matrix effect was measured by comparing
41
the peak response of sample spiked post-extraction (A) with that of the standard solution containing equivalent amounts of the compound (B). The ratio (A/B × 100) % was used to evaluate the matrix effect. The extraction recovery and matrix effect of I.S. were also evaluated using the same method. 2.5.5. Stability 2.5.5.1. Freeze and thaw stability. The effect of three freeze and thaw cycles on the stability of plasma samples containing tobramycin was determined by subjecting five aliquots of QC samples at low, mid and high concentration unextracted quality control samples to three freeze–thaw cycles. After completion of the three cycles, the samples were analyzed and the experimental concentrations were compared with the nominal values. 2.5.5.2. Long-term stability. Five aliquots of QC samples at low, mid and high concentration unextracted QC samples were stored at −20 ◦ C for 20 days. Then, the samples were processed and analyzed and the concentrations obtained were compared with the nominal values. 2.5.5.3. Short-term stability. Five aliquots of QC samples at low, mid and high concentration unextracted QC samples were kept at ambient temperature (25 ◦ C) for 4 h in order to determine the short-term stability of tobramycin in human plasma. Then, the samples were processed and analyzed. The concentrations obtained were compared with the nominal values of QC samples. 2.5.5.4. Post-preparation stability. In order to estimate the stability of tobramycin in the prepared sample, five aliquots of QC samples at low, mid and high concentration were kept in an autosampler maintained at room temperature for about 8 h. Then, the samples were analyzed and the concentrations obtained were compared with the nominal values. 2.6. Pharmacokinetic study The method was successfully applied to determine the pharmacokinetic study of tobramycin in 10 healthy Chinese volunteers. Each subject was intramuscular injected 1 mg/kg tobramycin. The pharmacokinetic study was approved by the local Ethics Committee and all volunteers gave their signed informed consent to participate in the study according to the principles of the Declaration of Helsinki. Blood samples were collected before and 0.25, 0.5, 0.75, 1.0, 1.5, 2.0, 3.0, 4.0, 5.0, 7.0, 9.0 and 12 h post-dosing. Samples were centrifuged and plasma was separated and stored at −80 ◦ C until analyzed. The maximum plasma concentration (Cmax ) and their time were noted directly. The elimination rate constant (ke ) was calculated by linear regression of the terminal points of the semi-log plot of plasma concentration against time. Elimination half-life (t1/2 ) was calculated using the formula t1/2 = 0.693/ke . The area under the plasma concentration–time curve (AUC0–t ) to the last measurable plasma concentration (Ct ) was calculated by the linear trapezoidal rule. The area under the plasma concentration–time curve to time infinity (AUC0–∞ ) was calculated as: AUC0–∞ = AUC0-t + Ct /ke . 3. Results and discussion 3.1. Selection of IS The best internal standard in LC–MS assay is a deuterated form of the analyte. In our laboratory, no deuterated tobramycin was available. Therefore, a compound being structurally or chemically similar to the analyte was considered. In LC–MS/MS the I.S. should
42
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
also have similar chromatographic and mass spectrometric behaviors to the analyte, and mimic the analyte in any sample preparation steps. Sisomycin was chosen as the internal standard for the assay because of its similarity of structure, retention time and ionization to tobramycin.
3.2. Chromatography and mass spectrometry It can be seen that all of the highly polar contents could not be retained on C18 column and eluted in dead time. But when using hydrophilic interaction chromatography (HILIC) for analysis, the fraction had a good separation on the column. Thus, HILIC will be an effective technique to analyze the highly polar compounds studies. HILIC is such a technique, where analyte retention is believed to be caused by partitioning of the analyte between a water-enriched layer of stagnant eluent on a hydrophilic stationary phase and a relatively hydrophobic bulk eluent, with the main components usually being 5–40%water in ACN. The use of water as the strongly eluting solvent gives HILIC a number of advantages over conventional normal phase chromatography (NPC); eluent preparation is less complicated since the need of total control over the solvent water content is omitted. Normal phase (NP) eluents are also nonpolar (often based on hexane) and polar analytes usually have a low solubility in these eluents. The interfacing with electrospray MS is also a problem with NP, since ionization is not easily achieved in totally organic, nonpolar eluents. The elution order in HILIC is more or less the opposite of that seen in RP separations, which means that HILIC works best for solutes that are problematic in RP. Therefore hydrophilic interaction chromatography (HILIC) on a silica column was chosen for the separation. The ionization of the analyte was affected by the composition of mobile phase. 5 mM ammonium acetate and 0.1% formic acid were employed to supply the ionic strength. A mixture of 5 mM ammonium acetate and 0.1% formic acid–acetonitrile (40:60) was finally adopted as the mobile phase. The LLOQ for tobramycin was 10.51 ng/mL. Due to the lower injection volume of 5 L, the on column sensitivity in our study (the quantity of drug injected on the column per injection) was 52.55 pg for tobramycin. It was much lower than those reported in the literatures, which were 200 ng/mL [15], 50 ng/mL [16], 150 ng/mL [17] and 100 ng/mL [18,19], respectively. Since the elimination rate constant (ke) was calculated by linear regression of the terminal points of the semi-log plot of plasma concentration against time and elimination half-life (t1/2 ) was calculated using the formula t1/2 = 0.693/ke, the lower the LLOQ was, the more accurate t1/2 was calculated. The total run time was 3.5 min per sample, which was lower than previously reported analysis times (longer than 10 min) [15,19]. This short analysis time indicated that our method better met the requirement of high sample throughput in bioanalysis. HILIC-MS/MS operation parameters were carefully optimized for determination of tobramycin. The mass spectrometer was tuned in both positive and negative ionization modes with ESI for tobramycin. Both signal intensity and ratio of signal to noise obtained in positive ionization mode were much greater than those in negative ionization mode. In the precursor ion full-scan spectra, the most abundant ions were protonated molecules [M+H]+ m/z 468.2 and 448.2 for tobramycin and I.S., respectively. The product ion scan spectra showed high abundance fragment ions at m/z 163.0 and 160.0 for tobramycin and I.S., respectively. Multiple reaction monitoring (MRM) using the precursor → product ion transitions of m/z 468.2 → m/z 163.0 and m/z 448.2 → m/z 160.0 was employed for quantification of tobramycin and I.S. respectively.
Table 1 Precision and accuracy for the determination of tobramycin in human plasma (intraday: n = 6; inter-day: n = 6 series per day, 3 days). Concentrations (ng/mL)
R.S.D (%)
Added
Found
Intra-day
Inter-day
Relative error (%)
10.51 26.28 105.1 840.8
10.65 26.94 108.2 888.7
4.6 2.3 1.8 1.4
11.6 8.6 3.4 4.0
1.3 2.5 3.0 5.7
3.3. Sample preparation Liquid–liquid extraction (LLE) and solid-phase extraction (SPE) are techniques often used in the preparation of biological samples for their ability to improve the sensitivity and robustness of assay. But tobramycin had a very high polarity, so it was impossible to extract it from biological fluids using liquid–liquid extraction method and SPE is too expensive. In the present experiment, a simple protein precipitation procedure was developed to reduce sample preparation time. No further concentration procedure was needed and the sample preparation procedure was simplified. To obtain high extraction efficiency, three different protein precipitation agents, acetonitrile, methanol and acetone were investigated. It was found that acetonitrile had a higher efficiency of precipitating. High extraction efficiency achieved too when this procedure was applied to I.S. Finally, this simple single-step acetonitrile protein precipitation was adopted. The protein precipitation sample preparation procedure was much simpler and less expensive than a previously used SPE method in the literature [19]. 3.4. Method validation 3.4.1. Selectivity Selectivity was assessed by comparing the chromatograms of six different batched of blank human plasma with the corresponding spiked plasma. As shown in Fig. 3A, no interference from endogenous substance was observed at the retention time of tobramycin and I.S. 3.4.2. Linearity and LLOQ The standard calibration curve for tobramycin was linear over the concentration range of 10.51–1051 ng/mL (r2 > 0.99) by using weighted least square linear regression analysis with a weigh factor of 1/x2 . The typical equations for the calibration curves for metformin and rosiglitazone were: y = 1.680 × 10−4 x + 5.660 × 10−5 , r = 0.9963. The lower limit of quantification for tobramycin was 10.51 ng/mL with precision and accuracy presented in Table 1 with RE within ±20% and RSD lower than 20%. A corresponding chromatogram is given in Fig. 3B 3.4.3. Precision and accuracy The data of intra-day and inter-day precision and accuracy for the method are listed in Table 1. The intra-day and inter-day precisions for low, mid and high QC levels of tobramycin were below 15%, with the accuracy within 1.3–5.7%. The precision and accuracy of the present method conform to the criteria for the analysis of biological samples according to the guidance of USFDA where the precision (RSD) determined at each concentration level is required not exceeding 15% and accuracy (RE) within ±15% of the actual value. 3.4.4. Extraction recovery and matrix effect The extraction recoveries of tobramycin from human plasma at three QC levels were all above 80% and the mean extraction
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
43
Fig. 3. Representative MRM chromatograms of tobramycin (peak 1) and sisomycin (peak 2) in human plasma samples. (A) A blank plasma samples; (B) a blank plasma sample spiked with tobramycin at the LLOQ of tobramycin 10.51 ng/mL, and sisomycin (1121 ng/mL); (C) a blank plasma sample spiked with tobramycin at 105.1 ng/mL and sisomycin (1121 ng/mL); the retention times of tobramycin and sisomycin were 2.52 min and 2.22 min, respectively.
44
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
Table 2 Stability of tobramycin in plasma samples (n = 5). Stability
26.28
105.1
Mean ± SD Short-term stability Long-term stability Freeze–thaw stability Post-preparative stability
26.81 26.88 26.96 26.35
± ± ± ±
0.6 1.2 1.3 2.3
840.8
RE (%)
Mean ± SD
2.0 0.3 0.3 −2.3
103.3 103.0 104.3 99.5
recovery of tobramycin was 85.3 ± 2.4%. The mean extraction recovery of sisomycin was 87.9 ± 1.2%. In terms of matrix effect, all the ratios defined as in Section 2 were between 85% and 115%, which means no matrix effect for tobramycin in this method. 3.4.5. Stability The stock solution of tobramycin and sisomycin was found to be stable at room temperature for 4 h and at 4 ◦ C for 30 days. The results from all stability tests presented in Table 2 demonstrated a good stability of tobramycin and sisomycin over all steps of the determination. The method is therefore proved to be applicable for routine analysis. 3.5. Method application The present method was successfully applied to the pharmacokinetic study of tobramycin after intramuscular injection in healthy volunteers. After the administration of tobramycin (1 mg/kg), Cmax was 5.52 ± 1.23 g/mL. Tmax was 0.51 ± 0.55 h. Plasma concentration declined with the t1/2 of 2.1 ± 0.32 h. The mean (±SD) AUC0–∞ for tobramycin in plasma was 15.13 ± 3.51 g h mL−1 . 4. Conclusion A sensitive, selective and rapid HILIC-MS/MS method for the determination of tobramycin in human plasma is described. Compared with the published methods, the method used a new HILIC column offering superior sensitivity with the LLOQ of 10.51 ng/mL, satisfactory selectivity and short run time of 3.5 min. This method can be used for the pharmacokinetic study of tobramycin in human. References [1] H.C. Neu, C.L. Dendush, Ototoxicity of tobramycin: a clinical overview, J. Infect. Dis. 134 (1976) S206. [2] N.E. Plaut, J.J. Schentag, W.J. Jusko, Aminoglycoside nephrotoxicity: comparative assessment in critically ill patients, J. Med. 10 (1979) 257.
± ± ± ±
9.9 8.8 9.3 8.0
RE (%)
Mean ± SD
−1.7 −0.3 1.3 −4.6
884.4 810.4 844.4 814.4
± ± ± ±
41.6 44.1 55.6 50.1
RE (%) 5.2 −8.4 4.2 −3.6
[3] C.R. Smith, J.J. Lipsky, O.L. Laskin, D.B. Hellman, Double-blind comparison of the nephrotoxicity and auditory toxicity of gentamicin and tobramycin, New Engl. J. Med. 302 (1980) 1106. [4] R.D. Moore, C.R. Smith, J.J. Lipsky, E.D. Mellits, P.S. Lietman, Risk factors for nephrotoxicity in patients treated with aminoglycosides, Ann. Intern. Med. 100 (1984) 352. [5] P. Noone, D.F. Beale, S.S. Pollock, Monitoring aminoglycoside use in patients with severely impaired renal function, Br. Med. J. 2 (1978) 4670. [6] D.A. Stead, Current methodologies for the analysis of aminoglycosides, J. Chromatogr. B 747 (2000) 69. [7] N. Isoherranen, S. Soback, Determination of gentamicins C(1), C(1a), and C(2) in plasma and urine by HPLC, Clin. Chem. 46 (2000) 837. [8] D.M. Barends, C.L. Zwaan, A. Hulshoff, Micro-determination of tobramycin in serum by high-performance liquid chromatography with ultraviolet detection, J. Chromatogr. 225 (1981) 417. [9] S.E. Back, I. Nilsson-Ehle, P. Nilsson-Ehle, Chemical assay, involving liquid chromatography, for aminoglycoside antibiotics in serum, Clin. Chem. 25 (1979) 1222. [10] C.H. Feng, S.J. Lin, H.L. Wu, S.H. Chen, Trace analysis of tobramycin in human plasma by derivatization and high-performance liquid chromatography with ultraviolet detection, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 780 (2002) 349. [11] P.M. Kabra, P.K. Bhatnagar, M.A. Nelson, J.H. Wall, L.J. Marton, Liquidchromatographic determination of tobramycin in serum with spectrophotometric detection, Clin. Chem. 29 (1983) 672. [12] D.B. Haughey, D.M. Janicke, M. Adelman, J.J. Schentag, High-pressure liquid chromatography analysis and single-dose disposition of tobramycin in human volunteers, Antimicrob. Agents Chemother. 17 (1980) 649. [13] H. Fonge, E. Kaale, C. Govaerts, K. Desmet, A. vanSchepdael, J. Hoog-martens, Bioanalysis of tobramycin for therapeutic drug monitoring by solid-phase extraction and capillary zone electrophoresis, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 810 (2004) 313. [14] J.P. Anhalt, S.D. Brown, High-performance liquid-chromatographic assay of aminoglycoside antibiotics in serum, Clin. Chem. 24 (1978) 1940. [15] M.X. Guo, L. Wrisley, E. Maygoo, Measurement of tobramycin by reversedphase high-performance liquid chromatography with mass spectrometry detection, Anal. Chim. Acta 571 (2006) 12. [16] M.A. Jonge, J.M. Bekkers, H.O. Straaten, R.W. Sparidans, E. Franssen, Simple and sensitive method for quantification of low tobramycin concentrations in human plasma using HPLC-MS/MS, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 862 (2008) 257. [17] B.G. Keevil, S.J. Lockhart, D.P. Cooper, Determination of tobramycin in serum using liquid chromatography–tandem mass spectrometry and comparison with a fluorescence polarisation assay, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 794 (2003) 329. [18] C.R. Roc, R.G. Roberto, G.F. Antonia, R.M. Manuel Angel, M.V. Jos Luis, Simultaneous analysis of antibiotics in biological samples by ultra high performance liquid chromatography–tandem mass spectrometry, J. Pharm. Biomed. Anal. 89 (2014) 203. [19] R. Oertel, V. Neumeister, W. Kirch, Hydrophilic interaction chromatography combined with tandem-mass spectrometry to determine six aminoglycosides in serum, J. Chromatogr. A 1058 (2004) 197–201.
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Hydrophilic interaction chromatography combined with tandem mass spectrometry method for the quantification of tobramycin in human plasma and its application in a pharmacokinetic study Lingyun Chen a , Huashan Chen b , Mei Shen a,∗ a b
Hygiene Detection Center, School of Public Health and Tropical Medicine, Southern Medical University, Guangzhou 510515, China State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Science, Guangzhou 510640, China
a r t i c l e
i n f o
Article history: Received 30 June 2014 Accepted 5 October 2014 Available online 12 October 2014 Keywords: Tobramycin HILIC-MS/MS Human plasma Tandem mass spectrometry Pharmacokinetics
a b s t r a c t A fast, sensitive and specific hydrophilic interaction chromatography combined with tandem mass spectrometry (HILIC-MS/MS) method was developed for the determination of tobramycin in human plasma. With sisomycin as internal standard, the analysis was carried out on a hilic column (150 mm × 2.1 mm, 3.5 m) using a mobile phase consisting of acetonitrile: 5 mM ammonium acetate and 0.1% formic acid (60:40, v/v). The detection was performed by tandem spectrometry via electrospray ionization (ESI). Linear calibration curves were obtained in the concentration range of 10.51–1051 ng/mL for tobramycin, with a lower limit of quantification of 10.51 ng/mL. The intra- and inter-day precision (RSD) values were below 15% and accuracy (RE) was 1.3–5.7% at all QC levels. The method was applicable to the clinical study of the pharmacokinetics of tobramycin in healthy volunteers. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Tobramycin is a broad spectrum aminoglycoside antibiotic and widely used for the treatment of infections caused by different bacteria. It is particularly useful for the treatment of P. aeruginosa in patients with cystic fibrosis. It has a narrow therapeutic range and monitoring of the drug is required to reduce serious side effects such as nephro and ototoxicity [1–4]. Dosage alterations based on the results of drug monitoring have been found to improve efficacy and minimize toxicity [5]. Analysis of tobramycin is quite challenging due to its physicochemical properties. The lack of a UV chromophore makes direct UV detection unfeasible. Therefore, traditionally, analysis of aminoglycosides has been performed using derivatization with ultraviolet, fluorescent or electrochemical detection [6–14]. However, derivatization methods require complicated sample preparation procedures. Moreover, the sensitivity of these techniques is relatively low. These may not meet the requirement of desired throughput speed and sensitivity in biosample analysis. An alternative for the detection of aminoglycosides is mass
∗ Corresponding author. Tel.: +86 20 6164 8564; fax: +86 20 6164 8966. E-mail address: [email protected] (L. Chen). http://dx.doi.org/10.1016/j.jchromb.2014.10.007 1570-0232/© 2014 Elsevier B.V. All rights reserved.
spectrometry with high selectivity and specificity. Thus, LC–MS [15] and LC–MS/MS [16,17] were used in the determination of tobramycin in biosamples with lower limit of quantification (LLOQ) of 200 ng/mL [15], 50 ng/mL [16], 150 ng/mL [17] and 100 ng/mL [18]. However, aminoglycosides are extremely hydrophilic compounds. Due to their highly polar characteristics, the use of simple chromatographic methods for separation are not applicable: aminoglycosides are positively charged at the pH range employed in reversed-phase HPLC and are not retained on conventional C18 bonded silica columns without an ion pairing reagent [16,18]. Ion-pair chromatography [17] has been described to prolong the retention of aminoglycosides. But electrospray ionization (ESI) MS detection of ion pairs is not ideal because the sensitivity of mass spectrometry will be reduced (due to suppression of ionization). An alternative for separation of hydrophilic compounds is hydrophilic interaction chromatography (HILIC). HILIC is a type of liquid chromatography that allows high-resolution separation of highly polar compounds. Only one HILIC-MS/MS method was used in the determination of tobramycin in serum [19]. However, this method needs a time-consuming and expensive sample – extraction. Moreover, this method has a long analysis time (10 min) and a high LLOQ (100 ng/mL). This paper describes a fast, selective and highly sensitive approach, which enables the determination of tobramycin at 10 ng/mL in plasma with good accuracy using hydrophilic
40
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
interaction chromatography combined with tandem mass spectrometry (HILIC-MS/MS) method. The total analysis time was 3.5 min. 2. Experimental 2.1. Reagents and chemicals Tobramycin reference standard (99.2% of purity) and sisomycin (99.3% of purity) (Fig. 1) were obtained from the National Institute for Control of Pharmaceutical and Biological Products (Beijing, PR China). Acetonitrile, ammonium acetate (HPLC grade) and formic acid (HPLC grade) were purchased from Dikma (11 Orchard Road, Suite 106, Lake Forest, CA 92630, USA). All other chemicals were of analytical grade and purchased from Guangzhou chemical reagent factory (Guangzhou, China). Water was purified by redistillation and filtered through 0.22 m membrane filter before use (Grade 1 water). 2.2. Apparatus and operation conditions 2.2.1. Liquid chromatography The chromatography was performed on Agilent 1200 system with autosampler and column oven enabling temperature control of analytical column. The Inertsil HILIC column (150 mm × 2.1 mm, 3.5 m) was employed. The column temperature was maintained at room temperature. The mobile phase consisted of acetonitrile: 5 mM ammonium acetate and 0.1% formic acid (60:40, v/v) at an isocratic flow rate of 0.30 mL/min. The injection volume was 5 L. 2.2.2. Mass spectrometry Detection was performed on a Sciex API 4000 Qtrap MS system equipped with a Turbo Ionspray interface. Mass spectral setting to operate in positive-ion mode (ESI+) were: ion source voltage: 5000 V; ion source temperature: 400 ◦ C; collision gas (N2 ): medium; curtain gas: 20 psi; nebulizer gas: 40 psi; auxiliary gas: 60 psi. Quantification was performed using multiple reaction monitoring (MRM) of the transitions of m/z 468.2 → m/z 163.0 for tobramycin and m/z 448.2 → m/z 160.0 for sisomycin, respectively. The product ion spectra of tobramycin and sisomycin are shown in Fig. 2. Data acquisition and processing were performed with the Analyst software 1.4.2.
Fig. 2. Product ion spectra of tobramycin (1) and sisomycin (2).
2.3. Preparation of standards and quality control samples Standard stock solutions of tobramycin and sisomycin were both prepared in water at the concentration of 105.1 g/mL and 112.1 g/mL. Then the stock solutions were serially diluted with methanol: water (1:1) to provide working standard solutions of desired concentrations. All the solutions were stored at 4 ◦ C.
Fig. 1. Chemical structures of tobramycin (1) and sisomycin (I.S.) (2).
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
Calibration standards were prepared by spiking 0.2 mL of blank human plasma with 400 L working standard solutions of tobramycin. The effective concentrations in standard plasma samples were 10.51, 21.02, 52.55, 105.1, 210.2, 525.5, 1051 ng/mL for tobramycin. One calibration curve was constructed on each analysis day using freshly prepared calibration standards. The quality control samples (QCs) were prepared with blank plasma at LLOQ, low, middle and high concentrations of 10.51, 26.28, 105.1, 840.8 ng/mL for tobramycin. The standards and quality controls were extracted on each analysis day with the same procedure for plasma samples as described below. 2.4. Plasma sample preparation To a 0.2 mL aliquot of plasma sample in 1.5 mL centrifuge tube, 50 L of internal standard (1121 ng/mL) and 400 L methanol:water (1:1) were added. The mixture was vortex-mixed thoroughly for 1 min and then centrifuged at 13,000 r.p.m. for 10 min. The supernatant was directly injected into the HILICMS/MS system. 2.5. Method validation Validation runs were conducted on three consecutive days. Each validation run consisted of a minimum of one set of calibration standards and six replicates of LLOQ and QC plasma samples at three concentrations. The results from LLOQ and QC plasma samples in three runs were used to evaluate the precision and accuracy of the method developed. 2.5.1. Selectivity Selectivity was studied by comparing chromatograms of six different batches of blank plasma obtained from six subjects with those of corresponding standard plasma samples spiked with tobramycin and sisomycin (1121 ng/mL). 2.5.2. Linearity and lower limit of quantification (LLOQ) Calibration curves were prepared by assaying standard plasma samples at seven concentrations of tobramycin ranging 10.51–1051 ng/mL. The linearity of each calibration curve was determined by plotting the peak area ratio (y) of tobramycin to sisomycin versus the nominal concentration (x) of tobramycin. The calibration curves were constructed by weighted (1/x2 ) least square linear regression. The lower limit of quantification, defined as the lowest concentration on the calibration curve, was validated using an LLOQ sample for which an acceptable accuracy (relative error, RE) within ±20% and a precision (relative standard deviation, RSD) below 20% were obtained. 2.5.3. Precision and accuracy For determination of the intra-day accuracy and precision, a replicate analysis of QC plasma samples of tobramycin was performed on the same day. The run consisted of a calibration curve and six replicates of each LLOQ, low, mid and high concentration quality control samples. The inter-day accuracy and precision were assessed by analysis of three batches on different days. The precision was expressed as the relative standard deviation (RSD) and the accuracy as the relative error (RE). 2.5.4. Extraction recovery and matrix effect The recovery was calculated by comparing the peak area of the tobramycin added into blank plasma and extracted using the protein precipitation procedure with those obtained from the compound spiked into post-extraction supernatant at three QC concentration levels. The matrix effect was measured by comparing
41
the peak response of sample spiked post-extraction (A) with that of the standard solution containing equivalent amounts of the compound (B). The ratio (A/B × 100) % was used to evaluate the matrix effect. The extraction recovery and matrix effect of I.S. were also evaluated using the same method. 2.5.5. Stability 2.5.5.1. Freeze and thaw stability. The effect of three freeze and thaw cycles on the stability of plasma samples containing tobramycin was determined by subjecting five aliquots of QC samples at low, mid and high concentration unextracted quality control samples to three freeze–thaw cycles. After completion of the three cycles, the samples were analyzed and the experimental concentrations were compared with the nominal values. 2.5.5.2. Long-term stability. Five aliquots of QC samples at low, mid and high concentration unextracted QC samples were stored at −20 ◦ C for 20 days. Then, the samples were processed and analyzed and the concentrations obtained were compared with the nominal values. 2.5.5.3. Short-term stability. Five aliquots of QC samples at low, mid and high concentration unextracted QC samples were kept at ambient temperature (25 ◦ C) for 4 h in order to determine the short-term stability of tobramycin in human plasma. Then, the samples were processed and analyzed. The concentrations obtained were compared with the nominal values of QC samples. 2.5.5.4. Post-preparation stability. In order to estimate the stability of tobramycin in the prepared sample, five aliquots of QC samples at low, mid and high concentration were kept in an autosampler maintained at room temperature for about 8 h. Then, the samples were analyzed and the concentrations obtained were compared with the nominal values. 2.6. Pharmacokinetic study The method was successfully applied to determine the pharmacokinetic study of tobramycin in 10 healthy Chinese volunteers. Each subject was intramuscular injected 1 mg/kg tobramycin. The pharmacokinetic study was approved by the local Ethics Committee and all volunteers gave their signed informed consent to participate in the study according to the principles of the Declaration of Helsinki. Blood samples were collected before and 0.25, 0.5, 0.75, 1.0, 1.5, 2.0, 3.0, 4.0, 5.0, 7.0, 9.0 and 12 h post-dosing. Samples were centrifuged and plasma was separated and stored at −80 ◦ C until analyzed. The maximum plasma concentration (Cmax ) and their time were noted directly. The elimination rate constant (ke ) was calculated by linear regression of the terminal points of the semi-log plot of plasma concentration against time. Elimination half-life (t1/2 ) was calculated using the formula t1/2 = 0.693/ke . The area under the plasma concentration–time curve (AUC0–t ) to the last measurable plasma concentration (Ct ) was calculated by the linear trapezoidal rule. The area under the plasma concentration–time curve to time infinity (AUC0–∞ ) was calculated as: AUC0–∞ = AUC0-t + Ct /ke . 3. Results and discussion 3.1. Selection of IS The best internal standard in LC–MS assay is a deuterated form of the analyte. In our laboratory, no deuterated tobramycin was available. Therefore, a compound being structurally or chemically similar to the analyte was considered. In LC–MS/MS the I.S. should
42
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
also have similar chromatographic and mass spectrometric behaviors to the analyte, and mimic the analyte in any sample preparation steps. Sisomycin was chosen as the internal standard for the assay because of its similarity of structure, retention time and ionization to tobramycin.
3.2. Chromatography and mass spectrometry It can be seen that all of the highly polar contents could not be retained on C18 column and eluted in dead time. But when using hydrophilic interaction chromatography (HILIC) for analysis, the fraction had a good separation on the column. Thus, HILIC will be an effective technique to analyze the highly polar compounds studies. HILIC is such a technique, where analyte retention is believed to be caused by partitioning of the analyte between a water-enriched layer of stagnant eluent on a hydrophilic stationary phase and a relatively hydrophobic bulk eluent, with the main components usually being 5–40%water in ACN. The use of water as the strongly eluting solvent gives HILIC a number of advantages over conventional normal phase chromatography (NPC); eluent preparation is less complicated since the need of total control over the solvent water content is omitted. Normal phase (NP) eluents are also nonpolar (often based on hexane) and polar analytes usually have a low solubility in these eluents. The interfacing with electrospray MS is also a problem with NP, since ionization is not easily achieved in totally organic, nonpolar eluents. The elution order in HILIC is more or less the opposite of that seen in RP separations, which means that HILIC works best for solutes that are problematic in RP. Therefore hydrophilic interaction chromatography (HILIC) on a silica column was chosen for the separation. The ionization of the analyte was affected by the composition of mobile phase. 5 mM ammonium acetate and 0.1% formic acid were employed to supply the ionic strength. A mixture of 5 mM ammonium acetate and 0.1% formic acid–acetonitrile (40:60) was finally adopted as the mobile phase. The LLOQ for tobramycin was 10.51 ng/mL. Due to the lower injection volume of 5 L, the on column sensitivity in our study (the quantity of drug injected on the column per injection) was 52.55 pg for tobramycin. It was much lower than those reported in the literatures, which were 200 ng/mL [15], 50 ng/mL [16], 150 ng/mL [17] and 100 ng/mL [18,19], respectively. Since the elimination rate constant (ke) was calculated by linear regression of the terminal points of the semi-log plot of plasma concentration against time and elimination half-life (t1/2 ) was calculated using the formula t1/2 = 0.693/ke, the lower the LLOQ was, the more accurate t1/2 was calculated. The total run time was 3.5 min per sample, which was lower than previously reported analysis times (longer than 10 min) [15,19]. This short analysis time indicated that our method better met the requirement of high sample throughput in bioanalysis. HILIC-MS/MS operation parameters were carefully optimized for determination of tobramycin. The mass spectrometer was tuned in both positive and negative ionization modes with ESI for tobramycin. Both signal intensity and ratio of signal to noise obtained in positive ionization mode were much greater than those in negative ionization mode. In the precursor ion full-scan spectra, the most abundant ions were protonated molecules [M+H]+ m/z 468.2 and 448.2 for tobramycin and I.S., respectively. The product ion scan spectra showed high abundance fragment ions at m/z 163.0 and 160.0 for tobramycin and I.S., respectively. Multiple reaction monitoring (MRM) using the precursor → product ion transitions of m/z 468.2 → m/z 163.0 and m/z 448.2 → m/z 160.0 was employed for quantification of tobramycin and I.S. respectively.
Table 1 Precision and accuracy for the determination of tobramycin in human plasma (intraday: n = 6; inter-day: n = 6 series per day, 3 days). Concentrations (ng/mL)
R.S.D (%)
Added
Found
Intra-day
Inter-day
Relative error (%)
10.51 26.28 105.1 840.8
10.65 26.94 108.2 888.7
4.6 2.3 1.8 1.4
11.6 8.6 3.4 4.0
1.3 2.5 3.0 5.7
3.3. Sample preparation Liquid–liquid extraction (LLE) and solid-phase extraction (SPE) are techniques often used in the preparation of biological samples for their ability to improve the sensitivity and robustness of assay. But tobramycin had a very high polarity, so it was impossible to extract it from biological fluids using liquid–liquid extraction method and SPE is too expensive. In the present experiment, a simple protein precipitation procedure was developed to reduce sample preparation time. No further concentration procedure was needed and the sample preparation procedure was simplified. To obtain high extraction efficiency, three different protein precipitation agents, acetonitrile, methanol and acetone were investigated. It was found that acetonitrile had a higher efficiency of precipitating. High extraction efficiency achieved too when this procedure was applied to I.S. Finally, this simple single-step acetonitrile protein precipitation was adopted. The protein precipitation sample preparation procedure was much simpler and less expensive than a previously used SPE method in the literature [19]. 3.4. Method validation 3.4.1. Selectivity Selectivity was assessed by comparing the chromatograms of six different batched of blank human plasma with the corresponding spiked plasma. As shown in Fig. 3A, no interference from endogenous substance was observed at the retention time of tobramycin and I.S. 3.4.2. Linearity and LLOQ The standard calibration curve for tobramycin was linear over the concentration range of 10.51–1051 ng/mL (r2 > 0.99) by using weighted least square linear regression analysis with a weigh factor of 1/x2 . The typical equations for the calibration curves for metformin and rosiglitazone were: y = 1.680 × 10−4 x + 5.660 × 10−5 , r = 0.9963. The lower limit of quantification for tobramycin was 10.51 ng/mL with precision and accuracy presented in Table 1 with RE within ±20% and RSD lower than 20%. A corresponding chromatogram is given in Fig. 3B 3.4.3. Precision and accuracy The data of intra-day and inter-day precision and accuracy for the method are listed in Table 1. The intra-day and inter-day precisions for low, mid and high QC levels of tobramycin were below 15%, with the accuracy within 1.3–5.7%. The precision and accuracy of the present method conform to the criteria for the analysis of biological samples according to the guidance of USFDA where the precision (RSD) determined at each concentration level is required not exceeding 15% and accuracy (RE) within ±15% of the actual value. 3.4.4. Extraction recovery and matrix effect The extraction recoveries of tobramycin from human plasma at three QC levels were all above 80% and the mean extraction
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
43
Fig. 3. Representative MRM chromatograms of tobramycin (peak 1) and sisomycin (peak 2) in human plasma samples. (A) A blank plasma samples; (B) a blank plasma sample spiked with tobramycin at the LLOQ of tobramycin 10.51 ng/mL, and sisomycin (1121 ng/mL); (C) a blank plasma sample spiked with tobramycin at 105.1 ng/mL and sisomycin (1121 ng/mL); the retention times of tobramycin and sisomycin were 2.52 min and 2.22 min, respectively.
44
L. Chen et al. / J. Chromatogr. B 973 (2014) 39–44
Table 2 Stability of tobramycin in plasma samples (n = 5). Stability
26.28
105.1
Mean ± SD Short-term stability Long-term stability Freeze–thaw stability Post-preparative stability
26.81 26.88 26.96 26.35
± ± ± ±
0.6 1.2 1.3 2.3
840.8
RE (%)
Mean ± SD
2.0 0.3 0.3 −2.3
103.3 103.0 104.3 99.5
recovery of tobramycin was 85.3 ± 2.4%. The mean extraction recovery of sisomycin was 87.9 ± 1.2%. In terms of matrix effect, all the ratios defined as in Section 2 were between 85% and 115%, which means no matrix effect for tobramycin in this method. 3.4.5. Stability The stock solution of tobramycin and sisomycin was found to be stable at room temperature for 4 h and at 4 ◦ C for 30 days. The results from all stability tests presented in Table 2 demonstrated a good stability of tobramycin and sisomycin over all steps of the determination. The method is therefore proved to be applicable for routine analysis. 3.5. Method application The present method was successfully applied to the pharmacokinetic study of tobramycin after intramuscular injection in healthy volunteers. After the administration of tobramycin (1 mg/kg), Cmax was 5.52 ± 1.23 g/mL. Tmax was 0.51 ± 0.55 h. Plasma concentration declined with the t1/2 of 2.1 ± 0.32 h. The mean (±SD) AUC0–∞ for tobramycin in plasma was 15.13 ± 3.51 g h mL−1 . 4. Conclusion A sensitive, selective and rapid HILIC-MS/MS method for the determination of tobramycin in human plasma is described. Compared with the published methods, the method used a new HILIC column offering superior sensitivity with the LLOQ of 10.51 ng/mL, satisfactory selectivity and short run time of 3.5 min. This method can be used for the pharmacokinetic study of tobramycin in human. References [1] H.C. Neu, C.L. Dendush, Ototoxicity of tobramycin: a clinical overview, J. Infect. Dis. 134 (1976) S206. [2] N.E. Plaut, J.J. Schentag, W.J. Jusko, Aminoglycoside nephrotoxicity: comparative assessment in critically ill patients, J. Med. 10 (1979) 257.
± ± ± ±
9.9 8.8 9.3 8.0
RE (%)
Mean ± SD
−1.7 −0.3 1.3 −4.6
884.4 810.4 844.4 814.4
± ± ± ±
41.6 44.1 55.6 50.1
RE (%) 5.2 −8.4 4.2 −3.6
[3] C.R. Smith, J.J. Lipsky, O.L. Laskin, D.B. Hellman, Double-blind comparison of the nephrotoxicity and auditory toxicity of gentamicin and tobramycin, New Engl. J. Med. 302 (1980) 1106. [4] R.D. Moore, C.R. Smith, J.J. Lipsky, E.D. Mellits, P.S. Lietman, Risk factors for nephrotoxicity in patients treated with aminoglycosides, Ann. Intern. Med. 100 (1984) 352. [5] P. Noone, D.F. Beale, S.S. Pollock, Monitoring aminoglycoside use in patients with severely impaired renal function, Br. Med. J. 2 (1978) 4670. [6] D.A. Stead, Current methodologies for the analysis of aminoglycosides, J. Chromatogr. B 747 (2000) 69. [7] N. Isoherranen, S. Soback, Determination of gentamicins C(1), C(1a), and C(2) in plasma and urine by HPLC, Clin. Chem. 46 (2000) 837. [8] D.M. Barends, C.L. Zwaan, A. Hulshoff, Micro-determination of tobramycin in serum by high-performance liquid chromatography with ultraviolet detection, J. Chromatogr. 225 (1981) 417. [9] S.E. Back, I. Nilsson-Ehle, P. Nilsson-Ehle, Chemical assay, involving liquid chromatography, for aminoglycoside antibiotics in serum, Clin. Chem. 25 (1979) 1222. [10] C.H. Feng, S.J. Lin, H.L. Wu, S.H. Chen, Trace analysis of tobramycin in human plasma by derivatization and high-performance liquid chromatography with ultraviolet detection, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 780 (2002) 349. [11] P.M. Kabra, P.K. Bhatnagar, M.A. Nelson, J.H. Wall, L.J. Marton, Liquidchromatographic determination of tobramycin in serum with spectrophotometric detection, Clin. Chem. 29 (1983) 672. [12] D.B. Haughey, D.M. Janicke, M. Adelman, J.J. Schentag, High-pressure liquid chromatography analysis and single-dose disposition of tobramycin in human volunteers, Antimicrob. Agents Chemother. 17 (1980) 649. [13] H. Fonge, E. Kaale, C. Govaerts, K. Desmet, A. vanSchepdael, J. Hoog-martens, Bioanalysis of tobramycin for therapeutic drug monitoring by solid-phase extraction and capillary zone electrophoresis, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 810 (2004) 313. [14] J.P. Anhalt, S.D. Brown, High-performance liquid-chromatographic assay of aminoglycoside antibiotics in serum, Clin. Chem. 24 (1978) 1940. [15] M.X. Guo, L. Wrisley, E. Maygoo, Measurement of tobramycin by reversedphase high-performance liquid chromatography with mass spectrometry detection, Anal. Chim. Acta 571 (2006) 12. [16] M.A. Jonge, J.M. Bekkers, H.O. Straaten, R.W. Sparidans, E. Franssen, Simple and sensitive method for quantification of low tobramycin concentrations in human plasma using HPLC-MS/MS, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 862 (2008) 257. [17] B.G. Keevil, S.J. Lockhart, D.P. Cooper, Determination of tobramycin in serum using liquid chromatography–tandem mass spectrometry and comparison with a fluorescence polarisation assay, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 794 (2003) 329. [18] C.R. Roc, R.G. Roberto, G.F. Antonia, R.M. Manuel Angel, M.V. Jos Luis, Simultaneous analysis of antibiotics in biological samples by ultra high performance liquid chromatography–tandem mass spectrometry, J. Pharm. Biomed. Anal. 89 (2014) 203. [19] R. Oertel, V. Neumeister, W. Kirch, Hydrophilic interaction chromatography combined with tandem-mass spectrometry to determine six aminoglycosides in serum, J. Chromatogr. A 1058 (2004) 197–201.