Sample Stacking
5.3. Results and discussion
5.3.1. Optimization of sample preparation method
5.3.1.1. Deproteinization method
Posaconazole is a highly hydrophobic drug with a protein-binding rate greater than 98%. The most frequently used additive to denature serum protein when measuring posaconazole is acetonitrile (ACN) [25, 28]. Although ACN provided a satisfactory extraction recovery (higher than 90%), the resulting solution showed high elution strength and could not be directly loaded into SPE cartridges. Compared to the ACN denaturation method, the urea denaturation method provided a similar recovery rate but with low elution strength. The urea-deproteinized solution could be directly loaded into SPE cartridges without any dilution steps; therefore, it was chosen as the protein-denaturing agent in this study.
144
5.3.1.2. Solid phase extraction (SPE) procedures
We applied the SPE procedures to reduce the matrix effect that was frequently encountered by biological samples in FASS system. In the stacking mode of the online preconcentration method, the amount of endogenous materials in the sample matrix significantly affected the stacking efficiency. In addition, ions in the plasma samples will introduce an injection bias during electrokinetic injection in the FASS analysis and hamper the quantification accuracy in the clinical assay. As displayed in Figure 5.2 (a), the posaconazole peak was very small if SPE procedure was not applied to clean up the plasma sample. The OASIS HLB cartridge was chosen as the extraction cartridge in this study. Two wash steps were employed in the SPE procedures. After sample loading, the salts were washed out with DI water followed by a MeOH solution to eliminate plasma interferences. Posaconazole is a highly hydrophobic compound (log P > 3), and it can be well retained by the reversed phase extraction sorbent. We optimized the MeOH percentage of the wash solution, and an 80% MeOH solution was selected as the optimum wash solution. The extraction recovery of the SPE method was 96.75 ± 3.02%
(n=3). Figure 5.2 (b) showed the performance of the optimum SPE method. Although the concentration factor for plasma samples in Figure 5.2 (a) and 5.2 (b) in sample
145
preprocessing step is the same, the signal intensity of posaconazole significantly increased after SPE pretreatment.
5.3.1.3. Sample filtration
It should be noted that filter types for sample filtration affected posoconazole recovery to a great extent. The filtration step is conducted after the redissolving step to ensure no particles were injected into the capillary. Although very few studies discussed the effect of filter types on sample recovery, we found that this parameter showed a significant effect on sample recovery for highly hydrophobic compounds, such as posaconazole, in this study. Typical materials for syringe filters are nylon, polyvinylidene difluoride (PVDF), Teflon (PTFE), polypropylene (PP) and cellulose derivative membranes; these materials were tested for posaconazole recovery. We found that the posaconazole and itraconazole (internal standard) recovery was lower than 40%
if we use hydrophobic filters or unmodified filters, such as nylon and PTFE. It is probably due to sample adsorption. Conversely, hydrophilic cellulose derivative material provided a much better recovery (>90%). In addition, the percentage of organic solvent in the sample solution during sample filtration also influenced the posaconazole recovery to a great extent in that a higher percentage of organic solvent in the sample
146
solution led to a higher sample recovery. In this study, 0.2 M formic acid in 95% MeOH was chosen as the sample solvent. The high percentage of MeOH in the sample solvent improved the recovery of posaconazole during sample filtration and also increased the difference in conductivity between the sample zone and the BGEs.
5.3.2. Analytical method development
Previous studies have revealed that the posaconazole concentrations in different patients could vary from 50 ng mL-1 to higher than 5,000 ng mL-1. In this study, conventional capillary zone electrophoresis (CZE) was tested for posaconazole analysis at first, but the sensitivity could not satisfy clinical need. FASS technique was therefore adapted to improve the detection sensitivity for plasma posaconazole. Posaconazole and itraconazole standard solution and drugs-spiked plasma sample were used to optimize the FASS method. Key parameters that affected the stacking efficiency included the composition of the BGEs and the sample matrix, the sample injection time, and the applied voltage. These parameters were optimized to achieve the highest sensitivity in the FASS system. The limits of detection (LOD) at a signal-to-noise ratio equal to three and the enhancement factor (EF) were determined to show the concentration efficiencies of FASS.
147
5.3.3. Effect of the sample matrix and the separation buffer
To ensure that the posaconazole was in its protonated form (pKa 3.6 and 4.6) [16, 32], an acidic buffer was used as the sample solution and BGE. The buffer ranges of formic acid, phosphoric acid and citric acid are all lower than the pKa value of posaconazole, and these acids were evaluated for their suitability as the buffer system for FASS. Citric acid showed a relatively high UV cut-off value, which would sacrifice detection sensitivity, and phosphoric acid generated a higher current, which would generate joule heat and lead to an unstable system at high concentration. Considering these issues, formic acid was chosen as the analytical buffer in this study.
Theoretically, the stacking efficiency is proportional to the conductivity difference
between the sample solution and BGE because of the much higher electric field being
distributed in the sample zone. Formic acid at concentrations of 0.75 M, 1.0 M, 1.25 M and 1.5 M was tested as the BGE solution, and the results are displayed in Figure 5.3 (a).
When the concentration of formic acid was higher than 1.0 M, the posaconazole peak intensity reached its maximum. The resolution between posaconazole and the internal standard, itraconazole, continuously increased with increasing formic acid concentrations in all tested ranges. When formic acid was at concentration of 1.5 M, the
148
electric current was higher than 100 A, and the system became unstable. The method’s
precision was also sacrificed under high current. Therefore, 1.25 M formic acid was chosen as the optimum buffer concentration in the BGE.
For basic compound analysis in stacking mode, a small amount of acid is recommended to be added into the sample matrix to enhance compound protonation and to improve reproducibility of the method [31]. When the formic acid concentration in the sample matrix was varied from 0.05 M to 0.3 M, the increase in formic acid concentration resulted in an increase of the peak intensity. When the concentration of formic acid in the sample matrix was lower than 0.2 M, the results were irreproducible because of the incomplete protonation of posaconazole especially when analyzing posaconazole spiked plasma samples. As the formic acid concentration increased, number of theoretical plates decreased because of the decreasing conductivity differences between the sample solution and BGE (Figure 5.3 (b)). Considering peak intensity, reproducibility and peak efficiency, 0.2 M of formic acid was chosen as the optimum concentration in the sample matrix.
The stacking efficiency has been shown to increase by adding water-miscible organic solvents into the sample matrix to enlarge the conductivity difference between the BGE and the sample plug [31]. Different concentrations (50% to 98%) of MeOH
149
were added to the sample matrix to test its effect on peak intensity. As shown in Figure 5.4, adding MeOH to the sample matrix significantly increased the peak intensity and peak area. This is because larger amount of posaconazole ions were injected into the capillary. This phenomenon is assumed to be caused by larger electric field being distributed in the sample matrix with increasing methanol percentage. Since the total length and the applied voltage didn’t change, the total electric field of the FASS system remained unchanged when changing the methanol percentage. Therefore, the electric field difference between the sample matrix and the background electrolyte is increased when adding higher percentage of methanol into the sample matrix. As posaconazole ions moved faster under the high electric field, more posaconazole ions were electrokinetically injected into the capillary. As a result, the peak area along with the peak intensity increased when the methanol percentage in the sample matrix was increased. When the percentage of MeOH was higher than 95%, the analytical results were irreproducible because of the decrease in protonation of posaconazole. Therefore, 95% MeOH was selected as the optimum percentage in the sample matrix.
150
5.3.4. Effect of the injection voltage and injection time
Theoretically, the injection voltage should be increased to reduce the injection time, assuming similar amount of sample is injected. The injection voltage was tested within 2 to 10 kV. Due to the high conductivity difference between the sample matrix and the separation buffer, excessive joule heat in the narrow part of the injection end could embrittle the capillary when the applied voltage was higher than 8 kV. Therefore, 8 kV was selected as the injection voltage. When a sample was injected into the capillary, the electric current remained constant for a while then gradually decreased. After the current reached 70% of the maximum current (approximately 48 seconds), the injection voltage was shut down. A longer injection time would lead to irreproducible results and even current disruption.
The separation voltage played a minor role on the separation result. To avoid too much joule heat, which can cause peak broadening and an irreproducible separation, the separation voltage was set at 25 kV. The electropherogram obtained under optimum separation conditions is displayed in Figure 5.6. When using 1.25 M formic acid as the BGE and 0.2 M formic acid in 95% (v/v) MeOH as the sample solution, the limit of detection (LOD) for posaconazole was 10 ng mL-1, with an analytical run time less than
151
5 minutes. Compared to the conventional CZE method, the FASS method improved the sensitivity by over 170-fold for the posaconazole analysis.
5.3.5. Method validation
To satisfy the clinical requirement, the developed method was validated within concentrations ranging from 30 to 10,000 ng mL-1. Itraconazole was used as the internal standard to improve the precision and accuracy of the method.
5.3.5.1. Linearity
The linearity of the method was tested within 30 to 10,000 ng mL-1. To obtain a better quantification accuracy at low concentrations, two calibration curves were generated (Figure 5.5). The calibration curves were y = 0.9108 x + 0.1515 (r2 = 0.9998) for posaconazole concentrations in the range of 30 to 800 ng mL-1, and y = 1.0090 x + 0.0106 (r2 = 0.9998) for posaconazole concentrations in the range of 800 to 10,000 ng mL-1.
152
5.3.5.2. Limit of detection (LOD) and limit of quantification (LOQ)
The limit of detection (LOD) was determined as the concentration when the signal-to-noise ratio (S/N) equaled 3. The limit of quantification (LOQ) was determined as the concentration when the signal-to-noise ratio (S/N) equaled 10. Under the optimized conditions, the LOD and LOQ of the posaconazole concentrations were 10 ng mL-1 and 30 ng mL-1, respectively.
5.3.5.3. Precision and accuracy
Run-to-run repeatability (intra-day, n = 6) and intermediate precision (inter-day, n
= 3) of the migration time and peak area ratios of posaconazole to itraconazole were tested. In terms of migration time, both repeatability and intermediate precision of posaconazole were within a 3.2 % relative standard deviation (RSD). The repeatability (intra-day, n = 6) and intermediate precision (inter-day, n = 3) of peak area ratios were tested at 30, 100, 800, 3,000 and 10,000 ng mL-1. The repeatability (n = 6) and intermediate precision (n = 3) at the tested ranges except LOQ, were within 7.2 % and 7.5% RSD, respectively. The accuracy of the method was studied by spiking posaconazole into blank plasma samples. The recoveries, except LOQ, were within 95.1
% and 106.4 % (n = 6). The method precision and accuracy were also tested at LOQ (30
153
ng mL-1). Both intraday precision and intermediate precision at the LOQ were lower than 17.5 % RSD. The recovery of posaconazole at LOQ was 93.8 % (n=6) (Table 5.1).
5.3.5.4. Selectivity
The selectivity of the method was tested by analyzing blank plasma obtained from six healthy volunteers. There was no endogenous material found at the migration time of posaconazole and the internal standard for six tested samples.
5.3.6. Determination of posaconazole in patient plasma
Plasma samples collected from one patient receiving posaconazole treatment was analyzed using the validated FASS method. Figure 5.7 displays the representative electropherogram, and the drug concentration was calculated to be 2.7 g mL-1. As displayed in Figure 5.7, no endogenous interferences or co-medications overlapped peaks of posaconazole or the internal standard. The results revealed that the developed method could be used for therapeutic drug monitoring (TDM) for patients undergoing posaconazole treatment.
154
The high morbidity and mortality associated with invasive fungal infections have increased the importance of improving treatment efficacy. Therapeutic drug monitoring of posaconazole has been proven to improve the outcome of treatment. Several methods have been developed for the determination of posaconazole concentrations in patient plasma. HPLC-UV is the most frequently used method to measure posaconazole concentrations. The LOQs of the HPLC-UV method were in the range of 50-620 ng mL-1 [22, 24-26]. One study applied the HPLC-LIF technique to improve the sensitivity of the method, and the LOQ was 100 ng mL-1 [23]. Another study applied the HPLC-MS technique for posaconazole quantification, and the LOQ was 31 ng mL-1 [23].
Several recent studies used LC-MS/MS method for posaconazole quantification. The LOQs of the LC-MS/MS method were in the range of 5-20 ng mL-1 [27-29]. As demonstrated in this study, the linear range of our developed FASS method was 30 ng mL-1 to 10,000 ng mL-1. The sensitivity of the current FASS method was superior to HPLC-UV or HPLC-LIF and comparable to HPLC-MS/MS method. The high sensitivity of the FASS method made it possible to analyze other noninvasive biofluids, such as urine and saliva, where posaconazole was found in low concentrations.
155
5.4. Conclusions
SPE-FASS analysis is an efficient, accurate and cost effective method to quantify posaconazole in human plasma. With the application of the optimized SPE procedures, the endogenous interferences and matrix effect in FASS system were greatly reduced.
Due to the highly hydrophobic nature of posaconazole (log P > 3), its recovery was found to be lower than 40 % if hydrophobic filters or unmodified filters, such as nylon and PTFE, were used. The recovery was much higher with the use of a cellulose derivative filter (>90%).
To ensure a large injection amount during electrokinetic injection, 0.2 M formic acid was added to the sample matrix to maintain the posaconazole in the protonated state. A solution consisting of 95 % MeOH was added to the sample matrix to increase the conductivity difference between the sample zone and the BGE. Compared to phosphoric acid, formic acid is relatively rarely used as the BGE in FASS, especially when CE is coupled with a UV detector instead of a mass spectrometer. In this study, formic acid was found to provide better system stability under high buffer concentrations, which is a big advantage when using the FASS method. With the use of 1.25 M formic acid as the BGE and 0.2 M formic acid in 95% (v/v) MeOH as the sample solution, the limit of detection (LOD) for posaconazole was 10 ng mL-1, and the analytical run time was less
156
than 5 minutes. The successful application of the developed method to determine posaconazole concentrations in patient sample has demonstrated its feasibility as an effective method for clinical use.
5.5. References
[1] K.A. Marr, R.A. Carter, F. Crippa, A. Wald, L. Corey, Epidemiology and outcome of mould infections in hematopoietic stem cell transplant recipients. Clinical Infectious Diseases, 34 (2002) 909-917.
[2] M.A. Pfaller, S. Messer, R.N. Jones, Activity of a new triazole, Sch 56592, compared with those of four other antifungal agents tested against clinical isolates of Candida spp and Saccharomyces cerevisiae. Antimicrobial Agents and Chemotherapy, 41 (1997) 233-235.
[3] H.K. Munayyer, P.A. Mann, A.S. Chau, T. Yarosh-Tomaine, J.R. Greene, R.S. Hare, L. Heimark, R.E. Palermo, D. Loebenberg, P.M. McNicholas, Posaconazole is a potent inhibitor of sterol 14 alpha-demethylation in yeasts and molds. Antimicrobial Agents and Chemotherapy, 48 (2004) 3690-3696.
157
[4] P. Pitisuttithum, R. Negroni, J.R. Graybill, B. Bustamante, P. Pappas, S. Chapman, R.S. Hare, C.J. Hardalo, Activity of posaconazole in the treatment of central nervous system fungal infections. Journal of Antimicrobial Chemotherapy, 56 (2005) 745-755.
[5] R. Herbrecht, Posaconazole: a potent, extended-spectrum triazole anti-fungal for the treatment of serious fungal infections. International Journal of Clinical Practice, 58 (2004) 612-624.
[6] R.N. Greenberg, K. Mullane, J.A.H. van Burik, I. Raad, M.J. Abzug, G. Anstead, R.
Herbrecht, A. Langston, K.A. Marr, G. Schiller, M. Schuster, J.R. Wingard, C.E.
Gonzalez, S.G. Revankar, G. Corcoran, R.J. Kryscio, R. Hare, Posaconazole as salvage therapy for zygomycosis. Antimicrobial Agents and Chemotherapy, 50 (2006) 126-133.
[7] J.A.H. van Burik, R.S. Hare, H.F. Solomon, M.L. Corrado, D.P. Kontoyiannis, Posaconazole is effective as salvage therapy in zygomycosis: A retrospective summary of 91 cases. Clinical Infectious Diseases, 42 (2006) E61-E65.
[8] O.A. Cornely, J. Maertens, D.J. Winston, J. Perfect, A.J. Ullmann, T.J. Walsh, D.
Helfgott, J. Holowiecki, D. Stockelberg, Y. Goh, M. Petrini, C. Hardalo, R. Suresh, D.
Angulo-Gonzalez, Posaconazole vs. fluconazole or itraconazole prophylaxis in patients with neutropenia. New England Journal of Medicine, 356 (2007) 348-359.
158
[9] A.J. Ullmann, J.H. Lipton, D.H. Vesole, P. Chandrasekar, A. Langston, S.R.
Tarantolo, H. Greinix, W. Morais de Azevedo, V. Reddy, N. Boparai, L. Pedicone, H.
Patino, S. Durrant, Posaconazole or fluconazole for prophylaxis in severe graft-versus-host disease. New England Journal of Medicine, 356 (2007) 335-347.
[10] M.E. Werner Christian Neubauer, Armin König, Stefanie Hieke, Manfred Jung, Hartmut Bertz, and Klaus Kümmerer, Therapeutic Drug Monitoring of Posaconazole in Hematology Patients: Experience with a New High-Performance Liquid Chromatography-Based Method. Antimicrobial Agents and Chemotherapy, 54 (2010) 4029-4032.
[11] G.R. Thompson, M.G. Rinaldi, G. Pennick, S.A. Dorsey, T.F. Patterson, J.S. Lewis, Posaconazole Therapeutic Drug Monitoring: a Reference Laboratory Experience.
Antimicrobial Agents and Chemotherapy, 53 (2009) 2223-2224.
[12] A.J. Ullmann, O.A. Cornely, A. Burchardt, R. Hachem, D.P. Kontoyiannis, K.
Topelt, R. Courtney, D. Wexler, G. Krishna, M. Martinho, G. Corcoran, I. Raad, Pharmacokinetics, safety, and efficacy of posaconazole in patients with persistent febrile neutropenia or refractory invasive fungal infection. Antimicrobial Agents and Chemotherapy, 50 (2006) 658-666.
159
[13] FDA, Noxafil (posaconazole) Drug Labeling Information September 2010, (2010) http://www.accessdata.fda.gov/drugsatfda_docs/label/2010/022003s022008,022027s022
001lbl.pdf.
[14] G. Krishna, L. Ma, D. Vickery, X. Yu, I. Wu, E. Power, E. Beresford, S. Komjathy, Effect of Varying Amounts of a Liquid Nutritional Supplement on the Pharmacokinetics of Posaconazole in Healthy Volunteers. Antimicrobial Agents and Chemotherapy, 53 (2009) 4749-4752.
[15] A. Sansone-Parsons, G. Krishna, A. Calzetta, D. Wexler, B. Kantesaria, M.A.
Rosenberg, M.A. Saltzman, Effect of a nutritional supplement on posaconazole pharmacokinetics following oral administration to healthy volunteers. Antimicrobial Agents and Chemotherapy, 50 (2006) 1881-1883.
[16] R. Courtney, D. Wexler, E. Radwanski, J. Lim, M. Laughlin, Effect of food on the relative bioavailability of two oral formulations of posaconazole in healthy adults.
British Journal of Clinical Pharmacology, 57 (2004) 218-222.
[17] G. Krishna, A. Moton, L. Ma, M.M. Medlock, J. McLeod, Pharmacokinetics and Absorption of Posaconazole Oral Suspension under Various Gastric Conditions in Healthy Volunteers. Antimicrobial Agents and Chemotherapy, 53 (2009) 958-966.
160
[18] D. Lebeaux, F. Lanternier, C. Elie, F. Suarez, A. Buzyn, J.P. Viard, M.E. Bougnoux, M. Lecuit, V. Jullien, O. Lortholary, Therapeutic Drug Monitoring of Posaconazole: a Monocentric Study with 54 Adults. Antimicrobial Agents and Chemotherapy, 53 (2009) 5224-5229.
[19] W.W. Hope, E.M. Billaud, J. Lestner, D.W. Denning, Therapeutic drug monitoring for triazoles. Current Opinion in Infectious Diseases, 21 (2008) 580-586.
[20] D. Smith, D. Andes, Therapeutic drug monitoring of antifungals: Pharmacokinetic and pharmacodynamic considerations. Therapeutic Drug Monitoring, 30 (2008) 167-172.
[21] R.J. Ekiert, J. Krzek, P. Talik, Chromatographic and electrophoretic techniques used in the analysis of triazole antifungal agents-a review. Talanta, 82 (2010) 1090-1100.
[22] H. Kim, P. Kumari, M. Laughlin, M.J. Hilbert, S.R. Indelicato, J. Lim, C.C. Lin, A.A. Nomeir, Use of high-performance liquid chromatographic and microbiological analyses for evaluating the presence or absence of active metabolites of the antifungal posaconazole in human plasma. Journal of Chromatography A, 987 (2003) 243-248.
161
[23] C. Muller, M. Arndt, C. Queckenberg, O.A. Cornely, M. Theisohn, HPLC analysis of the antifungal agent posaconazole in patients with haematological diseases. Mycoses, 49 (2006) 17-22.
[24] S. Chhun, E. Rey, A. Tran, O. Lortholary, G. Pons, V. Jullien, Simultaneous quantification of voriconazole and posaconazole in human plasma by high-performance liquid chromatography with ultra-violet detection. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, 852 (2007) 223-228.
[25] L. Baietto, A. D'Avolio, G. Ventimiglia, F.G. De Rosa, M. Siccardi, M. Simiele, M.
Sciandra, G. Di Perri, Development, Validation, and Routine Application of a High-Performance Liquid Chromatography Method Coupled with a Single Mass Detector for Quantification of Itraconazole, Voriconazole, and Posaconazole in Human Plasma. Antimicrobial Agents and Chemotherapy, 54 (2010) 3408-3413.
[26] W. Neubauer, A. Konig, R. Bolek, R. Trittler, M. Engelhardt, M. Jung, K.
Kummerer, Determination of the antifungal agent posaconazole in human serum by HPLC with parallel column-switching technique. Journal of Chromatography B-Analytical Technologies in the Biomedical and Life Sciences, 877 (2009) 2493-2498.
162
[27] J.M. Cunliffe, C.F. Noren, R.N. Hayes, R.P. Clement, J.X. Shen, A high-throughput LC-MS/MS method for the quantitation of posaconazole in human plasma:
Implementing fused core silica liquid chromatography. Journal of Pharmaceutical and Biomedical Analysis, 50 (2009) 46-52.
[28] L.A. Decosterd, B. Rochat, B. Pesse, T. Mercier, F. Tissot, N. Widmer, J. Bille, T.
Calandra, B. Zanolari, O. Marchetti, Multiplex Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry Method for Simultaneous Quantification in Human Plasma of Fluconazole, Itraconazole, Hydroxyitraconazole, Posaconazole, Voriconazole, Voriconazole-N-Oxide, Anidulafungin, and Caspofungin. Antimicrobial
Calandra, B. Zanolari, O. Marchetti, Multiplex Ultra-Performance Liquid Chromatography-Tandem Mass Spectrometry Method for Simultaneous Quantification in Human Plasma of Fluconazole, Itraconazole, Hydroxyitraconazole, Posaconazole, Voriconazole, Voriconazole-N-Oxide, Anidulafungin, and Caspofungin. Antimicrobial