An isotope dilution-liquid chromatography-tandem mass spectrometry (ID-LC-MS/MS)-based candidate reference measurement procedure for the quantification of carbamazepine in human serum and plasma

Chemicals and reagents

LC-MS grade methanol (CAS 67-56-1) and formic acid were (CAS 64-18-6) purchased from Biosolve (Valkenswaard, The Netherlands). CD₃OD (CAS 811-98-3, ACS reagent, ≥99.95 %), ammonium acetate (CAS 631-61-8, LC/MS grade), carbamazepine (CAS 298-46-4, Art. No. PHR1067, Lot No. LRAC1961) and tecnazene (CAS 117-18-0, Art. No. 40384, Lot No. BCBW6288) were purchased from Sigma Aldrich (Taufkirchen, Germany). Isopropanol (HPLC grade) was bought from Riedel-de Haën (Seelze, Germany). The internal standard (ISTD) [²H₈]-carbamazepine (CAS 1538624-35-9, Art. No. TRCC175844-1mg, Lot No. 24-MVI-44-1, Purity 96 %) was purchased from Brunschwig AG (Basel, Switzerland) and 10,11-dihydro-10,11-dihydroxycarbamazepine (CAS 58955-93-4, Art. No. D449040, Lot No. 1-NWW-177-2) was bought from TRC Canada (North York, Canada). Carbamazepine-10,11-epoxide (CAS 36507-30-9, Art. No. MM0076.01, Lot Nr. W988078) was bought from LGC Mikromol (Teddington, UK). 10,11-dihydro-10-hydroxycarbamazepine (CAS 29331-92-8, Art. No. 18467, Lot No. 0489614) was purchased from Cayman (Ann Arbor, Michigan US). 5 mm NMR tubes were obtained from Eurisotop GmbH (Hadfield, United Kingdom). Native human serum (Art. No. S1-Liter) was obtained from Merck (Darmstadt, Germany), TDM-free human serum (multi-individual pooled; surrogate matrix, ID No. 12095432001) from Roche Diagnostics GmbH (Mannheim, Germany). Native plasma matrix (Li-heparin, K₂-EDTA and K₃-EDTA) was obtained from anonymized, leftover patient samples and pools were prepared in accordance with the Declaration of Helsinki. Water was purified using a Millipore Milli Q 3 UV system from Merck (Darmstadt, Germany).

General requirements for laboratory equipment

Equipment was certified and calibrated by ISO-accredited calibration laboratories. The minimum sample weight for the microbalance used (XPR2, Mettler Toledo, Columbus, OH, USA) was determined according to the United States Pharmacopeial Convention (USP) guidelines (USP Chapters 41 and 1251). Direct displacement pipettes were used to measure organic solvents and serum. Volumetric glassware (Class A volumetric flasks) that fulfilled the criteria of ISO 1042 and USP were used to prepare stock and spike solutions.

qNMR for determination of the purity of the standard materials

qNMR experiments were performed on a Jeol 600 MHz NMR (Jeol Ltd, Tokyo, Japan) equipped with a Helium-cryoprobe head. For carbamazepine, single-pulse-¹H{¹³C}NMR was utilized for the quantitation (olefinic protons; δ=6.96 ppm; 2H; tecnazene as qNMR internal standard; CD₃OD as solvent) with an inter-scan delay of 70 s (Supplemental Figure 1). These protons were selected as the quantifiable resonance for carbamazepine because they are clearly separated, by δ=1 ppm, from the Dibenz[b,f]azepine which is commonly utilized for the synthesis of carbamazepine [32], therefore, even the hypothetical presence of Dibenz[b,f]azepine does not interfere with our qNMR methodology. Six individual experiments involving six individual weightings of analyte and internal standard were performed for the quantification of the absolute content of carbamazepine.

Further details about NMR acquisition and processing parameters are available in the Supplemental Material 2.

Preparation of calibrators and quality control samples

Two independent calibrator stock solutions were prepared by weighing 25 mg of carbamazepine in tin boats on a microbalance (XPR2, Mettler Toledo, Columbus, Ohio, USA) and dissolving in 5 mL DMSO using volumetric flasks. The concentrations of the stock solutions were calculated considering the purity of the reference material determined by qNMR (99.6 ± 0.2 %) and the amount weighed in, resulting in stock solutions with concentrations of 5.00 mg/mL. These stock solutions were further diluted with DMSO to prepare working solutions, which were used together with the stock solutions to prepare eight calibrator spike solutions. Final matrix-based calibrators, uniformly distributed from 0.800 to 18.0 μg/mL (3.39 to 76.2 μmol/L), were then prepared by a volumetric 1+99 dilution (v/v) of the calibrator spike solutions in human serum matrix.

Four levels of matrix-based quality control (QC) levels were prepared in the same manner as described for the calibrator levels using a third independent stock solution. The concentrations for the control levels were set at four critical control points: above the lower limit of quantitation, below and within the therapeutic reference range and at the laboratory alert level. The final concentrations of the QC levels were 1.20, 3.60, 8.00 and 16.0 μg/mL.

ISTD stock solution

For the preparation of the [²H₈]-carbamazepine ISTD stock solution, 1 mg was dissolved by volumetric addition of DMSO to obtain an ISTD stock solution with a concentration of 1,000 μg/mL. The ISTD stock solution was stored at −20 °C until further use. An ISTD working solution was freshly prepared by a two-fold dilution of the ISTD stock solution prior each sample preparation. Therefore, 168 µL of DMSO was mixed with 32 µL of [²H₈]-carbamazepine ISTD stock solution, followed by the addition of 7,800 µL Milli-Q water to obtain a final concentration of 4.00 μg/mL.

Sample preparation

Native human serum, TDM-free serum (surrogate serum), and plasma (Li-heparin, K₂-EDTA, and K₃-EDTA) were used as sample matrix. 100 µL of the ISTD working solution was pipetted into a 2 mL tube (Eppendorf Safe-Lock Tubes). This was followed by the addition of 50 µL of the sample specimen (native sample/calibrator/QC) was added. For protein precipitation, 1,000 µL 75 % methanol (v/v) was added. After centrifugation, a 50 µL aliquot of the supernatant was diluted 1+19 (v/v) with Milli-Q-water. This diluted solution was then further diluted with Milli-Q water in a second dilution step (1+39 [v/v]).

Liquid chromatography mass spectrometry

Chromatographic separation was performed on an Agilent 1290 Infinity II LC system (Santa Clara, CA, USA) equipped with a binary pump, a vacuum degasser, an autosampler at 7 °C, and a column compartment kept at 40 °C.

Carbamazepine and its main metabolite carbamazepine-10,11-epoxide as well as secondary metabolites (10,11-dihydro-10-hydroxycarbamazepine and 10,11-dihydro-10,11-dihydroxycarbamazepine) were separated using an Agilent Zorbax Eclipse XDB-C18 column (100 × 3 mm, 3.5 µm, Santa Clara, CA, USA). The mobile phases were composed of 2 mM ammonium acetate in Milli-Q water with 0.1 % formic acid (A) and methanol/2 mM ammonium acetate in Milli-Q-water 95+5 (v/v) with 0.1 % formic acid (B). The measurements were performed using a gradient with a total run time of 9 min at a flow rate of 0.6 mL/min. Contamination of the mass spectrometer was reduced using a divert valve, switching the eluent flow until 0.5 min and from 6.0 min to the waste.

Carbamazepine was detected in multiple reaction monitoring (MRM) mode using either an AB Sciex Triple Quad 6500+ (for method comparison study) or a Q-Trap 6500+ mass spectrometer (for validation experiments) (Framingham, Massachusetts, USA) with a Turbo V ion source operating in positive electrospray ionization mode (ESI + mode). For both systems the same settings were used. An ion spray voltage of 5,000 V and a temperature of 400 °C were applied. Nitrogen was used as curtain gas, collision gas, ion gas source 1, and ion gas source 2 and was set at 45, 11, 50, and 60 psi, respectively. For all analytes the collision cell entrance potential was set at 10 V and the collision cell exit potential at 15 V. A second specific mass transition (qualifier) was monitored to exclude interfering substances in native matrix samples by comparing the quantifier/qualifier ratios of neat System Suitability Test (SST) and native matrix samples. The ratios should not differ by more than 20 %. Table 1 provides an overview of the single reaction monitoring (SRM) transitions and MS settings.

System suitability test

Prior to each analysis, an SST was performed to check the sensitivity of the system, chromatographic performance, and potential carry-over. Data may only be collected if the SST has been passed. Therefore, a SST stock solution (1,000 mg/mL carbamazepine in DMSO) was freshly diluted with mobile phase A to get two samples (SST sample 1 and 2) with concentrations corresponding to the processed calibrator level 1 and 8. To pass the SST, signal-to-noise of the quantifier transition had to be ≥10 for SST sample 1. The retention time of both SST samples had to be within 3.9 ± 0.5 min. Moreover, the carry-over effect was examined by injecting SST sample 2 followed by the injection of two solvent blanks. The analyte peak area observed in the blank after the SST sample 2 injection had to be ≤20 % of the analyte peak area of the calibrator 1 level.

Calibration, structure of analytical series and data processing

The final calibration function was generated by measuring the calibration levels in increasing concentration at the beginning and the end of the analytical series. Carbamazepine was calibrated using a linear regression model (y=ax + b) by plotting the analyte and ISTD area ratios (y) against the analyte concentration (x). Data processing was performed using Analyst^® software (version 1.6.3 or higher) with the IntelliQuant algorithm. Carbamazepine and [²H₈]-carbamazepine peaks showed a retention time of 3.9 min and were integrated within a 30 s window. Peak integration included a smoothing factor of 7 and a peak and splitting factor of 5. The noise percentage was set to 80 % with a base sub-window of 0.8 min.

Method validation

Assay validation and determination of measurement uncertainty were performed based to existing validation guidelines such as Clinical & Laboratory Standards Institute’s C62A Liquid Chromatography-Mass Spectrometry Methods [33], the ICH guidance document Harmonised Tripartite Guideline Validation of Analytical Procedures: Text and Methodology Q2 (R1) [34] and the GUM [35]. Further details are published in Taibon et al. [36].

Traceability to SI units

Traceability to the SI unit of mass (kilogram), the most important parameter for a reference measurement method, has been established by the utilization of tecnazene as a qNMR internal standard. Tecnazene is a TraceCert CRM qNMR internal standard produced by SigmaAldrich which is directly traceable to the NIST PS1 (primary qNMR standard developed by NIST) as per the certificate of analysis attached with specific batch numbers. Furthermore, traceability to the SI unit of amount of substance (mole) is also included owing to NIST PS1, clearly mentioned by NIST in their publication about NIST PS1 [29]. Since Planckʼs constant is now the main parameter for defining kilogram and Avogadro’s constant for mole, qNMR methodology is best suited to fulfill both the traceability chains. Additionally, as per the latest IUPAC technical report qNMR has been stated as a potential primary reference measurement procedure ideally suited for the characterization of primary reference materials [30]. As per our aforementioned qNMR methodology, six individual experiments (Supplemental Figure 2) involving six individual weightings, yield a final target absolute content value of 99.6 ± 0.2 % (k=1; Supplemental Table 1 and Figure 2). The expanded uncertainty, k=2, can be obtained by multiplying the above mentioned uncertainty with a coverage factor of 2.

Selectivity

Baseline separation of carbamazepine (eluting at 3.8 min) and its enzymatically mono- and dioxidated metabolites carbamazepine-10,11-epoxide, 10,11-dihydro-10-hydroxycarbamazepine, and 10,11-dihydro-10,11-dihydroxycarbamazepine was achieved on a reversed phase column (Agilent Zorbax Eclipse XDB-C18) showing resolutions greater than 10.0 in all matrices tested (see Figure 1). Moreover, the chromatogram of different matrices (pools of analyte-free native human serum, surrogate serum, and human Li-heparin plasma) showed no signals at the expected retention time of carbamazepine. The tested ISTD did not contain any unlabeled analyte and is therefore suitable for the intended use (see Figure 2).

Figure 1:

LC-MS/MS read outs of a native human serum sample containing 4.00 μg/mL of each carbamazepine (peak 1), carb ama zepine-10,11-epoxide (peak 2), 10,11-dihydro-10-hydroxycarbamazepine (peak 3), and 10,11-dihydro-10,11-dihydroxycarbamazepine (peak 4). SRM ion traces of (A) carbamazepine (peak 1); peak 5 demonstrates the in-source dehydration of 10,11-dihydro-10-hydroxycarbamazepine to carbamazepine. (B) Carbamazepine-10,11-epoxide (peak 2); peak 6 shows in-source dehydration of 10,11-dihydro-10,11-Dihydroxycarbamazepine to oxcarbazepine. (C) 10,11-Dihydro-10-hydroxycarbamazepine. (D) 10,11-Dihydro-10,11-dihydroxycarbamazepine.

Figure 2:

Carbamazepine LC-MS/MS derived analytical readouts. (A) Chromatogram of a matrix blank showing the analyte SRM ion trace (left) and the ISTD SRM ion trace (right). (B) Chromatogram of the lowest calibrator level with a concentration of 0.800 μg/mL carbamazepine spiked in native serum; analyte (left) and ISTD (right). (C) Patient sample with a concentration of 1.97 μg/mL, showing peaks of carbamazepine (peak 2) and the in-source dehydration of 10,11-dihydro-10-hydroxycarbamazepine to carbamazepine (peak 1); analyte (left) and ISTD (right). ISTD, internal standard.

When measuring patient samples, the carbamazepine ion traces show additional peaks at the retention time of the metabolite 10,11-dihydro-10-hydroxycarbamazepine (Figure 1). This signal can be attributed to the in-source water loss of 10,11-dihydro-10-hydroxycarbamazepine to carbamazepine (Figure 3) as previously reported in GC-MS and LC-MS [23, 41, 42]. If not separated chromatographically from carbamazepine, the contribution of this instrumental artefact would lead to an overestimation of the carbamazepine sample concentration. Hence, chromatographic separation of carbamazepine from its oxidized metabolite 10,11-dihydro-10-hydroxycarbamazepine is a necessity of any diagnostic method and consequently also an RMP to guarantee an accurate quantification of carbamazepine (Figure 2).

Figure 3:

Schematic overview of the in-source dehydration of 10,11-dihydro-10-hydroxycarbamazepine to carbamazepine.

Matrix effect and specificity

Matrix dependent effects were avoided by a sample preparation protocol with a high dilution after protein precipitation. This was demonstrated by the post-column infusion experiments as no change in the ionization field was observed at the expected retention times in native human serum, surrogate serum, and native plasma matrices (Li-heparin, K₂-EDTA and K₃-EDTA plasma).

In addition, MEs were investigated by comparing slopes and coefficients of determination of calibration curves in different matrices. The carbamazepine calibration showed slopes (n=6, sample preparations) of 0.334 (95% CI 0.331 to 0.338) for neat solution, 0.337 (95% CI 0.335 to 0.339) for native serum, 0.336 (95% CI 0.335 to 0.338) for surrogate serum, and 0.338 (95% CI 0.336 to 0.340) for native Li-heparin plasma. The coefficients of determination were 1.00 regardless of the matrix used for calibration. The 95% CIs of the slopes overlap, which leads to the assumption that they are not significantly different. Furthermore, matrix-based samples were evaluated against neat calibrators as standards. Relative bias for all matrices ranged from −1.6 to 1.5 % (n=6, sample preparations) with CVs of ≤2.9 %.

The absence of any MEs was further proven comparing analyte peak areas, ISTD peak areas as well as area ratios of spiked serum samples with spiked neat samples. Mean peak areas ranged from 96 to 100 % for carbamazepine and [²H₈]-carbamazepine in all tested matrices. The corresponding area ratios were between 98 and 101 % (Table 2). Hence, the absence of a ME can be confirmed.

Linearity

Linearity was demonstrated by analyzing six native serum calibration curves with an extended measuring range of 20 % at both lower and higher concentrations, resulting in a measuring range from 0.640 to 21.6 μg/mL. The residuals of carbamazepine were randomly distributed in a linear regression model and the correlation coefficient was 1.00 for all individual calibration curves.

Based on the selected regression model, the dilution series was evaluated and showed a linear dependence with a correlation coefficient of 1.00. The relative deviation (n=6, sample preparations) ranged from −1.2 to 3.2 % with a CV of ≤1.7 %.

Lower limit of the measuring interval and limit of detection

The LLMI was determined using samples spiked with the concentration of the lowest calibrator concentration (0.800 μg/mL). The relative bias showed a deviation of −0.4 % and the CV was 1.0 %. The LOD of carbamazepine was estimated to be 0.115 μg/mL.

Precision and accuracy

Precision was evaluated in a multi-day validation experiment using spiked (low, mid, high) and native samples in the therapeutic range. Two different operators prepared each level in triplicate on one day over five different days and the samples were injected twice (n=60, measurements). To assess the overall variability of the methods variability components as between injections, between preparations, between calibrations, and between days were determined using an ANOVA-based variance component analysis. Intermediate precision CV was found to be ≤1.6 % and repeatability CV ranged from 0.8 to 1.3 % across all concentration levels and the measurement variability of patient samples was found to be indistinguishable from spiked levels (Table 3).

Trueness was demonstrated using spiked serum samples and two high-concentrated native human serum samples (20.0 and 24.0 μg/mL) which were diluted with native human serum prior to the sample preparation. The mean bias (n=6, sample preparations, two operators) of native human serum samples ranged from −0.1 to 0.6 %. Li-heparin plasma samples showed a mean bias between −0.3 and −0.1 %. The mean bias of diluted samples (dilution 1 and 2) ranged from 0.4 to 1.9 % (Table 4).

Stability

Stability of processed samples at 7 °C was verified for 6 days. The recoveries of processed carbamazepine samples ranged from 101 to 105 %. The stability of spiked calibrator and QC samples stored at −20 °C was shown for 34 days. The recoveries ranged from 100 to 105 %.

Equivalence of results between the JCTLM listed RMP and the candidate RMP

Equivalence to the already existing RMP from Taibon et al. [26] was proven performing a method comparison study with a total of 126 samples, including 62 native serum, 26 native plasma patient samples, 8 native patient pools, and 30 spiked native patient samples. All samples were analyzed at site 1 using the JCTLM listed RMP whereas the newly developed candidate RMP was performed at site 2. The sample preparation of Taibon et al. had to be slightly adjusted due to insufficient sample volume: all volumes used during sample preparation were reduced to three quarters of the original volumes. Two samples were highlighted as outliers using the LORELIA (local reliability) outlier test [43]. These samples were therefore not considered for the method comparison and n=124 samples were evaluated. Passing–Bablok regression analysis showed a good agreement between the JCTLM listed RMP and the candidate RMP and resulted in a regression equation with a slope of 1.01 (95% CI 1.00 to 1.02) and an intercept of −0.01 (95% CI −0.07 to 0.06) (Figure 4A). The Pearson correlation coefficient was found to be ≥0.996. The Bland Altman plot showed a 2S agreement of 7.4 % (lower limit CI interval from −7.6 to −5.3 %, upper limit CI interval from 7.1 to 9.4 %) (Figure 4B). The mean bias in the patient cohort was 0.9 % with a 95% CI interval ranging from 0.3 to 1.6 %.

Figure 4:

Results from the patient sample based carbamazepine method comparison study of the JCTLM listed and the modernized RMP. (A) Passing–Bablok regression plot including the Pearson regression analysis for the method comparison study of the RMPs (n=124 samples) resulted in a regression equation with a slope of 1.01 (95 % CI 1.00 to 1.02) and an intercept of −0.01 (95 % CI −0.07 to 0.06). The Pearson correlation value was ≥0.999. (B) Bland–Altman plot for the method comparison study of the RMPs (n=126 samples). The bias was 0.9 % (95 % CI interval from 0.3 to 1.6 %) and the 2S interval of the relative difference was 7.4 % (lower limit CI interval from −7.6 to −5.3 %, upper limit CI interval from 7.1 to 9.4 %).

Additionally, a precision experiment was performed at site 2. The resulting CVs of ≤1.3 % and ≤0.9 % for intermediate precision and repeatability are very good comparable to the results from site 1, showing that the method is transferable and meets the requirements independent of the laboratory equipment or personnel.

Uncertainty of results

Total measurement uncertainties of serum samples for carbamazepine for single measurements and for target value assignment (n=6, three measurements on two days) ranged from 1.4 to 1.8 % and from 1.0 to 1.3 %, respectively, independent of the concentration level of the sample (Table 5 and 6). Consequently, the expanded measurement uncertainties ranged from 2.7 to 3.6 % for single measurements and from 1.9 to 2.6 % for multiple measurements thus meeting the requirements defined in the introduction. The derived combined uncertainty is multiplied by a coverage factor of k=2 to obtain an expanded uncertainty that represents a confidence level of 95 %, assuming normal distribution.