Abstract
Objective
To develop and validate an LC‐M/SMS method for the determination of tacrolimus in human whole blood.
Method
The LC–MS/MS method for the determination of tacrolimus in whole blood was developed and validated according to the guidelines. Concentrations of TAC in 100 kidney transplant patients measured by LC–MS/MS were compared with CMIA using correlation analysis and Bland–Altman plots.
Results
The method had a total chromatographic run time of 5 min. The calibration curves were linear over the range of 0.5–100.0 ng/mL with a lower limit of quantification of 1 ng/mL. The intra‐ and interday accuracy was within the range of 93.3%–109.2% and 96.0%–108.4%, respectively, with precision ranging from 0.8 to 9.4%. The mean extraction recoveries of TAC ranged from 102.6 to 107.8%. The mean concentrations of TAC in whole blood of kidney transplant patients measured by the two assays were different at 1, 3 months and all time points (p < 0.001), but no significant difference was observed at 6 months (p = 0.094). The correlation of data was good with the correlation coefficients (r 2) of 0.7581, 0.8811, 0.8777, and 0.8077, respectively. Passing–Bablok regression analysis demonstrated good correlations with r 2 values higher than 0.88 between TAC levels measured by LC–MS/MS and CMIA. Using Bland–Altman plots yielded average biases of 1.29, 0.79, 0.11, and 0.65 ng/mL at 1, 3, and 6 months and all time points.
Conclusion
The LC–MS/MS method was validated for the accurate determination of TAC in human whole blood. The comparison of tacrolimus concentrations measured by the LC–MS/MS with CMIA showed a good correlation and agreement of two methods, suggesting LC–MS/MS should be used routinely to monitor TAC concentrations in kidney transplant patients.
Keywords: CMIA, human whole blood, kidney transplant, LC–MS/MS, tacrolimus
1. INTRODUCTION
Tacrolimus (TAC, FK506) is a macrolide antibiotic with immunosuppressive properties isolated from the culture of Streptomyces. Tacrolimus is a calcineurin inhibitor (CNI); its main pharmacological effect is to effectively inhibit the activation of T lymphocytes and bind to endogenous cell receptors to form an immunophilin complex, thereby exerting pharmacological effects. 1 Compared with cyclosporine (CsA), in vivo and in vitro experiments have confirmed that TAC has 50–100 times the immune activity of CsA in vitro. 2 , 3 It can effectively prevent acute rejection after liver and kidney transplantation and reverse refractory rejection. Currently, TAC has been widely used in the treatment of antirejection and autoimmune system‐related diseases in the liver, kidney, heart, pancreas, and other organ transplants. Tacrolimus is a narrow therapeutic index drug with a high variability of pharmacokinetics among individuals, therapeutic drug monitoring of Tacrolimus is recommended. 4 Controlling the blood level of TAC plays a key role in the clinical follow‐up of transplant recipients. To optimize Tacrolimus therapy, drug's dose adjustment based on monitoring tacrolimus systemic exposure is essential for each patient. In our clinical practice, tacrolimus blood concentrations are optimally maintained within the therapeutic interval of 4–12 ng/mL in kidney transplantation according to guidelines. 5 Therefore, proper clinical analytical tools are needed.
In current clinical practice, a number of assay methods are used to monitor TAC, for example, immunoassays (MEIA, ELISA, and CMIA), and liquid chromatography–tandem mass spectrometry (LC–MS/MS). It has been reported that the concentration of TAC in the blood estimated by immunoassays may be higher than the corresponding values determined by LC–MS/MS. 6 , 7 The higher levels obtained using TAC immunoassays may be owed to cross‐reactivity with some metabolites of parent drugs. 8 The LC–MS/MS method would be specific to the parent compound while maintaining the accuracy and precision of immunoassays. However, this skilled method is difficult to implement in clinical practice because it requires personnel expertise and high‐cost analytical equipment. 9 Owing to automation and easy operation in clinical settings, CMIA is more commonly used to monitor tacrolimus, although it may be less specific in certain conditions.
This study intends to establish a fast, sensitive, and specific LC–MS/MS method to determine the concentration of TAC in whole blood in comparison with CMIA method using specimens from kidney transplant patients.
2. MATERIALS AND METHODS
2.1. Chemicals and reagents
LC–MS grade methanol, acetonitrile, ammonium acetate, and zinc sulfate heptahydrate were purchased from Merck (Darmstadt, Germany). TAC (QT193031186; 98,6%) and Roxithromycin (QT159060618; 95,5%) (internal standard) were obtained from the National Institute of Drug Quality Control (Hanoi, Vietnam).
2.2. Instrumentation
CMIA procedure was performed according to the manufacturer's instruction (Abbott Laboratories, Abbott Park, IL, USA). LC–MS/MS analysis of TAC was conducted using a Waters Xevo TQD liquid chromatography‐tandem mass spectrometer, equipped with an electrospray ionization source (ESI) and Waters Empower™ chromatography software or MassLynx mass spectrometry software, which controls the system.
2.3. LC–MS/MS conditions
Quantification was analyzed using MRM (Multiple Reaction Monitoring) of the transitions of m/z 821.5 → 768.5 for TAC and m/z 837.8 → 158.2 for Roxithromycin in positive ion mode under the conditions as follows.
2.3.1. LC conditions
Tacrolimus and Roxithromycin were separated using gradient elution on an ACQUITY UPLC® BHE C18 column (2.1 mm x 100 mm; 1.7 μm) kept at 25°C. The mobile phase consisted of solvent A (Acetonitrile containing 1% formic acid) and solvent B (Ammonium acetate 5 mM). The gradient started at 60% solvent A and 40% solvent B for 1.25 min, then changed to 98% solvent A and 2% solvent B for 0.01 min, and maintained for another 1.74 min. At 3.00 min, the mobile phase conditions were changed back to 60% solvent A and 40% solvent B for 0.01 min and maintained for another 1.99 min.
2.3.2. MS/MS conditions
The MS/MS conditions of analysis were as follows: Capillary voltage of 3.25 kV, cone voltage: 55 V. The desolvation temperature was set at 350°C, and desolvation gas flow and cone gas were 800 L/h and 10 L/h, respectively.
2.4. Preparation of standards and quality control samples
Stock solutions of TAC and Roxithromycin were accurately prepared in acetonitrile and methanol at 250 μg/mL and then stored at −70°C. The calibration standards were prepared by diluting stock solutions in human whole blood to concentrations of 0,5 ng/mL, 1; 5, 10; 25; 50; 75, and 100 ng/mL. Four quality control (QC) samples were prepared at 5; 10; 50; and 75 ng/mL.
2.5. Sample pretreatment
Each blood sample (100 μL) combined with 10 μL of the working solution of the internal standard (500 ng/mL) was precipitated using 200 μL of methanol/1.125 M ZnSO4 in water (66/34, v/v). The mixture was vortexed for 30 s and left for 5 min at room temperature, followed by centrifugation at 13,000 rpm for 10 min at 4°C. A supernatant of 5 μL was injected into the LC–MS/MS instruments using an autosampler operating at 20°C.
2.6. Method validation
The method development and validation were performed in accordance with the guidelines. 10 , 11 Method specificity was assessed by analyzing six different blank whole‐blood samples. The acceptance criteria were that at least five of six lots should have ≤20% area response in comparison to the mean response of the lower limit of quantification (LLOQ) sample for each matrix. The lower limit of quantification (LLOQ) was evaluated singularly in six separate batches with a nominal concentration of 1.0 ng/mL. Linearity was determined by analyzing five different calibration curves with concentrations ranging from 0.5 to 100 ng/mL on five separate days. The precision and accuracy were evaluated by analyzing six replicates of each QC sample on a single day and three replicates of each QC sample over five different days. Extraction recovery was determined by comparing the peak areas of extracted QC samples in whole blood with equivalent concentrations in the mobile phase.
The matrix effect (ME) was calculated using low‐quality control (LQC) and (high‐quality control) HQC samples by comparing the peak areas of two QC samples in whole blood from healthy subjects with those at the same concentration in the mobile phase. Six different lots of whole blood from healthy subjects were used to assess the matrix effects, and the ratio (peak areas in whole‐blood/peak areas in mobile phase) was defined as the matrix factor (MF) for investigating ME.
The stability of TAC in human whole blood was assessed by analyzing the LQC and HQC samples using six replicates during a variety of storage and process conditions. The short‐term stability was determined by analyzing these QC samples after being stored at room temperature for 5 h. The freeze and thaw conditions include three freezes at 70°C and a thaw at 37°C cycles. The postpreparative stability was assessed by analyzing these samples after protein precipitation and keeping them in the autosampler for 24 h. Long‐term stability was evaluated at −70°C for 2 months. The acceptable criteria for stability were for the measured concentration to be 15% of the known concentration.
2.7. Method comparison
A total of 100 kidney transplant patients at Military Hospital 103, Vietnam Military Medical University (Hanoi, Vietnam) treated by a standard immunosuppressive regimen consisted of tacrolimus, mycophenolate mofetil, and prednisolone with bodyweight‐based tacrolimus initial dose of 0.1 mg/kg/d and then adjusted dose to achieve a therapeutic target trough level of 4–12 ng/mL after transplantation according to the current guideline. 5 Clinical and laboratory data are presented in Table 1. Concentrations of TAC in whole blood were analyzed by chemiluminescent microparticle immunoassay (CMIA) on Architect i2000 system (Abbott Diagnostics Laboratories, Abbott, United States) for routine therapeutic drug monitoring and residual whole blood samples were stored in ‐70°C until use for LC–MS/MS analysis. The drug‐free whole blood samples from healthy donors used in this study were provided Military Hospital 103, Vietnam Military Medical University. The study protocol was approved by the local ethics committee (protocol code: 177 and date of approval: 25 November 2022).
TABLE 1.
Clinical and laboratory data of renal transplant recipients (n = 100).
| Characteristics | Value |
|---|---|
| Age (year, mean ± SD) (Min–Max) | 40.02 ± 9.18 (23–65) |
| Gender, n (%) | |
| Male | 72 (72.0%) |
| Female | 28 (28.0%) |
| Type of donor n (%) | |
| Living related | 15 (15.0%) |
| Living unrelated | 85 (85.0%) |
| Acute rejection, n (%) | |
| Yes | 2 (2.0%) |
| No | 98 (98.0%) |
| HLA mismatch (Mean ± SD) (Min–Max) | 3.12 ± 1.13 (0–5) |
| Height (cm), mean ± SD (Min–Max) | 162.96 ± 7.08 (145–180) |
| Body weight (kg), mean ± SD (Min–Max) | 54.45 ± 9.58 (36–89) |
| BMI (kg/m2), mean ± SD (Min–Max) | 20.31 ± 2.65 (15.00–29.06) |
| Creatinine at month 1 (μmol/L), mean ± SD (Min–Max) | 102.48 ± 33.11 (54.23–245.10) |
| Creatinine at month 3 (μmol/L), mean ± SD (Min–Max) | 103.31 ± 23.36 (60.65–200.98) |
| Creatinine at month 6 (μmol/L), mean ± SD (Min–Max) | 103.84 ± 24.98 (73.51–231.94) |
Abbreviations: BMI, body mass index; HLA, human leukocyte antigen; SD, standard deviation.
2.8. Statistical analyses
All data were analyzed using with Excel (Microsoft), SPSS 22.0 (IBM) and GraphPad Prism 8.0 (GraphPad software). The nonparametric Mann–Whitney U test was used to compare mean concentrations obtained with both assays. Passing–Bablok regression, Pearson's correlation analysis, and linear regression were used to assess the overall correlation of two methods. Bland–Altman plots were used to evaluate the agreement between both assays. 12 , 13 , 14 p values of <0.05 were statistically significant.
3. RESULTS
3.1. Method optimization
To optimize mass spectrometry conditions, TAC standard solution at a concentration of 100 ng/mL was injected directly into the mass spectrometer. ESI (+) was selected and the parent ion of [M+H]+ was fragmented to obtain the daughter ion. For quantification, the daughter ion with the highest intensity was selected. Mass parameters were determined using Water's Intellistart software and are shown in Table 2. The retention times of TAC and IS were 4.13 and 4.12 min, respectively.
TABLE 2.
Mass parameters of tacrolimus and IS.
| MS/MS conditions | TAC | IS |
|---|---|---|
| Mode of analysis | ESI (+) | ESI (+) |
| Capillary voltage (kV) | 3.25 | 3.5 |
| Cone voltage (V) | 55 | 55 |
| Desolvation temperature (°C) | 350 | 350 |
| Desolvation gas (L/H) | 800 | 800 |
| Cone gas (L/H) | 10 | 20 |
| Collision energy (V) | 22 | 30 |
| Parent ion (Dalton) | 821.5 | 837.8 |
| Product ion (Dalton) | 768.5 | 158.2 |
Abbreviations: ESI, electrospray ionization source; IS, internal standard; TAC, Tacrolimus.
3.2. Specificity
The chromatograms of the blank whole blood samples, the TAC standard and IS in whole blood samples are shown in Table 3. Figure 1 shows that peak of TAC and IS were well separated with no significant direct interfering peaks from endogenous matrix components being observed at the retention times of TAC. The response of peak area ratios of TAC and IS was less than 20% and 5% compared with the blank sample, which means that the method meets the acceptance criteria of selectivity and specificity, respectively.
TABLE 3.
Specificity and lower limit of quantification of tacrolimus.
| Replicate no | BL/LLOQ | BL/IS | Accuracy (%) |
|---|---|---|---|
| 1 | 0.1248 | 0.0018 | 100.0 |
| 2 | 0.1222 | 0.0013 | 100.0 |
| 3 | 0.1407 | 0.0019 | 100.0 |
| 4 | 0.1302 | 0.0017 | 100.0 |
| 5 | 0.1407 | 0.0023 | 90.0 |
| 6 | 0.1558 | 0.0022 | 90.0 |
| Mean | 0.1357 | 0.0018 | 96.7 |
| CV% | 5.3 |
Abbreviations: BL, Blank; CV, Coefficient of variation; IS, Internal standard; LLOQ, Lower limit of quantification.
FIGURE 1.
Chromatograms of tacrolimus and IS in whole blood.
3.3. Lower limit of quantification
The results in Table 3 show that the response of peak area ratios in the blank sample was <20% compared with the LLOQ sample, which means that the response of peak area ratios in the LLOQ sample was five times greater than that in the blank sample. At a concentration of 1 ng/mL, the method meets the acceptance criteria of accuracy (97% of the actual concentration) and precision (CV: 5.3%). Therefore, this concentration meets the requirement for analysis of TAC.
3.4. Linearity
The calibration curve was calculated using the response of peak area ratios (y) versus TAC concentration (x) and regressed with a linear 1/x2 weighted relation. Figure 2 shows that eight‐point calibration curves of TAC were linear over a concentration range from 0.5 to 100 ng/mL according to the regression equation of the form Y = aX + b with R 2 > 0.99. As shown in Table 4, the TAC concentration back‐calculated from the equation of the regression analysis was in the range of 92,2%‐ 109,5% and within the acceptance criteria of 85%–115% and 80%–120% for LLOQ. These results show that the calibration curve meets the acceptance criteria for the generation of acceptable data for the concentrations of the analyte in the samples during the validation for TAC in whole blood.
FIGURE 2.
Calibration curves of tacrolimus in whole blood.
TABLE 4.
Summary of calibration curves with back‐calculated concentration of tacrolimus.
| CC | CC1 | CC2 | CC3 | CC4 | CC5 | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| NC (ng/mL) | Accuracy (%) | NC (ng/mL) | Accuracy (%) | NC (ng/mL) | Accuracy (%) | NC (ng/mL) | Accuracy (%) | NC (ng/mL) | Accuracy (%) | |
| SC1 | 0.6 | 102.7 | 0.5 | 95.2 | 0.6 | 100.1 | 0.6 | 99.4 | 0.5 | 101.3 |
| SC2 | 1.0 | 95.4 | 0.9 | 109.5 | 1.0 | 99.1 | 1.0 | 100.8 | 0.9 | 97.3 |
| SC3 | 4.7 | 102.1 | 5.1 | 95.8 | 5.1 | 103.7 | 5.4 | 104.7 | 5.0 | 102.8 |
| SC4 | 9.4 | 95.0 | 9.9 | 96.9 | 10.2 | 98.4 | 11.0 | 92.2 | 10.5 | 98.6 |
| SC5 | 23.7 | 99.3 | 26.3 | 100.0 | 26.7 | 103.0 | 27.8 | 100.6 | 25.7 | 102.2 |
| SC6 | 48.6 | 103.0 | 52.4 | 102.7 | 52.9 | 101.2 | 58.7 | 100.4 | 55.9 | 100.0 |
| SC7 | 74.3 | 101.3 | 82.6 | 100.9 | 82.9 | 97.3 | 94.8 | 104.1 | 87.0 | 97.8 |
| SC8 | 113.3 | 101.1 | 114.7 | 98.9 | 115.6 | 97.2 | 130.7 | 97.7 | 124.6 | 100.1 |
| Slope | 0.0451 | 0.0399 | 0.0403 | 0.0436 | 0.0355 | |||||
| Intercept | 0.0107 | 0.0066 | 0.0007 | 0.0027 | 0.0006 | |||||
| r 2 | 0.998 | 0.999 | 1.000 | 0.999 | 1.000 | |||||
Abbreviations: CC, Calibration curve; NC, Nominal concentration; SC, Standard curve.
3.5. Intra‐ and inter‐day precision and accuracy
The accuracy and precision results are presented in Table 5. The intra‐ and interday accuracy was within the range of 93.3%–109.2% and 94.1%–108.4%, respectively, and within the acceptance criteria of 85%–115.%. The intraday and interday precision met the acceptance criteria with a CV ranging from 0.8 to 7.5%. Thus, the method meets the standards for the analysis of TAC in biological fluids.
TABLE 5.
Intra‐ and interday precision and accuracy of tacrolimus.
| Quality control (ng/mL) | n | Mean ± SD | CV (%) | Accuracy (%) |
|---|---|---|---|---|
| Intraday precision and accuracy | ||||
| LLOQ | 6 | 0.9 ± 0.1 | 5.5 | 93.3 |
| LQC | 6 | 5.1 ± 0.1 | 2.4 | 101.3 |
| SQC | 6 | 10.0 ± 0.1 | 1.2 | 99.7 |
| MQC | 6 | 52.8 ± 0.4 | 0.8 | 105.6 |
| HQC | 6 | 81.9 ± 0.6 | 0.9 | 109.2 |
| Interday precision and accuracy | ||||
| LLOQ | 6 | 0.9 ± 0.1 | 7.5 | 96 |
| LQC | 6 | 4.9 ± 0.2 | 4.6 | 94.1 |
| SQC | 6 | 9.8 ± 0.3 | 3.7 | 98.5 |
| MQC | 6 | 51.8 ± 2.1 | 3.2 | 98.2 |
| HQC | 6 | 81.3 ± 3.0 | 3.8 | 108.4 |
Abbreviations: BL, Blank; CV, Coefficient of variation; HQC, High‐quality control; LLOQ, Lower limit of quantification; LQC, Low‐quality control; MQC, Medium‐quality control; SD, Standard Deviation; SQC, Supplement quality control.
3.6. Recovery and matrix effect
The results of the recovery in Table 6 showed that the CV% of each concentration calculated according to the response of the peak area of TAC was less than 15%. The mean extraction recoveries of TAC were within the acceptable range of 30%–110% with a CV < 15% and this method was appropriate and met the recovery criteria.
TABLE 6.
Recovery and matrix effect of tacrolimus.
| Quality control (ng/mL) | n | CV (%) | Recovery (%) | |
|---|---|---|---|---|
| In solvent | In plasma | |||
| Recovery | ||||
| LQC | 6 | 2.4 | 4.6 | 106.6 |
| SQC | 6 | 2.1 | 1.5 | 104.0 |
| MQC | 6 | 1.4 | 3.1 | 107.8 |
| HQC | 6 | 5.3 | 2.1 | 108.1 |
| IS | 6 | 4.2 | 4.9 | 103.0 |
| Matrix effect | ||||
| LQC | 6 | 5.4 | 99.2 | |
| HQC | 6 | 7.4 | 108.6 | |
Abbreviations: CV, Coefficient of variation; HQC, High‐quality control; LQC, Low‐quality control; MQC, Medium quality control; SQC, Supplement quality control.
The matrix value effect in Table 6 was assessed by using chromatographically screened human whole blood. The CV for LQC and HQC samples observed was 5.4% and 7.4%, whereas the accuracy found at these levels was 99.2% and 108.6%, respectively. These results indicate that the matrix effects were acceptable and would not impact the reproducibility of the analysis.
3.7. Stability
The results in Table 7 show that the CV at room temperature, in the autosampler and after three freeze–thaw cycles of the LQC were 2.5%, 2.4%, and 4.6%, respectively, and of the HQC were 0.9%, 0.9%, and 1.9%, respectively. The CV in these conditions was less than 15% of the acceptance criteria.
TABLE 7.
Stability data of tacrolimus.
| Quality control (ng/mL) | n | Mean ± SD | CV (%) | Accuracy (%) |
|---|---|---|---|---|
| Room temperature | ||||
| LQC | 6 | 4.6 ± 0.1 | 2.5 | 96.2 |
| HQC | 6 | 69.8 ± 0.4 | 0.9 | 91.4 |
| Autosampler | ||||
| LQC | 6 | 4.8 ± 0.1 | 2.4 | 95.4 |
| HQC | 6 | 80.2 ± 0.3 | 0.9 | 97.9 |
| Freeze–thaw stability | ||||
| LQC | 6 | 5.1 ± 0.2 | 4.6 | 103.8 |
| HQC | 6 | 84.8 ± 0.8 | 1.9 | 101.8 |
| Long‐term stability | ||||
| LQC | 6 | 4.9 ± 0.1 | 2.4 | 96.4 |
| HQC | 6 | 83.3 ± 1.6 | 1.9 | 101.7 |
Abbreviations: CV, Coefficient of variation; HQC, High‐quality control; LQC, Low‐quality control; SD, Standard Deviation; SQC, Supplement quality control.
In terms of long‐term stability, the CV of LQC and HQC samples after storage at ‐70°C for 2 months was 2.4% and 1.9%, respectively. These findings show that Tacrolimus was stable in human whole blood at various temperatures and storage conditions.
3.8. Method comparison
We evaluated comparison of TAC concentrations measured by the Architect i2000 assay and data obtained with LC–MS/MS on whole blood samples collected from kidney transplant patients treated with TAC at different time points of 1, 3 and 6 months. As shown in Table 8 and Figure 3, the mean levels of TAC measured by LC–MS/MS were significantly lower than those measured by CMIA at 1, 3 months and at all time points (p < 0.001), whereas there was no significant difference as compared values of TAC determined by two methods at time point of 6 months (p = 0.094). In addition, a good correlation of the two methods was observed using linear regression analysis. The correlation equations of two measurements at 1, 3, 6 months and all time points were y = 0.6921x +1.4481, R 2 = 0.7581; y = 0.8525x +0.4541, R 2 = 0.8811; R 2 = 0.8777, y = 0.8197x +1.5101 and R 2 = 0.8077, y = 0.7615x + 1.3426, respectively. Furthermore, the agreement between two assays was evaluated by Bland–Altman difference plots, showed average biases of 1.29 ng/mL (±1.96 SD: −2.46‐5.04 ng/mL) at 1 month, 0.79 ng/mL (±1.96 SD: −1.38‐2.97 ng/mL) at 3 months, −0.11 ng/mL (±1.96 SD: −2.63‐2.40 ng/mL) at 6 months and 0.65 ng/mL, (±1.96 SD: −2.44‐2.75 ng/mL) at all time points. In addition, the result of Passing–Bablok regression analysis was presented in Table 9. Spearman correlation coefficients were from 0.887 to 0.939 for analyses. No systematic differences between two methods as shown in Figure 4.
TABLE 8.
Comparison of mean TAC concentrations at each time point between CMIA and LC–MS/MS.
| Time point | CMIA, mean ± SD (Min–Max) | LC–MS/MS, mean ± SD (Min–Max) | p |
|---|---|---|---|
| Month 1 | 8.9 ± 3.84 (2.1–21.2) | 7.60 ± 3.05 (2.2–15.3) | <0.001 |
| Month 3 | 8.51 ± 3.2 (3.5–16.7) | 7.72 ± 2.91 (2.9–16.5) | <0.001 |
| Month 6 | 7.75 ± 3.61 (2.6–27.6) | 7.86 ± 3.16 (2.8–26.6) | 0.094 |
| All time points | 8.38 ± 3.58 (2.1–27.6) | 7.73 ± 3.04 (2.2–26.6) | <0.001 |
Abbreviations: CMIA, chemiluminescent microparticle immunoassay; LC–MS/MS, liquid chromatography–tandem mass spectrometry; SD, Standard Deviation.
FIGURE 3.
Comparison of mean TAC C0 concentrations at each time point between CMIA and LC–MS/MS.
TABLE 9.
Regression analysis by Passing–Bablok.
| Time points | Passing–Bablok regression equation | Intercept (95% CI) | Slope (95% CI) | r (p) | Cusum test |
|---|---|---|---|---|---|
| Month 1 | y = 0.158 + 0.830.x | 0.158 (−0.430–0.834) | 0.830 (0.755–0.900) | 0.897 (<0.001) | p = 0.53 |
| Month 3 | y = 0.0900 + 0.900.x | 0.090 (−0.415–0.583) | 0.900 (0.833–0.963) | 0.939 (<0.001) | p = 0.96 |
| Month 6 | y = 0.558 + 0.954.x | 0.558 (−0.081–1.094) | 0.954 (0.875–1.042) | 0.890 (<0.001) | p = 0.70 |
| All time | y = 0.357 + 0.875.x | 0.375 (0.024–0.838) | 0.875 (0.824–0.918) | 0.887 (<0.001) | p = 0.43 |
FIGURE 4.
Passing–Bablok regression analysis for comparing tacrolimus concentrations measured by LC–MS/MS and CMIA methods.
4. DISCUSSION
Several LC–MS/MS methods have been reported for the quantitation of tacrolimus in whole blood by various sample preparation solutions such as methanol in 0.1 M ammonium bicarbonate, 15 ethanol–ethanediol–water, zinc sulfate solution, and acetone, 16 zinc sulfate, and acetonitrile. 17 , 18 , 19 In the present study, a simple protein precipitation method for routine monitoring was performed by methanol/1.125 M ZnSO4 aquos solution, 19 , 20 , 21 , 22 for a total of about 15 min resulted in a clear supernatant giving chromatograms with no interfering compounds present. Method development and validation showed lower limit of quantification of 1 ng/mL. Elution of 4,12 min allowed a turnaround time of 5 min, thus making the method suitable for routine TAC monitoring in clinical practice. All QC samples and calibration samples fulfilled FDA criteria. The use of stable isotope‐labeled (SIL) analogs, compared with the use of structurally related compounds as internal standards, has been widely shown to reduce matrix effects and give reproducible and accurate recoveries in LC–MS/MS assays. The application of SILs for LC–MS/MS analysis has proven to reduce variations in mass spectrometry results, such as ionization issues, and has also improved the accuracy and precision for the analysis of both small and large molecules. 23 In this study, we used Roxithromycin as the internal standard based on the following reasons: Tacrolimus and Roxithromycin are both macrolides, have similar physicochemical properties. In addition, Roxithromycin's ionization reaction and fragmentation pattern is similar to Tacrolimus. Roxithromycin MRM transitions did not interfere with tacrolimus MRMs, and Roxithromycin has an identical chromatographic retention, acceptable mass differentiation, and retention of the mass differential in the daughter ion to Tacrolimus. Roxithromycin is also of high purity, co‐elutes with Tacrolimus, and has little to no scrambling or cross‐talk. Furthermore, Roxithromycin is accessible in our lab and it has been used as an internal standard in the development of an LC–MS/MS method for the determination of Clarithromycin in plasma. 24
The CMIA method has been widely used in Vietnam for the monitoring of whole‐blood Tacrolimus concentration in transplanted patients. Previous studies have reported a comparison of Tacrolimus concentrations in transplant patients between CMIA and other methods. 19 , 25 , 26 The difference in data obtained by CMIA and LC–MS/MS was found an overestimation of TAC concentrations measured by CMIA associated with the cross‐reactivity of the antibody with metabolites of TAC present in the patient's whole blood. 8 In this study, we analyzed comparative measurements with a large number of samples from kidney transplant recipients at three time points of follow‐up. Comparative data in our laboratory showed a good correlation between LC–MS/MS and CMIA (Figure 5A–D). However, the correlation coefficient (0.7581) at month 1 was lower, but the overall correlation coefficient observed in this study was comparable with previously reported studies. 27 , 28 In addition, Passing–Bablok regression analysis demonstrated good correlations with r 2 values higher than 0.88 between TAC levels measured by LC–MS/MS and CMIA for all three time points (Table 9 and Figure 4). Bland–Altman plots gave positive biases of 1.29 ng/mL in samples at 1 month (Figure 6A), 0.79 ng/mL at 3 months (Figure 6B), and 0.65 ng/mL at all time points (Figure 6D). This result is supported by the manufacturer's data on package insert supplied by Abbott Architect as well as previously published studies. 8 , 29 , 30 Recent guidelines recommended either immunoassays or LC–MS/MS for determination of TAC in whole blood. 5 Although immunoassays were performed on automated systems and had high throughput, but their major drawbacks were nonspecific for parent drugs. It has been shown that the cross‐reactivity of immunoassay to TAC metabolites resulted in higher concentrations than true values observed in patient samples. 29 LC–MS/MS with high selectivity and sensitivity proved to be a promising alternative method for therapeutic drug monitoring in clinical laboratories equipped LC–MS/MS instrument. 31
FIGURE 5.
Comparisons of tacrolimus concentrations in whole blood of kidney transplant patients measured by LC–MS/MS and CMIA methods.
FIGURE 6.
Bland–Altman bias plots for comparing tacrolimus concentrations measured by LC–MS/MS and CMIA methods. The average biases (solid line) and limits of agreement (square line) are defined as mean ± 1.96 SD.
5. CONCLUSION
We developed and validated a simple, rapid, sensitive, specific, reproducible, and high‐throughput LC–MS/MS method to quantify TAC in human whole blood. The LC–MS/MS method applied for the accurate determination of TAC in human whole blood from kidney transplant patients. The comparison of Tacrolimus concentrations measured by the LC/MS/MS with CMIA showed a good correlation and agreement between the two methods, suggesting LC–MS/MS should be used for routine monitoring of TAC concentrations in kidney transplant patients.
AUTHOR CONTRIBUTIONS
T.B.H, N.M.D, H.X.S: conceived and designed, supervised study, drafting of the manuscript and critical revision of the manuscript. C.V.M, V.Q.H, V.D.D, D.T.T.H, T.T.N, N.H.B, N.V.K, N.M.P: performed experiments, acquisition of data, analysis and interpretation of data. T.Q.K, N.V.D, P.V.T: material support, acquisition of data and commented on the manuscript. P.Q.T, L.V.T, D.D.L, N.V.T: critical revision of the manuscript.
CONFLICT OF INTEREST STATEMENT
The authors have no conflicts of interest to disclose.
ACKNOWLEDGEMENTS
The authors thank all patients who participated in this study. This study was supported by a grant from the Vietnam National Foundation for Science and Technology Development (NAFOSTED, grant number 04/2020/TN).
Hieu TB, Dung NM, Toan PQ, et al. Determination of tacrolimus in human whole blood in kidney transplant recipients using a rapid and specific LC–MS/MS method. J Clin Lab Anal. 2023;37:e24958. doi: 10.1002/jcla.24958
Tran Ba Hieu, Nguyen Manh Dung and Pham Quoc Toan contributed equally to this work and shared as co‐first authors.
Contributor Information
Tran Ba Hieu, Email: manhhieu86@gmail.com.
Hoang Xuan Su, Email: hoangxuansu@vmmu.edu.vn.
DATA AVAILABILITY STATEMENT
All data in this study can be obtained directly from the corresponding author.
REFERENCES
- 1. Jørgensen KA, Koefoed‐Nielsen PB, Karamperis N. Calcineurin phosphatase activity and immunosuppression. A review on the role of calcineurin phosphatase activity and the immunosuppressive effect of cyclosporin a and tacrolimus. Scand J Immunol. 2003;57(2):93‐98. [DOI] [PubMed] [Google Scholar]
- 2. Cirillo R, Triggiani M, Siri L, et al. Cyclosporin A rapidly inhibits mediator release from human basophils presumably by interacting with cyclophilin. J Immunol. 1990;144(10):3891‐3897. [PubMed] [Google Scholar]
- 3. Geba GP, Ptak W, Askenase PW. Topical tacrolimus and cyclosporin A differentially inhibit early and late effector phases of cutaneous delayed‐type and immunoglobulin E hypersensitivity. Immunology. 2001;104(2):235‐242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Johnston A. Equivalence and interchangeability of narrow therapeutic index drugs in organ transplantation. Eur J Hosp Pharm. 2013;20(5):302‐307. doi: 10.1136/ejhpharm-2012-000258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Brunet M, Van Gelder T, Åsberg A, et al. Therapeutic drug monitoring of tacrolimus‐personalized therapy: second consensus report. Ther Drug Monit. 2019;41(3):261‐307. [DOI] [PubMed] [Google Scholar]
- 6. Armstrong VW, Schuetz E, Zhang Q, et al. Modified pentamer formation assay for measurement of tacrolimus and its active metabolites: comparison with liquid chromatography–tandem mass spectrometry and microparticle enzyme‐linked immunoassay (MEIA‐II). Clin Chem. 1998;44(12):2516‐2523. [PubMed] [Google Scholar]
- 7. Cogill JL, Taylor PJ, Westley IS, Morris RG, Lynch SV, Johnson AG. Evaluation of the tacrolimus II microparticle enzyme immunoassay (MEIA II) in liver and renal transplant recipients. Clin Chem. 1998;44(9):1942‐1946. [PubMed] [Google Scholar]
- 8. Wallemacq P, Goffinet JS, O'Morchoe S, et al. Multi‐site analytical evaluation of the Abbott ARCHITECT tacrolimus assay. Ther Drug Monit. 2009;31(2):198‐204. [DOI] [PubMed] [Google Scholar]
- 9. Aucella F, Lauriola V, Vecchione G, Tiscia GL, Grandone E. Liquid chromatography–tandem mass spectrometry method as the golden standard for therapeutic drug monitoring in renal transplant. J Pharm Biomed Anal. 2013;86:123‐126. [DOI] [PubMed] [Google Scholar]
- 10. Kaza M, Karaźniewicz‐ Lada M, Kosicka K, Siemiątkowska A, Rudzki PJ. Bioanalytical method validation: new FDA guidance vs. EMA guideline. Better or worse? J Pharm Biomed Anal. 2019;165:381‐385. [DOI] [PubMed] [Google Scholar]
- 11. European Medicines Agency . Guideline on Bioanalytical Method Validation. 2012. https://www.ema.europa.eu/en/documents/scientific‐guideline/guideline‐bioanalytical‐method‐validation_en.pdf [DOI] [PubMed]
- 12. Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res. 1999;8(2):135‐160. [DOI] [PubMed] [Google Scholar]
- 13. Bland JM, Altman D. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;327(8476):307‐310. [PubMed] [Google Scholar]
- 14. Altman DG, Bland JM. Measurement in medicine: the analysis of method comparison studies. J R Stat Soc Series D. 1983;32(3):307‐317. [Google Scholar]
- 15. Napoli KL. 12‐hour area under the curve cyclosporine concentrations determined by a validated liquid chromatography‐mass spectrometry procedure compared with fluorescence polarization immunoassay reveals sirolimus effect on cyclosporine pharmacokinetics. Ther Drug Monit. 2006;28(6):726‐736. [DOI] [PubMed] [Google Scholar]
- 16. Holt DW, Lee T, Jones K, Johnston A. Validation of an assay for routine monitoring of sirolimus using HPLC with mass spectrometric detection. Clin Chem. 2000;46(8):1179‐1183. [PubMed] [Google Scholar]
- 17. Keevil BG, McCann SJ, Cooper DP, Morris MR. Evaluation of a rapid micro‐scale assay for tacrolimus by liquid chromatography‐tandem mass spectrometry. Ann Clin Biochem. 2002;39(5):487‐492. [DOI] [PubMed] [Google Scholar]
- 18. Taylor PJ, Salm P, Lynch SV, Pillans PI. Simultaneous quantification of tacrolimus and sirolimus, in human blood, by high‐performance liquid chromatography–tandem mass spectrometry. Ther Drug Monit. 2000;22(5):608‐612. [DOI] [PubMed] [Google Scholar]
- 19. Mei S, Wang J, Chen D, et al. Simultaneous determination of cyclosporine and tacrolimus in human whole blood by ultra‐high performance liquid chromatography tandem mass spectrometry and comparison with a chemiluminescence microparticle immunoassay. J Chromatogr B. 2018;1087:36‐42. [DOI] [PubMed] [Google Scholar]
- 20. Vogeser M, Fleischer C, Meiser B, Groetzner J, Spöhrer U, Seidel D. Quantification of sirolimus by liquid chromatography‐tandem mass spectrometry using on‐line solid‐phase extraction. Clin Chem Lab Med. 2002;40:40‐45. [DOI] [PubMed] [Google Scholar]
- 21. Annesley TM, Clayton L. Simple extraction protocol for analysis of immunosuppressant drugs in whole blood. Clin Chem. 2004;50(10):1845‐1848. [DOI] [PubMed] [Google Scholar]
- 22. Karapirli M, Kizilgun M, Yesilyurt O, et al. Simultaneous determination of cyclosporine a, tacrolimus, sirolimus, and everolimus in whole‐blood samples by LC‐MS/MS. Scientific World Journal. 2012;2012:1‐8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Nageswara Rao R. Stable labeled isotopes as internal standards: a critical review. Mod Appl Pharm Pharmacol. 2017;1:1‐4. [Google Scholar]
- 24. Li W, Rettig J, Jiang X, Francisco DT, Naidong W. Liquid chromatographic‐electrospray tandem mass spectrometric determination of clarithromycin in human plasma. Biomed Chromatogr. 2006;20(11):1242‐1251. [DOI] [PubMed] [Google Scholar]
- 25. Bazin C, Guinedor A, Barau C, et al. Evaluation of the Architect® tacrolimus assay in kidney, liver, and heart transplant recipients. J Pharm Biomed Anal. 2010;53(4):997‐1002. [DOI] [PubMed] [Google Scholar]
- 26. Marubashi S, Nagano H, Kobayashi S, et al. Evaluation of a new immunoassay for therapeutic drug monitoring of tacrolimus in adult liver transplant recipients. J Clin Pharmacol. 2010;50(6):705‐709. [DOI] [PubMed] [Google Scholar]
- 27. Kaneko T, Fujioka T, Suzuki Y, et al. Comparison of whole‐blood tacrolimus concentrations measured by different immunoassay systems. J Clin Lab Anal. 2018;32(9):e22587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Tszyrsznic W, Borowiec A, Pawlowska E, et al. Two rapid ultra performance liquid chromatography/tandem mass spectrometry (UPLC/MS/MS) methods with common sample pretreatment for therapeutic drug monitoring of immunosuppressants compared to immunoassay. J Chromatogr B Analyt Technol Biomed Life Sci. 2013;928:9‐15. doi: 10.1016/j.jchromb.2013.03.014 [DOI] [PubMed] [Google Scholar]
- 29. Shigematsu T, Suetsugu K, Yamamoto N, Tsuchiya Y, Masuda S. Comparison of 4 commercial immunoassays used in measuring the concentration of tacrolimus in blood and their cross‐reactivity to its metabolites. Ther Drug Monit. 2020;42(3):400‐406. [DOI] [PubMed] [Google Scholar]
- 30. Abbott Laboratories , Architect® Tacrolimus kit (Package Insert). 2015.
- 31. Dubbelboer IR, Pohanka A, Said R, Rosenborg S, Beck O. Quantification of tacrolimus and three demethylated metabolites in human whole blood using LC–ESI–MS/MS. Ther Drug Monit. 2012;34(2):134‐142. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data in this study can be obtained directly from the corresponding author.