Immunosuppressive Drugs in Whole Blood

Authors: Pierre Picard, Jean Lacoursière and Serge Auger
Themes: High-Throughput, Whole blood, Luxon-MS/MS
Télécharger le PDF

Introduction

Over the last 10 years, the quantitation of immunosuppressive drugs has been subject to continuous improvements in analytical methods to optimize cost, time, and accuracy of results. The transition from immunoassay to liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) has significantly improved all these criteria but has not reduced the analytical time below a minute. The Laser Diode Thermal Desorption (LDTD) represents a technological breakthrough that removes the chromatographic step and significantly increases the analytical throughput for the quantitation of Tacrolimus (TAC), Sirolimus (SIR), Everolimus (EVE) and Cyclosporin A (CsA) in the field of clinical analysis.

Our goal for this application note is to use an automated sample preparation method for an immunosuppressive drug panel in whole blood using a single operation in LDTD-MS/MS.

LDTD-MS/MS offers specificity combined with an ultra-fast analysis for an unrivaled analytical method. To develop this application, we focused on performing an automated sample preparation. Four drugs are analyzed simultaneously with quantitative results obtained in less than 9 seconds per sample.

Sample Preparation Method

 

Automated Sample Extraction

An automated process (Figure 1) is used to extract the samples using the following conditions:

 

Azeo
Figure 1 – Azeo Liquid Handler

 

LDTD®-MS/MS Parameters

LDTD

Model: Luxon S-960, Phytronix

Carrier gas: 3 L/min (air + infusion of NH4OH at 3 µL/min)

Laser pattern:

 

MS/MS

MS model: Q-Trap System® 5500, Sciex

Scan Time: 20 msec

Total run time: 9 seconds per sample

Ionization: APCI

Analysis Method: MRM mode

 

Table 1 – Positive MRM transitions for LDTD-MS/MS

Transition CE
Sirolimus 931.5 → 864.55 25
Sirolimus-d3 934.5 → 864.55 25
Everolimus 975.6 → 908.55 25
Everolimus-d4 979.6 → 912.55 25
Tacrolimus 821.4 → 768.45 25
Ascomycin (IS) 809.4 → 756.45 25
Cyclosporin-A 1219.9 → 1184.9 20
Cyclosporin-D (IS) 1233.9 → 1198.9 20

 

Results and Discussion

 

Linearity

For each analytical run, calibrators and QC samples were freshly prepared. 100 µL of working solution were spiked in 900 µL of drug-free whole blood to achieve a calibration standard ranging from 2.25 to 50 ng/mL for TAC, SIR and EVE. For CsA, calibration standards from 25 to 1000 ng/mL were obtained. Figure 4 shows a typical calibration curve obtained for Tacrolimus. All equations for the calibration curves were weighted according to 1/x least-square linear and regression correlation coefficients (r) obtained were between 0.99635 and 0.99950.

 

tacrolimus-curve
Figure 2 – Tacrolimus calibration curve

 

Precision and Accuracy

Spiked samples at QC level (QC-L: 6 ng/mL, QC-M:20 ng/mL and QC-H: 25 ng/mL) for TAC, SIR and EVE. For CsA, QC levels (QC-L: 125 ng/mL, QC-M: 500 ng/mL and QC-H: 800 ng/mL) are used to validate the precision and accuracy of the method. The peak area against the internal standard (IS) ratio was used to normalize the signal. Replicate extractions are deposited on a LazWell™ plate and dried before analysis.

 

Following acceptance criteria were used:

 

For the inter-run precision experiment, each fortified sample set is analyzed in sextuplicate on five different runs. Table 2 shows the inter-run precision results. Accuracy within 15% of nominal value and a %CV below 15% are reached. Similar results are obtained for the other drugs.

Table 2 – Inter-Run Precision and Accuracy for TAC

Tacrolimus QC-L QC-M QC-H
Conc (ng/mL) 6 20 25
N 30 30 30
Mean (ng/mL) 5.5 20.5 25.0
%CV 5.9 2.1 3.0
%Nom 91.4 102.4 100.1

 

For the intra-run precision experiment, each fortified sample is extracted and analyzed (6 replicates). Table 3 shows the results. Accuracy within 15% of nominal value and a %CV below 15% are reached. Similar results are obtained for the other drugs.

Table 3 – Intra-Run Precision and Accuracy for TAC

Tacrolimus QC-L QC-M QC-H
Conc (ng/mL) 6 20 25
N 6 6 6
Mean (ng/mL) 5.4 20.3 24.9
%CV 2.1 2.0 1.7
%Nom 90.6 101.3 99.7

 

Matrix Effect Evaluation

The variability from different sources of matrix was evaluated in blood samples of six individual donors spiked at low quality control (QC-L) levels. The inter-individual variation in matrix effect did not affect the precision and accuracy of the assay. All results are within the acceptance criteria. Tacrolimus results are reported in Table 4. Similar results are obtained for the other drugs.

Table 4 – Multi-matrix evaluation results

Analytes Sample Conc.
(ng/mL)
N Mean
(ng/mL)
Precision
(%)
Accuracy
(%)
TAC
 
 
 
 
 
M1 6 6 5.7 1.1 95.0
M2 6 6 5.2 1.7 86.7
M3 6 6 5.8 2.2 96.7
M4 6 6 5.6 3.0 93.3
M5 6 6 5.5 3.3 91.7
M6 6 6 5.8 4.0 96.7

 

Cross-Validation Study

Correlation plots between concentrations obtained for real samples using both LDTD-MS/MS and LC-MS/MS methods were drawn to evaluate r2 values. All correlation plots exceeded 0.97 r2 values. Bland-Altman plots are graphical representations of the method agreement between LC and LDTD. Evenly scattered points on the y-axis indicate that the methods agree while the contrary would indicate that one method constantly under or over-evaluates drug concentration. Bland-Altman plots show that the average percent difference between both methods remained below 5% for all drugs, suggesting that neither method is biased towards under or over-evaluating the drug concentration of a sample. Tacrolimus results are showed in Figure 3 and Figure 4. Similar results are obtained for the other drugs.

 

tacrolimus-correlation
Figure 3 – Tacrolimus correlation plot

 

tacrolimus-bland-altman
Figure 4 – Tacrolimus Bland-Altman plot

 

Conclusion

Luxon Ion Source® combined to Sciex Q-Trap 5500 mass spectrometer system allows ultra-fast (9 seconds per sample) analysis of immunosuppressive drugs in whole blood using a simple and automated sample preparation method.