HPLC and liquid chromatography-tandem mass spectrometry (LC-MS/MS) have been used to measure the concentrations of 6-TGN and 6-MMPN in RBCs [51112]. As the metabolites are unstable and can be altered during sample preparation and storage, control of every pre-analytical step is important to ensure reliable assay results. However, to our knowledge, stability of thiopurine metabolites has rarely been studied [1314]. For the first time in Korea, we developed and validated an LC-MS/MS method for the quantification of 6-TGN and 6-MMPN in RBCs. Additionally, we investigated the stability of thiopurine metabolites under various storage conditions to identify appropriate sample processing and storage conditions.

We prepared stock solutions of 3.0 mM 6-TGN and 30 mM 6-MMPN using standard materials (Sigma-Aldrich, St. Louis, MO, USA) and a stock solution of internal standards (IS) containing 1 mg/mL of each of 6-TG-¹³C₂¹⁵N and 6-MMP-d₃ (Toronto Research Chemicals, Toronto, Canada) in methanol. These solutions were stored at −70℃. We diluted the stock solutions in 50 mM NaOH to prepare calibrators. We prepared the calibrators and control samples using whole blood (matrix-matched) spiked with the prepared drug standard solutions. Then, we validated the accuracy of our method for both metabolites. Six concentrations of calibrators (0.5, 2, 4, 6, 8, 10 µmol/L for 6-TGN and 5, 20, 40, 60, 80, 100 µmol/L for 6-MMPN) and quality-control materials at two concentrations for each analyte (2.5 µmol/L and 8.0 µmol/L for 6-TGN, and 25.0 µmol/L and 80.0 µmol/L for 6-MMPN) were measured during each analysis.

Residual EDTA-treated whole blood samples from patients under thiopurine treatment (N=4) were centrifuged (4,000 rpm, 10 minutes), and the plasma and buffy coat were discarded. This study was approved by the Institutional Review Board of Samsung Medical center, Seoul, Korea (IRB No. SMC-04-106-014). Two washing steps were performed, and the RBC count was determined using an automatic hematological cell counting device, XE-2100 (Sysmex Corp. Kobe, Japan), after diluting 100 µL of RBC three times in 200 µL of saline. The isolated RBCs were stored at −70℃ until further analysis. After thawing the stored samples, 50 µL of each sample was transferred into a tube containing 200 µL of 0.2 M D,L-dithiothreitol (DTT) and 30 µL of working internal standards (0.5 mg/L for 6-MMPN and 0.125 mg/L for 6-TGN). After vortexing twice, the sample was rapidly deproteinized by the addition of 40 µL of 70% perchloric acid (GFS Chemicals, Columbus, Ohio, USA). After vortexing at the highest frequency, the deproteinized samples were centrifuged at 15,000 rpm for 10 minutes. The supernatant (100 µL) was transferred to a new tube that was sealed well and heated for one hour at 100℃ to hydrolyze thiopurine nucleotides into their individual bases. After cooling, a 5-µL sample aliquot in 200 µL of mobile phase A of the chromatograph (water containing 2 mM ammonium acetate and 0.1% formic acid) was injected into the column as described below.

Chemical analyses were performed on an API 4000 tandem mass spectrometer (AB Sciex, Foster City, CA, USA) equipped with an Agilent 1200 HPLC system (Agilent Technologies Inc., Santa Clara, CA, USA) and an electrospray ionization (ESI) source operated in positive-ion detection mode. Chromatographic separation of 6-TGN and 6-MMPN was conducted on an X-bridge Shield RP18 (Imtackt, Portland, OR, USA; 1.7 µm, 2.1×150 mm). The flow rate was 0.5 mL/min. Quantitative analysis was performed in multiple reaction-monitoring mode (m/z 168.0→150.9 for 6-TGN, 158.0→110.0 for 6-MMP, 171.0→154.0 for 6-TG-¹³C₂¹⁵N, 161.1→110.1 for 6-MMP-d₃), with a total running time of 3 minutes for each sample. Quantification was performed using the ratio of the integrated peak areas of 6-TGN and 6-MMPN to that of the IS, and calculations were performed using Analyst MD software (AB Sciex, Foster City, CA, USA) and a calibration curve with a weighting factor of 1/x.

The chromatogram showed no interfering peaks for 6-TGN and 6-MMPN. For the IS, retention times were 0.94 minutes for 6-TG-¹³C₂¹⁵N and 1.19 minutes for 6-MMP-d₃ (Fig. 1). The within-run (N=5) and between-run (N=10) precision and accuracy were determined at different concentrations of 6-TGN and 6-MMPN (2.5 µmol/L and 8.0 µmol/L for 6-TGN and 25.0 µmol/L and 80.0 µmol/L for 6-MMPN). Within- and between-run imprecisions evaluated by using quality control materials demonstrated coefficient of variation (CV) values less than 10% (see Supplemental Data Table S1). The linear assay range was 0.1–10 µmol/L and 0.5–100 µmol/L for 6-TGN and 6-MMPN, respectively (R²>0.999 for both 6-TGN and 6-MMPN). The lower limit of quantitation (LLOQ) was defined as the lowest concentration with a signal to noise ratio of >10% and <20% of the CV. The LLOQ was 0.1 µmol/L for 6-TGN and 0.5 µmol/L for 6-MMPN. Matrix effects were evaluated by comparing the average peak areas of post-extraction blank samples spiked with 6-TGN and 6-MMPN to those of pure standard solutions of the corresponding concentrations. Samples were analyzed in quadruplicate at each concentration. The matrix effect ranged from 67.7% to 70.0% for 6-TGN and from 111.0% to 114.1% for 6-MMPN. The extraction recovery of 6-TGN and 6-MMPN was determined. The mean values of extraction recovery at two concentrations were 71.0% and 75.0% for 6-TGN, and 102.2% and 96.4% for 6-MMPN. Low extraction recovery of 6-TGN (~60%) was reported in a previous study [15].

Freeze-and-thaw stability was tested at two concentrations of 6-TGN and 6-MMPN (2.5 µmol/L and 8.0 µmol/L for 6-TGN and 25.0 µmol/L and 80.0 µmol/L for 6-MMPN) four times repeatedly. The materials were subjected to three freeze-and-thaw cycles (−70℃) with none of the materials showing significant changes in concentration; the decrease in concentration materials was less than 5% for both materials.

Short-term stability was investigated by analyzing two concentrations of each of 6-TGN and 6-MMPN (2.5 µmol/L and 8.0 µmol/L for 6-TGN and 25.0 µmol/L and 80.0 µmol/L for 6-MMPN) stored at 25℃ and 4℃ for 0.5, 1, 2, and 4 hours. There was no significant concentration change in either 6-TGN or 6-MMPN at either temperature after storage for 4 hours (Fig. 2). Long-term stability was investigated by analyzing samples stored at −70℃ and −20℃ until assayed (7, 14, 21, 28, 90, and 180 days). When stored at −70℃ for 180 days, 6-TGN decreased to approximately 5% of the initial concentrations (2.5 µmol/L and 8.0 µmol/L), but when stored at −20℃ for 180 days, 6-TGN decreased by 30% compared with the initial concentrations (2.5 µmol/L and 8.0 µmol/L). 6-MMPN decreased by approximately 10% of the initial concentration after 180 days at both temperatures and was rather stable (Fig. 2). Previously, Pike et al [14] reported that 6-TGN concentration decreased by about 12% after six months of storage at −80℃ and that 6-MMPN showed almost no change in concentration after six months. Therefore, samples for long-term storage should be kept at a temperature lower than −70℃ to prevent loss of 6-TGN.

Because whole blood samples cannot be pretreated immediately in a clinical laboratory, it is necessary to evaluate the stability of the thiopurine metabolites in the whole blood. To this end, we stored pooled patient blood samples (N=4) at room temperature (25℃) in the whole blood state and measured 6-TGN and 6-MMPN concentrations for up to seven days. For each concentration, four repeated measurements were conducted. Compared with the initial concentrations, 6-TGN decreased by more than 20% on day 2, and 6-MMPN decreased more than 40% on day 4 (Fig. 3). Although the degree of degradation is somewhat different, Pike et al [14] also found that the 6-TGN concentration decreased by 2–4% per day at room temperature in the whole blood state. Additionally, they found that the loss of 6-TGN was 1% after one week and 12% after 24 weeks at 80℃, indicating that 6-TGN is reasonably stable in whole blood, and 6-MMPN did not show a significant change when stored in the whole blood state for 24 weeks [14]. 6-TGN and 6-MMPN were relatively stable at 4℃ without pretreatment. However, the 6-TGN concentration decreased by about 20% after four days of storage at 4℃. Therefore, pretreatment should be carried out within four days when samples are stored at 4℃.

In conclusion, we developed and validated a reliable LC-MS/MS method for quantification of RBC 6-TGN and 6-MMPN and comprehensively analyzed their stability. Based on the stability test results, we recommend −70℃ for long-term storage of pretreatment samples. The whole blood sample should be preprocessed as soon as possible.