A fast LC-MS/MS assay for methotrexate monitoring in plasma: validation, comparison to FPIA and application in the setting of carboxypeptidase therapy

Abstract

High-dose methotrexate remains a mainstay in the treatment of acute lymphoblastic leukaemia, osteosarcoma and non-Hodgkin lymphoma. Therapeutic drug monitoring of plasma MTX is important to monitor efficacy and adverse events. The authors aimed at developing a liquid chromatography tandem mass spectrometry (LC-MS/MS) method with online extraction to determine MTX and 7-OH-MTX in plasma for therapeutic drug monitoring. The analysis combined straightforward sample preparation, consisting of protein precipitation with methanol/ZnSO4, with a 4-minute run time consisting of an on-line enrichment by a flush/back-flush cycle (Poros column, R1/20, 2.1 mm × 30 mm) before the second dimension chromatography (Phenomenex Luna 5 μm Phenyl Hexyl, 2 mm × 50 mm column). Samples were analysed using an HPLC Agilent 1200 Series and ABSciex API 3200. The electrospray was operated in positive ionization mode monitoring the following mass transitions: m/z 455.11 → 308.3 for MTX, 471.14 → 324.3 for 7-OH-MTX and m/z 459.1 → 312.3 for internal standard (MTX¹³C²H₃). The method was linear up to 50 μmol L⁻¹, and intra-day and inter-day quality control CVs were below 8.3% for MTX and 11.71% for 7-OH-MTX. Average recovery was 24% for MTX and 57% for 7-OH-MTX. The lower limit of quantitation was 25 nmol L⁻¹ for the 2 analytes. For MTX and 7-OH-MTX the standard line slope CV percentage was <3% and the slope difference <6%, indicating that our analytical method is almost free from a significant relative matrix effect. Method comparison with the Abbott TDx fluorescent polarization immunoassay (FPIA) showed excellent agreement: LC-MS/MS = 0.0011 + 1.0334 (FPIA). Because of antibody cross-reactivity between DAMPA and MTX, none of the immunoassays can be used after carboxypeptidase administration. The goal of our work was to develop a specific LC-MS/MS method to monitor both MTX and 7-OH-MTX plasma concentrations within the clinically relevant range. It's expected that the LC-MS/MS method for MTX monitoring after carboxypeptidase administration will be very rarely used since it concerns only exceptional cases. Therefore, the geographically balanced distribution of University Hospital able to ensure follow-up of these patients, within 24 hours of collection, can draw a reliable solution for the security of the patients. We develop a fast and reliable LC-MS/MS method for both routine TDM of MTX as in the setting of carboxypeptidase therapy.

---

Introduction

Methotrexate (MTX) is the main folate antagonist and one of the most widely used antimetabolites in cancer chemotherapy. Folates are essential for DNA synthesis and cell division. MTX follows folate pathway: MTX is actively taken up into the cells by the reduced folate carrier and then converted to polyglutamates, which are retained in the cells and inhibit dihydrofolate reductase which converts first the folate to dihydrofolate and then to tetrahydrofolate.^1–3

High-dose methotrexate (HDMTX) remains a mainstay in the treatment of acute lymphoblastic leukaemia, osteosarcoma and non-Hodgkin lymphoma. HDMTX is defined as MTX dose >1 g m⁻² administered by intravenous infusion in many different doses and schedules.⁴ For example to treat children's osteosarcoma, a MTX dose of 8–12 g m⁻² is administered by 3 hour intravenous infusion.⁵

MTX is poorly metabolized; only a few percentage of doses are converted to 7-hydroxy-methotrexate (7-OH-MTX), and most of the dose is found unchanged in urine.⁴ Common unwanted effects include bone marrow depression and mucositis, but acute nephrotoxicity induced by MTX and 7-OH-MTX precipitations in renal tubules is dreadful. However, toxicity incidence can be alleviated through diverse measures. HDMTX-induced nephrotoxicity is prevented through substantial hydration and alkalinisation of urine to enhance elimination.^6,7 Folinic acid rescue, to restore intracellular folate concentration, is guided by plasma creatinine and MTX concentration monitoring. Nevertheless, renal impairment causes delayed MTX elimination and ineffective folinic acid rescue. As a last resort, hydrolysis of MTX and 7-OH-MTX by carboxypeptidase (CPDG2) can be used. A single administration of this enzyme can inactivate more than 95% of MTX to 2,4-diamino-N10-methylpteroic acid (DAMPA) and around 50% of 7-OH-MTX⁸ (Fig. 1). In France, CPDG2 administration is recommended for patients whose plasma MTX concentration is >10 μmol L⁻¹ 48 hours after the beginning of infusion or for patients whose plasma creatinine is >1.5 times the basal level and plasma MTX concentration is >3 μmol L⁻¹ 48 hours after the beginning of infusion.^7,9,10

Fig. 1 MTX metabolism action of aldehyde oxidase and carboxypeptidase-G2 on MTX and 7-OH-MTX.

Thus, MTX concentration monitoring is crucial for detection of a poor MTX elimination as quickly as possible. Plasma MTX concentrations are routinely monitored by fluorescence polarization immunoassay (FPIA) after HDMTX administration. Unfortunately, MTX concentrations following carboxypeptidase administration can only be monitored by a chromatographic method, because of DAMPA cross-reactivity with antibodies used by immunoassays.^11–13 In order to monitor MTX concentrations after CPDG2 administration, we have developed a liquid chromatography-tandem mass spectrometry method (LC-MS/MS) for MTX and 7-OH-MTX measurements.

While HPLC methods were published in the past year, only a few authors describe a method able to quantify rapidly both MTX and 7-OH-MTX.^14–16 Most of the above-mentioned methods have not been developed in human patients' plasma, with a wide range of possible cross-reactivities, and a wide amplitude of concentration.^14,17 Begas et al. developed an SPE offline-LC-MS/MS with specificity and sensibility close to our method but for MTX alone and with a run of 16 min (Fig. 2).¹⁶

Fig. 2 Timetable of SPE and HPLC mobile phase flow rate and ten-port switching valve position programming. Connections and positions of the column-switching valve for on-line extraction step from 0.0 to 1.2 min (A). Analytes elution, transfer to the HPLC column and analysis from 1.2 to 4.0 min (B). SPE-online washing step from 2.0 to 4.0 min (C).

The goal of our method was to develop a specific LC-MS/MS method to monitor both MTX and 7-OH-MTX plasma concentrations within the clinically relevant range.

Materials and methods

Patients

All patient samples tested in this work derived from an ongoing drug-monitoring program and were reported in accordance with ethical guidelines. Indeed an informed consent was not required.

Chemicals, reagents and standard solutions

All solvents and reagents were of HPLC-grade and were purchased from VWR International (Fontenay-sous-Bois, France). MTX solution was purchased from Mylan (Saint Priest, France), DAMPA was purchased from Schircks laboratories (Jona, Switzerland), 7-OH-MTX was purchased from Alphachimica laboratories (Châtenay-Malabry, France) and ¹³C²H₃-MTX, used as the internal standard (IS), was purchased from Alsachim (ILLKIRCH, France).

MTX plasma multilevel calibrators and quality controls (QC) were provided by Abbott Diagnostics (Abbott Park, IL) and were used for fluorescent polarization immunoassay. Homemade quality controls and plasma multilevel calibrators, containing both MTX and 7-OH-MTX, were used for all LC-MS/MS assays.

Protein precipitation solution was a mixture of methanol and 0.2 M ZnSO₄ (80 : 20, vol/vol) containing IS (1 μmol L⁻¹) and stored at −20 °C. The different lots of drug-free plasma samples originated from our laboratory.

LC-MS/MS assay

Protein precipitation was carried out in 1.5 mL polypropylene tubes (Eppendorf, Le Pecq, France). A volume of 50 μL of calibrator, QC or patient sample was mixed with 100 μL of precipitation solution. The mixture was vortex-mixed for 5 min, and centrifuged at 15 300 g at 4 °C for 10 min. Subsequently the supernatant was transferred into a polypropylene tube with pierceable membrane screw caps, and 20 μL were injected into the chromatographic system.

The instrument setup is shown in Fig. 2. The chromatographic system consists of Agilent 1200 Series components (Palo Alto, USA) including a binary pump, an isocratic pump, a column oven, and an auto-sampler. The hardware configuration included an ABSciex API 3200 (Toronto, Canada) equipped with a turboionspray source. Two-dimension chromatography was performed. First dimension chromatography is an on-line enrichment performed by a perfusion column (POROS R1/20, 2.1 mm × 30 mm, ABSciex, Foster City, USA). The binary pump supplied water–ammonium acetate (10 mM)–acetic acid (0.1%), for enrichment of analytes and IS on the Poros column. A back-flush elution was performed using the isocratic pump (methanol–ammonium acetate (10 mM)–acetic acid (0.1%)–formic acid (0.2%)). The second dimension chromatography was performed using a Phenomenex Luna 5 μm Phenyl Hexyl, 2 mm × 50 mm column housed in an oven at 60 °C (Torrance, USA). Positive ion electrospray, schedule MRM mode was used for analytes and IS (Table 1). The procedure was performed using the Analyst 1.5.2 software package.

Table 1 Settings for mass/charge (m/z) transitions^a

Compound	Precursor ion (m/z)	Product ion (m/z)	RT (min)	EP	CEP	CE (eV)	CXP
a Retention times (RT), entrance potential (EP), cell entrance potential (CEP), collision energy (CE) cell exit potential (CXP) for API 3200, and for the phenyl–hexyl HPLC column.
MTX	455.11	308.3	1.7	8.5	18	27	4
7-OH-MTX	471.14	324.3	1.7	6	20	19	6
^13C^2H3-MTX	458.10	312.3	1.7	10.5	16	41	10

---

Fluorescent polarization immunoassay (FPIA)

The assay is performed according to manufacturers' instructions (Abbott Diagnostic, Abbott Park, IL). Calibrators, controls, or patient plasma samples (100 μL) were transferred into reaction cells and submitted to a TD_xFL_x instrument for automated analysis. The range of the methotrexate calibration curve was 0.00 to 1.00 μmol L⁻¹, and samples containing methotrexate at higher concentrations were diluted. The limit of quantitation was 0.03 μmol L⁻¹.

Standard line slope assays for matrix effect assessment

Stock solutions in water were prepared as follows: MTX and 7-OH-MTX at a concentration of 1000 μmol L⁻¹. These stock solutions were diluted with an adequate volume of drug-free plasma samples. Five different lots of plasma sample were used for spiking all standard curve samples for the post-extraction addition assay.

LC-MS/MS validation procedures

Precision, accuracy and limit of quantification. Plasma extracts spiked with MTX and 7-OH-MTX standards were used to determine precision and accuracy. Triplicates of each QC were repeatedly analysed in a single series the same day to assess the intra-day performance, and daily over three days for the inter-day performance. The precision of the method was measured by the coefficient of variation (%CV) and required to be <15%. The accuracy was expressed by 100 − [(mean observed concentration)/(spiked concentration)] × 100.

Triplicates of a dilution of the low QC determined the lowest limit of quantification (LLOQ) of MTX and 7-OH-MTX. The lowest concentration that can be measured with an imprecision and inaccuracy below <20% each defines the LLOQ.

Selectivity and “cross talk”. The selectivity of the method was evaluated by monitoring all analytes and IS m/z transitions for MTX and 7-OH-MTX free human plasma sample from six different sources to determine the presence or absence of endogenous peaks at all the transitions used to monitor analytes and IS.

The “cross-talk” between MRM transition used for monitoring analytes and IS were evaluated by analysing of MTX or 7-OH-MTX calibrator samples spiked with MTX or 7-OH-MTX or DAMPA.

Extraction recovery. The recovery was determined by comparing the absolute peak area obtained from a standard plasma extracted according the relevant procedure versus a blank human plasma extract spiked after extraction with the same amount of molecules (0.05; 0.25; 0.5; 1 and 10 μmol L⁻¹ for MTX and 7-OH-MTX).

Matrix effect. The European Medicine Agency (EMA) and Food and Drug Administration (FDA) both indicate that to ensure that precision, selectivity and sensitivity are not compromised, assessment of the matrix effects is needed.^18,19 As proposed by Matuszewski and requested by EMA, the matrix effects have been investigated in five different sources of EDTA plasma samples.^18,20 The linearity of each calibration curve was confirmed by plotting the peak area ratio of the drug to IS versus drug concentration. Sample concentrations were calculated from the equation y = b + ax, as determined by the weighted linear regression of the standard line. Slopes of standard lines were determined as above. The matrix effect was evaluated by comparison of the CV values and the slope difference of standard line slopes in five different lots of plasma samples for each analyte.

Stability

Manufacturers provided the stability studies of analytes and IS. The storage stability of human plasma and the influence of the freeze–thaw cycle were also examined by processing a set of QC and patient samples.

Data analysis, interpretation and statistics

Chromatographic data processing was performed using the Analyst 1.5.2 software package (ABSciex, Foster City, USA). Linear regression analyses and statistical analyses were performed with Prism software (Graphpad Software, La Jolla, USA). The Passing and Bablok test was used for method comparison.

Results

LC-MS/MS method validation

This method was validated in human plasma over the concentration range of 0.05 to 50 μmol L⁻¹ for MTX and 7-OH-MTX. The accuracy and precision, determined for both intra- and inter-runs, are summarized in Table 2. Intra-day and inter-day QC CVs were below 8.3% for MTX and 11.71% for 7-OH-MTX. The intra-day accuracy was found to be between −0.05% and +1.61% for MTX and −1.1% and 11.0% for 7-OH-MTX of the nominal concentration while inter-day performance was between 3.92% and 4.87% for MTX and <1.58% for 7-OH-MTX. The accuracy and precision at the LLOQ are summarized in Table 2.The LLOQ were 0.025 μmol L⁻¹ for the 2 analytes, with accuracy and CV <20% of the nominal concentration. −20 °C frozen MTX QC, 7-OH-MTX QC and patient samples remained stable over 1 month. Fig. 3 shows the representative chromatogram obtained from the standard.

Table 2 Accuracy and precision^a

Analyte	QC concentration (μmol L^−1)	Intra-day (n = 3)		Inter-day (n = 3)
		CV (%)	Mean accuracy (%)	CV (%)	Mean accuracy (%)
a The following abbreviations were used: QC: quality control, LLOQ: lower limit of quantification, CV: coefficient of variation.
MTX	0.025 (LLOQ)	6.11	12.68	16.54	9.84
	0.08	4.64	1.61	7.6	4.87
	0.80	3.52	−0.05	6.64	4.58
	25	7.22	2.28	8.3	3.92
7-OH-MTX	0.025 (LLOQ)	3.46	13.4	4.37	6.12
	0.08	1.02	11.0	11.71	1.13
	0.80	2.27	−1.10	5.33	1.58
	20	4.50	6.32	3.01	0.25

---

Fig. 3 Representative chromatogram example chromatograms obtained by schedule MRM for MTX, 7-OH-MTX and internal standard. Extracted ion chromatograms of plasma extracts spiked at 0.25 μmol L⁻¹ for MTX (white) and 7-OH-MTX (grey) and 1 μmol L⁻¹ for internal standard.

Selectivity and cross-talk

Assessment of selectivity needs to be confirmed in the presence of metabolites of the analytes. Some metabolites may be converted into the parent drug during sample preparation and undergo partial fragmentation in the ion sources at high temperatures, giving the same molecular ion as for the parent drug. Because methotrexate polyglutamate forms were only found in red or white blood cells, the main metabolites of MTX found in plasma are 7-OH-MTX and DAMPA after carboxypeptidase infusion.

In this work, the cross-talk phenomenon among MS/MS channels was evaluated injecting MTX, 7-OH-MTX and DAMPA separately, at the concentrations of 10 μmol L⁻¹, and monitoring the response in the other channels including the IS one. Samples spiked with the IS alone were also injected to monitor its interference potential on other drug channels. From these experiments no “cross-talk” was observed.

Moreover, the assay selectivity and interferences were also assessed by analysing extracts from the five different blank plasma samples used to assess the matrix effect; endogenous peaks at the retention time of the analytes of interest were not observed in any of the plasma samples evaluated.

Extraction recovery

Extraction recovery was evaluated for all analytes using standards spiked at the concentration mentioned in Materials and Methods. The mean recovery of MTX and 7-OH-MTX were 24 ± 3% and 57 ± 14%, respectively. Extraction of analytes was consistent over the entire range of the standard curve used (Table 3).

Table 3 Extraction recovery percentage

Concentrations (μmol L^−1)	MTX	7-OH-MTX	^13C^2H3-MTX (mean)
0.05	26	75	25
0.25	24	47	28
0.5	23	61	26
1	21	41	25
10	28	55	28
Mean ± SD	24 ± 3	57 ± 14	26 ± 1

---

Matrix effect assessment

Standard line slope method. The precision values of standard line slopes constructed in five different lots of plasma samples for MTX and 7-OH-MTX are listed in Table 4. The %CV of standard line slopes constructed did not exceed the value of 2.5%. The difference between the lowest and highest slope values obtained in the five different lots did not exceed 6%. These results confirm the absence of the relative matrix effect of our method.^18,20

Table 4 Percentage and inter individual matrix effects, assay performance

Analyte (IS)	Slopes CV (%)	% Slope difference	Measurement performance		Peak area CV (%)
			% Precision CV range	% Accuracy mean ± SD	Analyte (range)	Internal standard
MTX	2.38	5.70%	5.61–10.96	100.7 ± 7.95	3.04–10.98	7
7-OH-MTX	2.47	5.85%	2.87–9.29	100.6 ± 6.72	5.31–14.21

---

Influence of the matrix effect on precision and accuracy. In order to evaluate the influence of the matrix effect on overall method variability, we determined the CV assay range and accuracy by calculating, with one of the standard lines, each point of other calibration curves as an unknown. The precision value for all concentrations used to construct the five standard curves and calculated with the five different curves did not exceed 11% for MTX and 7-OH-MTX. Accuracy values were within 100.7 ± 7.95% for MTX and 100.6 ± 6.72% for MTX and 7-OH-MTX (Table 4).

MTX assay method comparison

We compared FPIA versus LC-MS/MS analysis for 276 patient plasma samples. A highly significant correlation was found between FPIA and LC-MS/MS with a correlation coefficient of r = 0.991 (95% confidence interval from 0.987 to 0.993). The Passing and Bablok method comparison test (P&B) showed no difference between the two methods with no significant deviation from linearity (Fig. 4A). Fig. 4B shows the Bland–Altman systematic difference between the two methods. The average bias was 0.1 μmol L⁻¹, with a 95% confidence interval from −4 to 4.2.

Fig. 4 Method comparison (A) Passing and Bablok regression plot of MTX samples measured with fluorescent polarization immunoassay (FPIA) vs. liquid chromatography tandem mass spectrometry (LC-MS/MS) (n = 276). (B) Bland and Altman plot for MTX measured by FPIA vs. LC-MS/MS (n = 276). The difference between LC-MS/MS and the reference method (FPIA) is plotted against the average of the reference method.

Application to MTX delayed elimination and carboxypeptidase rescue

On May 10, 2011, a 9 years old young boy, suffering from osteoblastic osteosarcoma, was hospitalized for a first cure of neo-adjuvant chemotherapy based on HD-MTX. 12 g of MTX were infused over 4 h. MTX plasma concentration monitored by FPIA indicated 76 μmol L⁻¹ and 32 μmol L⁻¹, respectively, 24 and 30 hours after administration. In view of this delayed elimination, it was decided to treat the patient with carboxypeptidase G2. 6 and 30 hours after CPDG2 administration, the MTX plasma concentration monitored by LC-MS/MS was 0.14 μmol L⁻¹. Although salvage therapy was effective, this delayed MTX elimination caused serious disturbance in renal and hepatic functions. As shown in Fig. 5A, the level of serum creatinine increased rapidly after MTX infusion, peaked at H42, and slowly decreased to reach normal values on day 26. The consequences of MTX intoxication on liver function were delayed (Fig. 5B). Hepatic cytolysis reached a peak around 20 ± 2 days after, with 1276, 627 and 147 IU L⁻¹ for aspartate aminotransferase (reference range 0–39 IU L⁻¹), alanine aminotransferase (reference range 0–51 IU L⁻¹), and γ-GT (reference range 0–19 IU L⁻¹), respectively.

Fig. 5 Clinical application in the setting of delayed MTX elimination treated by carboxypeptidase-G2 The H0 corresponds to the high-dose MTX infusion; H42 corresponds to the CPDG2 infusion. (A) The black bars and grey bars correspond, respectively, to the MTX and 7-OH-MTX plasma concentrations monitored by the LC-MS/MS method described above (left Y-axis). Serum creatinine concentration evolution is represented by open circles on the panel (right Y-axis); the reference ranges are illustrated by the dashed lines. (B) Normalized hepatic enzymes aspartate aminotransferase (ASAT, open circles), alanine aminotransferase (ALAT, open squares) and gamma glutamyl transpeptidase (γGT, open diamonds). Normalization was calculated as the ratio of the measured value and the upper value of the reference range (illustrated by the dashed line at 1 N).

As expected, CPDG2 enabled significant reduction in the MTX plasma concentration and probably attenuation of renal impact. Strangely, carboxypeptidase performances seem to be less effective on 7-OH-MTX than on MTX. 33% of 7-OH-MTX plasma concentrations remained intact where more than 99% of MTX were destroyed. This result seems in line with the only human study to date.⁸ Nevertheless, this raises questions about in vivo carboxypeptidase activity on 7-OH-MTX; an in vitro study could confirm this observation.

Discussion

We describe the performance of an LC-MS/MS method for simultaneous quantification of plasma MTX and 7-OH-MTX concentrations. The main objective of this paper was to proceed to a complete validation of our method in order to be able to monitor MTX after carboxypeptidase infusion.

Assessment of intra- and inter-day variability was ≤12% for all concentrations tested and <17% for LLOQ (Table 2). The method accuracy was found to be within ±15% for intra-run and inter-run. No “cross-talk” from metabolites or endogenous compounds was observed. For all analytes, the signal in the blank matrix was widely inferior to the first standard curve point as well as the LLOQ.

The overall performance of our method was a LLOQ of 25 nmol mL⁻¹ for MTX and 7-OH-MTX, with an initial plasma sample volume of 50 μL. These performances are consistent with TDM of MTX in both children and adults.^9,21,22

For MTX and 7-OH-MTX, the standard line slope CV percentage was <3% and the slope difference <6%, indicating that our analytical method is almost free from a significant relative matrix effect. These results seem to point to the absence of a significant relative matrix effect, as described by Matuszewski et al.^18,20

HDMTX has been associated with severe toxicities in as much as 10% of patients, with a mortality rate estimated between 4.6 and 6% in the 1970s.⁶ Renal dysfunction induced by HDMTX has been estimated to occur in 1.8% of patients; among them 4.4% would die.⁶ MTX-induced renal dysfunction seems to be mediated by the precipitation of MTX and its metabolites in the renal tubules and probably via a direct toxic effect of MTX on the renal tubules. If MTX is poorly soluble at acidic pH, 7-OH-MTX and DAMPA are six- to tenfold less soluble than MTX.⁶ Increasing the urine pH higher than 7.0 results in a greater solubility of these molecules. This finding underlies the recommendation of intravenous hydration and urine alkalinisation. Nonetheless, MTX-induced nephrotoxicity can appear, and a vicious cycle is then initiated: impaired renal function leads to even more delayed clearance of MTX, with further impairment of renal function and high sustained MTX blood levels. In this case, folinic acid rescue is ineffective, with an enhancement of MTX's other toxicities, especially bone marrow suppression and mucositis. Serum creatinine and plasma MTX monitoring are essential for early diagnosis of MTX-induced nephrotoxicity. Patients whose, 48 hours after the beginning of the HDMTX infusion, MTX concentration is >10 μmol L⁻¹ or plasma creatinine is >1.5 times the basal level plus plasma MTX concentration is >3 μmol L⁻¹ required carboxypeptidase administration.^6,9,21 When CPDG2 is infused, initial dramatic reductions of plasma MTX concentrations (>97%) are achieved, and at the same time the DAMPA concentration reaches a peak. Unfortunately, because of antibody cross-reactivity between DAMPA and MTX, none of the immunoassays can be used to monitor MTX concentration within 48 h after CPDG2 administration. After 48 h, because the DAMPA half-life is around 5 to 9 hours, the immunoassay can be used again. In the meantime, MTX concentration can only be reliable by a chromatographic method. In our hospital, HDMTX-induced toxicities in patients with delayed clearance due to impaired renal function are exceptional: only 3 patients required CPDG2 administration in the past five years.

Despite the striking efficacy of carboxypeptidase, MTX monitoring is required in this patient because it's the sole early criterion for assessing the real importance of the efficiency. Indeed, Widemann et al. emphasized that recovery of renal function, defined as serum creatinine within normal limits, occurred at a median of 22 days.⁶

Thus, because of delayed clinical improvement and lack of simple and early biological markers able to assess carboxypeptidase efficiency, there is a feeling of helplessness. This may lead us to consider additional CPDG2 dose, which did not result in additional benefits, neither for MTX plasma concentration additional decreases nor for clinical outcome, and is financially misguided (close to 28 000 € per administration).^7,9

It's expected that this LC-MS/MS method for MTX monitoring after carboxypeptidase administration will be very rarely used since it concerns exceptional cases only. Nevertheless, given the seriousness of the disease, it seemed to be rational and justified to monitor, as quickly as possible, MTX residue. Obviously, because of the nonavailability of suitable equipment, lack of experience in the use of analytical techniques such as chromatography and the exceptional nature of recourse to this technique, a number of laboratories can't offer this monitoring. Therefore, the geographically balanced distribution of University Hospital able to ensure follow-up of these patients, within 24 hours of collection, can draw a satisfactory, if not optimal, solution for the security of the patients.

Conclusion

FPIA is used by a number of laboratories to follow MTX plasma concentrations. However, because of antibody cross-reactivity between DAMPA and MTX, none of the immunoassays can be used after carboxypeptidase administration. Given the seriousness of the MTX delayed elimination, it seemed to be rational and justified to monitor, as quickly as possible, MTX residue. It's why we needed to develop a specific, simple and fast LC-MS/MS method to monitor MTX plasma concentration within the clinically relevant LOQ (25 nmol L⁻¹ to 50 μmol L⁻¹).

References

1. I. D. Goldman and L. H. Matherly, The cellular pharmacology of methotrexate, Pharmacol. Ther., 1985, 28, 77–102.
2. M. G. Rots, et al. Role of folylpolyglutamate synthetase and folylpolyglutamate hydrolase in methotrexate accumulation and polyglutamylation in childhood leukemia, Blood, 1999, 93, 1677–1683.
3. T. A. de Beaumais and E. Jacqz-Aigrain, Intracellular Disposition of Methotrexate in Acute Lymphoblastic Leukemia in Children, Curr. Drug Metab., 2012, 13, 822–834.
4. J. D. Borsi, E. Sagen, I. Romslo and P. J. Moe, Comparative study on the pharmacokinetics of 7-hydroxy-methotrexate after administration of methotrexate in the dose range of 0.5–33.6 g m⁻² to children with acute lymphoblastic leukemia, Med. Pediatr. Oncol., 1990, 18, 217–224.
5. J. Ritter and S. S. Bielack, Osteosarcoma, Ann. Oncol., 2010, 21, vii320–vii325.
6. B. C. Widemann and P. C. Adamson, Understanding and Managing Methotrexate Nephrotoxicity, Oncologist, 2006, 11, 694–703.
7. B. C. Widemann, et al. Glucarpidase, Leucovorin, and Thymidine for High-Dose Methotrexate-Induced Renal Dysfunction: Clinical and Pharmacologic Factors Affecting Outcome, J. Clin. Oncol., 2010, 28, 3979–3986.
8. B. C. Widemann, et al. Carboxypeptidase-G2, thymidine, and leucovorin rescue in cancer patients with methotrexate-induced renal dysfunction, J. Clin. Oncol., 1997, 15, 2125–2134.
9. H. Blasco, et al. The need for a better definition of therapeutic indications of carboxypeptidase in delayed elimination of methotrexate, Therapie, 2008, 63, 19–28.
10. S. Schwartz, et al. Glucarpidase (Carboxypeptidase G2) Intervention in Adult and Elderly Cancer Patients with Renal Dysfunction and Delayed Methotrexate Elimination After High-Dose Methotrexate Therapy, Oncologist, 2007, 12, 1299–1308.
11. V. S. Kumar, T. Law & M. Kellogg, in Clinical Applications of Mass Spectrometry, 2010, pp. 359–363.
12. K. Fotoohi, T. Skärby, S. Söderhäll, C. Peterson and F. Albertioni, Interference of 7-hydroxymethotrexate with the determination of methotrexate in plasma samples from children with acute lymphoblastic leukemia employing routine clinical assays, J. Chromatogr., B: Anal. Technol. Biomed. Life Sci., 2005, 817, 139–144.
13. E. Klapkova, et al. The influence of 7-OH methotrexate metabolite on clinical relevance of methotrexate determination, Clin. Lab., 2011, 57, 599–606.
14. G. Rule, M. Chapple and J. Henion, A 384-Well Solid-Phase Extraction for LC/MS/MS Determination of Methotrexate and Its 7-Hydroxy Metabolite in Human Urine and Plasma, Anal. Chem., 2001, 73, 439–443.
15. P. Guo, et al. Determination of methotrexate and its major metabolite 7-hydroxymethotrexate in mouse plasma and brain tissue by liquid chromatography-tandem mass spectrometry, J. Pharm. Biomed. Anal., 2007, 43, 1789–1795.
16. E. Begas, et al. Simple and Reliable HPLC Method for the Monitoring of Methotrexate in Osteosarcoma Patients, J. Chromatogr. Sci., 2013 DOI:10.1093/chromsci/bmt081.
17. E. Den Boer, et al. A U-HPLC-ESI-MS/MS-based stable isotope dilution method for the detection and quantitation of methotrexate in plasma, Ther. Drug Monit., 2012, 34, 432–439.
18. Committee for Medicinal Products for Human Use (CHMP), Guideline on Bioanalytical Method Validation, EMA, http://www.ema.europa.eu/ema/pages/includes/document/open_document.jsp?webContentId=WC500109686.
19. Food And Drug Administration, Guidance for Industry on Bioanalytical Method Validation, Federal Register, 23 May 2001, vol. 66(100), p. 28526.
20. B. K. Matuszewski, Standard line slopes as a measure of a relative matrix effect in quantitative HPLC-MS bioanalysis, J. Chromatogr., B: Anal. Technol. Biomed. Life Sci., 2006, 830, 293–300.
21. C. Le Guellec, H. Blasco, I. Benz and A. Hulin, Therapeutic drug monitoring of methotrexate after its administration in high-dose protocols, Therapie, 2010, 65, 163–169.
22. L. Holmboe, A. M. Andersen, L. Mørkrid, L. Slørdal and K. S. Hall, High dose methotrexate chemotherapy: pharmacokinetics, folate and toxicity in osteosarcoma patients, Br. J. Clin. Pharmacol., 2012, 73, 106–114.

---