Accepted Manuscript Title: Determination of irinotecan and SN38 in human plasma by TurboFlowTM liquid chromatography-tandem mass spectrometry Author: P. Herviou D. Richard L. Roche J. Pinguet F. Libert A. Eschalier X. Durando N. Authier PII: DOI: Reference:

S0731-7085(15)30215-6 http://dx.doi.org/doi:10.1016/j.jpba.2015.10.044 PBA 10321

To appear in:

Journal of Pharmaceutical and Biomedical Analysis

Received date: Revised date: Accepted date:

3-7-2015 28-10-2015 31-10-2015

Please cite this article as: P.Herviou, D.Richard, L.Roche, J.Pinguet, F.Libert, A.Eschalier, X.Durando, N.Authier, Determination of irinotecan and SN38 in human plasma by TurboFlowrmTM liquid chromatography-tandem mass spectrometry, Journal of Pharmaceutical and Biomedical Analysis http://dx.doi.org/10.1016/j.jpba.2015.10.044 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Determination of irinotecan and SN38 in human plasma by TurboFlow™ liquid chromatography–tandem mass spectrometry P.HERVIOU1,2,3* [email protected], D.RICHARD1,2, L.ROCHE1,2 J.PINGUET1,2, F.LIBERT1,2,4 , A.ESCHALIER1,2,4, X.DURANDO3,5,6, N. AUTHIER1,2,4

1

CHU Clermont-Ferrand, Service de Pharmacologie Médicale, BP69, F-63003

ClermontFerrand. 2

INSERM, U 1107, Neuro-Dol, F-63000 Clermont-Ferrand. Centre Jean Perrin, F-63011 Clermont-Ferrand. 4 Clermont Université, faculté de Médecine, laboratoire de pharmacologie médicale, BP 10448, F-63000 Clermont-Ferrand. 5 CREAT EA 7283, Université d’Auvergne, F-63000 Clermont-Ferrand. 6 Centre d'Investigation Clinique, Inserm U501, F-63000 Clermont-Ferrand. 3

*

Corresponding author at: Service de Pharmacologie Médicale, CHU Clermont-Ferrand, 58 rue Montalembert, 63003 Clermont-Ferrand, France. Tel.: + 33 (0)473751822, fax: +33 (0)473751823, INSERM, U 1107, Neuro-Dol, Faculté de Médecine, 28 place Henri Dunant, 63000 Clermont-Ferrand, France. Centre Jean Perrin, Division de Recherche Clinique, 58 rue Montalembert, BP 392, 63011 Clermont-Ferrand cedex 1, France. Tel.: +33 (0)463663358, fax: +33 (0)473278410.

1

Highlights • • • •

We developed TurboFlow™ LC-MS/MS method for quantify irinotecan and SN38. Turbulent Flow technology provides short time preparation. Full validation procedure certified to EMA and French COFRAC standard. This method is applicable and used for pharmacokinetic studies in clinical trial.

2 Abstract Irinotecan is a cytotoxic agent used in the treatment of metastatic colorectal cancer. Irinotecan is a prodrug when is converted in vivo to an active metabolite SN38, which has potent pharmacological activity. SN38 is then inactivated and excreted as SN38-glucuronide. Highperformance liquid chromatography–mass spectrometry is a widely used bioanalysis technique that can be coupled to the turbulent-flow extraction line to shorten preparation time. A technique was developed to quantify irinotecan and its metabolite by liquid chromatography– tandem mass spectrometry coupled with a turbulent-flow online extraction method. Assays were performed on 100 µL of plasma after protein precipitation. The supernatant is injected directly into the extraction column, transferred to the chromatographic column, and analyzed by tandem mass spectrometry. Linearity, reproducibility and repeatability of the method were validated on a concentration range of 25–2500 ng/mL for irinotecan and 5–500 ng/mL for SN38. For the low limit of quantification of irinotecan and SN38, precision is 6.31% and 8.73%, and accuracy is 84.0% and 91.8%, respectively. The SN38-glucuronide determination protocol included a hydrolyzation step. This method was successfully used to quantify irinotecan, SN38 and SN38-G in human plasma in a clinical trial.

Keywords: irinotecan; high-turbulence liquid chromatography; LC-MS/MS; human plasma 4 1. Introduction Irinotecan is a semi-synthetic derivative of camptothecin that achieves good anti-tumor activity by inhibiting topoisomerase I [1], inducing DNA transcription termination and leading to cell death. This molecule is a prodrug which is used in to the treatment of metastatic colorectal cancer (mCRC), where it is generally used in combination with 5fluorouracil and targeted therapies (cetuximab or bevacizumab). The metabolism of irinotecan is mainly hepatic with biliary elimination. It is oxidized by cytochrome P450 3A (CYP3A) to inactive metabolites 7-ethyl-10-[4-N-(5-aminopentanoic 3

acid)-1-piperidino]-carbonyloxycamptothecin

(APC)

and

7-ethyl-10-[4-amino-(1-

piperidino)carbonyloxycamptothecin (NPC) [2]. Irinotecan is cleaved by carboxylesterase (CES) to 7ethyl-10-hydroxycamptothecin (SN38), a pharmacologically active metabolite 100 to 1000 times more active than the parent drug [3]. SN38 plays a major role in the inhibition of topoisomerase I [4,5] but it also causes major side effects (diarrhea and neutropenia) following irinotecan administration. SN38 is removed by glucuronidation in the liver, mainly by uridine diphosphate glucuronosyltransferases including UGT1A1, which transforms it into an inactive SN38-glucuronide (SN38-G). This combination renders SN38 both inactive and hydrophilic, thus facilitating biliary elimination [6]. Different patients show different rates of SN38 glucuronidation, which explains the interindividual variation in pharmacokinetic parameters of SN38 and toxicties observed after administration of irinotecan. Indeed, genetic polymorphism in the gene promoter (UGT1A1*28) causes stronger downregulation of the enzyme associated with circulating SN38 concentrations lowest than in patients carrying UGT1A1*1 [7]. Prevalance of this mutation is 7.7–8.8% for genotype *28/*28, 41.9–45.6% for genotype *1/*28 and 45.6–50.5% for genotype *1/*1 [8]. A pharmacokinetic study led on 20 patients treated with irinotecan at a dose of 300 mg/m² every 3 weeks found that circulating SN38 rates are higher for carriers of allele *28 [9]. The area under curve (AUC)

ratio SN38-G/SN38 indicates low glucuronidation rates for *1/*28 and *28/*28 patients (p = 0.001). These molecules need to be measured to yield information on the patient’s metabolism and the potential interindividual variability involved. Therapeutic drug monitoring of irinotecan can be divided into two strands, genetic tests to determine the adapted dose, and pharmacokinetic assays to validate the effective circulating concentrations of irinotecan and SN38. Several quantification methods have been developed using high-performance liquid chromatography (HPLC) with fluorimetric/UV detection [10–18] or mass spectrometry/tandem mass spectrometry (LC-MS or LC-MS/MS) after solid or liquid extraction [1,2,5,19–24], and one with online extraction [25], but never with TurbulentFlow® technology. This system is an application of turbulent-flow chromatography columns and automated online sample preparation that enables dramatically shorter sample pretreatment times. The extraction column is composed by a macroporous structure and an adsorbent with large particle size to binds small molecule like drugs, and adapted to high flow rate [26]. A designed loop (focus mode), whose composition and transfer time have been optimized, enables the retained analytes to subsequently get eluted from the extraction column onto the analytical column for chromatographic separation. This online extraction method eliminates the need for timeconsuming off-line sample extraction procedures, thus significantly increasing productivity. This technology is already used for other kinds of application like pharmaceuticals drugs [27,28]. The aim of this study was to develop, validate and apply an LC-MS/MS method coupled with TurboFlow™ technology to quantify irinotecan, SN38 and SN38-G for therapeutic drug monitoring. The optimal circulating concentration of irinotecan established by this method will then be studied in pharmacokinetic assays on human plasma.

5

2. Material and Methods 2.1. Chemicals and reagents All chemical were analytical-grade or high-quality standard. Irinotecan (purity ≥ 97%), SN38 (purity ≥ 98%) and camptothecin (purity ≥ 90%) were from Sigma-Aldrich (Saint-Quentin-Fallavier, France). SN38-glucuronide (purity = 98%) was from Toronto Research Chemicals (North York, ON, Canada). Acetonitrile and acetone were from Carlo Erba Reagents (Val-de-Reuil, France). Formic acid, dimethylsulfoxide, ammonium acetate, ammonium formate, perchloric acid and Tris-HCl were from Fluka Analytical (SigmaAldrich, Saint-Quentin-Fallavier, France). ß-glucuronidase from Helix pomatia was from Sigma-Aldrich (Saint-Quentin-Fallavier, France). Ammonia, isopropanol and methanol were from VWR International (Pessac, France). High-performance aqueous mobile phases were prepared on a Maxima station (Millipore, Molsheim, France). Hemoglobin, bilirubin and intralipid used for matrix effect were from Sigma-Aldrich (Saint-Quentin-Fallavier, France). The plasma used for the development of this method was provided from Etablissement Français du Sang (Saint Etienne, France).

2.2. Preparation of stock solutions, calibration standards and quality control samples Standard stock solutions of irinotecan and camptothecin were prepared in methanol at a concentration of 1 g/L and stored at -20°C. Standard stock solution of SN38 was prepared at 1 g/L in MeOH/ dimethylsulfoxide (65:35, v/v) and stored at -20°C [1]. Working solutions were prepared by diluting standard stock solution in methanol. For calibration standards, working solutions were freshly spiked in plasma to achieve 6 calibration standards at concentration of 25, 50, 100, 500, 1000 and 2500 ng/mL for irinotecan and 5, 10, 20, 100, 200 and 500 ng/mL for SN38. Four quality-control samples (QC) were freshly and separately prepared in plasma and assayed each day samples were analyzed. QC1 (lower limit of quantification, LLOQ) is at

6

25 and 5 ng/mL, QC2 at 75 and 15 ng/mL, QC3 at 750 and 150 ng/mL and QC4 (upper limit of quantification, ULOQ) at 2500 and 500 ng/mL for irinotecan and SN38, respectively. For the calibration curve used to determine SN38-G after the step of hydrolization, working solutions were freshly spiked in plasma to obtain 6 calibration standards at concentration of 5, 10, 20, 100, 200 and 500 ng/mL for SN38. Four QC samples were also prepared, QC1 at 5 ng/mL, QC2 at 15 ng/mL, QC3 at 150 ng/mL and QC4 at 500 ng/mL.

2.3. Sample preparation Fresh blood samples were centrifuged at 3400 rpm for 10 minutes at 4°C and plasma was frozen at -20°C until analysis [1]. For the determination of irinotecan and SN38, protein precipitation is performed on 100 µL of plasma by adding 10 µL of camptothecin (IS) working solution (100 µg/mL) and 10 µL of perchloric acid 30% for deproteinization in polypropylene microtubes. After vortex-mixing for 1 minute followed by centrifugation at 11000 rpm for 10 minutes at 4°C, the supernatant was transferred to insert and 20µL was analyzed by LC-MS/MS. For the determination of SN38-G, enzymatic hydrolysis with ßglucuronidase is carried out just before the deproteinization. The same sample is analyzed hydrolyzed and non-hydrolyzed in order to identify the amount of free SN38 and the glucuronide form using specific calibration curves. The hydrolysis step was based on Zhang’s protocol [5]. Ten µL of IS solution is added to 100 µL of plasma sample. After addition of 95 µL of Tris-HCl buffer and 5 µL of ß-glucuronidase, the sample is incubated for 1 hour at 56°C then deproteinized and analyzed. 2.4. LC-MS/MS system and conditions The LC-MS/MS set-up consisted of a TSQ Quantum Ultra™ (Thermo Fisher Scientific, San Jose, United States) coupled to a Transcend TLX-1 system (Thermo Fisher Scientific). The TurboFlow™ method with automated online sample preparation was performed on a 0.5 × 50

7

mm TurboFlow Cyclone P column (Thermo Fisher Scientific). The analytical column was a 50 × 2.1 mm, 3 µm particle-size Thermo Scientific Hypersil GOLD column. The TLX-1 system is composed of a CTC PAL autosampler at 8°C with an injection DLW system, two binary pumps (Agilent Technologies® 1200 series binary pump SL) and two selective mobile phases (CTC Analytics, Switzerland). One loading pump is dedicated to the extraction line and the other eluting pump is used for the chromatographic separation. During the loading phase (step 1-2), the sample is injected onto the extraction column using ammonium acetate buffer (pH9; 10 mM). After 0.58 minutes, the compounds are eluted from the extraction column by a focus mode (step 3) with H2O 0.1% formic acid-acetonitrile 0.1% formic acid (80:20, v/v) contained in the transfer loop, then out onto the analytical column where they are then eluted with a ramp from 30% to 100% of acetonitrile 0.1% formic acid, in 100 seconds (step 5). The last four steps (6-7-8-9) are column rinsing, loading and equilibration of the two columns ready for the next injection. Total run time is 10.92 minutes (Table 1). A triple-quadrupole mass spectrometer was operated in positive ion mode with a HESI II source. The parameters of the source were used with the following settings: spray voltage 5000 V, vaporizer temperature 300°C, capillary temperature 350°C, sheath gas and auxiliary gas (nitrogen) flow rate at 15 and 10 psi, respectively. MS/MS was acquired in multiple reaction monitoring (MRM) mode in Q1 and Q3. Q2 collision gas (argon) pressure was set to 1.5 mTorr. Determination of Q2 potential settings and MS/MS transitions (Q1 and Q3) was carried out by direct infusion of each compound solution at a concentration of 100 ng/mL diluted in water/acetonitrile (50:50, v/v) with a resolution of 0.7 FWHM. MRM quantification transitions are as follows: irinotecan m/z 587.3→124.1; SN38 m/z 393.2→349.3; IS camptothecin m/z 349.1→305.1. Integration of peak areas, calculation of peak areas ratio, standard calibration curves and determination of drug concentrations were performed on TraceFinder® version 3.1

8

(Thermo Fisher Scientific). Table 2 recaps the selected quantification and confirmation transitions, % ratios and collision energies for irinotecan, SN38 and IS.

2.5. Validation procedure The liquid chromatography–mass spectrometry validation program comprised linearity, accuracy, precision, limit of detection (LOD), limit of quantification (LOQ), matrix effects, carryover, extraction recovery, stability and specificity for each compound in human plasma. All validated parameters allow analyses to be certified to EMA (European Medicines Agency) [29] and the French COFRAC (ISO 15189) standards.

2.5.1. Linearity, accuracy, precision and dilution integrity Calibration curves were calculated using the ratio of peak area of analyte to internal standard with a 1/X weighted quadratic regression. Accuracy describes the closeness of the test results obtained by the method to the nominal concentration, and precision describes the closeness of repeated individual measures of the analyte. Both accuracy and precision should be evaluated within-run (intraday) and between-run (interday). Within-run accuracy was determined by analyzing 5 plasma samples per level at four concentrations (QC1, QC2, QC3 and QC4). Between-run accuracy and precision were evaluated based on 5 replicates of QC samples performed on 5 different days. To validate accuracy, mean concentration should be within 15% of the nominal values for the QC samples, except for true LOQ (QC1) which should be within 20%. To validate precision, CV value should not exceed 20% for LOQ and 15% for other measures. The concentration of each sample was determined using calibration standards prepared on the same day [29]. Calibration standards concentration are 0, 25, 50, 100, 500, 1000 and 2500 ng/mL for irinotecan and 0, 5, 10, 20, 100, 200 and 500 ng/mL for SN38.

9

Dilution integrity was demonstrated by spiked the matrix with working solution at QC3 level and diluted in half in matrix, on 5 replicates. Accuracy and precision were validated if CV values not exceed 15%.

2.5.2. LOD, LOQ The theoretical limits of detection (LOD) and quantification (LOQ) were calculated by measuring the analytical background response of 5 drug-free extracts from plasma using the maximum sensitivity allowed by the system. Signal-to-noise ratio was used to determine LOD and was estimated as the concentration of each molecule in 5 samples that generated a peak area at least 3 times higher than background. Theoretical lower LOQ (LLOQ) was considered as 5 times the standard deviation of the 5 blank samples analyzed using the maximum sensitivity allowed by system. For the validation, real LLOQ is determined by the lowest concentration level that can be quantified with acceptable accuracy and precision (n=5) [29].

2.5.3. Matrix effects Ion suppression was performed by directly infusing quality control (QC3 level) into the mass spectrometer during the chromatographic analysis of blank plasma extracts from 6 different sources (all from Etablissement Français du Sang, France). The signal for each molecule should remain undisturbed at the retention time of the analyte. Matrix effect between plasma and water was performed by comparing values obtained with 4 levels of quality control (LLOQ, two ‘medium’, LLOQ), spiked in water or in plasma, on three days. For each level, percentage of matrix effect was calculated. Matrix effects were also validated between serum, EDTA plasma, and heparinate plasma at two levels of quality control and also on hemolytic (hemoglobin 1 g/L), icteric (bilirubin 100 mg/L) and lipemic (intralipid 1g/L) matrix, again with two levels of quality control (QC1 and QC4).

10

To evaluate the matrix factor (MF), 6 different blank plasma were extracted and then spiked with the working solution of analyte and IS at 2 level control (QC1 and QC3). Peak area of the analyte in spiked plasma post-extraction was compared to those spiked in water. The ratio is defined as MF. The IS-normalized MF was also calculated by dividing the MF of the analyte by the MF of the IS. The CV of the IS-normalised MF calculated should be not greater than 15%.

2.5.4. Carryover and extraction recovery To study carryover, blank samples were injected after the calibration standard at the ULOQ on 4 consecutive days. Carryover in the blank sample after the high-concentration standard should not be greater than 20% of the LLOQ [29]. Extraction recoveries were calculated by comparing values obtained with 4 levels of extracted and non-extracted quality control (LLOQ, two ‘medium’, ULOQ) on three days. Extracted quality control uses the TurboFlow™ technology, whereas non-extracted quality controls are directly injected with standard chromatographic system.

2.5.5. Stability and specificity The matrix stability of irinotecan and SN38 was tested at -20°C (after two freeze-thaw cycles) for 7 and 14 days at two levels of control (LLOQ and high level). To validate stability of stock solution after 1 month, QC samples (LLOQ, two ‘medium’, ULOQ) were realized with new stock solution, and calibration curve was realized with old stock solution. Storage stability of molecules in the reconstitution solution was assessed by reinjection of QC samples (LLOQ and high level) after remaining in the autosampler (8 °C) for 24 h following initial injection. The stability of analytes in matrix was also tested at short term, at room temperature. Two levels of control (LLOQ and high level) were spiked in matrix, and left at room temperature for 1, 3

11

and 6 hours. The concentration of analytes after each time point was compared to the nominal concentration. Accuracy and precision were validated if CV values not exceed 15%. The specificity of the analytical method was determined by analyzing a control spiked in a complex matrix with more molecules: acetaminophen, amikacin, amiodaron, amitriptyline, caffeine, carbamazepin, chloramphenicol, cortisol, cyclosporine, desipramin, digoxin, disopyramide, estriol, ethosuximide, flecainide, gentamicin, haloperidol, imipramine, lidocaine, lithium, methotrexate, N-acetylprocainamide, netylmicin, nortriptyline, phenobarbital, phenytoin, primidone, procainamide, propranolol, quinidine, salicylate, T3 and T4, theophyline, tobramycine, TSH, valproate and vancomycin (Liquichek™ Therapeutic Drug Monitoring 724-MSDS, Biorad, Marnes-la-Coquette, France). Acceptance criterion for precision was 80% for level QC2. For the interference test, plasma was spiked with working solution of irinotecan, SN38 or camptothecin, at the QC2 level. To identify back-conversion from SN38-G to SN38 in sample and in ion source, working solution of SN38-G was spiked plasma at CQ level (LLOQ and high level) and analyzed in the same time of stability test of irinotecan and SN38.

2.6. Pharmacokinetics studies For pharmacokinetics studies, blood samples were obtained from a patient at the first course of treatment by the FOLFIRI (irinotecan–5-fluorouracil–folinic acid) regimen. The samples were taken at 0, 30, 60, 90 min, 2, 4, 8, 12 and 24 hours after the start of irinotecan infusion. Irinotecan dose was 370 mg/m². Plasma was obtained by centrifugation at 3400 rpm for 10 min at 4°C and stored at -20°C until analysis. The plasma irinotecan concentration curves were used to determine the peak plasma concentration (Cmax), area under the curve (AUC) and half-life (T½) of irinotecan. This clinical study was approved by Ethics Committee (Comité de Protection des Personnes Sud-Est 6, Clermont-Ferrand, France) and was declared

12

and approved by the Agence nationale de Sécurité du Médicament et des produits de santé (ANSM, France).

3. Results and Discussion 3.1. Optimization of chromatographic conditions Loading phase and TLX column were the first parameters optimized. Several Turboflow columns were tested (all Thermo Fisher Scientific): polymeric apolar phase (Cyclone, Cyclone P), cationic/anionic polymeric phase (Cyclone MCX, Cyclone MAX), silica apolar phase (C18XL and C18PXL). To enhance compound extraction, a mobile loading phase covering a wide pH range (pH 3 to pH 9, 0% to 100%) and all combinations were tested. Cyclone P extracted the highest amount of compound with ammonium acetate (pH9; 10 mM). The second optimization step focused on loop composition (focus mode) and influence of loading flow rate during analytes transfer onto the chromatographic column. The best result was obtained with an elution with water 0.1% formic acid-acetonitrile 0.1% formic acid (80/20, v/v), at 0.2 mL/min for 100 seconds. The next step was to optimize the chromatographic separation. Several analytical columns were tested with different stationary phase (all Thermo Fisher Scientific):Hypersil Gold PFP, Hypersil Gold C18 1.9 µm and 3 µm and Betasil Phenyl Hexyl. Phase composition was tested at the start of the chromatographic run (100%, 95% or 90% of ammonium formate 0.1% formic acid (pH3; 10 mM) on the eluting pump) and over the full course of the run. The last step was to optimize gradient of the mobile phase and the time window of the ramp program. The selected mobile phase was able to discriminate between analytes. Percentage of organic/aqueous phase was optimized to yield the best analyte separation with satisfactory peak performance and high sensitivity. The resulting chromatograms (data not shown) demonstrated the specificity of the assay with a clean elution peak and no observed interference in the region

13

of elution. In the final method, molecules are separated onto Hypersil Gold 3 µm and eluted with a ramp from 30% to 100% of acetonitrile 0.1% formic acid for 100 s. Retention times were 4.57 min for irinotecan, 4.67 for SN38 and 4.68 for camptothecin. It was not possible to quantify SN38-G with this online extraction system as the glucuronide is not retained on the TurboFlow™ column chosen for the analysis of other molecules (Cyclone P). SN38-G was therefore quantified indirectly by enzymatic hydrolysis. The SN38-G concentration was determined by subtraction between SN38 concentration after enzymatic-hydrolysis and before. This method has been already published by Zhang [5] and used in analysis of patient’s samples.

3.2. Optimization of mass spectrometry detection MS/MS conditions were optimized for irinotecan, SN38 and campthotecin (IS) through infusion to ascertain precursor ion then select the product ions in MRM mode. Protonated molecule [M+H]+ were the most abundant ions, with m/z 587.3 for irinotecan, m/z 383.2 for SN38 and 349.1 for IS. Different MS parameters were evaluated to optimize the compound ionization process. Parameters such as spray voltage, vaporizer temperature, capillary temperature, sheath gas and auxiliary gas flow rate were optimized to obtain the best resolution, selectivity and sensitivity for irinotecan and SN38. Also, the best result were obtain with spray voltage at 5000 V, vaporizer temperature at 300°C, capillary temperature at 350°C, and sheath gas and auxiliary flow rate at 15 and 10 psi respectively, in positive mode. Mass spectra of irinotecan and SN38 were given in Figure 1, with collision energy at 30 eV. The transition m/z 587.3124.1 for irinotecan, m/z 393.2349.3 for SN38 and m/z 349.1305.1 for IS where choose for MRM quantification transitions. There is no in-source fragmentation for these two molecules.

14

3.3. Extraction from biological matrices A simple protein precipitation procedure was successfully applied to extract analytes from plasma. Several deproteinizing agents were compared, i.e. perchloric acid 30%, methanol, acetonitrile, and a 50:50 (v/v) mix of methanol/acetonitrile. The best result, based on the presence of molecule and the property of supernatant, was obtained with 10 µL of perchloric acid 30% added to 100 µL of plasma sample.

3.4. Validation of the method 3.4.1. Linearity, accuracy, precision, LOQ, LOD and dilution integrity The method was validated in human plasma for irinotecan and SN38 at 6 concentrations ranging from 25 to 2500 ng/mL and 5 to 500 ng/mL, respectively. Calibration curve was linear with average correlation coefficients (r²) greater than 0.9950 in plasma. With 100-µL samples, the LOQs are 25 and 5 ng/mL and the theoretical LODs are 3.97 and 1.41 ng/mL, respectively. For irinotecan at the LOQ, the intraday precision is 6.31 % and intraday accuracy is 84.0 %. For SN38, intraday precision is 8.73% and intraday accuracy is 91.8 %. Intraday and interday accuracy and precision values are given in Table 3. To validate dilution integrity, diluted to half QC3 level is analyzed in 5 replicates, and accuracy and CV precision values are less than 15% both for irinotecan and SN38 (Table 3). A representative chromatogram for plasma samples of irinotecan and SN38 at the LOQ and blank plasma are presented in Figure 2. 3.4.2. Matrix effect, carryover and extraction recovery No ion suppression was observed at the retention times of irinotecan and SN38 during the infusion of CQ3 sample and the chromatographic blank extract matrices (Figure 3). The whole validation was performed in 6 different lots plasma, in order to be the most possible in the same condition of patient’s analysis. Endogenous compounds present in these different matrices do not interfere with the quantification of irinotecan and SN38. No matrix effect was

15

identified between quality control spiked in plasma or water, with mean percentage of matrix effect at 98.4% for irinotecan and 106% for SN38. Moreover, there were no significant differences between the values of controls spiked in hemolytic, lipemic or icteric matrices compared to controls spiked in normal plasma. Results of mean matrix factor (MF) for irinotecan and SN38 are 1.16 and 5.20 at QC1 level, and 7.05 and 2.19 at QC3 level, calculated in 6 different plasma lots. Overall IS-normalised MF, in a same conditions, are 0.25 and 1.21 for irinotecan and SN38 at QC1 level and 0.61 and 3.23 at QC3 level. CV obtained between ISnormalised MF are greater than 15 %, with values for CQ1 level at 11.9 for irinotecan and 12.6 for SN38, and 14.7 and 8.3 at QC3 level. Extraction recovery was obtained after comparing three direct injections and three TurboFlow™ injections of each compound at four concentrations, spiked in water. Extraction recoveries are 110% for irinotecan and 88.4% for SN38. Carryover was less than 20% for each compound. There was no evidence of carryover causing elevated measurements of the drugs in the lowest calibrator or quality controls.

3.4.3. Stability and specificity For the stability test, accuracy values at LLOQ and high-level control were less than 15%, indicating no degradation of irinotecan and SN38 in plasma during two freeze-thaw cycles at 20°C (at day 1, day 7 and day 14). Fourteen days is the maximum delay for analyze patient’s sample in the context of therapeutic drug monitoring. Stock solution is stable during 1 month. The 24-h reinjection measurements were consistent with the initial run, allowing samples extracted from human plasma to be reanalyzed on the following day when necessary. Short term stability in human plasma at room temperature was assessed up to 6 hours. All analytes demonstrated stability (15% difference from prepared extracted and the nominal

16

concentration) (Table 4). For the specificity analysis, medium QC level (QC2) was spiked in complex matrix with several molecules (multiparametric control, molecules listed in section 2.5.5). Accuracy is 95.4% for irinotecan and 98.5% for SN38. This technique is able to differentiate and quantitate analytes in the presence of other molecule in the sample. No significant interference or overestimation in values of CQ levels of SN38 proved absence of backconversion from SN38G to SN38 during the successive steps of freeze-thaw cycles.

3.4.5. Analysis of patient samples The validated LC-MS/MS method was applied to assay the clinical pharmacokinetics of irinotecan, SN38 and SN38-G in a plasma sample from patient treated with irinotecan. For example, for one patient treated with 370 mg/m² of irinotecan, pharmacokinetic curves of irinotecan, SN38 and SN38-G are given in Figure 4. The Cmax of irinotecan (3150 ng/mL) was achieved at 90 minutes after the start of infusion. As these result is upper the high level of calibration curve of irinotecan, the sample was previously diluted. To values contained in a range of concentration, no dilution is necessary. T½ of irinotecan is about 5 hours 30 min. AUC0-24h of irinotecan is 15.8 µg.h/mL. Tmax of SN38 is 60 minutes, it’s 90 minutes for SN38G. At these times, Cmax of SN38 and SN38-G are 35 ng/mL and 241 ng/mL, respectively. SN38 and SN38-G AUC0-12h are 232 ng.h/mL and 1523 ng.h/mL

4. Conclusion This paper presents a novel method for rapid simultaneous quantification of irinotecan and its main metabolite SN38. The method couples high performance chromatography with tandem mass spectrometry and is validated here as, specific, accurate, precise, and easy to implement in clinical practice. It employs TurboFlow™ technology, enabling fast preparation

17

and analysis on low sample volume, while giving robust results to short deadlines. Turbulentflow chromatography thus emerges as an excellent tool for on-line pretreatment of blood samples. Combined with LC-MS/MS, it offers a very fast, robust and economical analytical platform for pharmacokinetic studies.

18

References [1] S. Ragot, P. Marquet, F. Lachâtre, A. Rousseau, E. Lacassie, J.M. Gaulier, et al., Sensitive determination of irinotecan (CPT-11) and its active metabolite SN-38 in human serum using liquid chromatography-electrospray mass spectrometry, J. Chromatogr. B. Biomed. Sci. App. 736 (1999) 175–184. [2] F. D’Esposito, B.N. Tattam, I. Ramzan, M. Murray, A liquid chromatography/electrospray ionization mass spectrometry (LC-MS/MS) assay for the determination of irinotecan (CPT-11) and its two major metabolites in human liver microsomal incubations and human plasma samples, J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 875 (2008) 522–530. doi:10.1016/j.jchromb.2008.10.011. [3] S. Poujol, F. Pinguet, F. Malosse, C. Astre, M. Ychou, S. Culine, et al., Sensitive HPLCfluorescence method for irinotecan and four major metabolites in human plasma and saliva: application to pharmacokinetic studies, Clin. Chem. 49 (2003) 1900–1908. [4] Y. Kawato, M. Aonuma, Y. Hirota, H. Kuga, K. Sato, Intracellular roles of SN-38, a metabolite of the camptothecin derivative CPT-11, in the antitumor effect of CPT-11, Cancer Res. 51 (1991) 4187–4191. [5] W. Zhang, G.E. Dutschman, X. Li, M. Ye, Y.-C. Cheng, Quantitation of Irinotecan and its two major metabolites using a liquid chromatography-electrospray ionization tandem mass spectrometric, J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 877 (2009) 3038–3044. doi:10.1016/j.jchromb.2009.07.025. [6] M. Ramesh, P. Ahlawat, N.R. Srinivas, Irinotecan and its active metabolite, SN-38: review of bioanalytical methods and recent update from clinical pharmacology perspectives, Biomed. Chromatogr. BMC. 24 (2010) 104–123. doi:10.1002/bmc.1345. [7] G. Toffoli, E. Cecchin, G. Gasparini, M. D’Andrea, G. Azzarello, U. Basso, et al., Genotype-driven phase I study of irinotecan administered in combination with fluorouracil/leucovorin in patients with metastatic colorectal cancer, J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 28 (2010) 866–871. doi:10.1200/JCO.2009.23.6125. [8] G. Toffoli, E. Cecchin, G. Corona, A. Russo, A. Buonadonna, M. D’Andrea, et al., The role of UGT1A1*28 polymorphism in the pharmacodynamics and pharmacokinetics of irinotecan in patients with metastatic colorectal cancer, J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 24 (2006) 3061–3068. doi:10.1200/JCO.2005.05.5400. [9] L. Iyer, S. Das, L. Janisch, M. Wen, J. Ramírez, T. Karrison, et al., UGT1A1*28 polymorphism as a determinant of irinotecan disposition and toxicity, Pharmacogenomics J. 2 (2002) 43–47. [10] T. Bansal, A. Awasthi, M. Jaggi, R.K. Khar, S. Talegaonkar, Development and validation of reversed phase liquid chromatographic method utilizing ultraviolet detection for quantification of irinotecan (CPT-11) and its active metabolite, SN-38, in rat plasma and bile samples: application to pharmacokinetic studies, Talanta. 76 (2008) 1015–1021. doi:10.1016/j.talanta.2008.04.058. [11] M.-T. Baylatry, A.-C. Joly, J.-P. Pelage, L. Bengrine-Lefevre, J.-L. Prugnaud, A. Laurent, et al., Simple liquid chromatography method for the quantification of irinotecan and SN38 in sheep plasma: application to in vivo pharmacokinetics after pulmonary artery chemoembolization using drug eluting beads, J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 878 (2010) 738–742. doi:10.1016/j.jchromb.2010.01.017.

19

[12] F.A. de Jong, R.H.J. Mathijssen, P. de Bruijn, W.J. Loos, J. Verweij, A. Sparreboom, Determination of irinotecan (CPT-11) and SN-38 in human whole blood and red blood cells by liquid chromatography with fluorescence detection, J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 795 (2003) 383–388. [13] W. Guo, A. Ahmad, S. Khan, F. Dahhani, Y.F. Wang, I. Ahmad, Determination by liquid chromatography with fluorescence detection of total 7-ethyl-10-hydroxycamptothecin (SN-38) in beagle dog plasma after intravenous administration of liposome-based SN-38 (LE-SN38), J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 791 (2003) 85–92. [14] Z.-P. Hu, X.-X. Yang, X. Chen, E. Chan, W. Duan, S.-F. Zhou, Simultaneous determination of irinotecan (CPT-11) and SN-38 in tissue culture media and cancer cells by high performance liquid chromatography: application to cellular metabolism and accumulation studies, J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 850 (2007) 575–580. doi:10.1016/j.jchromb.2006.12.056. [15] J. Escoriaza, A. Aldaz, C. Castellanos, E. Calvo, J. Giráldez, Simple and rapid determination of irinotecan and its metabolite SN-38 in plasma by high-performance liquid-chromatography: application to clinical pharmacokinetic studies, J. Chromatogr. B. Biomed. Sci. App. 740 (2000) 159–168. [16] D.F. Chollet, L. Goumaz, A. Renard, G. Montay, L. Vernillet, V. Arnera, et al., Simultaneous determination of the lactone and carboxylate forms of the camptothecin derivative CPT-11 and its metabolite SN-38 in plasma by high-performance liquid chromatography, J. Chromatogr. B. Biomed. Sci. App. 718 (1998) 163–175. [17] T.S. Owens, H. Dodds, K. Fricke, S.K. Hanna, K.R. Crews, High-performance liquid chromatographic assay with fluorescence detection for the simultaneous measurement of carboxylate and lactone forms of irinotecan and three metabolites in human plasma, J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 788 (2003) 65–74. [18] N.E. Schoemaker, H. Rosing, S. Jansen, J.H.M. Schellens, J.H. Beijnen, Highperformance liquid chromatographic analysis of the anticancer drug irinotecan (CPT-11) and its active metabolite SN-38 in human plasma, Ther. Drug Monit. 25 (2003) 120– 124. [19] S. Bardin, W. Guo, J.L. Johnson, S. Khan, A. Ahmad, J.X. Duggan, et al., Liquid chromatographic-tandem mass spectrometric assay for the simultaneous quantification of Camptosar and its metabolite SN-38 in mouse plasma and tissues, J. Chromatogr. A. 1073 (2005) 249–255. [20] X. Chen, C.J. Peer, R. Alfaro, T. Tian, S.D. Spencer, W.D. Figg, Quantification of irinotecan, SN38, and SN38G in human and porcine plasma by ultra high-performance liquid chromatography-tandem mass spectrometry and its application to hepatic chemoembolization, J. Pharm. Biomed. Anal. 62 (2012) 140–148. doi:10.1016/j.jpba.2012.01.008. [21] S. Khan, A. Ahmad, W. Guo, Y.-F. Wang, A. Abu-Qare, I. Ahmad, A simple and sensitive LC/MS/MS assay for 7-ethyl-10-hydroxycamptothecin (SN-38) in mouse plasma and tissues: application to pharmacokinetic study of liposome entrapped SN-38 (LE-SN38), J. Pharm. Biomed. Anal. 37 (2005) 135–142. doi:10.1016/j.jpba.2004.09.053. [22] J. Zhou, S. Gao, F. Zhang, B. Jiang, Q. Zhan, F. Cai, et al., Liquid chromatography– tandem mass spectrometry method for simultaneous determination of seven commonly used anticancer drugs in human plasma, J. Chromatogr. B. 906 (2012) 1–8. doi:10.1016/j.jchromb.2012.07.033. [23] E. Marangon, B. Posocco, E. Mazzega, G. Toffoli, Development and validation of a highperformance liquid chromatography-tandem mass spectrometry method for the

20

[24]

[25]

[26]

[27]

[28]

[29]

simultaneous determination of irinotecan and its main metabolites in human plasma and its application in a clinical pharmacokinetic study, PloS One. 10 (2015) e0118194. doi:10.1371/journal.pone.0118194. D.J. Park, J.H. Won, A.R. Cho, H.J. Yun, J.H. Heo, T.H. Hwhang, et al., Determination of irinotecan and its metabolite SN-38 in rabbit plasma and tumors using a validated method of tandem mass spectrometry coupled with liquid chromatography, J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 962 (2014) 147–152. doi:10.1016/j.jchromb.2014.05.042. G. Ahn, D.M. Park, J.W. Park, H.-Y. Kim, J.-Y. Cho, S. Rhee, et al., A rapid, simple and reliable HPLC-triple quadrupole tandem mass spectrometer method for a simultaneous quantification of irinotecan and its active metabolite 7-ethyl-10-hydroxycamptothecin (SN38) in mouse plasma, Biomed. Chromatogr. BMC. 28 (2014) 919–922. doi:10.1002/bmc.3134. U. Ceglarek, J. Lembcke, G.M. Fiedler, M. Werner, H. Witzigmann, J.P. Hauss, et al., Rapid simultaneous quantification of immunosuppressants in transplant patients by turbulent flow chromatography combined with tandem mass spectrometry, Clin. Chim. Acta Int. J. Clin. Chem. 346 (2004) 181–190. doi:10.1016/j.cccn.2004.03.017. P.C. Dolder, M.E. Liechti, K.M. Rentsch, Development and validation of a rapid turboflow LC-MS/MS method for the quantification of LSD and 2-oxo-3-hydroxy LSD in serum and urine samples of emergency toxicological cases, Anal. Bioanal. Chem. 407 (2015) 1577–1584. doi:10.1007/s00216-014-8388-1. R. López-Serna, M. Petrović, D. Barceló, Direct analysis of pharmaceuticals, their metabolites and transformation products in environmental waters using on-line TurboFlowTM chromatography-liquid chromatography-tandem mass spectrometry, J. Chromatogr. A. 1252 (2012) 115–129. doi:10.1016/j.chroma.2012.06.078. Guideline Bioanalytical method validation WC500109686.pdf, (n.d.). http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2011/08/ WC500109686.pdf (accessed October 17, 2014).

21

Figure Captions Figure 1: Mass spectra for ion m/z 587.3 for irinotecan and ion m/z 383.2 for SN38 at 30 eV. The fragmention at m/z 249 for SN38 is not shown in the mass spectrum because it requires, for its formation, a higher collision energy than the other fragments.

22

Figure 2: LC–MS/MS MRM chromatogram of the lower limit of quantification of a sample containing irinotecan, SN38 and IS camptothecin (A) and blank plasma (B).

23

Figure 3: Chromatograms of blank extract with post-column infusion of medium-level control (750 ng/mL for irinotecan and 150 ng/mL for SN38).Retention times of irinotecan (4.57 min) and SN38 (4.67 min) are represented by black lines.

24

Tables Table 1: Chromatographic conditions for the loading and eluting pump. Total run time is 10.92 minutes. Steps 1 and 2: sample loading, step 3: transfer of analytes onto the analytical column, steps 4 and 5: elution, step 6 and 7: washing the two columns, step 8: loading transfer loop (loading pump) and washing the analytical column, step 9: system equilibration. LOADING PUMP Time

ELUTING PUMP

Flow

Flow

Step

Grad

A/B

%A

%B

Tee

Loop

Grad

A/B

%A

%B

(min)

(mL/min)

(mL/min)

1 2 3

0.00 0.50 0.58

1.50 0.10 0.20

Step Step Step

A1/B1 A1/B1 A1/B1 -

100 100 100

=== === T

out out in

0.50 0.40 0.50

Step Step Step

A1/B1 95 A1/B1 95 A1/B1 95

5 5 5

4 5 6

2.25 3.08 4.75

1.50 1.50 1.50

Step Step Step

A1/B1 A2/B2 100 A2/B1 -

100 100

=== === ===

in in in

0.50 0.50 0.50

Step Ramp Step

A1/B1 70 A1/B1 A1/B1 -

30 100 100

7 8 9

5.75 6.75 8.58

1.50 1.50 1.50

Step Step Step

A1/B2 A1/B2 80 A1/B1 -

100 20 100

=== === ===

in in out

0.50 0.50 0.50

Step Step Step

A2/B1 60 A2/B1 10 A1/B1 95

40 90 5

LOADING: A1 0.1% formic acid in water; A2 isopropanol, acetonitrile and acetone (20/30/50, v/v/v); B1 ammonium acetate (pH9; 10 mM); B2 acetonitrile 0.1% formic acid ELUTING: A1 ammonium formate pH 3; 10 mM); A2 isopropanol, acetonitrile and acetone (20/30/50, v/v/v); B1 acetonitrile 0.1% formic acid

25

Table 2: Selected reaction monitoring parameters

Confirmation product ion 1 Product Collision Relative ion (m/z) energy (V) abundance (%)

Precursor ion (m/z)

Quantification product ion (m/z)

Collision energy (V)

Irinotecan

587.30

124.1

34

167.1

37

69

SN38

393.20

349.3

25

249.2

49

17

Camptothecin

349.14

305.1

43

219.1

20

184

Confirmation product ion 2 Product Collision Relative ion (m/z) energy (V) abundance (%) Irinotecan SN38 Camptothecin

195.2 293.1 248.1

28 33 29

50 10 41

26

Table 3: Results of quantitative method validation in plasma. QC3 diluted correspond to QC3 level diluted in half, in matrix. QC1 is the LLOQ and QC4 the high level. Irinotecan SN38 QC1 QC2 QC3 QC4 QC1 QC2 QC3 QC4 Nominal concentration (ng/mL) 25 75 750 2500 5 15 150 500 Intra day (n=5) Mean concentration (ng/mL) Accuracy (%) CV Precision (%)

21.0 84.0 6.31

67.1 89.4 7.58

738 98.4 4.07

2342 93.7 4.53

5.41 91.8 8.73

15.1 99.4 11.7

152 98.8 4.02

573 85.4 1.64

Interday (n=5) Mean concentration (ng/mL) Accuracy (%) CV Precision (%)

24.1 96.3 7.62

70.6 94.1 12.2

777 96.4 12.1

2247 89.9 7.06

4.88 97.6 7.24

13.5 89.8 12.6

155 96.5 7.53

487 97.5 11.6

Mean concentration (ng/mL) Accuracy (%) CV Precision (%)

QC3 diluted (n=5)

QC3 diluted (n=5)

692 92.3 4.08

168 87.9 4.39

For precision, coefficient of variation (CV) between mean concentration and the nominal values should be within 15%, except for true LLOQ (QC1) which should be within 20%.

27

Table 4: Stabilities of irinotecan and SN38 in plasma. QC1 is the LLOQ and QC4 the high level. Irinotecan SN38 QC1 QC4 QC1 QC4 Nominal concentration (ng/mL) 25 2500 5 500 Short term stability 1 h Concentration (ng/mL) Accuracy (%) 3 h Concentration (ng/mL) Accuracy (%) 6 h Concentration (ng/mL) Accuracy (%)

27.8 88.6 23.9 99.9 25.2 99.3

2596 96.1 2695 92.2 2579 96.9

4.58 91.6 5.69 99.8 4.61 92.2

487 97.5 491 98.1 540 90.1

Freeze-thaw stability Day 1 Concentration (ng/mL) Accuracy (%) Day 7 Concentration (ng/mL) Accuracy (%) Day 14 Concentration (ng/mL) Accuracy (%)

28.7 85.1 24.5 97.8 22.3 89.3

2536 98.6 2514 99.4 2139 85.6

5.39 92.2 4.98 99.6 4.33 86.6

514 97.1 497 99.4 476 95.1

Re-injection stability 24 h Concentration (ng/mL) Accuracy (%)

28.3 86.8

2819 87.2

5.56 88.8

509 98.2

28

Determination of irinotecan and SN38 in human plasma by TurboFlow™ liquid chromatography-tandem mass spectrometry.

Irinotecan is a cytotoxic agent used in the treatment of metastatic colorectal cancer. Irinotecan is a prodrug when is converted in vivo to an active ...
566B Sizes 2 Downloads 17 Views