Determination of Crizotinib in Mouse Tissues by LC-MS/MS and Its Application to a Tissue Distribution Study

Toxicity induced by crizotinib, a small-molecule tyrosine kinase inhibitor, is a significant clinical issue during treatment. A tissue distribution study is required to explore the organs affected by this molecule. In this study, a simple liquid chromatography tandem mass spectrometry method was developed and validated for the determination of crizotinib in various mouse tissues. Mouse tissue homogenates were processed by protein precipitation with methanol, and apatinib was chosen as the internal standard. The analytes were separated on a Phenomenex Kinetex C18 (50 mm × 2.1 mm, 2.6 μm) column with gradient elution using methanol and 0.3% formic acid water solution. Tandem mass spectrometric detection was conducted using multiple reaction monitoring via an electrospray ionization source in the positive mode. The monitored ion transitions were m/z 450.1 ⟶ 260.2 for crizotinib and m/z 398.2 ⟶ 212.0 for apatinib. The problem of the severe carryover effect was successfully resolved. The method was validated and applied to a tissue distribution study of crizotinib in mice, which was reported for the first time. The results of the study showed that the main target organs of crizotinib were the lung, liver, and spleen, and a high concentration of crizotinib was found in the gastrointestinal tract. This study offers a reliable method for quantifying crizotinib and provides a basis for further research on crizotinib toxicity.


Introduction
Crizotinib (Figure 1(a)) is an oral multitargeted receptor tyrosine kinase inhibitor that targets anaplastic lymphoma kinase (ALK), ROS proto-oncogene 1 (ROS1), and hepatocyte growth factor MET receptor tyrosine kinases [1,2]. Crizotinib was approved by the FDA as a first-line treatment of ALKrearranged nonsmall cell lung cancer (NSCLC) in 2011 [1] and was approved in the United States and European Union for the therapy of patients with ROS1-positive advanced NSCLC in 2016 [3]. To date, crizotinib remains the only approved, standard first-line therapy for ROS1-rearranged NSCLC [4,5] and is highly effective in the clinic [6]. Furthermore, it represents a clinically meaningful treatment option for NSCLC with MET alterations [2]. Compared with chemotherapy, crizotinib showed a better therapeutic effect in ALK-positive lung cancer [7]. erefore, crizotinib is used globally for NSCLC treatment. However, the toxicity induced by crizotinib is a challenging clinical problem during the treatment of NSCLC [8][9][10], although crizotinib is safer and more effective than chemotherapy. To explore the mechanism of its toxicity and preventive measures, it is necessary to investigate the tissue distribution of crizotinib at toxic doses.
An accurate and reliable bioanalytical method is needed to detect crizotinib in different biosamples for tissue distribution. To our knowledge, several liquid chromatography tandem mass spectrometry (LC-MS/MS) methods have been developed for the determination of crizotinib in human plasma for therapeutic drug monitoring [11][12][13][14] or pharmacokinetic studies [15]. In addition, Qiu et al. [16] and Sparidans et al. [17] reported two LC-MS/MS methods for the quantification of crizotinib in rat or mouse plasma and applied it to pharmacokinetic studies. However, no successful method for the quantitative determination of crizotinib in mouse tissues has been reported thus far. Moreover, there have been no reports on the tissue distribution of crizotinib at toxic doses in mice.
In this study, a simple liquid chromatography tandem mass spectrometry method was developed and validated for the determination of crizotinib in various mouse tissues. e method was successfully applied to the tissue distribution of crizotinib in mice after the oral administration of crizotinib.

Chemicals and Reagents.
Crizotinib (purity >99%) for the analytical method and dosage administration was purchased from Dalian Meilun Biology Technology Co., Ltd. (Dalian, China). e reference standards of apatinib ( Figure 1(b), internal standard (IS) purity ≥99%) and formic acid (chromatographic grade) were purchased from Aladdin Industrial Corporation (Shanghai, China). Sodium carboxymethylcellulose (CMC-Na) was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Methanol, acetonitrile, and isopropanol were all HPLC grade and were purchased from Fisher Scientific (Shanghai, China). Deionized water was purified using a Milli-Q system (Millipore, Milford, MA, USA).

Animals.
Institute of Cancer Research mice (20 ± 2 g, half male and half female) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). ey were kept in an environmentally controlled breeding room for one week before starting the experiments and were fed standard laboratory food and water. e mice were fasted with free access to water for at least 10 h before the administration of the drugs. Animal studies were performed in accordance with the Guide for the Care National Institutes of Health. Ethical approval for the animal experiment was obtained from the Animal Ethics Committee of the First Affiliated Hospital of Zhengzhou University (Zhengzhou, China). e mass spectrometer detector was set as multiple reaction monitoring (MRM) in the positive ion mode. e MRM fragmentation transitions were m/z 450.1 ⟶ m/z 260.1 for crizotinib and m/z 398.2 ⟶ m/z 212.0 for IS. e dwell time for each transition was 100 ms. e instrument was operated with the ion spray voltage set at 5.0 kV and the heater gas temperature set to 500°C. e nebulizer gas (gas (1)) and heater gas (gas (2)) were both maintained at 50 psi. In addition, a curtain gas of 30 psi and a collision gas of medium were used. e mass parameters for both crizotinib and IS were optimized as follows: entrance potential (EP) was 10 V; collision cell exit potential (CXP) was 6.5 V; and collision energy (CE) was 34 V. e declustering potential (DP) for crizotinib and IS was set to 110.5 V and 118 V, respectively. Data acquisition was conducted using Analyst 1.6.2 software (Applied Biosystems, USA).

Preparation of Calibration Standards and Quality Control
Samples.
e stock solutions of crizotinib and IS were prepared by dissolving accurately weighed amounts of the reference substances in methanol to yield concentrations of 1.0 mg/mL. Working solutions of crizotinib at the desired concentration for the preparation of calibration standards and quality control (QC) samples were obtained by serially diluting stock solutions using methanol. e 400 ng/mL working solution was prepared in methanol. All solutions were stored at −20°C until use.
Calibration standards and QC samples were prepared by spiking the appropriate volumes of the working solutions with blank (drug-free) mouse tissue homogenates. e blank tissue homogenates were prepared with the liver, lung, kidney, spleen, fat, stomach, large intestine, small intestine,  International Journal of Analytical Chemistry heart, muscle, and brain, respectively. e preparation method was the same as the method adopted in Section 2.7. e final concentration levels of the standard curve samples were 20-8000 ng/mL. e QC samples were prepared at concentrations of 60, 600, and 6400 ng/mL for crizotinib.

Sample Preparation.
After adding 50 μL aliquots of mouse tissue homogenate to a 1.5 mL centrifuge tube, 5 μL of IS solution (400 ng/mL) was added, and the mixture was vortexed for 10 s. en, 200 μL methanol was added to the mixture for protein precipitation and extraction of the analytes, vortexmixed for 3 min, and centrifuged at 4°C at 14000 rpm for 10 min. e supernatant was separated and mixed with an equal volume of 60% (v/v) methanol-water solution. Finally, the mixture was transferred into an autosampler vial, and 1 μL was injected for UPLC-MS/MS analysis.

Method Validation.
Method validations were performed using the UPLC-MS/MS method according to the guidelines set by the United States Food and Drug Administration (FDA) [18] and the Chinese Pharmacopoeia Commission [19] for bioanalytical method validation. e partial method validation could be performed when changes in the matrix within species were occurred in bioanalytical method [18]. In this study, the full-method validations, including selectivity, linearity, precision and accuracy, matrix effect, recovery, carryover effect, stability, and dilution integrity, were performed on liver and small intestine samples. Partial method validations were conducted for the other tissue samples.

Tissue Distribution Study.
irty-six mice were randomly divided into six groups (six mice per group, half male and half female). After fasting overnight, the mice were administered a single dose of 500 mg/kg crizotinib through oral gavage. e crizotinib was prepared in 0.5% carboxymethyl cellulose sodium (CMC-Na) to make a suspension of 50 mg/mL. e mice were euthanized by decapitation, and different tissues were harvested at each time point. Tissues including the brain, heart, liver, lung, kidney, spleen, fat, muscle, stomach, large intestine, and small intestine were dissected, washed with saline, and blotted with filter paper. Tissue samples were homogenized in ice-cold saline (1 : 50, w/v, g/mL for the liver, lung, kidney, spleen, fat, stomach, large intestine, and small intestine; 1 : 30, w/v for the heart and muscle; and 1 : 10, w/v for the brain). ese tissue homogenates were stored at −80°C, allowed to thaw at room temperature, and vortex-mixed before processing.
In this study, the animal equivalent dose was calculated according to the guidelines set by the US FDA [20]. e equivalent dose of crizotinib in mice was 102.5 mg/kg, and it was obtained by multiplying the dose in humans (500 mg/day, assumed 60 kg human) by the body surface area conversion factor (12.3). e toxic dosage (500 mg/ kg) of crizotinib in mice was about five times that of the equivalent dose of crizotinib in mice. e toxic dosage has been investigated and verified in another study (data not shown).

Optimization of Mass Spectrometry Conditions.
To optimize the mass spectrometric conditions, the standard solutions of crizotinib and IS were injected into the mass spectrometer.
e positive-ion ESI mode was performed, considering that nitrogen is present in the chemical structures of crizotinib. In the Q1 full-scan mass spectra, the predominant ion of crizotinib was the protonated molecule ion [M + H] + at m/z 450.1.
e MRM mode was chosen because of its better specificity and sensitivity. Interestingly, the product ion mass spectrum of crizotinib showed that the main fragment ion was m/z 260.2 with low CE (approximately 34 eV, Figure 2

Optimization of Liquid Chromatographic Conditions.
e chromatographic conditions, including the mobile phase composition and chromatographic column types, were optimized to achieve good chromatographic peak shapes, resolution, and proper retention time for the analytes according to the methods reported before [21][22][23]. After many tests, methanol with 0.3% formic acid using a gradient elution was chosen as the mobile phase. e gradient program was set as follows at first: 0 to 0.2 min, 10% A; 0.2 to 0.5 min, 10 to 60% A; 0.5 to 1.9 min, 60% A; 1.9 to 2.2 min, 60 to 10% A; and 2.2 to 4.0 min, 10% A. e column used for this method was an Acquity UPLC HSS T3 column (1.8 μm, 100 mm × 2.1 mm i.d., Waters). Under that chromatographic conditions, crizotinib and IS exhibited good chromatographic behavior.
Unfortunately, the carryover effect was detected for crizotinib under the initial chromatographic conditions. Carryover is a ubiquitous problem in the LC-MS bioanalytical method, and it limits accurate quantitation. Studies have demonstrated that carryover may be caused by the autosampler or LC column [24][25][26]. In this study, the source of carryover was examined using a "duplicated" solvent gradient method based on a previous study [26]. e "duplicated" solvent gradient includes two gradients for one analysis, which means that an injection is eluted by the first gradient, and the second gradient is performed without sample injection. e results (Figure 3) showed that the peak (6.55 min) obtained by the second gradient was lower than that (2.04 min) obtained by the first gradient, which indicated that the carryover was caused by both the autosampler and column. erefore, possible solutions need to be tested to eliminate the carryover effect from the autosampler and column. We tried several needle rinse solvents (acetonitrile, methanol, and isopropanol) individually as well as with various combinations of these solvents. Finally, a methanol/ acetonitrile/isopropanol (1 : 1 : 1, v/v/v) solution containing 5% formic acid was selected as the needle rinse solvent to clean the autosampler system. Furthermore, a wash step with a high organic phase ratio (100% methanol) added to the initial gradient program was performed to clean the column and reduce carryover from the column. e final gradient program was the one mentioned in Section 2.3. Although the carryover effect was improved in some way, it still could not meet the requirements of the bioanalytical method. en, other kinds of columns, including Agela Venusil MP C 18 column and Kinetex C 18 column, were tested. Ultimately, a Kinetex C 18 column (2.6 μm, 50 mm × 2.1 mm, i.d., Phenomenex, USA) was chosen which provided the minimum carryover effect for crizotinib.

Selectivity.
e typical MRM chromatograms of crizotinib and IS in mouse liver and small intestine samples are shown in Figure 4  substances was observed at the retention times of crizotinib and the IS which were 1.31 and 1.40 min, respectively.

Linearity and Carryover.
e typical regression equations of the calibration curves were calculated by a weighted (1/x 2 ) least-squares linear regression analysis. e calibration curves, correlation coefficients (r), and linear ranges of crizotinib in tissue samples are listed in Table 2.
e calibration curves were linear over the concentration ranges of 20-8000 ng/mL (r > 0.99, n � 8) for crizotinib. The concentration ranges of crizotinib in this study were chosen according to the crizotinib concentration of the tissues in  our pilot study. For carryover, the mass response of the blank sample at the retention time of crizotinib was less than 20% of the peak area of the lower limit of quantitation (LLOQ) sample, and no carryover effect was observed for IS.

Accuracy and Precision.
e intra-and interday precision and accuracy were calculated using one-way analysis of variance for the LLOQ and QC samples. e accuracy was demonstrated as the relative error (RE, %) of the samples, and the precision was indicated as the relative standard deviation (RSD, %). e accuracy and precision data for the determination of crizotinib in mouse tissues are shown in Table 3. e results were all within the acceptable variability limits, which demonstrated that good accuracy and precision were obtained under the current method.

Matrix Effect and Extraction Recovery.
e matrix effect was evaluated by IS-normalized matrix factors (MFs) using QC samples at the low QC (LQC) and high QC (HQC) level. e IS-normalized MF was calculated by dividing the MF of the analyte by the MF of the IS [19,22]. e ISnormalized MF of six different batches of biological matrices at two concentration levels of crizotinib was 106.5 ± 2.2% and 101.6 ± 2.7% for the mouse liver homogenate and 102.5 ± 2.7% and 96.8 ± 3.4% for the small intestine homogenate, respectively. e results indicated that matrix effects for crizotinib and IS under the current method conditions were negligible, or the matrix effects of analytes could be compensated by the IS. e extraction recoveries of crizotinib at concentrations of 60 and 6400 ng/mL were 99.1 ± 3.0% and 88.6 ± 7.3% in the mouse liver homogenate and 75.9 ± 2.1% and 86.6 ± 3.6% for the mouse small intestine homogenate, respectively. e recovery of crizotinib in the mouse small intestine homogenate at concentration of 60 ng/mL was lower than that of the mouse liver homogenate. However, the extent of the recovery of crizotinib in the mouse small intestine sample was consistent and reproducible. erefore, the recovery could meet the requirement of quantitative analysis for biological samples. e recoveries of IS at the concentration of 40 ng/mL in the mouse liver homogenate and the small intestine homogenate were 94.2 ± 7.2% and 92.2 ± 3.6%, respectively.

Stability.
e stability results of crizotinib are summarized in Table 4. e results revealed that crizotinib was stable in the matrices under different storage conditions. is method can be used for routine analysis.

Dilution
Integrity. e dilution integrity experiments of crizotinib were conducted at two concentration (low and high) levels. e intraprecision was lower than 10% in terms of the relative standard deviation, and the accuracy was within ± 10% in terms of the relative error for crizotinib. e intraprecision and accuracy of the dilution test were within the acceptable criteria. e results indicated that crizotinib could be assayed reliably by 5-fold diluting with the blank tissue matrix when the crizotinib concentration in samples exceeded the linear ranges of standard curves.

Tissue Distribution Studies.
ough the FDA and European Medicines Agency (EMA) recommend that crizotinib can be taken with or without food [27], in order to minimize the impact of the gastrointestinal tract and food, the mice in this study were treated with fasted condition. After intragastric administration of 500 mg/kg crizotinib, crizotinib was widely distributed into the tested tissues. e tendency graphs of crizotinib in various mouse tissues are shown in Figures 5 and 6. e highest concentration levels of crizotinib observed in each of the tissues and the time to reach them were 50.29 ± 19.34 μg/g for the heart and 239.3 ± 142.0 μg/g for the small intestine at 2 h postdosing; 556.4 ± 408.6 μg/g for the stomach, 239.9 ± 66.3 μg/g for the liver, 153.0 ± 104.0 μg/g for the kidney, 244.4 ± 357.9 μg/g for the spleen, 321.2 ± 449.2 μg/g for the large intestine, and 173.3 ± 112.7 μg/g for the fat at 4 h postdosing; 612.4 ± 166.6 μg/g for the lung and 41.39 ± 12.50 μg/g for the muscle at 8 h postdosing; and 2.4 ± 1.1 μg/g for the brain at 15 h postdosing. A high concentration of crizotinib was found in the gastrointestinal tract, and the results showed that the main target organs of crizotinib were the lung, liver, and spleen. A small amount of crizotinib could penetrate the blood-brain barrier. e concentrations of crizotinib in different tissues after oral administration of crizotinib (500 mg/kg) in mice have vast standard deviations, which may be caused by large differences among individuals in mice. In addition, the bimodal phenomenon of crizotinib was observed in the tissue distribution tendency graphs of some mouse tissues. e bimodal phenomenon of crizotinib might be related to the enterohepatic circulation, but we could not find the reference to support this speculation. Besides, there was no bimodal phenomenon of crizotinib in plasma according to reference [17] reported before.
erefore, further studies are required to explain this phenomenon. One limitation of the present study was that

Conclusion
In this study, a reliable LC-MS/MS method was developed and validated for the quantitative determination of crizotinib in mouse tissue samples. e method was successfully applied to a tissue distribution study of crizotinib at toxic doses in mice, which has not been reported. e results suggested that crizotinib was well distributed in the tissues detected in this study, and it was quickly eliminated in tissues, indicating that there was no long-term accumulation of crizotinib in mouse tissues. Additionally, crizotinib could cross the blood-brain barrier and was distributed in the brain tissue. e lung, liver, and spleen were the main target organs of crizotinib in the mice at the dosage of this study.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare no conflicts of interest.