Development and validation of an LC-MS/MS assay for the quantification of cintirorgon (LYC-55716) in mouse plasma and tissue homogenates

Cintirorgon (LYC-55716) is a promising first-in-class antitumor agent as a ROR γ agonist in the treatment against various types of cancer. To support preclinical mouse studies


Introduction
Cintirorgon (LYC-55716) is a promising first-in-class immunotherapeutic agent in cancer treatment as mono-and combination therapy.It has successfully undergone phase 1/2 A studies as monotherapy in adults with locally advanced or metastasized cancer, such as non-small cell lung cancer, renal cell carcinoma, ovarian carcinoma, and urothelial carcinoma [1].Currently, a phase 1b combination therapy of cintirorgon and pembrolizumab trial is ongoing to assess the safety and tolerability in adults with metastatic non-small cell lung cancer [2].
Cintirorgon has agonistic properties to the retinoic acid receptor-related orphan receptor γ (RORγ), a transcription factor encoded by the RORC genes [3].This receptor modulates Type 17 effector T cells that contribute to antitumor responses by concurrently increasing immune activation and decreasing immune suppression.Increased immune activation by RORγ is established by enhancing multiple processes that play a role in tumor resistance, such as the number of effector T cells in tumors, the cytolytic activity, and the co-stimulatory molecules (CD137, CD27, CD226).The production of pro-inflammatory cytokines, chemokines and prolonged survival of anti-tumor lymphocytes are also increased by RORγ activation.On the other hand, immune suppression is decreased by reduction of regulatory T cells, CD39 and CD73 enzymes that generate extracellular adenosine, PD-1, and other co-inhibitory molecules e.g.TIM3, LAG3, or TIGIT [3][4][5][6].

]] ]] ]]]]]]
To the best of our knowledge, there are no bioanalytical methods for the analysis of cintirorgon reported so far.Given the importance of mouse knockout and tumor models in the development of anticancer drugs like cintirorgon, we developed our assay for various mouse matrices.To support preclinical pharmacokinetic (PK) studies in mice, an LC-MS/MS method was developed for its quantification in mouse plasma and eight different homogenates (brain, kidney, liver, lung, spleen, small intestine, small intestine contents, and testis).Since preclinical PK studies often generate many samples, a simple and fast method was developed using a 96-well format with protein precipitation as sample pretreatment.Full method validation was performed in mouse plasma and partial validation for all eight other matrices in compliance with the FDA and EMA international guidelines [7,8].The method was successfully applied in a preclinical study with cintirorgon in mice, demonstrating its applicability.

Analytical instruments
The LC-MS/MS equipment consisted of a DGU-14A degasser, a CTO-10Avp column oven, a Sil-HTc autosampler, two LC10-ADvp-μ pumps (all from Shimadzu, Kyoto, Japan), and a Finnigan TSQ Quantum Discovery Max triple quadrupole mass spectrometer with electrospray ionization (Thermo Electron, Waltham, MA, USA).The MS was controlled using the Finnigan Xcalibur software (version 1.4, Thermo Electron).The same software was used to record data.

LC-MS/MS conditions
Partial-loop injections (10 µL) were done on a Polaris 3 C18-A column (50 × 2 mm, d p = 3 µm, average pore diameter = 10 nm, Varian, Middelburg, The Netherlands) with a corresponding precolumn (10 × 2 mm).The column temperature was maintained at 40 °C and the sample rack compartment at 4 °C.A gradient program with a flow rate of 0.5 mL/min was used with 0.2% formic acid and 0.2% ammonium hydroxide in water as eluent A and methanol as eluent B. After injection, the percentage of methanol was increased linearly from 70% to 90% (v/v) during 1.5 min.Next, the column was flushed with 100% (v/v) methanol for 0.3 min and finally, the column was reconditioned at the starting conditions (70% (v/v) B) for 1.2 min resulting in a total run time of 3 min.The whole eluate was then transferred into the electrospray probe between 0.7 and 2.2 min after injection by switching the MS divert valve.Positive electrospray settings of the assay were a 5,000 V spray voltage and a 320 °C capillary temperature.The nitrogen sheath, ion sweep, and auxiliary gasses were set at 60, 0 and 25 arbitrary units, respectively.The skimmer voltage was set off.The SRM mode was used with argon and the collision gas was set at 1.7 mTorr.Compound-dependent parameters are reported in Table 1.Mass resolutions were set at 0.7 full width at half height (unit resolution) for both separating quadrupoles.

Stock and working solutions
The IS working solution of 200 ng/mL was prepared by dilution of the reference standard in methanol to a concentration of 1 mg/mL and it was then further diluted in acetonitrile to obtain the intended concentration of 200 ng/mL.This IS working solution also functioned as precipitation solvent.Stock solutions were prepared by diluting the cintirorgon reference standard in methanol to a concentration of 0.2 mg/mL.For the preparation of the calibration standards, the stock solution was further diluted to a concentration of 4,000 ng/mL in mouse lithium heparin plasma, which was used as the highest calibrator.The remaining calibrators were prepared by subsequent dilution of the highest calibrator in mouse lithium heparin plasma to obtain a total of eight calibrators with concentrations of 2, 4, 20, 40, 200, 400, 2,000, and 4,000 ng/mL.Least-squares weighted linear regression using the ratio of the peak areas of cintirorgon and the internal standard (IS), monensin, was employed to define the calibration with a weighing factor of 1/x 2 , where x is the analyte concentration.Quality control (QC) samples in mouse lithium heparin plasma were prepared from a separate stock solution at 3,000 (QC-H), 100 (QC-M), 10 (QC-L), and 5 (QC-LLOQ) ng/mL cintirorgon.

Sample preparation
Ten µL of plasma samples were pipetted into a 200-µL polypropylene 96-well plate with conical bottom and 20 µL of 200 ng/mL IS in acetonitrile was added.The 96-well plates were closed with a silicone mat and were vortex mixed for 10 s at maximum capacity.Next, the plate was centrifuged for 5 min at 2643×g and 20 µL of the supernatant was transferred to a 1-mL 96-deep well plate with round bottom.After adding 200 µL of 50% (v/v) methanol in water, samples were mixed by gentle manual shaking.The prepared plate was placed in the autosampler rack without the silicone mat for injection.Finally, 10 µL of the final sample was injected onto the analytical column for analysis.

Bioanalytical method validation
Full validation was performed for mouse plasma and partial validation for eight homogenates using the latest United States Food and Drug Administration (FDA) and European Medicines Agency (EMA) guidelines for bioanalytical method validation [7,8].

Calibration
Calibration standards were prepared in mouse lithium heparin plasma.A double blank (no IS), a blank (IS only), and eight non-zero calibration standards were analyzed in three independent runs in duplicate (n = 6).A linear regression of 1/x 2 was used to express the area ratio between cintirorgon/IS against the corresponding cintirorgon concentration (x).At least 75% of the non-zero calibration standards should be within ± 15% deviation of the nominal concentration to accept the analytical run.

Accuracy and precision
The assay performance was determined in three independent runs in 6-fold (n = 18) at QC-LLOQ, QC-L, QC-M, and QC-H level in mouse lithium heparin plasma.For the eight homogenates, the performance was determined at QC-M level only (n = 18).The accuracy should be between 85%-115% (80-120% for the LLOQ) accuracy to meet the criteria.The overall inter-and intraday precision was considered acceptable if variations were within ± 15% standard deviation ( ± 20% for the LLOQ) [7,8].

Selectivity
Singular analysis of six individual blank mouse lithium heparin plasma and four individual mouse homogenates was used to assess selectivity and specificity at LLOQ level.The response from each spiked sample and its corresponding double blank were compared.The interference in the double blank samples at the retention time of the analyte and IS should be < 20% and < 5%, respectively [7,8].

Recovery and matrix effect
The matrix effect (n = 4, at QC-L, QC-M, and QC-H level), relative matrix effect (n = 6, at QC-L and QC-H level), and extraction recoveries (n = 4, at QC-L, QC-M, and QC-H level) were determined for cintirorgon in mouse lithium heparin plasma.For the homogenates, the relative matrix effect (n = 4) was determined at QC-L and QC-H levels.To determine the matrix effect, two sets of samples were prepared, a matrix present and a matrix absent set of samples.Matrix present samples were prepared by spiking 20 µL blank extracts with 10 µL working solution in 50% methanol at QC-L, QC-M, or QC-H level.Matrix absent samples were prepared by spiking 10 µL working solution in 50% methanol at QC-L, QC-M, or QC-H level to 10 µL acetonitrile and then diluting it with 200 µL 50% methanol.The absolute area response of the matrix present samples should be between 85%-115% of the matrix absent samples [7,8].The relative matrix effect was determined in individual mouse plasma and individual homogenates by dividing the area ratio of the matrix present samples with the area ratio of the matrix absent samples.The area ratio of the matrix present samples should be within 85-115% of the matrix absent samples.The extraction recoveries were calculated by dividing the area ratio of the extracted (and diluted) spiked plasma samples by the area ratio of the matrix present samples.

Incurred samples reanalysis
Seven mouse homogenates (there were no study samples of lungs) from six different mice and the last time point of the plasma samples from of the same mice were reanalyzed after 1 week to verify the reliability of the reported cintirorgon concentrations in the study samples.The difference between the reanalyzed and original samples concentrations was divided by the mean concentration of the two results.Two thirds of the reanalyzed results should be within ± 20% of the mean to meet the criteria.

Pharmacokinetic study in mice
For the preclinical study, 50 mg/mL cintirorgon was prepared in dimethyl sulfoxide (DMSO), which was diluted in polysorbate 80/ ethanol (1:1, v/v) and then further diluted in 5% glucose water (w/v) to obtain a final concentration of 4 mg/mL cintirorgon.Thus, the concentrations of DMSO, polysorbate 80, ethanol, and glucose in the dosing solution in water were 8%, 4%, 4%, and 4.2% (v/v/v/w), respectively.The housing and handling of the 6 wild-type (FVB/NRj) male mice were according to the institutional guidelines complying with the Dutch and EU legislation.After oral administration of cintirorgon, blood was sampled via the tail vein at t = 0.05, 0.125, 0.25, 0.5, and 1 h, using heparin-containing tubes (Sarstedt, Nümbrecht, Germany).At t = 2 h, the mice were anesthetized with isoflurane and blood was again collected in heparin-containing tubes, but this time by cardiac puncture.The mice were then sacrificed by cervical dislocation to collect the brain, kidney, liver, spleen, small intestine, and testis.The small intestine was separated from the small intestine content, which was also collected, and washed with cold saline to remove any remaining feces.Plasma was isolated from the whole blood samples by centrifugation at 9,000 ×g for 6 min at 4 °C.The brain, kidney, liver, spleen, small intestine, small intestine content, and testis were weighed and 1, 2, 3, 1, 3, 2, and 1 mL of 2% (w/v) bovine serum albumin was added, respectively.As a final step, the samples were homogenized with the Fast Prep-24™ 5 G (MP Biomedicals, Santa Ana, CA, USA) homogenizing device.All samples were stored at −30 °C until analysis.PK parameters were calculated using the PKSolver add-in for excel [9].

Method development
Preclinical PK studies in mice usually generate many samples of which a majority has low sample volume.For this reason, a simple and fast 96-well format was used for the protein precipitation step which also enables working with low sample volumes.Because of the lack of a commercially available stable isotopically-labeled IS, several compounds (e.g.monensin, salinomycin, and diclofenac) were tested as IS alternatives [10,11].The retention time of monensin was most similar to that of cintirorgon, with a difference of only 0.2 min, and was thus selected for further use as IS (Fig. 1).
ESI-mass spectrometric parameters for cintirorgon were optimized by direct infusion.Abundant Q1 peaks for cintirorgon after positive ionization were the ammonium and sodium adducts (Fig. 2A) in ammonium hydroxide solution and only the sodium adduct in formic acid solution, after negative ionization the protonated molecule was found.Collision-induced dissociation (CID) revealed the most abundant product ions for the ammonium adduct Fig. 2B).The two most abundant product ions of this precursor were selected for quantification.For monensin, ESI-MS/MS optimization was described previously by our group resulting in the use of the sodium adduct and its CID product for quantification [11].
For chromatographic optimization, alkaline versus acidic conditions were compared with the latter showing better peak shapes.This might be caused by free silanol groups that become protonated in an acidic environment.However, since better MS responses were observed when ammonia was present to facilitate detection of the ammonium adduct, ammonium hydroxide was added to the acidic mobile phase with formic acid.Next, acetonitrile was compared to methanol as organic modifier.Better responses were observed with methanol.This was probably because of the proton-donating properties of methanol at high concentrations and cintirorgon eluting at a concentration of 70% methanol.
At first, injection volumes of 30 µL were used but the accuracy and precision at the initial LLOQ level of 2 ng/mL did not meet the criteria for the homogenates for some tissues (Supplemental Table 1).In contrast, the performance in mouse lithium heparin plasma with injection volumes of 30 µL was acceptable.The poor performance in the homogenates was most likely because of matrix effects.It was therefore decided to lower the injection volume from 30 to 10 µL, exposing less matrix onto the system which resolved the issue.However, because of the decreased sensitivity, it was necessary to increase the LLOQ from 2 to 5 ng/mL, as 2 ng/mL failed to meet the criteria (Table 2).For the sake of simplicity of the method, the LLOQ was increased for all matrices as no results below LLOQ were found for cintirorgon concentrations in the study samples.However, to meet the criteria for the selectivity experiments, it was necessary to increase the LLOQ of the brain homogenates further from 5 to 10 ng/mL.This decreased the deviation from the nominal concentration from −22.6% to −11.6%.

Calibration
Calibration curves were linear within the validated concentration range of 2-4,000 ng/mL where a linear regression of 1/x 2 was used to express the area ratio cintirorgon/IS against the corresponding cintirorgon concentration (x).At least 75% of the non-zero calibration standards were within ± 15% standard deviation.The regression parameters were y = −0.0004( ± 0.0048) +0.0028 ( ± 0.0003)x, with y being the area ratio cintirorgon/IS and x being the cintirorgon concentration in ng/mL.The regression coefficient was r = 0.9942 ( ± 0.0013).

Accuracy and precision
Results of the accuracy and precision experiments are reported in Table 2.All four concentration levels tested in mouse lithium heparin plasma and the single concentration level in all eight homogenates were within ± 15% ( ± 20% at LLOQ level) deviation, as required by the guidelines [7,8].

Selectivity
The selectivity of this method was considered acceptable as the responses in the double blank samples at the retention time of cintirorgon and the IS in all matrices were ≤4.67% and ≤ 0.49% of the corresponding LLOQ (5 ng/mL) responses, respectively.The average measured concentration in the plasma and homogenates that were spiked at LLOQ level were all within ± 20% deviation, except for the brain (Table 3).To meet the criteria, the LLOQ for the brain homogenates was increased from 5 to 10 ng/mL, which improved the deviation from −22.6% to −11.6%.The highest deviation was found in the liver homogenates and was −15.8%.

Recovery and matrix effect
Matrix effects in mouse lithium heparin plasma at QC-L, QC-M, and QC-H level were between 100.2%−109.0%.The extraction recoveries at the same levels were between 90.6%−112.4% (Table 4).The relative matrix effect of the 9 different tested matrices at QC-L and QC-H level were between 91.6%−107.5% (Table 5).Based on these results, (relative) matrix effects and recovery losses were considered negligible.

Stability
Stability results of cintirorgon in all tested conditions are reported in Table 6.In plasma, QC-L and QC-H levels were tested.At least 24 h at room temperature and 9-month storage at 30 °C showed cintirorgon to be stable.At least three freeze (−30 °C)/thaw (room temperature) cycles were acceptable, and the recovered responses of the final extracts after 9 days at 4 °C were within 85-115%.Lastly, the recovered responses of the homogenates at QC-M level at room temperature after 5 h were also within 85-115%.

Incurred samples reanalysis
The acceptance criteria for the singular reanalysis of the plasma samples (n = 6) and homogenates (n = 42) were met.Only 1 out of 6 reanalyzed plasma samples exceeded the ± 20% requirement.For the homogenates, none of the reanalyzed samples exceeded the requirements except for the small intestine and small intestine contents.Interestingly, these homogenates had higher cintirorgon concentrations than initially measured.During analysis of the study samples, there was an extra observed peak with a lower retention time than cintirorgon.It was hypothesized that this extra peak was a metabolite and that it was enzymatically hydrolyzed to form   cintirorgon.An experiment where the small intestine and its contents were stored separately for 4 h at three different conditions (at −30 °C, room temperature, and 37 °C) showed results in line with this hypothesis.As shown in Figure 3, slightly higher cintirorgon concentrations were measured for the samples that were stored at room temperature compared to the ones that were stored at −30 °C.At 37 °C, cintirorgon concentrations were significantly increased in most samples.Since glucuronidation is an important pathway in drug metabolism, it was tested if the extra peak originated from cintirorgon-glucuronide that undergoes in-source fragmentation and would therefore be observed in the cintirorgon mass transition.
A product-ion spectrum was recorded at the retention time of the extra peak between 1.1 and 1.14 min (Fig. 4 and 5).The mass of cintirorgon-glucuronide ammonium adduct (m/z 797.3), cintirorgon ammonium adduct (m/z 621.2), and the product ion (m/z 395.2) were all present in the spectrum which supports the hypothesis that the extra peak is indeed cintirorgon-glucuronide. (Fig. 5).

Preclinical application
The applicability of the method was successfully shown in a PK study with mice.A representative plasma PK curve up to 2 h after 40 mg/kg cintirorgon oral administration is shown in Figure 6. Figure 7 shows the calculated cintirorgon concentrations in the tissues at t = 2 h.The T max was reached at 0.86 ± 0.56 h, the C max was 4109 ± 2033 ng/mL, and the AUC 0-2 h was 5,355 ± 1,906 ng•mL −1 •h −1 .The elimination phase parameters need further attention and are therefore not mentioned here.In this study, the peak concentration of cintirorgon (4,109 ± 2,033 ng/mL) in mice with oral 20 mg/kg cintirorgon were of the same order as seen in patients treated

Fig. 1 .
Fig. 1.Representative LC-MS/MS chromatograms in mouse lithium heparin plasma of a double blank (red) and cintirorgon (black) at LLOQ level (A) and a double blank (red) and IS (black) at the concentration used for quantification (B, 200 ng/mL).

Fig. 3 .
Fig. 3. Cintirorgon concentrations (ng/mL) in small intestine (A) and small intestine content (B) of six individual mice after 4-h storage at three different conditions.

Fig. 4 .
Fig. 4. Chromatogram of the small intestine content from a wild-type mouse where an extra peak was observed.

Table 1
LC-MS/MS parameters of individual compounds.

Table 2
Assay performance data for cintirorgon in mouse lithium heparin plasma and mouse homogenates (n = 18 on 3 different days).

Table 3
Selectivity data in mouse plasma and homogenates.

Table 4
Matrix effect and recovery data in mouse lithium heparin plasma.