2.1. Determination of CK41016 Using Ultra Performance Liquid Chromatography-Tandem Mass Spectrometer (UPLC-MS/MS)
In this study, an improved sensitive and selective UPLC-MS/MS method was developed for determining the analyte in biological samples. Chromatographic condition, sample preparation method, and mass spectrometric parameters, such as capillary voltage, collision energy, desolvation temperature, ion source temperature, and flow rates of desolvation and cone gases were optimized to determine CK41016 and the internal standard (amlodipine, IS). Full-scan product mass spectra of CK41016 and IS are shown in
Figure 1. Multiple reaction monitoring (MRM) transitions for CK41016 and IS were at
m/z 342.1→255.1 and 409.1→238.1, respectively. For CK41016, the daughter ion could be detected at
m/z 255.1 and 144.1. We used
m/z 255.1 for the daughter ion due to its high intensity. Various collision energies were examined to obtain the best abundance of peak. When 20 eV of collision energy was used, the highest sensitivity of CK41016 was obtained. Therefore, 20 eV was chosen as the collision energy for CK41016.
Considering the selectivity and effects of the coeluting peak for CK41016, we tried various mobile phases, such as water, acidic, and a buffer solution to optimize the liquid chromatography method. When 0.1% formic acid in water and 0.1% formic acid in acetonitrile were used, the intensity was higher than in other conditions. In addition, tetrahydrofuran was added to mobile phase A and B to avoid interference with detection of CK41016. To achieve stable base line and better peak shape, we also tested various columns with different sizes and compounds bonded to silica, including HALO C18 column (50 mm × 2.1 mm, 2 μm particle size, Advanced Materials Technology, Wilmington, DE, USA); Kinetex C18 column (50 mm × 2.1 mm, 1.7 μm particle size, Phenomenex, Torrance, CA, USA); Capcell core C18 column (50 mm × 2.1 mm, 2.7 μm particle size, Shiseido, Osaka, Japan); Unison UK-C18 column (50 mm × 2.0 mm, 3 μm particle size, Imtakt Corp., Tokyo, Japan); and Kinetex biphenyl column (2.1 mm × 50 mm, 1.7 µm particle size, Phenomenex, Torrance, CA, USA). Kinetex biphenyl column exhibited the best sensitivity, selectivity, symmetric peak shape, and intensity. The retention time was 2.35 min for CK41016 and 1.18 min for IS.
For sample preparation, we compared the protein precipitation (PP) method using methanol and acetonitrile and the liquid-liquid extraction (LLE) method using ethyl acetate, ethyl ether, methyl-tert-butyl ether, and methylene chloride to determine the most optimized sample preparation method. PP was not selected due to lower selectivity and recovery than LLE. According to the results of each method, the extraction using methylene chloride was better than other preparation methods in sensitivity, recovery, and intensity at a low concentration. For this reason, a simple LLE method using methylene chloride was adopted for the determination of CK41016.
2.2. Method Validation Result
Method validation was conducted according to the FDA Guidance for Industry: Bioanalytical Method Validation [
26].
2.2.1. Selectivity and Specificity
The representative chromatograms of the blank plasma (A), spiked rat plasma with CK41016 at lower limit of quantification (LLOQ) of 0.1 ng/mL and the IS (20 ng/mL) in extracted blank plasma (B), standard solution of CK41016 (0.1 ng/mL) and IS (C), and rat plasma taken at 5 min after eye drop administration of 75.21 μg/kg CK41016 (D) are shown in
Figure 2. As seen in the figure, the matrix effect in plasma and tissues of rats and rabbits did not exceed 15%, and no significant chromatographic interferences were observed with CK41016 and the IS at their retention times in drug-free plasma and tissues.
2.2.2. Calibration Curves and LLOQ
The calibration curves of CK41016 showed a good linearity over the concentration range of 0.1–200 ng/mL with a correlation coefficient for all standard curves exceeding 0.992 at each batch (
n = 5). The linear regression equations of the calibration curves for CK41016 are represented in
Table 1, with
y as the analyte peak-area ratio to the IS and
x (ng/mL) as the analyte concentrations in plasma (or organ tissue). The LLOQ of CK41016 was 0.1 ng/mL. The UPLC-MS/MS analysis in this study provided a sufficient LLOQ for further PK study after the intravenous and eye drop administration of CK41016 in rats and rabbits.
2.2.3. Precision and Accuracy
The intra-batch precision and accuracy at low, middle, and high quality control (QC) samples were in the ranges 2.52–5.02% and 95.33–103.93% (12.46% and 100.33% for the LLOQ), respectively. The inter-batch precision and accuracy at low, middle, and high QC samples were in the ranges 2.94–4.45% and 96.44–100.40% (1.80% and 102.11% for the LLOQ), respectively. These results indicate that all the values were within the acceptable range of ±15% for QC samples and ±20% for LLOQ, and this method showed satisfactory precision, accuracy, and reproducibility.
2.2.4. Sensitivity
The LLOQ for this quantitative method is set as 0.1 ng/mL. The inter- and intra-batch results showed that LLOQ concentrations (n = 5) were within 100.33–102.11% of the theoretical value. Thus, it can be assured that the sensitivity for this quantitation method is appropriately set.
2.2.5. Recovery and Matrix Effect
The extraction recovery of CK41016 from rat plasma was 88.16% ± 11.36%, whereas the recovery of the IS was 92.48% ± 6.50%. The mean matrix effects for CK41016 at low and high concentrations were 0.99–1.00, which indicated negligible suppression or enhancement.
2.2.6. Stability
The QC samples (low and high concentration) of CK41016 were stable at room temperature (25 °C) for 24 h without any significant degradation, and the relative errors (REs) were within 12.19% for CK41016. Moreover, CK41016 was also considered stable after three freeze-thaw cycles in rat plasma (RE was less than 6.43%). The stock solutions of CK41016 and the IS stored at −20 °C in methanol were stable for a month, and the plasma samples stored at −80 °C were stable for three months, with a variation of <12.16%. All the results showed that CK41016 was stable under the different storage conditions.
2.2.7. Incurred Sample Reanalysis
This assay was further evaluated using four rat samples in an incurred sample reanalysis (ISR). Two-third (67%) of the repeated sample results should be within 20% for small molecules and 30% for large molecules. The results of all the reanalyzed samples were within 10% [
26].
All results were within the criteria of the FDA guidance. Thus, the developed bioanalytical method for determining CK41016 was successfully validated.
2.3. PK Study in Rats
The validated method was applied to a PK study of CK41016 in rats after IV or eye drop administration. The plasma concentration–time curves of CK41016 after IV or eye drop administration are shown in
Figure 3. The concentration–time profiles of CK41016 were adequately described by the two-compartment model without lag time using the WinNonlin
® software (version 8.1, Certara
TM, Princeton, NJ, USA) program. The estimated results of PK parameters for CK41016 of each group are summarized in
Table 2.
Diagnostic plots for the final PK model of CK41016 after eye drop administration in rats, including the observed versus model-predicted concentration (A), individual weighted residual (IWRES) versus time (B), or model-predicted concentration (C) plots, are shown in
Figure 4.
As seen from
Table 2, the mean elimination half-life (t
1/2) was approximately 206.83 min after IV administration. The AUC
0–∞, area under plasma concentration-time curve from time 0 to infinity, significantly increased in a dose disproportional manner from 12.58 μg·min/mL for 350.80 μg/kg to 76.25 μg·min/mL for 905.50 μg/kg. There was a significant (
p < 0.05) difference in clearance (CL), which decreased nonlinearly from 27.88 to 11.88 mL/min/kg with the increasing dose. This result might be due to saturation of elimination by the enzyme-dependent clearance mechanism for the drug. After eye drop administration, the T
max was 4.30–4.40 min, suggesting that CK41016 was quickly absorbed into the body after administration of the eye drop.
Due to nonlinear PKs of CK41016 in rats, AUCIV, area under plasma concentration-time curve after IV administration, data of 350.80 μg/kg was used for the calculation of bioavailability (F). The F value of the CK41016 eye drop in rats was 31.53% for 37.14 μg/kg and 28.18% for 75.21 μg/kg.
The tissue distribution of CK41016 was evaluated at 720 min after IV administration and at two timepoints (60 and 360 min) after eye drop administration in rats. The results are shown in
Figure 5.
After eye drop administration in rats, CK41016 was rapidly distributed in tissues through systemic circulation. The tissue-to-plasma partition coefficient (Kp) value of CK41016 in almost all rat tissues has increased over timepoints; however, it was markedly higher in the eyeballs than that in any other tissues. This is probably due to the fact that the instilled dose of CK41016 was in direct contact with the eyes. CK41016 was mainly distributed in the vitreous humor of the eyeballs. Significant differences in Kp values at two timepoints were observed in the eyeball tissues. Since AMD is a disease affecting the posterior segment of the eye, the treatment target areas of this disease are retina and choroid. Therefore, the drug for the treatment of AMD should be delivered to the back of the eye. The Kp values in vitreous humor, retina, and choroid were over 1, suggesting that CK41016 might have been delivered and well-distributed in the target areas up to 360 min after eye drop administration.
2.4. PK Study in Rabbits
To investigate species differences, a PK study of CK41016 was evaluated in rabbits after IV or eye drop administration. Plasma concentration–time curves of CK41016 after IV or eye drop administration are shown in
Figure 6. Similar to the results in rats, concentration–time profiles of CK41016 were adequately described by the two-compartment model without lag time using the WinNonlin
® software program. The estimated results of the PK parameters for CK41016 of each group are summarized in
Table 3.
Diagnostic plots for the final PK model of CK41016 after eye drop administration in rabbits, including the observed versus model-predicted concentration (A), individual weighted residual (IWRES) versus time (B) or model-predicted concentration (C) plots, are shown in
Figure 7.
As shown in
Table 3, the mean elimination half-life was approximately 104.38 or 88.45 min, and T
max was 15.39 min after eye drop administration. The absolute eye drop F of CK41016 in rabbits was 58.72%.
The tissue distribution of CK41016 was evaluated at 240 min after IV or eye drop administration in rabbits. Results are shown in
Figure 8.
After eye drop administration in rabbits, CK41016 was distributed in major organs through systemic circulation, as well as the eyeball. Kp values of CK41016 in cornea, vitreous humor, retina, choroid, brain, lung, kidney, and small intestine were more than 1, indicating that CK41016 was well-distributed in these tissues at 240 min after eye drop administration. This suggests that CK41016 might have reached the posterior segment of the eyeball at 240 min after eye drop administration.
According to previous reports on the routes of drug delivery after eye drop administration [
21,
27], the ocular distribution of a drug is carried out via three major pathways after topical instillation: tear turnover, anterior (cornea/conjunctiva) region, and nasolacrimal drainage. Furthermore, there are four ways for a drug to reach the posterior of the eye. The drug will have to either (1) go through the cornea and aqueous chamber, penetrate the lens/iris to reach the vitreous before finally getting to the retina, (2) diffuse through the conjunctiva, sclera, and choroid to arrive at the retina, (3) go through horizontal diffusion from the cornea to the conjunctiva, or (4) go through the nasolacrimal drainage. The tissue distribution of CK41016 after eye drops in rats and rabbits shows that CK41016 was distributed to the cornea, vitreous humor, retina, and choroid but not to the aqueous humor. Based on the result, it can be hypothesized that the ocular penetration of CK41016 was carried out via a route similar to the pathway (2) from the report mentioned previously [
27].
The F and tissue distribution in rabbits after eye drop administrations were different from those of rats. Although the highest distributed tissue was eyes in both rabbits and rats, the distribution trend within the eyeballs of rabbits was different from that of rats. After the eye drop administration in rats or rabbits, results showed that CK41016 reached the eyeballs in both species. However, less amounts of the drug were delivered to the retina of rabbits than in rats. This discrepancy might be due to the species differences. The corneal diameter and the anterior tissue volume of the rabbits are 2- and 18-fold larger than that of the rats, respectively [
28]. This could explain why more CK41016 was distributed in the cornea of the anterior segment but less in the posterior tissue in rabbits than in rats.