Determination of pharmacokinetics and tissue distribution of a novel lutetium-labeled PSMA-targeted ligand, 177Lu-DOTA-PSMA-GUL, in rats by using LC–MS/MS

Prostate specific membrane antigen (PSMA) is known to be overexpressed in prostate cancer cells, providing as a diagnostic and therapeutic target for prostate cancer. A lutetium-labeled PSMA targeted ligand, 177Lu-DOTA-PSMA-GUL is a novel radiopharmaceutical, which has been developed for the treatment of prostate cancer. While the GUL domain of 177Lu-DOTA-PSMA-GUL binds to the antigen, the beta-emitting radioisotope, 177Lu-labeled DOTA, interacts with prostate cancer cells. However, the in vivo pharmacokinetics of intact 177Lu-DOTA-PSMA-GUL has never been characterized. This study aimed to evaluate the pharmacokinetics and tissue distribution of the radiopharmaceutical in rats by using its stable isotope-labeled analog, 175Lu-DOTA-PSMA-GUL. A sensitive liquid chromatography-tandem mass spectrometry (LC–MS/MS) analysis of 175Lu-DOTA-PSMA-GUL was developed and validated. Following intravenous injection, the plasma concentration–time profiles of 175Lu-DOTA-PSMA-GUL showed a multi-exponential decline with the average elimination half-life of 0.30 to 0.33 h. Systemic exposure increased with the dose and renal excretion is the major elimination route. Tissue distribution of 175Lu-DOTA-PSMA-GUL was most substantial in kidneys, followed by the prostate. The developed LC–MS/MS assay and the in vivo pharmacokinetic data of 175Lu-DOTA-PSMA-GUL would provide helpful information for further clinical studies to be developed as a novel therapeutic agent for prostate cancer.

The electrospray ionization (ESI) source was operated in positive mode, and the mass spectrometer was operated in the multiple reaction monitoring (MRM) mode. The observed MRM transitions and mass spectrometry settings are summarized in Supplementary Table 1. Preparation of stock solutions, calibration standards, and quality control samples. Stock solutions. The stock solutions of 175 Lu-DOTA-PSMA-GUL were prepared by diluting 2.1 mg/mL synthesized solution in methanol to 400 μg/mL. The stock solutions of NOTA-PSMA-GUL (internal standard 1, IS1) and esomeprazole (internal standard 2, IS2) were prepared by separately dissolving 10 mg of each in 10 mL of methanol (1 mg/mL).
Calibration standards and quality control samples. For drug analysis in the plasma, calibration curves were constructed by spiking 50 μL of working stock solutions to blank plasma (50 μL each) to provide 175 Lu-DOTA-PSMA-GUL concentrations at 20,000, 10,000, 5000, 1000, 500, 100, 50, and 20 ng/mL. The plasma was spiked with 50 μL of IS1 solution and 150 μL of methanol and mixed on a vortex mixer. The mixture was then centrifuged for 10 min at 4,000 rpm (3220 × g), and 100 μL of the supernatant was transferred to a plastic vial. After 100 μL of distilled water was added to the supernatant, the mixture was vortex-mixed for 10 min and 5 μL of the mixture was injected onto the LC-MS/MS. Quality control (QC) samples were prepared by spiking the working drug solutions to blank rat plasma to provide high concentration QC (16,000 ng/mL), middle concentration QC (8,000 ng/mL), low concentration QC (80 ng/mL) and lower limit of quantification (LLOQ) QC (20 ng/mL).
Similarly, calibration standards and QC samples were prepared for drug analysis in urine, feces, and twelve different tissues. Calibration ranges were 100-20,000 ng/mL for urine, 100-5000 ng/mL for feces and tissues. High, middle, and low QC sample concentrations were 16,000, 8000, and 400 ng/mL for urine, 4000, 1600, and 400 ng/mL for feces and tissue matrices.
Sample preparation. For plasma samples, NOTA-PSMA-GUL (internal standard 1, IS1) solution 50 μL was added to 50 μL of the rat plasma. As a precipitation solvent, 200 μL of methanol was added, and the mixture was mixed on a vortex mixer for 10 min, followed by centrifugation for 10 min at 4000 rpm (3220 × g). After taking 100 μL of the supernatant, 100 μL of distilled water was added, vortex-mixed for 10 min. Finally, 5 μL of the prepared mixture was injected onto the LC-MS/MS. Since several plasma samples showed concentrations above the ULOQ, those samples were diluted 10-or 20-fold for analysis.
For urine and fecal homogenate samples, the working IS2 solution 50 μL was added to 50 μL of the homogenate samples. The samples were precipitated with methanol (900 μL) on a vortex mixer for 10 min, followed by centrifugation for 10 min at 4000 rpm (3220 × g). After taking 100 μL of the supernatant, 100 μL of distilled water was added, vortex-mixed for 10 min.
For tissue homogenate samples, the working IS2 solution 50 μL was added to 50 μL of tissue homogenate samples. The samples were precipitated with methanol (400 μL) on a vortex mixer for 10 min, followed by www.nature.com/scientificreports/ centrifugation for 10 min at 4,000 rpm (3220 × g). After taking 200 μL of the supernatant, 200 μL of distilled water was added to the supernatant and centrifuged for 10 min again. After the second centrifugation, 100 μL of supernatant and the same volume of distilled water was mixed for 10 min. Finally, 5 μL of the prepared mixture was injected onto the LC-MS/MS.
In vivo pharmacokinetic studies in rats. Animals. Male Sprague-Dawley rats (7 weeks, 190-210 g; DBL co., Eumsung, Korea) were kept in plastic cages with free access to a standard diet (Youngbio, Seong-nam, Korea) and water. All experiments were performed in accordance with the relevant guidelines and regulations. The animal study protocol was approved by the Institutional Animal Care and Use Committee of Sungkyunkwan University (SKKUIACUC2018-07-25-1). Studies involving animals are reported in accordance with ARRIVE guidelines (https:// arriv eguid elines. org).
Pharmacokinetics of 175 175 Lu-DOTA-PSMA-GUL was administered by i.v. injection as a loading dose (LD) and i.v. infusion for 3 h to achieve the target C ss . The i.v. bolus LD and i.v. infusion rates (K 0 ) were calculated by LD = C ss,target ·V ss and K 0 = C ss,target ·CL, respectively 24 . The volume of distribution (V ss , 0.20 L/kg) and clearance (CL, 607.55 mL/h/kg) of 175 Lu-DOTA-PSMA-GUL were obtained from the i.v. injection study. The calculated LD was 0.60 mg/kg and 1.20 mg/kg for the target C ss of 3,000 ng/mL and 6,000 ng/mL, respectively. The calculated K 0 was 1.82 mg/h/kg and 3.65 mg/h/ kg for the target C ss of 3,000 ng/mL and 6,000 ng/mL, respectively. Blood samples were collected at 1.5, 2, 2.5, and 3 h during i.v. infusion, and centrifuged at 3,200 × g for 10 min. At the end of the infusion, rats were sacrificed, and brain, lung, heart, spleen, small intestine, stomach, kidney, liver, prostate, fat, muscle, and testis were excised and immediately homogenized in normal saline. All samples were stored at -70 °C until analysis.
Non-compartmental analysis. The plasma concentration-time data were analyzed by non-compartmental method using Phoenix ® WinNonlin ® (Pharsight, NC, USA). The fraction of 175 Lu-DOTA-PSMA-GUL excreted into urine (F urine ) and feces (F feces ) were calculated by the ratio of the total amount of drug excreted in the urine and feces to the fraction of the dose, respectively. The tissue-to-plasma partition coefficients (K P ) were calculated as the tissue-to-plasma concentration ratios.
Dose proportionality. Dose proportionality was tested for C max , AUC all , and AUC inf based on power model. Assuming the natural logarithm of the pharmacokinetic parameter is linearly related to the natural logarithm of dose as in the following equation: ln(PK parameter) = β 0 + β 1 × ln(dose), the slope coefficient (β 1 ) and its twosided 95% confidence intervals (CI) were estimated. Statistical analysis. The data were statistically tested by the unpaired t-test to compare between two means and by one-way analysis of variance (ANOVA) followed by scheffe or games-howell post hoc test. The statistical significance level was set at p < 0.05. All the statistical analyses were performed by using IBM® SPSS® Statistics 26 (IBM, Armonk, NY, USA). The sample preparation and chromatographic conditions were optimized to increase the sensitivity of the analyte. The recovery was significantly enhanced when methanol was used as a precipitation solvent compared to acetonitrile. Thus, a single step protein precipitation with methanol was developed for the extraction of both analyte and internal standard (IS) from the biological samples. NOTA-PSMA-GUL was initially used as IS for the development of analytical method in the plasma. However, when the method was applied to other biological matrices, the recovery of NOTA-PSMA-GUL was poor (< 60%) and variable depending on the matrices. Thus, we have tried other compounds, including rebamipide, ketoprofen, diclofenac, aceclofenac, and esomeprazole, as www.nature.com/scientificreports/ an internal standard. Since esomeprazole showed higher recovery and acceptable reproducibility, it was selected as an internal standard for urine, feces, and tissue samples. For chromatographic separation, Agilent Zorbax SB-Aq column (100 × 2.1 mm, i.d., 3.5 μm) resulted in the best chromatographic separation with improved sensitivity after various reverse phase columns were examined. The presence of formic acid in the mobile phase significantly increased the peak response. The use of methanol as an organic solvent was also beneficial compared to acetonitrile. Finally, the gradient profile was optimized to provide the best sensitivity of 175 Lu-DOTA-PSMA-GUL with the minimal interference.

Development of LC-MS/MS assay for
Validation of LC-MS/MS assay. The LC-MS/MS assay was validated according to the FDA guidance 25 .
Representative MRM chromatograms of 175 Lu-DOTA-PSMA-GUL and IS are shown in Fig. 2. Examination of blank sample, zero sample and other calibrators showed no interfering peaks at the retention times corresponding to the analyte and internal standards in all matrices, indicating the specificity of the assay. The calibration curves were linear over 175 Lu-DOTA-PSMA-GUL concentration ranges of 20-20,000 ng/mL for plasma, 100-20,000 ng/mL for urine, and 100-5,000 ng/mL for feces and all tissues with R 2 > 0.998. The lower limit of quantification (LLOQ) was 20 ng/mL in the plasma and 100 ng/mL in other biological matrices, including urine, feces, and tissue samples. The signal-to-noise (S/N) ratios at LLOQ concentration were over 11.7 for all tested biological matrices. The intra-and inter-day accuracy and precision determined by assaying LLOQ, low, medium, and high QC samples in the plasma, urine, feces, and twelve tissues are shown in Supplementary Table 2. The intra-and interday accuracy ranged from 90.96 to 106.82% in all the rat biological matrices. The intra-and inter-day precisions were under 10.22% and 9.22%, respectively. The intra-day and inter-day accuracy and precision were all within the ranges recommended by the FDA 25 . www.nature.com/scientificreports/ Analytical validation data including stability and process efficiency are shown in Supplementary Tables 3 and  4. Compared with the theoretical concentrations, 175 Lu-DOTA-PSMA-GUL concentrations in each biological matrix did not deviate significantly after storage under four different conditions including short-term (room temperature for 4 h), long-term (-70 °C for two weeks), freeze-thaw (4 cycles), and autosampler (4 °C for 24 h) (Supplementary Table 3). These data indicates that 175 Lu-DOTA-PSMA-GUL was stable in each biological matrix under the sample collection and assay conditions. The process efficiency in each biological matrix is generally consistent (Supplementary Table 4). Although the process efficiency of 175 Lu-DOTA-PSMA-GUL was found to be lower in several tissues such as lung, small intestine, kidney, and liver, they were consistent and reproducible as indicated by the small CV%. FDA guidance also indicated that "Recovery need not be 100 percent, but the extent of the recovery of an analyte and of the ISs should be consistent and reproducible. " 25 For plasma samples, the dilution integrity was evaluated for a 10-and 20-fold dilution of QC samples at 40,000 ng/mL, which was two times higher than ULOQ, with blank matrix. The accuracy for dilution integrity was 105.56% and 102.21%, while precision was 1.03% and 2.39% for tenfold and 20-fold dilution, respectively (n = 3, each).

Pharmacokinetics of 175 Lu-DOTA-PSMA-GUL in rats.
A drug solution of 175 Lu-DOTA-PSMA-GUL was prepared immediately before drug administration and tested by HPLC-UV. The purity of 175 Lu-DOTA-PSMA-GUL was > 95%. The peak response of 175 Lu-DOTA-PSMA-GUL was consistent (97.1 ± 8.0%) compared to the reference, indicating that 175 Lu-DOTA-PSMA-GUL was uniformly prepared. Figure 3 shows the average plasma concentration of 175 Lu-DOTA-PSMA-GUL vs. time profiles after a single i.v. bolus injection of 175 Lu-DOTA-PSMA-GUL (1, 2, and 5 mg/kg). The non-compartmental pharmacokinetic parameters of 175 Lu-DOTA-PSMA-GUL are presented in Table 1. As shown in Fig. 3, the plasma concentration of 175 Lu-DOTA-PSMA-GUL showed multi-exponential decline following i.v. injection in rats. The average  www.nature.com/scientificreports/ PSMA-GUL was assessed at steady state following i.v. infusion at two dose levels that aimed to obtain C ss,target of 3,000 ng/mL and 6,000 ng/mL. An i.v. bolus loading dose was also administered to rapidly achieve the target steady state concentration. Average plasma 175 Lu-DOTA-PSMA-GUL concentrations determined following simultaneous i.v. bolus injection (loading bolus dose 0.60 and 1.20 mg/kg) and continuous i.v. infusion (infusion rate 1.82 and 3.65 mg/h/ kg) are shown in Fig. 4. Plasma concentrations of 175 Lu-DOTA-PSMA-GUL appeared stable at approximately 3,000 ng/mL and 6,000 ng/mL from 1.5 h following i.v. infusion. These data indicate that the steady-state has been reached, and the intended steady-state plasma concentrations have been achieved.
Tissue distribution. After confirming that the steady state has been achieved, tissue samples were collected to assess the tissue distribution of 175 Lu-DOTA-PSMA-GUL. The average tissue concentrations of 175 Lu-DOTA-PSMA-GUL and the respective K P in twelve different tissues at steady state are shown in Fig. 5 and Table 2. The tissue concentrations are presented in ng/g tissue, which was calculated as the measured concentration in the tissue homogenate by LC-MS/MS (ng/mL) divided by the density of tissue homogenate (g tissue/mL). Tissue concentration of 175 Lu-DOTA-PSMA-GUL was the highest in the kidneys, followed by the prostate, lung, liver, small intestine, heart, spleen, testis, muscle, fat, and stomach. 175 Lu-DOTA-PSMA-GUL was not detected in the brain. Accordingly, the average K P values of kidneys and prostate were 2.58 and 1.20, which are greater than 1.0 www.nature.com/scientificreports/ and significantly higher than those of other tissues, indicating that 175 Lu-DOTA-PSMA-GUL is highly distributed into these tissues. The average K P for other tissues was less than 0.34.

Discussion
This study aimed to evaluate the pharmacokinetics of an intact PSMA-targeted ligand in rats. The introduction of an isotopically labeled compound combined with mass spectrometry allowed highly sensitive and specific bioanalysis as well as a full characterization of the pharmacokinetics. A robust quantification method of 175 Lu-DOTA-PSMA-GUL has been developed using LC-MS/MS to analyze the intact 175 Lu-DOTA-PSMA-GUL in the biological fluids, including plasma, excreta, and 12 different tissues. Although sensitive mass spectrometry-based analytical methods have become essential for any pharmacokinetic studies for radio-inactive compounds, there have been limited attempts to actively utilize LC-MS/MS for radioactive ligands. Conventional analytical methods such as LC-MS/MS for the intact compound might be either infeasible due to the radioactivity of the radioligands or not demanding due to the presence of alternatives such as radioactivity and imaging techniques. Nevertheless, the radioactivity assays and imaging techniques are primarily indirect and may result in inaccurate results because they cannot discern the radionuclide located in the intact drug or in the metabolites 20 . Therefore, we utilized a stable isotope-labeled analog, 175 Lu-DOTA-PSMA-GUL, combined with LC-MS/MS to characterize the pharmacokinetic behavior of a PSMA-targeted radioligand, 177 Lu-DOTA-PSMA-GUL. www.nature.com/scientificreports/ By using the LC-MS/MS, the pharmacokinetic profiles and excretion of 175 Lu-DOTA-PSMA-GUL were first examined after i.v. bolus injection in rats. Following i.v. injection, 175 Lu-DOTA-PSMA-GUL exhibited multiexponential decline with a relatively short elimination half-life (t 1/2 ) of 0.30-0.33 h. The rapid clearance from the systemic circulation of PSMA ligands has already been recognized 10,26 . Our data also showed that the systemic exposure represented by C 0 , AUC all , and AUC inf was dose-proportional over the doses of 177 Lu-DOTA-PSMA-GUL from 1 to 5 mg/kg. Over 50% of the administered 175 Lu-DOTA-PSMA-GUL was excreted into the urine, suggesting that renal excretion is the primary elimination route of 175 Lu-DOTA-PSMA-GUL, whereas the excretion into feces was insignificant. These results are in good agreement with the clinical reports that the majority of 177 Lu-PSMA-617 was eliminated by urine (approximately 50%) and only 1-5% by fecal excretion in patients with PSMA-positive tumor phenotype 9 .
The distribution of 175 Lu-DOTA-PSMA-GUL was assessed at two steady state conditions after intravenous infusions of 175 Lu-DOTA-PSMA-GUL in rats. Although PSMA targeted radioligands are reported to be highly distributed to PSMA expressing tumors and normal organs like kidneys via imaging technology 9,11,[16][17][18][19] , direct experimental evidence of their distribution into various tissues has not been reported. Moreover, in the present study, the in vivo tissue distribution of 175 Lu-DOTA-PSMA-GUL has been determined at steady states (Fig. 4). At steady state, the rate of drug input, i.e., intravenous infusion, is equal to the rate of drug output, i.e., elimination, resulting in no net change in the amount of drug in the body. Thus, steady state conditions allow evaluating the true extent of tissue distribution when the plasma and tissue concentration do not change, which may be affected by confounding factors otherwise.
From the plasma concentration-time data obtained after i.v. injection, the average volume of distribution (V ss ) was estimated to be low, i.e., 0.20 L/kg (Table 1), indicating that the distribution of 175 Lu-DOTA-PSMA-GUL into tissues may be limited. Indeed, the observed concentrations of 175 Lu-DOTA-PSMA-GUL in all tested tissues except the kidneys and prostate were lower than half of the plasma concentration ( Table 2). The limited tissue distribution of 175 Lu-DOTA-PSMA-GUL is likely due to its large molecular weight, preventing the passage across the cell membrane. Nevertheless, our data clearly showed that the distribution of 175 Lu-DOTA-PSMA-GUL into the kidneys and prostate was significantly higher than other tissues (Table 2 and Fig. 5). The kidneys and prostate are the well-known tissues where PSMA is expressed 5 . The high distribution in the kidneys may also be attributed to that the primary elimination route of 175 Lu-DOTA-PSMA-GUL is urinary excretion. As presented in Table 1, more than half of the 175 Lu-DOTA-PSMA-GUL is excreted into the urine. On the other hand, the high K P of 175 Lu-DOTA-PSMA-GUL in the prostate may be mainly associated with the prostate targeting property of 175 Lu-DOTA-PSMA-GUL. The GUL domain of 175 Lu-DOTA-PSMA-GUL binds explicitly to PSMA that is overexpressed in the prostate cancer cells, providing the prostate targeting property 10 . These tissue distribution data indicate that 177 Lu-DOTA-PSMA-GUL may be specific and effective for the treatment of prostate cancer.
In summary, the pharmacokinetics of a PSMA-targeted ligand, 175 Lu-DOTA-PSMA-GUL, including excretion and tissue distribution, are characterized for the first time by applying LC-MS/MS. The present data indicate that 175 Lu-DOTA-PSMA-GUL exhibited linear pharmacokinetics over the dose range from 1 to 5 mg/kg and renal excretion is the primary elimination route. The prostate-specific targeting property of 175 Lu-DOTA-PSMA-GUL was also demonstrated. The developed LC-MS/MS assay and the in vivo pharmacokinetic data would provide useful information for the further development of 177 Lu-DOTA-PSMA-GUL and other structural analogs as a novel therapeutic agent for the treatment of prostate cancer.  175 Lu-DOTA-PSMA-GUL and tissue to plasma partition coefficient (K P ) in tissues obtained following simultaneous intravenous bolus injection (loading bolus dose 0.60 and 1.20 mg/ kg) plus continuous intravenous infusion (infusion rate 1.82 and 3.65 mg/h/kg) in rats (n = 5, each). *Plasma concentrations are in ng/mL; BLOQ, below the lower limit of quantification. www.nature.com/scientificreports/

Data availability
The datasets generated and analyzed during this study are available from the corresponding author upon reasonable request.