A Validated LC–MS/MS Method for the Determination of Perospirone in Human Plasma and Its Pharmacokinetic Application in Healthy Volunteers Authors: Benkai Qin, Huina Zhang, Hangyu Zhao, Yuchen Su, Yifu Kong, Shengshuo Zhang, Luwei Wang, Yonghua Yu, Jie Gu, Tingting Lou Published: 2026 ABSTRACT Perospirone hydrochloride hydrate is an atypical antipsychotic widely used in the management of schizophrenia. Reliable quantification of its plasma concentrations remains essential for pharmacokinetic characterization and dose optimization. A rapid, specific, and sensitive liquid chromatography–tandem mass spectrometry (LC–MS/MS) method was developed and fully validated for the quantification of perospirone in human EDTA–K2 plasma using perospirone-d 8 as the internal standard (IS). Chromatographic separation was achieved on a Kinetex C18 analytical column (100 Å, 50 mm × 2.1 mm, 5 μm, 100 Å). The mobile phase consisted of an aqueous solution of 0.1% formic acid and 1 mM ammonium formate (MPA) and a solution of 95% acetonitrile containing 0.1% formic acid and 1 mM ammonium formate (MPB), delivered at a flow rate of 0.300 mL/min. The injection volume was 10 μL. Mass spectrometric detection was performed using electrospray ionization in positive-ion mode with multiple reaction monitoring (MRM). The method exhibited linearity over a concentration range of 0.0200–12.0 ng/mL. The validated method was successfully applied to fasting and postprandial pharmacokinetic studies following oral administration of perospirone hydrochloride tablets. The method demonstrates strong performance characteristics and practical utility, supporting its application in pharmacokinetic investigations, bioequivalence assessment, and clinical drug monitoring. Introduction Perospirone hydrochloride hydrate, a second-generation atypical antipsychotic, has been shown to exhibit a high affinity for the dopamine D2 receptor and to effectively alleviate both positive and negative symptoms of schizophrenia. Unlike many first-generation antipsychotics, perospirone has been associated with more favorable effects on cognitive function, a critical factor in the treatment of schizophrenia. These characteristics position perospirone as a promising therapeutic option, offering enhanced management of the disorder with a lower risk of cognitive impairment, which is a common and debilitating adverse effect associated with older antipsychotic agents. The distinctive pharmacological profile of perospirone, characterized by balanced activity at dopaminergic and serotonergic receptors, has been suggested to contribute to its clinical efficacy and improved tolerability, thereby supporting its growing role in psychiatric practice. This compound is formulated as an oral tablet and is marketed under the trade name of Perospirone Hydrochloride Hydrate Tablets. The drug was developed by a Japanese pharmaceutical company and was first introduced into clinical use in 2001. The chemical name of this compound is cis-N-[4-[4-(1,2-benzisoxazole-3-yl)-1-piperazinyl] butyl]­cyclohexane-1,2-dicarboximide hydrochloride, and its molecular formula is C21H26FN3O2·HCl. This molecular structure reflects a complex pharmacological profile, incorporating both a benzisoxazole moiety and a piperazine ring, which are known to facilitate interactions with multiple central nervous system receptors.. The specific molecular configuration is considered to underlie its targeted therapeutic activity, thereby distinguishing it from other compounds within the same pharmacological class. In addition to its D2 antagonism, perospirone’s 5-HT2 receptor antagonistic effects contribute to its reduced incidence of extrapyramidal side effects, which are commonly associated with first-generation antipsychotics. Perospirone’s dual receptor-blocking profile allows it to achieve balanced dopaminergic and serotonergic modulation, enhancing its efficacy while minimizing side effects related to cognitive function, a key concern in the management of schizophrenia. Furthermore, electrophysiological studies have demonstrated that perospirone can reverse the inhibition of dopamine neuron firing induced by dopamine agonists, suggesting a direct action on the autoreceptors. This action is likely to be a key contributor to its ability to normalize dopaminergic transmission and improve cognitive outcomes, which are significant advantages over older neuroleptics. The drug demonstrates moderate bioavailability due to significant first-pass metabolism, which is primarily mediated by liver enzymes CYP3A4 and CYP2D6. Its plasma protein binding is approximately 80%, which suggests a potential for drug interactions, particularly with other compounds that affect cytochrome P450 enzymes. The drug is effective not only for alleviating acute psychotic episodes but also in preventing relapse in chronic schizophrenia. Furthermore, its ability to improve cognitive deficits has made it a promising treatment for patients with cognitive impairments commonly associated with schizophrenia. Perospirone’s relatively favorable side-effect profile makes it a preferred choice for long-term therapy in psychiatric settings. Perospirone is administered orally as tablets, with the hydrochloride hydrate being the most commonly prescribed salt form. Its pharmacodynamic profile demonstrates a linear dose–response relationship, with clinical efficacy closely correlated with systemic plasma concentrations. Clinical evidence indicates that therapeutic effectiveness is achieved when plasma perospirone concentrations reach or exceed 1.17 ng/mL, supporting the relevance of concentration–effect relationships in its clinical use. The oral bioavailability of perospirone is sufficient to support once-daily dosing in routine clinical practice; however, interindividual variability in absorption and systemic exposure may occur, largely due to differences in metabolic capacity among patients. Consequently, the development of a robust, rapid, and sensitive analytical method for the quantification of perospirone in blood is of considerable importance for therapeutic drug monitoring (TDM). Such an approach facilitates individualized dose optimization, improves therapeutic outcomes, and enhances the overall safety and effectiveness of the perospirone-based treatment regimens. Previous studies have indicated that perospirone identification and quantification primarily relied on analytical techniques including high-performance liquid chromatography (HPLC), ultraviolet spectroscopy, mass spectrometry, and liquid chromatography-based methods. Among these, liquid chromatography–tandem mass spectrometry (LC–MS/MS) has been extensively validated and widely applied in bioanalytical research owing to its superior sensitivity, selectivity, and robustness. An earlier study reported the use of liquid chromatography–single quadrupole mass spectrometry (LC–SQMS) for the determination of perospirone in human plasma. However, the reported lower limit of quantification (LLOQ) was 0.10 ng/mL, indicating a relatively narrow detection range and limited sensitivity. Consequently, this approach was considered for the accurate quantification of trace levels of perospirone in human plasma. In addition, a high-performance liquid chromatography column switching method coupled with fluorescence detection has been reported for perospirone analysis. Although this technique provided acceptable selectivity, it was analytically complex, required specialized instrumentation, and involved prolonged analytical run times, thereby limiting its applicability in routine clinical analysis and large-scale bioequivalence studies. In the present study, a novel, sensitive, and reliable LC–MS/MS method was developed and fully validated for the quantification of perospirone in human EDTA–K2 plasma. LC–MS/MS demonstrated several key advantages for perospirone quantification in biological matrices. First, enhanced analytical sensitivity was achieved through multiple reaction monitoring (MRM), whereby improved signal-to-noise ratios and selective monitoring of precursor-to-product ion transitions enabled reliable detection of low-abundance analytes. Second, high analytical specificity was achieved. In contrast to conventional high-performance liquid chromatography (HPLC), which relies primarily on retention time, LC–MS/MS is based on mass-to-charge ratios and characteristic fragmentation patterns, thereby minimizing matrix interference and enhancing analyte discrimination. Third, a substantially lower limit of quantification (LLOQ) was achieved, 0.0200 ng/mL, markedly lower than that reported for previous LC-based methods. This level of sensitivity was essential for pharmacokinetic studies and therapeutic drug monitoring (TDM), particularly for analytes present at low plasma concentrations. Fourth, rapid analysis and high-throughput performance were achieved, as complete chromatographic separation was not required. In the present study, both perospirone and the internal standard (IS) eluted at approximately 1.00 min, with total run times of 2–3 min, thereby enabling high-throughput analysis for large-scale clinical and bioequivalence studies. Fifth, robust tolerance to matrix effects was demonstrated across normal, hemolyzed, and hyperlipidemic plasma samples, reflecting the inherent selectivity of mass-based detection in complex biological matrices. Furthermore, the validated method was successfully applied to human pharmacokinetic studies of perospirone hydrochloride tablets, demonstrating its suitability for clinical pharmacokinetic evaluation, therapeutic drug monitoring, and bioequivalence assessment. Experimental Process Materials The reference standard, perospirone hydrochloride (salt coefficient 1.085; batch no. 25-03-2802; purity 95.11%; correction factor 0.877), was purchased from SINCO PHARMACHEM. The internal standard, perospirone-d 8 hydrochloride (salt coefficient 1.084; batch no. 23-12-1518; purity 100.00%; correction factor 0.923), was also obtained from SINCO PHARMACHEM. The reference formulation of perospirone hydrochloride tablets was supplied by Sumitomo Pharma Co., Ltd. Reagents Methanol (HPLC grade) was obtained from Merck, and acetonitrile (LC–MS grade) was purchased from Ferton. Ethyl acetate (HPLC grade) and sodium hydroxide (analytical reagent grade) were supplied by Sinopharm Chemical Reagent Co., Ltd. Formic acid (ACS grade) and ammonium formate (analytical reagent grade) were purchased from Aladdin. Ultrapure water was generated using an H2O-I-2-UV-T integrated water purification system (Sartorius, Germany). Instrumentation Liquid chromatographic analyses were performed using a Shimadzu liquid chromatography system coupled to a Triple Quad 5500 mass spectrometer (Applied Biosystems/Sciex). Additional equipment included an MX-S oscillator (Scilogex), a DG-2500R water bath (Shanghai Bajiu Industry), a 5810R centrifuge (Eppendorf), a CyBi-SELMA liquid handling workstation (96/1000 μL; Analytik Jena), analytical balances (MSA6.6S-0CE and CPA225D, Sartorius), an ultrasonic cleaner (GT SONIC-D20, GoodUltrasonic), a nitrogen evaporator (MD200-1A, Aosheng, Hangzhou), and a thermo shaker. Mass Spectrometric Conditions Mass spectrometric detection was performed using electrospray ionization in positive-ion mode with multiple reaction monitoring (MRM). The optimized source parameters are as follows: ion spray voltage, 5500 V; curtain gas (CUR), 40; nebulizer gas (GS1), 50; auxiliary gas (GS2), 55; source temperature (TEM), 550 °C; interface heater, ON; entrance potential (EP), 10 V; collision cell exit potential (CXP), 13 V; and collision gas (CAD), 8. The MRM transition monitored for perospirone was m/z 427.4 → 177.2, while that for the internal standard perospirone-d 8 was m/z 435.4 → 177.2. Liquid Chromatographic Conditions Chromatographic separation was achieved using a Kinetex C18 analytical column (50 mm × 2.1 mm, 5 μm, 100 Å; Finova, 00B-4605-AN). The mobile phase consisted of an aqueous solution of 0.1% formic acid and 1 mM ammonium formate (mobile phase A) and a solution of 0.1% formic acid and 1 mM ammonium formate in 95% acetonitrile (mobile phase B). The column temperature was maintained at 35 °C, and the autosampler temperature was set at 5 °C. A 70% acetonitrile solution was used as the needle wash solvent, with a rinsing volume of 1000 μL and a rinsing speed of 2.1 mL/min. The rinse mode was applied both before and after aspiration using the rinse-port-only method, with rinse and dip times of 2 s each. The flow rate was maintained at 0.300 mL/min, and the injection volume was 10 μL. Gradient elution was performed as follows: 68% mobile phase A from 0 to 0.8 min, linear transition to 95% mobile phase B from 1.3 to 1.7 min, and followed by re-equilibration to 68% mobile phase A from 1.8 to 2.6 min. The retention times of perospirone and perospirone-d 8 were approximately 1.00 min. Preparation of Stock and Working Solutions for Calibration Standards and Quality Control (QC) Samples Stock Solutions After adjustment of the weighed amount according to the correction factor (0.877), the perospirone reference standard was accurately weighed and completely dissolved in methanol to prepare a primary stock solution at a concentration of 1,000,000.00 ng/mL. The stock solution was serially diluted with 50% methanol to obtain working solutions for calibration standards and quality control (QC) samples. The stock solutions were stored in amber glass vials, while the working solutions were stored in colorless polypropylene vials. All solutions were stored at −20 °C. Working Solutions The perospirone stock solution was serially diluted with 50% methanol to prepare the corresponding calibration standard and QC working solutions. The concentrations of the calibration standard working solutions were 0.400, 0.800, 2.00, 10.0, 40.0, 80.0, 200, and 240 ng/mL. The concentrations of the QC working solutions were 0.400, 1.20, 12.0, 120, 180, and 480 ng/mL. All working solutions were stored in colorless polypropylene vials at −20 °C Preparation of Stock and Working Solutions for the Internal Standard (IS) Stock Solution An accurately weighed amount of perospirone-d 8 (1.00 mg, equivalent to 0.923 mg after correction using the correction factor of 0.923) was completely dissolved in 923 μL of methanol to prepare an internal standard stock solution with a final concentration of 1,000,000.00 ng/mL. The stock solution was stored in an amber glass vial at −20 °C. Working Solution The internal standard stock solution was diluted with 50% methanol to obtain a working solution at 2.00 ng/mL. The working solution was stored in a colorless polypropylene vial at −20 °C. Preparation of Calibration Standards Calibration standards were freshly prepared on the day of analysis. An aliquot of 20.0 μL of the appropriate calibration working solution was added to 380 μL of blank plasma matrix or an equivalent dilution volume and mixed thoroughly. The final concentrations of the calibration standards were 0.0200, 0.0400, 0.100, 0.500, 2.00, 4.00, 10.0, and 12.0 ng/mL. Calibration standards prepared immediately prior to analysis were compliant with the requirements of International Council for Harmonization (ICH) M10 and FDA bioanalytical guidelines. This practice ensured that the true concentration of the analyte was accurately represented at the time of analysis, thereby minimizing errors arising from chemical instability, adsorption, or degradation. Such standards are commonly employed for analytes with limited stability and in studies requiring a high degree of quantitative accuracy. They are particularly well suited for LC–MS/MS analyses involving complex biological matrices, such as plasma. Similar analytical strategies have been widely adopted in bioanalytical method development. Preparation of Quality Control Samples Quality control samples were freshly prepared and analyzed on the same day. An aliquot of 20.0 μL of the corresponding QC working solution was added to 380 μL of blank plasma matrix or an equivalent dilution volume and mixed thoroughly to obtain final QC concentrations of lower limit of quantification quality control (LLOQ QC) (0.0200 ng/mL), low-quality control (LQC) (0.0600 ng/mL), GMQC (0.600 ng/mL), MQC (6.00 ng/mL), and high-quality control (HQC) (9.00 ng/mL). The dilution QC (DQC) sample was prepared at a concentration of 24.0 ng/mL. The fresh preparation and same-day analysis of QC samples are common and highly recommended practices, particularly in LC–MS/MS analyses involving biological matrices. This approach ensures accurate representation of the analyte concentration and prevailing matrix conditions, thereby minimizing analytical errors arising from chemical degradation, adsorption, matrix instability, and freeze–thaw effects. Consequently, this practice is commonly employed in studies involving unstable analytes or complex biological matrices, as well as in applications requiring high analytical precision. , Sample Preparation An aliquot of 100 μL of each sample was transferred into the corresponding well of a 96-well plate. For calibration standards, QC samples, test samples, control blanks (QCO), and Equil/SST samples (if applicable), 50 μL of the internal standard working solution was added. For double-blank (DB) samples and upper limit of quantification (ULOQ) samples without an internal standard (if applicable), 50 μL of 50% methanol was added instead. Subsequently, 50 μL of 0.1 M aqueous sodium hydroxide solution was added, and the mixture was vortex-mixed thoroughly. Ethyl acetate (800 μL) was then added, followed by mixing using a multitube vortex mixer at 2200 rpm for 10 min. The samples were centrifuged at 3220g at 4 °C for 5 min. An aliquot of 400 μL of the organic supernatant was transferred to a new 96-well plate and evaporated to dryness under a gentle stream of nitrogen at room temperature. The dried residues were reconstituted with 200 μL of 30% acetonitrile, after which the plate was sealed and shaken for 10 min to ensure complete dissolution prior to LC–MS/MS analysis. Method Validation Method validation was conducted in accordance with the International Council for Harmonization (ICH) Quality Guideline Q2­(R2): Validation of Analytical Procedures and Multidisciplinary Guideline M10: Bioanalytical Method Validation and Study Sample Analysis. Calibration Procedure The injection sequence for each analytical batch was generated by using Watson LIMS (version 7.5) and imported into Analyst software (version 1.6.3). Chromatographic data were acquired and processed using Analyst (version 1.6.3), in which the peak areas of the analyte and internal standard were obtained through automatic integration. The peak area data were subsequently imported into Watson LIMS (version 7.5) for regression analysis by using a weighting factor of 1/x 2 to calculate sample concentrations. Descriptive statistics, including mean, standard deviation, accuracy, and precision for each analytical batch, were calculated by using both Watson LIMS (version 7.5) and Excel (WPS Office). Concentrations were reported in ng/mL to three significant figures. Accuracy (% bias) and precision (% coefficient of variation (CV)) were reported as percentages to one decimal place. Peak areas were retained as integers, retention times were reported to three decimal places, and peak area ratios were reported to six decimal places. Verification of Stock Solution Preparation To verify the accuracy of stock solution preparation, six replicate injections of two independently prepared diluted stock solutions were analyzed using the internal standard for calibration. The ratio of the mean peak areas obtained from the two solutions was compared. The precision (%CV) of the peak area ratio was required to be ≤10.0%; otherwise, the analysis was repeated. If the deviation of the mean peak area ratio between the two solutions was within ±5.0%, the stock solution preparation was considered acceptable. One stock solution was used to prepare calibration standards, and the other was used to prepare quality control samples; calibration standards and QC samples were required to be prepared from different stock solutions. If the deviation exceeded ±5.0%, both solutions were rediluted and reanalyzed. If the deviation remained outside ±5.0% upon repeat analysis, a third stock solution was prepared, and all three solutions were evaluated. When three or more stock solutions were tested, the two solutions with deviations within ±5.0% were selected for preparation of calibration standards and QC samples. The independent preparation of quality control (QC) samples and calibration standards using separate stock solutions, in compliance with ICH M10 and FDA bioanalytical guidelines, was implemented to minimize analytical error, reduce bias, and ensure independent verification. This approach is widely adopted in bioanalytical and chemical analyses to enhance method reliability and data integrity. , Verification of Working Solution Preparation To assess the accuracy of working solution preparation, six replicate injections of the LLOQ QC working solution and the LLOQ calibration standard working solution were analyzed using the internal standard for calibration, and the ratio of the mean peak areas obtained from the two solutions was compared. System Suitability System suitability was evaluated by injecting six replicates of the samples at the LLOQ concentration. For each injection, the signal-to-noise (S/N) ratio for both the analyte and the internal standard was required to be ≥5. The %CV of the peak area ratio and retention time for both the analyte and the internal standard was required to be ≤15.0%. Calibration Curve Performance Calibration curves were constructed using freshly prepared standards at eight concentration levels: 0.0200, 0.0400, 0.100, 0.500, 2.00, 4.00, 10.0, and 12.0 ng/mL. The backcalculated concentration of each calibration standard was required to be within ±15.0% of the nominal value, except for the LLOQ, for which a deviation of ±20.0% was acceptable. No more than 25% of calibration standards could be excluded, and at least 50% of the calibration standards at each concentration level were required to meet the acceptance criteria. The correlation coefficient (R 2) of the calibration curve was required to be ≥0.98. Within-Batch Accuracy and Precision Within-batch precision and accuracy were evaluated using six replicates at each of the following QC levels: LLOQ QC (0.0200 ng/mL), low QC (LQC, 0.0600 ng/mL), geometric mean QC (GMQC, 0.600 ng/mL), medium QC (MQC, 6.00 ng/mL), and high QC (HQC, 9.00 ng/mL). Precision was expressed as %CV, and accuracy was expressed as the percentage deviation of the measured concentration (or mean measured concentration) from the nominal concentration. Precision was required to be ≤15.0% (≤0.0% for LLOQ QC), and accuracy was required to be within ±15.0% (±20.0% for LLOQ QC). Between-Batch Accuracy and Precision Between-batch accuracy and precision were evaluated using QC samples prepared independently in three analytical batches (n = 6 per batch) at the same concentration levels as those used for within-batch assessment (LLOQ QC, LQC, GMQC, MQC, and HQC). The deviation from nominal concentration was required to be within ±15.0% (±20.0% for LLOQ QC), and precision was required to be ≤15.0% (≤20.0% for LLOQ QC). Lower Limit of Quantitation (LLOQ) Six LLOQ samples prepared in mixed blank plasma matrix were analyzed. The %CV of the measured concentrations was required to be ≤20.0%, the mean measured concentration was required to be within ±20.0% of the nominal concentration, and at least 50% of the individual samples were required to have deviations within ±20.0% of the nominal value. Selectivity Selectivity was evaluated by comparing chromatograms obtained from blank human plasma samples to those from LLOQ calibration standards. The peak area of the analyte in blank plasma was required to be less than 20.0% of that observed in the LLOQ sample, and the peak area of the internal standard was required to be less than 5.0% of that observed in the LLOQ sample. Specificity Interference of the IS with the Analyte Potential interference of the internal standard with the analyte was evaluated by comparing blank plasma samples spiked with the internal standard to LLOQ calibration standards. The mean peak area observed in the analyte channel was required to be ≤20.0% of the mean analyte peak area in the LLOQ calibration standards of the same analytical batch. Interference of the Analyte with the IS Potential interference of the analyte with the internal standard was evaluated using samples (n = 3) prepared by spiking blank plasma with the analyte at the ULOQ concentration level. Matrix Effect Matrix effects were evaluated using six independent sources of normal human plasma (nonhemolytic, nonhyperlipidemic), one lot of hemolytic plasma, and one lot of hyperlipidemic plasma. Matrix effects were assessed at the LQC and HQC levels, with three replicates per level. The mean measured concentration was required to be within ±15.0% of the nominal concentration, with precision (%CV) ≤15.0%. Additionally, at least 50% of QC samples at each concentration level were required to meet the ±15.0% acceptance criterion. Extraction Recovery Extraction recovery was evaluated by comparing the mean peak areas of extracted LQC, MQC, and HQC samples (n = 6) with those of postextraction spiked samples at corresponding concentrations. The precision (%CV) of the recovery at each concentration level was required to be ≤15.0%. Dilution Integrity Samples prepared at a concentration above the upper limit of quantitation (DQC, 24.0 ng/mL) were diluted tenfold with blank plasma to bring concentrations within the calibration range (n = 6). The mean measured concentration was required to be within ±15.0% of the nominal value, with at least two-thirds of individual samples meeting the same criterion. Precision (%CV) was required to be ≤15.0%. Stability The stability of perospirone stock and working solutions was evaluated at room temperature (28 h for stock solutions and 26 h for working solutions) and under long-term storage conditions (41 days at −20 °C). The long-term stability and freeze–thaw stability (five cycles) of perospirone in human EDTA–K2 plasma were assessed at −20 and −70 °C. In addition, the room-temperature stability of perospirone in human whole blood under nonlight-protected conditions was evaluated. For QC samples at each concentration level, the mean measured concentration was required to be within ±15.0% of the nominal concentration, and precision was required to be ≤15.0%. Pharmacokinetic Study Design The pharmacokinetic study of piroxicam hydrochloride tablets (8 mg; Sumitomo Pharmaceuticals, Dai Nippon) was conducted in healthy Chinese subjects in accordance with the Guiding Principles for Human Bioequivalence Studies of Generic Chemical Drugs Using Pharmacokinetic Parameters as End Points. The study employed a single-center, randomized, open-label, four-period, two-sequence, and fully replicated crossover design under both fasting and postprandial conditions. Each study period involved the administration of a single oral dose of 8 mg, with a washout interval of 7 days between consecutive periods. Subjects were randomly assigned to two equal-sequence groups. In each period, the study drug was administered orally at 240 mL of water. Subjects were prohibited from drinking water for 1 h before and 1 h after dosing. Under fasting conditions, subjects fasted overnight for at least 10 h prior to drug administration. Under postprandial conditions, subjects consumed a standardized high-fat meal within 30 min, and the study drug was administered with 240 mL of water 30 min after the start of the meal. Subject inclusion criteria were defined in accordance with the bioequivalence study guidelines issued by the National Medical Products Administration (NMPA). For the fasting study, 64 subjects were screened, and 28 subjects were randomized. Venous blood samples were collected at 18 time points: predose (0 h) and at 0.25, 0.5, 0.75, 1, 1.33, 1.67, 2, 2.33, 2.67, 3, 3.5, 4, 6, 8, 12, 24, and 48 h postdose. At each time point, approximately 4 mL of blood was collected in EDTA–K2 anticoagulant tubes. When an indwelling venous catheter was used, approximately 0.5 mL of blood was discarded prior to each sample collection. All 28 subjects were included in the full analysis set (FAS). Among these subjects, 21 (75.0%) were male and 7 (25.0%) were female. Twenty-six subjects (92.9%) were Han Chinese, one (3.6%) was Miao, and one (3.6%) was Manchu. The mean age was 35.21 ± 11.31 years, the mean height was 166.72 ± 8.91 cm, the mean body weight was 66.23 ± 9.76 kg, and the mean body mass index (BMI) was 23.78 ± 2.57 kg/m2. For the postprandial study, 77 subjects were screened, and 36 subjects were randomized. Of these, 35 subjects completed the study, and one subject withdrew prematurely. Venous blood samples were collected at 19 time points: predose (0 h) and at 0.25, 0.5, 0.75, 1, 1.33, 1.67, 2, 2.33, 2.67, 3, 3.5, 4, 5, 6, 8, 12, 24, and 48 h postdose. At each time point, approximately 4 mL of blood was collected in EDTA–K2 anticoagulant tubes. Thirty-five subjects were included in the postprandial FAS. Among these subjects, 24 (68.6%) were male and 11 (31.4%) were female. Thirty-four subjects (97.1%) were Han Chinese, and one subject (2.9%) was She. The mean age was 34.06 ± 9.23 years, the mean height was 165.09 ± 7.87 cm, the mean body weight was 62.33 ± 8.40 kg, and the mean BMI was 22.81 ± 1.95 kg/m2. This clinical study was approved by the Ethics Committee of the Affiliated Provincial Hospital of Fuzhou University (approval date: June 23, 2025; approval no.: 2025-062-01). All subjects provided written informed consent prior to their participation. Pharmacokinetic Analysis The study was conducted in accordance with the Declaration of Helsinki and relevant regulatory requirements and was approved by the Ethics Committee of the Affiliated Provincial Hospital of Fuzhou University. Subject baseline characteristics and safety data were analyzed by using SAS software (version 9.4). Pharmacokinetic parameters were calculated using a noncompartmental analysis (NCA) approach. The selection of an appropriate pharmacokinetic modeling approachwhether NCA or a compartmental or population pharmacokinetic (PopPK) modelis determined by the aims of the investigation, the richness and structure of the available data, and the drug-specific pharmacokinetic characteristics. In the present study, an NCA approach was selected in preference to compartmental or population pharmacokinetic models because of its methodological simplicity and minimal reliance on structural assumptions. As a model-independent technique, NCA does not require predefined hypotheses regarding drug distribution or elimination kinetics. Instead, it derives pharmacokinetic parameters from the statistical moments of the plasma concentration–time profile, making it especially well suited for single-dose studies or when detailed mechanistic characterization of the drug’s disposition is not necessary. Moreover, noncompartmental analysis yields key pharmacokinetic parameters, including the area under the concentration–time curve (AUC), maximum plasma concentration (C max), time to reach C max (T max), elimination half-life (T 1/2), and apparent clearance (CL), which are generally adequate for characterizing systemic drug exposure and elimination behavior. NCA is particularly appropriate when the available data set is limited to plasma concentration measurements and lacks additional informationsuch as tissue distribution datathat would be necessary to support more complex compartmental or population-based pharmacokinetic modeling approaches. Notably, noncompartmental analysis demands substantially fewer resources than the development and validation of compartmental or population pharmacokinetic models, rendering it a more economical and time-efficient approach. This advantage makes NCA particularly suitable for early-phase investigations or for studies conducted under constraints of limited time, data availability, or analytical resources. Based on bioequivalence study (BES) requirements, the primary pharmacokinetic parameters, including maximum plasma concentration (C max), area under the plasma concentration–time curve from time zero to the last quantifiable concentration (AUC0–t ,), and area under the plasma concentration–time curve from time zero to infinity (AUC0–∞), were logarithmically transformed and analyzed using analysis of variance (ANOVA). C max was defined as the observed maximum plasma concentration. AUC0– t was calculated using the trapezoidal method and represented the area under the concentration–time curve from time zero to the last quantifiable concentration. AUC0–∞ represented the area under the concentration–time curve extrapolated to infinity. T max, the time to reach C max, was analyzed as an observed value. The elimination half-life (T 1/2) was defined as the time required for the plasma concentration to decrease by 50%, and λ z represented the apparent terminal elimination rate constant. Results and Discussion Verification of Stock Solution Preparation The results of the stock solution verification are summarized in Table . For all tested stock solutions, the precision of the peak area ratios of perospirone to the internal standard was ≤10.0%, satisfying the predefined acceptance criteria. The maximum deviation between the two independently prepared stock solutions was 4.0%, which was within the allowable range of ±5.0%. 1Verification of Stock Solution Preparationno.sample IDratio of the perospirone to IS in the standard curveratio of the perospirone to IS in QCA11.4289941.39904821.4368931.41764531.4411711.41339741.4553741.40994351.4430181.39686061.4237741.405934mean1.4382041.407138SD0.0111490.008124%CV0.80.6deviation2.2B11.4668231.42404421.4938851.38267631.4361231.45056941.4541321.42219851.4894561.41013561.5168801.421335mean1.4762171.418493SD0.0294090.022064%CV2.01.6deviation4.0C11.4336611.39159721.4667181.40100931.4093731.40361341.4502731.39553251.4461851.45429661.4400331.434160mean1.4410411.413368SD0.0191280.025089%CV1.31.8deviation1.9 Verification of Stock Solution Preparation These results demonstrate that the preparation of the perospirone stock solutions was accurate and reproducible. Consequently, the stock solutions were deemed suitable for the preparation of calibration standards and quality control samples used in subsequent method validation and pharmacokinetic analyses. Verification of Working Solution Preparation The results of the working solution verification are summarized in Table . For all evaluated working solutions, the precision of the peak area ratios of perospirone to the internal standard was ≤10.0%, meeting the predefined acceptance criteria. The deviations between the working solutions prepared for calibration standards and quality control samples ranged from −1.8 to 3.2%, which were well within the allowable limit of ±10%. 2Verification of Working Solution Preparationno.sample IDratio of the perospirone to IS in the standard curveratio of the perospirone to IS in QCA10.0716910.07370520.0754400.07397630.0747480.07281140.0758910.07490450.0758620.07423960.0755980.075080mean0.0748720.074119SD0.0016130.000831%CV2.21.1deviation1.0B10.0687690.06963120.0697670.06786330.0697110.06717440.0735640.06841050.0693960.06700960.0703540.068034mean0.0702600.068020SD0.0016990.000950%CV2.41.4deviation3.2C10.0703340.07332120.0723310.07246230.0710330.07295940.0703250.07291550.0711790.07200760.0714930.070823mean0.0711160.072415SD0.0007570.000902%CV1.11.2deviation–1.8 Verification of Working Solution Preparation These findings indicate that the preparation of the perospirone working solutions was accurate and reproducible. Accordingly, the working solutions were considered suitable for preparing calibration standards and quality control samples for subsequent method validation and pharmacokinetic analyses. System Suitability As summarized in Table , the signal-to-noise ratios for both perospirone and the internal standard exceeded 5 for all evaluated samples. In addition, the coefficients of variation (%CV) for both peak area ratios and retention times were ≤15.0%, satisfying the predefined system suitability acceptance criteria. 3System Suitability of Perospironesample n S/N of perospironeRT of perospironeRT of perospirone CV%S/N of ISIS RTCV% of IS RTarea ratioarea ratio CV%A1331.0480.513771.0400.30.0136856.02311.05215191.0420.0140333291.05316101.0410.0129404371.05715861.0460.0130005301.06413551.0380.0131006391.05011611.0410.015063B1381.0400.730171.0390.60.0131311.22331.05341711.0410.0127233451.04824871.0340.0130244381.05934051.0440.0127745371.06033811.0500.0128016391.05728411.0480.012936C1300.9551.116340.9620.60.0138448.12320.98715460.9610.0170913340.98013630.9740.0141794370.97115660.9690.0152925280.97912760.9640.0140776420.97313510.9730.014834D1161.0740.415981.0610.20.0188258.72121.07915531.0640.0152003141.06714481.0610.0148944121.07118651.0630.0159505151.07216531.0650.0159696131.07114481.0620.016812E1421.0541.925891.0420.20.0154275.22421.04628681.0440.0167963421.06429821.0450.0166744421.05632661.0400.0157685411.03925641.0460.0162896441.06026271.0470.014617F1771.0120.438231.0110.50.0141514.32801.02243251.0140.0145613811.01443531.0010.0155724681.01142901.0120.0153175831.01538221.0160.0153136731.01243701.0060.014138G1361.0610.632211.0550.50.0144336.12311.05437551.0540.0133133391.06641101.0440.0156034321.04831371.0560.0135055361.06034851.0440.0143696331.06037331.0500.013569H1420.9420.624790.9340.40.0130955.92420.93527510.9300.0138973430.94829240.9330.0147244520.93826630.9250.0149745450.94724400.9280.0145746380.93826370.9350.013087 System Suitability of Perospirone These results confirm that the LC–MS/MS system was stable and reliable and that the analytical system was suitable for the continuous analysis of both solution-based and extracted plasma matrix samples. Calibration Curve Performance and LLOQ As shown in Figure , a linear relationship between perospirone concentration and detector response was observed in human plasma over the concentration range of 0.0200–12.00 ng/mL (0.0200, 0.0400, 0.100, 0.500, 2.00, 4.00, 10.0, and 12.00 ng/mL). At each concentration level, at least 50% of the calibration standards met the predefined acceptance criteria, with concentration deviations within ±15.0% of nominal values (±20.0% at the lower limit of quantitation). Standard Curve of perospirone. The correlation coefficients (R 2) of the calibration curves ranged from 0.9974 to 0.9996, demonstrating excellent linearity across all analytical batches (Table ). The representative chromatograms obtained at the LLOQ level are listed in Figure . All LLOQ chromatograms met the established acceptance criteria for selectivity, signal intensity, and peak shape. Chromatogram of LLOQ. 4Calibration Curve Parameters for Perospironesampleslopeintercept R 2 LLOQ (ng/mL)ULOQ (ng/mL)standard curve10.74180.00110.99960.020012.0 y = 0.7418x + 0.001120.74960.00130.99920.020012.0 y = 0.7496x + 0.001330.69880.00110.99740.020012.0 y = 0.6988x + 0.001140.75020.00050.99950.020012.0 y = 0.7502x + 0.000550.73750.00130.99920.020012.0 y = 0.7375x + 0.001360.74580.00100.99920.020012.0 y = 0.7458x + 0.001070.71760.00100.99820.020012.0 y = 0.7176x + 0.001080.68820.00140.99950.020012.0 y = 0.6882x + 0.0014mean0.72870.00110.9990NANANASD0.02420.00030.0008NANANA%CV3.3NA0.1NANANA n 888NANANA Calibration Curve Parameters for Perospirone Accuracy and Precision The intrabatch and interbatch accuracy and precision results for perospirone are summarized in Table . For intrabatch quality control samples, the maximum precision (%CV) across all concentration levels, excluding the LLOQ QC, was 4.5%, with accuracy deviations ranging from −3.8 to 3.7%. At the LLOQ QC level, the maximum intrabatch precision was 14.2%, and accuracy deviations ranged from 0.0 to 5.5%. 5Intrabatch and Interbatch Accuracy and Precision for PerospironesampleLLOQ QCdeviation (%)LQCdeviation (%)GMQCdeviation (%)MQCdeviation (%)HQCdeviation (%)10.02073.50.06000.00.6050.85.91–1.58.91–1.00.0188–6.00.06386.30.6132.25.88–2.08.92–0.90.02031.50.06040.70.594–1.05.90–1.79.020.20.0172–14.00.0594–1.00.6142.35.89–1.88.89–1.20.022713.50.06569.30.599–0.26.101.78.88–1.30.025628.00.06244.00.598–0.36.010.28.69–3.4mean0.0209NA0.0619NA0.604NA5.95NA8.89NASD of intrabatch0.00296NA0.00244NA0.00828NA0.088NA0.108NA%CV of intrabatch14.2NA3.9NA1.4NA1.5NA1.2NAdeviation of intrabatch (%)4.5NA3.2NA0.7NA–0.8NA–1.2NA n 6NA6NA6NA6NA6NA20.0187–6.50.0572–4.70.6000.06.122.08.84–1.80.0196–2.00.0571–4.80.6223.76.305.09.000.00.0197–1.50.0579–3.50.6183.06.172.88.86–1.60.02178.50.0584–2.70.6111.86.132.28.68–3.60.02073.50.0567–5.50.6203.36.142.38.64–4.00.0195–2.50.0586–2.30.6386.36.030.58.90–1.1mean0.0200NA0.0577NA0.618NA6.15NA8.82NASD of intrabatch0.00106NA0.000766NA0.0126NA0.0880NA0.136NA%CV of intrabatch5.3NA1.3NA2.0NA1.4NA1.5NAdeviation of intrabatch (%)0.0NA–3.8NA3.0NA2.5NA–2.0NA n 6NA6NA6NA6NA6NA30.02147.00.06172.80.6101.75.89–1.88.89–1.20.02000.00.06284.70.6142.35.74–4.38.63–4.10.022512.50.067312.20.6172.85.80–3.38.77–2.60.02115.50.0599–0.20.6203.35.93–1.28.90–1.10.02105.00.0597–0.50.6050.85.97–0.59.010.10.02063.00.06183.00.6233.85.97–0.58.87–1.4mean0.0211NA0.0622NA0.615NA5.88NA8.85NASD of intrabatch0.000839NA0.00277NA0.00662NA0.0946NA0.130NA%CV of intrabatch4.0NA4.5NA1.1NA1.6NA1.5NAdeviation of intrabatch (%)5.5NA3.7NA2.5NA–2.0NA–1.7NA n 6NA6NA6NA6NA6NAmean of interbatch0.0207NA0.0606NA0.612NA5.99NA8.85NASD of interbatch0.00183NA0.00296NA0.0109NA0.144NA0.121NA%CV of interbatch8.8NA4.9NA1.8NA2.4NA1.4NAdeviation of interbatch (%)3.5NA1.0NA2.0NA–0.2NA–1.7NA n 18NA18NA18NA18NA18NA Intrabatch and Interbatch Accuracy and Precision for Perospirone For interbatch evaluation, the maximum precision for all QC concentration levels, excluding the LLOQ QC, was 4.9%, with accuracy deviations ranging from −1.7 to 0.0%. At the LLOQ QC level, the maximum interbatch precision was 8.8%, with accuracy deviations of 3.5%. All intrabatch and interbatch accuracy and precision values met the predefined acceptance criteria, with deviations within ±15.0% for QC samples and within ±20.0% for the LLOQ QC. These results demonstrate that the developed LC–MS/MS method provides robust and reproducible accuracy and precision across analytical batches. Selectivity The selectivity results are summarized in Table . The blank plasma matrix exhibited a maximum interference of 4.9% with the analyte and 0.1% with the internal standard. For all evaluated samples, the peak area of the analyte in blank plasma was less than 20.0% of that observed in the corresponding LLOQ samples, and the peak area of the internal standard did not exceed 5.0% of that observed in the LLOQ samples. 6Interference of Blank Plasma Matrix with Perospirone and the Internal Standard perospironeperospirone-d 8 LLOQLLOQ LLOQLLOQ samplearea of the blank matrixareaareainterference %area of the blank matrixareaareainterference %1218451044484.964280,685282,9530.021022.31900.13591.32890.141212.73670.151012.34090.16501.1500.07701.62180.18922.11420.1 Interference of Blank Plasma Matrix with Perospirone and the Internal Standard These results demonstrate that endogenous components in human plasma did not cause significant interference with the detection of perospirone or the internal standard, confirming the adequate selectivity of the developed LC–MS/MS method. Specificity The specificity results are summarized in Tables and . The internal standard exhibited a maximum interference of 3.8% with the analyte, while no interference of the analyte with the internal standard was observed (0.0%). All of the results met the predefined acceptance criteria. 7Interference of the Internal Standard with the Analyte perospirone LLOQ-1LLOQ-2 samplearea of the blank matrixareaRT (min)areaRT (min)interference (%)18045101.05444481.0521.821423.23942.14601.35741.761383.171403.181703.891872.66895 Interference of the Internal Standard with the Analyte 8Interference of the Analyte with the Internal Standard perospirone-d 8 LLOQ-1LLOQ-2 samplearea of ULOQ without ISareaRT (min)areaRT (min)interference (%)1227280,6581.055282,9531.0410.02030 Interference of the Analyte with the Internal Standard For the assessment of the interference of the internal standard on the analyte, the mean peak area detected in the analyte channel in blank plasma spiked with an internal standard was less than 20.0% of the corresponding analyte peak area observed in the LLOQ calibration samples of the same analytical batch. For the assessment of interference of the analyte on the internal standard, the peak area of the internal standard in samples spiked with the analyte at the ULOQ level did not exceed 5.0% of that observed in the LLOQ samples. These results confirm that neither the internal standard nor the analyte caused significant interference with each other, demonstrating the adequate specificity of the developed LC–MS/MS method. Matrix Effect The matrix effect results obtained from normal, hyperlipidemic, and hemolyzed plasma are summarized in Tables and . At both the low-quality control (LQC) and high-quality control (HQC) concentration levels, the mean measured concentrations for all plasma matrices deviated from the nominal concentrations by no more than ±15.0%, and the corresponding precision (%CV) values were ≤15.0%. 9Matrix Effect Evaluation at the LQC Levelsample n LQCmeanSDCV%accuracy deviationaverage deviation of accuracyA10.06020.06140.001442.30.32.320.06101.730.06305.0B10.06060.06170.002724.41.02.820.0597–0.530.06488.0C10.05990.06140.004126.7–0.22.320.0583–2.830.066110.2D10.06180.06190.001252.03.03.220.06325.330.06071.2E10.06330.06310.0008621.45.55.220.06396.530.06223.7F10.05950.06120.002554.2–0.82.020.0599–0.230.06416.8G10.06000.06000.0002520.40.00.020.0597–0.530.06020.3H10.05980.06230.002784.5–0.33.820.06183.030.06538.8 Matrix Effect Evaluation at the LQC Level 10Matrix Effect Evaluation at the HQC Levelsample n HQCmeanSDCV%accuracy deviationaverage deviation of accuracyA19.029.100.1121.20.21.129.232.639.060.7B18.858.970.1441.6–1.7–0.328.93–0.839.131.4C18.788.930.1551.7–2.4–0.829.091.038.92–0.9D18.979.020.1011.1–0.30.228.96–0.439.141.6E18.919.120.1792.0–1.01.329.222.439.222.4F18.768.830.07510.9–2.7–1.928.91–1.038.83–1.9G18.979.130.1501.6–0.31.429.141.639.273.0H18.678.810.1241.4–3.7–2.128.88–1.338.89–1.2 Matrix Effect Evaluation at the HQC Level Across all evaluated plasma types, including normal, hyperlipidemic, and hemolyzed matrices, the acceptance criteria for the matrix effect assessment were consistently met. These results indicate that endogenous matrix components did not cause significant ion suppression or enhancement under the established analytical conditions, demonstrating that matrix effects had no meaningful impact on the quantification of perospirone in the human plasma. Extraction Recovery The extraction recovery results are summarized in Table . The mean extraction recovery of the internal standard was 84.5% with a coefficient of variation (%CV) of 1.5%. For perospirone, the precision (%CV) of the extraction recovery at each evaluated concentration level was less than 15.0%, satisfying the predefined acceptance criteria. 11Extraction Recovery of Perospirone and the Internal StandardsampleHQCMQCLQC1arearecovery ratearearecovery ratearearecovery rate22,502,72685.31,670,50183.118,61888.132,523,06385.21,691,86485.917,59984.842,539,19884.91,701,97885.217,60684.852,552,20485.31,685,71185.717,60584.762,574,70884.31,690,91084.618,05485.2mean2,588,86387.41,677,42588.218,11688.9SD2,546,79485.41,686,39885.517,93386.1%CV32,1001.0511,2001.684111.90average recovery rate85.485.586.1total recovery rate85.7total variation %CV0.4 Extraction Recovery of Perospirone and the Internal Standard These results indicate that the sample preparation procedure yielded consistent, reproducible extraction efficiencies for both perospirone and the internal standard. Dilution Integrity Dilution integrity was evaluated using a dilution factor of 10, and the results for perospirone are summarized in Table . All diluted quality control samples showed concentration deviations within ±15.0% of the nominal value, and the corresponding precision (%CV) values were also ≤15.0%. 12Dilution Integrity Results for Perospirone at a Tenfold Dilutionsampledetected concentration of DQC (ng/mL) (actual concentration was 24.0 ng/mL)deviation (%)124.93.7223.3–2.9322.9–4.6423.5–2.1523.9–0.4623.8–0.8average23.7NASD0.682NA%CV1.5NAaccuracy%103.5NAdeviation%3.5NA Dilution Integrity Results for Perospirone at a Tenfold Dilution These results demonstrate that samples with concentrations above the upper limit of quantitation can be reliably diluted and accurately quantified by using the developed LC–MS/MS method, confirming the acceptability of the dilution procedure. Stability The stability evaluation results are summarized in Table . Perospirone stock solutions in methanol and working solutions in 50% methanol were stable for 28 and 26 h at room temperature, respectively, and for 41 days when stored at −20 °C. Perospirone in the human whole blood matrix remained stable for 2 h at room temperature without light protection. 13Summary of Stability Evaluation for Perospirone under Various Storage and Handling Conditionssampleconditionsstabilitystock solution (in MeOH)room temperature28 h–20 °C41 daysworking solution (in 50%MeOH)room temperature26 h–20 °C41 daysanalytes in the whole blood matrixroom temperature (not protected from light)2 hanalytes in the plasma matrixroom temperature29 hlong-term stability (−20 °C)40 dayslong-term stability (−70 °C)83 daysfreeze–thaw stability (freeze at –70 °C, thaw at room temperature)5 cyclesprepared samplesroom temperature3 h2–8 °C172 hautosampler (5 °C)183 h Summary of Stability Evaluation for Perospirone under Various Storage and Handling Conditions In human plasma, perospirone was stable for 29 h at room temperature, for 40 days at −20 °C, and for 83 days at −70 °C. The analyte also demonstrated stability following five freeze–thaw cycles (freezing at −70 °C and thawing at room temperature). Prepared samples were stable for 3 h at room temperature, for 172 h when stored at 2–8 °C, and for 183 h in the autosampler maintained at 5 °C. All stability results met the predefined acceptance criteria, indicating that perospirone exhibited an adequate stability under all tested storage and handling conditions relevant to routine sample analysis. Pharmacokinetic Study The validated LC–MS/MS method was successfully applied to determine plasma concentrations of perospirone following single-dose oral administration under fasting and postprandial conditions. The main pharmacokinetic parameters are summarized in Tables and . 14Pharmacokinetic Parameters of Perospirone Following Single-Dose Administration under Fasting Conditions mean ± SD (%CV)parametersreference (N period = 56) T max (h)0.75(0.25,1.67) C max (ng/mL)5.33 ± 3.98(74.68)AUC0–t (h·ng/mL)8.71 ± 6.11(70.13)AUC0–∞ (h·ng/mL)9.08 ± 6.25(68.88)λ z (h–1)0.12 ± 0.06(44.67) T 1/2 (h)6.81 ± 3.03(44.49)AUC_%Extrap (%)4.86 ± 1.82(37.43)CL/F (L/h)1480.87 ± 1134.27(76.60)V z /F (L)11302.31 ± 5303.36(46.92)a T max is expressed by the median (minimum and maximum). Pharmacokinetic Parameters of Perospirone Following Single-Dose Administration under Fasting Conditions T max is expressed by the median (minimum and maximum). 15Pharmacokinetic Parameters of Perospirone Following Single-Dose Administration under Postprandial Conditions mean ± SD (%CV)parametersreference (N period = 70) T max (h)1.34(0.25,5.00) C max (ng/mL)4.84 ± 2.96(61.09)AUC0–t (h·ng/mL)13.30 ± 7.78(58.48)AUC0–∞ (h·ng/mL)13.73 ± 7.83(56.99)λ z (h–1)0.10 ± 0.03(29.55) T 1/2 (h)7.87 ± 2.25(28.64)AUC_%Extrap (%)4.27 ± 3.12(73.14)CL/F (L/h)866.55 ± 677.42(78.17)V z /F (L)8770.30 ± 5095.66(58.10)a T max is expressed by the median (minimum, maximum). Pharmacokinetic Parameters of Perospirone Following Single-Dose Administration under Postprandial Conditions T max is expressed by the median (minimum, maximum). Under fasting conditions, perospirone reached a mean maximum plasma concentration (C max) of 5.33 ± 3.89 ng/mL, while under postprandial conditions, the mean C max was 4.84 ± 2.96 ng/mL. The area under the plasma concentration–time curve extrapolated to infinity (AUC0–∞) was 9.08 ± 6.25 ng·h/mL following fasting administration and 13.73 ± 7.83 ng·h/mL following postprandial administration. The mean elimination half-life (T 1/2) of perospirone was 6.81 ± 3.03 h under fasting conditions and 7.87 ± 2.25 h under postprandial conditions. The pharmacokinetic parameters obtained under both dosing conditions are presented in Tables and . Food intake markedly modulates the pharmacokinetic profile of perospirone, predominantly affecting absorption kinetics rather than elimination. The observed prolongation of T max, accompanied by a concomitant, modest reduction in C max, under postprandial conditions is consistent with delayed gastric emptying and reduced absorption rate. In contrast, systemic exposure increased substantially, as reflected by a 50% elevation in AUC, indicating enhanced oral bioavailability, potentially driven by improved solubility and attenuation of presystemic metabolism. Although a slight increase in apparent T 1/2 was noted in the fed state, this is unlikely to reflect a true change in elimination. Rather, it may be attributable to absorption–elimination overlap or improved characterization of the terminal phase. Consistently, the apparent reductions in CL/F and V z /F further support a food-induced increase in bioavailability. Collectively, these findings underscore a clinically relevant food effect, characterized by delayed absorption but enhanced systemic exposure, with minimal impact on intrinsic elimination processes. Figures and demonstrate the relationship between plasma concentration and time for perospirone in the human body under fasting and under postprandial conditions. Fasting test: mean plasma concentration–time curve of perospirone and semilogarithmic plasma concentration–time curve. Postprandial test: mean plasma concentration–time curve of perospirone and semilogarithmic plasma concentration–time curve. Discussion The comprehensive validation results demonstrate that the developed LC–MS/MS method is robust, reliable, and well suited for the quantitative determination of perospirone in human plasma. The method employed liquid–liquid extraction for sample pretreatment, with the stable isotope-labeled compound perospirone-d 8 used as the internal standard to ensure analytical accuracy and reproducibility. Detection was performed using electrospray ionization in positive-ion mode coupled with multiple reaction monitoring (MRM), which provided high selectivity and sensitivity for perospirone quantification. Quantitation was based on the peak area ratio of perospirone to the internal standard, which enabled accurate quantification over a validated concentration range of 0.0200 ng/mL (lower limit of quantitation, LLOQ) to 12.0 ng/mL (upper limit of quantitation, ULOQ). The method required only 100 μL of human EDTA–K2 plasma per sample, making it particularly suitable for pharmacokinetic and bioequivalence studies where sample volume may be limited. Furthermore, validation of a tenfold dilution protocol confirmed that samples exceeding the ULOQ could be reliably quantified without compromising precision or accuracy. The validated method demonstrated excellent linearity, high sensitivity, and high specificity across the entire calibration range, consistently meeting regulatory acceptance criteria for accuracy and precision. These performance characteristics confirm that the method is suitable for routine bioanalytical applications in human plasma. Application of the method to a clinical bioequivalence study further verified its practical utility. Following oral administration of an 8 mg dose of perospirone hydrochloride tablets in healthy volunteers, the observed maximum plasma concentration (C max) was 2.79 ± 0.78 ng/mL. Considering the administered dose, the validated analytical range was sufficient to encompass the plasma concentration profile of perospirone throughout its absorption, distribution, and elimination phases. Despite these advantages, several limitations were identified. Extensive sample pretreatment procedures, including protein precipitation and liquid–liquid extraction, were required to remove endogenous interferences, thereby increasing methodological complexity, introducing potential sources of variability, and prolonging method development. In addition, LC–MS/MS instrumentation was associated with substantial acquisition and maintenance costs and required highly trained personnel, thereby limiting its accessibility in routine or resource-limited laboratory settings. The use of stable isotope-labeled internal standards (e.g., perospirone-d 8) was necessary to ensure analytical accuracy, further increasing cost and operational complexity. Finally, the highly targeted nature of multiple reaction monitoring (MRM) acquisition provided limited structural information compared with full-scan mass spectrometry, reducing its suitability for metabolite identification or nontargeted screening applications. Collectively, these findings demonstrate that LC–MS/MS provides a highly sensitive, selective, and efficient analytical platform for perospirone quantification. It represents a reliable approach for the determination of perospirone in human plasma and is well suited for pharmacokinetic investigations and bioequivalence evaluations of perospirone hydrochloride formulations. Conclusion A highly sensitive and selective liquid chromatography–tandem mass spectrometry (LC–MS/MS) method was successfully developed and fully validated for the quantification of perospirone in human EDTA–K2 plasma. The method demonstrated excellent linearity over a concentration range of 0.0200–12.0 ng/mL, with robust accuracy and precision across all validation parameters, and a short analytical run time of 1.0 min. All analytical batches met the predefined acceptance criteria, confirming the reliability and reproducibility of the assay. Stability assessments showed that perospirone remained stable in human plasma for up to 29 h at room temperature, further supporting the method’s suitability for routine clinical sample analysis. The validated analytical method was successfully applied to pharmacokinetic evaluations of perospirone hydrochloride tablets following single-dose administration under both fasting and postprandial conditions. Owing to its high sensitivity, excellent selectivity, and robust analytical performance, this assay is well suited for clinical pharmacokinetic investigations and bioequivalence assessments of perospirone hydrochloride formulations. Notably, the method features a low lower limit of quantitation and a wide dynamic range, enabling reliable determination of perospirone at trace concentrations in human plasma. The improved analytical capability substantially enhances the precision of therapeutic drug monitoring, thereby facilitating rational dose optimization to enhance clinical efficacy and safety, reducing drug-related toxicity, and mitigating the risk of treatment failure or the emergence of drug resistance.