The test formulation, TU soft capsules, was supplied by manufactured by Zhejiang Medicine Co., Ltd. Xinchang Pharmaceutical Factory (Bath NO.: 1052202, 40 mg), while the reference formulation, Andriol Testocaps^®, was provided by Catalent France Beinheim S.A. (Bath NO.: T023247, 40 mg). All study drugs were offered by Xinchang Pharmaceutical Factory.

This clinical trial was carried out at the Clinical Trial Research Center of Xinxiang Central Hospital. According to the requirements of the sponsor, the clinical trial was registered at chinadrugtrials.org.cn (CTR20223150, registered on 14 December 2022), Which was an authoritative registration authority widely recognized under Chinese laws and regulations. The study protocol and its amendments met the Good Clinical Practice guidelines and Declaration of Helsinki. The study protocol was approved by the Ethics Committee of Xinxiang Central Hospital (approval number: 2022-075). All participants willingly consented to take part in this study and provided written informed consent.

The clinical trial recruited Chinese healthy naturally postmenopausal females who were in good health and aged between 45 and 65 years (inclusive), with cessation of menses for≥12 months, and the BMI range was 18–28 kg/m² (inclusive). Postmenopausal women were chosen as surrogates for hypogonadal men due to their stable, low endogenous testosterone levels, which minimize confounding fluctuations inherent in male subjects. This selection aligns with regulatory guidance for endogenous compound bioequivalence studies. Female participants had a minimum weight of 45 kg and a total testosterone level ≤0.75 ng/mL. The participants underwent a comprehensive evaluation. Participants who satisfied the eligibility criteria were included, whereas those who fulfilled any of the exclusion criteria were not recruited. Additional details regarding the criteria for including and excluding individuals from the study can be found in the Supplementary Material.

Healthy participants were randomly assigned to either the T-R-T-R sequence group or the R-T-R-T sequence group in a 1:1 ratio. Participants received oral administration of 40 mg TU soft capsules (T/R) under postprandial conditions on Day 1/Day 6/Day 11/Day 16. On the first day of each dosing cycle, participants, after an overnight fast of at least 10 h, initiate the consumption of a high-fat meal (800–1000 calories) 30 min prior to medication administration. Subsequently, participants orally take 40 mg of Andriol Testocaps^® or TU according to the schedule. The washout period between periods was set to be no less than 5 days.

To compare the bioequivalence of TU and Andriol Testocaps^®, a single-center, open-label, randomized, single-dose, four-period, fully repeated crossover design was conducted. The determination of the sample size was based on prior reports indicating that the intra-participant variability (CV_W%) of testosterone undecanoate C_max and AUC ranged from 30% to 50% (Yin et al., 2012; Schnabel et al., 2007). This study employed a four-period fully replicated crossover design with pharmacokinetic parameters (AUC, C_max) as primary analysis indices. Assuming α = 0.05, β = 0.2, Intra-CV = 30%, geometric mean ratio (GMR) of test/reference formulation = 0.90, and bioequivalence interval of 80.00%–125.00%, PASS software calculated a minimum sample size of 40 participants. Considering a certain dropout rate, 48 participants were ultimately enrolled in the postprandial trial.

Blood samples were collected in each study period. For sample collection, vacuum collection tubes pre-cooled in an ice-water bath containing K₂-ethylenediaminetetraacetic (EDTA-K₂) anticoagulant were used, with blood drawn at 24 time points: 2.0 h before dosing (−2.0 h), 1.0h before dosing (−1.0 h), 0 h (dosing time), and 1.0 h, 2.0 h, 3.0 h, 3.5 h, 4.0 h, 4.5 h, 5.0 h, 5.5 h, 6.0 h, 6.5 h, 7.0 h, 7.5 h, 8.0 h, 9.0 h, 10.0 h, 11.0 h, 12.0 h, 13.0 h, 14.0 h, 16.0 h, 24.0 h after dosing. Biologic sample collection and processing were conducted under ice-bath white light conditions. The collected whole blood was centrifuged at 2 °C–8 °C with a relative centrifugal force (RCF) of 1700 g for 10 min to separate plasma. The resulting plasma supernatant was stored at −80 °C prior to analysis.

The plasma concentrations of Testosterone undecanoate and Testosterone were measured with a validated liquid chromatography with tandem mass spectrometry (LC-MS/MS) method developed by Shanghai Xihua Scientific Co. Ltd. Testosterone undecanoate, Testosterone and their internal standards (D₃-Testosterone undecanoate and Testosterone-¹³C₃) were isolated from human plasma using solid-phase extraction with C₁₈ (Cleanert ODS C₁₈, Agela). Elution was performed using acetonitrile. The eluate was evaporated to dryness, and the residue subsequently dissolved in methanol and water 60:40 (V:V). For calibration samples, the internal standard solution was vortexed with surrogate matrix and working solutions. The concentration range of the calibration curve used for sample analysis was 0.300–120 ng/mL for Testosterone undecanoate and 0.0750–15.0 ng/mL for Testosterone.

Bioanalysis was performed on a Shimadzu LC-30AD system with an Applied Biosystems Triple Quadrupole 6500+ mass spectrometer equipped with an APCI source. The data were recorded using the Analyst1.7.2 software (Applied Biosystems, USA). Chromatographic separation was achieved on an Poroshell 120 EC-C18 column (100 × 2.1 mm, 2.7 μm; Agilent). The mobile phase consisted ofwater with 0.1% formic acid and 2 mM ammonium acetate (A) and methanol (B), and the flow rate was 0.50 mL/min. Quantification was performed in positive ion mode through the multiple reaction monitoring of the ion pairs of 457.2/271.2 for Testosterone undecanoate, 460.2/274.2 for D₃-Testosterone undecanoate, 289.2/97.2 for Testosterone and 292.2/100.2 for Testosterone-¹³C₃.

Vital signs (including auricular temperature, pulse rate, and sitting blood pressure) were measured at 1.0 h before dosing and at 2.0 ± 0.5, 6.0 ± 0.5, 12.0 ± 0.5, and 24.0 ± 1.0 h post-dosing in each study period. Researchers promptly documented adverse events (AEs) throughout the trial. For participants in Period 4, safety assessments (physical examination, vital sign reevaluation, 12-lead electrocardiogram, and laboratory tests) were conducted on the day of pharmacokinetic blood sampling completion. A telephone follow-up was performed 7 ± 2 days after study discharge to inquire about subsequent AEs and document their outcomes. All AEs observed during the trial were followed until resolution.

Pharmacokinetic parameters of testosterone undecanoate—including C_max (maximum plasma concentration), AUC_0-t, AUC 0 − ∞ , t_max, λ_z (elimination rate constant), t_1/2, and AUC_{_%Extrap}—were calculated using Phoenix WinNonlin 8.3 software via non-compartmental pharmacokinetic analysis. Descriptive statistics were generated for all parameters using SAS 9.4. For test and reference formulations, the natural logarithm-transformed C_max, AUC_0-t, and AUC 0 − ∞ values underwent mixed-effects model analysis of variance. GMRs of the two formulations were computed at the 90% confidence level, with ratios converted back to linear scale via antilogarithmic transformation. A statistical significance threshold of P < 0.05 was applied.

Equivalence criteria were evaluated based on reference formulation CV_w%. If CV_w% < 0.294 (i.e., CV_w% < 30%), the average bioequivalence (ABE) method was applied, requiring the 90% confidence interval (CI) of GMR for C_max, AUC_0-t, and AUC 0 − ∞ to fall within 80.00%–125.00%. For CV_w% ≥ 30%, the reference-scaled average bioequivalence (RSABE) method was implemented, mandating simultaneous fulfillment of two conditions: (1) GMR upper confidence bound ≤ pre-specified upper limit (calculated via FDA-recommended formula) and (2) GMR point estimate within 80.00%–125.00%. Both testosterone undecanoate (primary analyte) and testosterone (supportive analyte) adhered to these equivalence criteria.

A total of 127 participants were screened, and 48 participants (all female) were ultimately enrolled, including 47 Han Chinese and 1 Miao ethnic participant. The mean (±SD) age was 53.67 ± 3.33 years, with average height of 158.40 ± 5.16 cm, weight of 60.08 ± 6.07 kg, and body mass index (BMI) of 23.93 ± 1.89 kg/m², as shown in Table 1. The study employed a four-period replicate crossover design with two treatment sequences: T-R-T-R and R-T-R-T, each comprising 24 participants. All enrolled participants completed the trial without premature withdrawal, adhering to the scheduled drug administration protocol.

Figure 1 presents the mean plasma concentration-time profiles of testosterone undecanoate under postprandial conditions for both test and reference formulations. The corresponding pharmacokinetic parameters are summarized in Table 2. For testosterone undecanoate, the primary pharmacokinetic parameters (C_max, AUC_0-t, AUC 0 − ∞ ) were analyzed using a mixed-effects model after natural logarithm transformation. The model incorporated the effects of dosing sequence, formulation factors, and dosing periods to compare their influences on parameter variability.

The results indicated no statistically significant differences (P > 0.05) in C_max, AUC_0-t and AUC 0 − ∞ between the test and reference formulations for both dosing sequence and formulation factors. However, significant differences (P < 0.05) were observed across dosing periods. Detailed pharmacokinetic analyses of testosterone (baseline-corrected and non-corrected) are provided in Supplementary Material S1, S2.

As shown in Table 3, the within-participant coefficient of variation (S_WR) for the reference formulation’s C_max of testosterone undecanoate was 0.4801 (>0.294), indicating CV_W% of 50.91% (>30%). Therefore, reference-scaled average bioequivalence (RSABE) was applied for bioequivalence evaluation with a regulatory acceptance range of −0.1329 (<0). The geometric mean ratio (GMR) of C_max between the test and reference formulations was 102.20%, falling within the 80.00%–125.00% range.

For the AUC_0-t and AUC 0 − ∞ parameters of testosterone undecanoate, the S_WR values (0.2871 and 0.2838) were <0.294, corresponding to CV_W% values of 29.31% and 28.96% (<30%), respectively. Thus, average bioequivalence (ABE) criteria were applied. The GMRs and 90% confidence intervals for AUC_0-t and AUC 0 − ∞ were 99.85% (92.82%–107.41%) and 99.79% (92.90%–107.20%), both within the 80.00%–125.00% range. Sensitivity analyses of testosterone undecanoate confirmed consistency with the primary analysis, supporting bioequivalence between the test and reference formulations (Supplementary Material S3). Bioequivalence was also demonstrated for both baseline-corrected and non-corrected testosterone levels (Supplementary Material S4, S5).

As shown in the study, a total of 7 participants experienced 8 AEs with an incidence rate of 14.6% (7/48), all occurring during the test formulation period. Additionally, 5 participants reported 6 adverse drug reactions (ADRs) with an incidence rate of 10.4% (5/48), also exclusively during the test formulation administration phase. The most common AE was urine leukocyte positivity.

All adverse events were classified as Grade 1 severity, with no serious adverse events (SAEs), serious adverse drug reactions (SADRs), AEs/ADRs leading to withdrawal, or irreversible outcomes observed. Notably, all 8 AEs resolved spontaneously without sequelae under non-intervention conditions, demonstrating full reversibility. These findings collectively indicate comparable acute tolerability for both the test and reference formulations of testosterone undecanoate soft capsules when administered postprandially to healthy participants. Detailed AE summaries are provided in Table 4.