Assessment of Pharmacokinetics and Safety with Bioequivalence of the Nitroglycerin Sublingual Tablets of Two Formulations in Chinese Healthy Subjects: A Bioequivalence Study

Abstract

Background and Objective

Nitroglycerin, a cornerstone therapy for acute angina pectoris, achieves rapid symptom relief through sublingual administration by bypassing hepatic first-pass metabolism. This study aimed to investigate the pharmacokinetics (PK), bioequivalence, and safety profiles between a test (T) formulation and a reference (R) formulation of nitroglycerin sublingual tablets in healthy volunteers (HVs).

Methods

In this single-center, randomized, open, single-dose, two-part formulations, four-cycle, two-sequence complete repeat crossover design, fasting-dose bioequivalence study, HVs (n = 36) were 1:1 divided into two groups (T-R-T-R and R-T-R-T) and received 0.6 mg of nitroglycerin sublingual with a 3 d washout. Venous blood was collected 3 h after each administration. Plasma levels of nitroglycerin were analyzed using a high-performance liquid chromatography–tandem mass spectrometry (HPLC-MS/MS) technique, and PK parameters were calculated using noncompartmental methods.

Results

For nitroglycerin, the bioequivalence between the test and reference formulations was assessed using the Reference-Scaled Average Bioequivalence (RSABE) method for the maximum plasma concentration (C_max), area under the curve (AUC) from t = 0 to infinity (AUC_0–∞) and AUC from t = 0 to the final measurable concentration (AUC_0–t). Results showed that the least squares geometric mean ratios (T/R) of C_max, AUC_0–t, and AUC_0–∞ for nitroglycerin (94.62%, 89.92%, and 89.44%, respectively) were in the range of 80.00–125.00%, and the upper limit of unilateral 95% confidence interval (CI) of C_max, AUC_0–t, and AUC_0–∞ for nitroglycerin (−0.06, −0.03, and −0.03, respectively) were less than 0. Safety profiles were comparable between formulations, with no serious adverse events (AE) reported.

Conclusions

The study confirmed bioequivalence between the test and reference formulations, demonstrating equivalent absorption rates (C_max) and extents (AUC). Rapid therapeutic plasma levels were achieved via sublingual administration, aligning with nitroglycerin’s clinical need for prompt angina relief. These findings, combined with favorable safety data, support the test formulation as a therapeutically equivalent alternative.

Key Points

For the highly variable drug nitroglycerin, this study employed the RSABE method to ensure the scientific validity and reliability of the bioequivalence assessment between the test and reference formulations.

Both formulations achieved rapid absorption via sublingual administration and demonstrated consistent safety profiles, which is essential for emergency use in managing acute angina attacks.

Introduction

Nitroglycerin an organic nitrate vasodilator, has served as a cornerstone therapy for acute angina pectoris since its clinical introduction in 1847 [1]. The drug acts quickly by releasing nitric oxide and activating guanylate cyclase, which subsequently promotes smooth muscle relaxation dependent on cyclic guanosine monophosphate [2, 3]. This rapid mechanism facilitates essential coronary vasodilation within minutes of administration. While nitroglycerin’s therapeutic applications extend to hypertensive crises, heart failure, and pulmonary edema [4, 5], its paramount clinical value lies in the emergency management of angina attacks [6, 7], where sublingual administration remains the gold standard owing to unique PK advantages: achieving rapid peak plasma concentration within 3 min of administration [8] and avoidance of hepatic first-pass metabolism [9]. A clinical study on patients with acute angina treated with nitroglycerin also shows that nitroglycerin sublingual formulations have become the core treatment method for acute angina pectoris owing to their rapid onset, simple operation, predictable efficacy, and low risk of tolerance [10].

The growing global burden of cardiovascular diseases has intensified demand for cost-effective nitroglycerin generics. Given this clinical urgency, as a first-line emergency medication for acute angina, nitroglycerin exhibits critical time-dependent therapeutic characteristics [11]. Any variation in C_max or time to C_max (T_max) between generic formulations may potentially delay optimal treatment intervention. Furthermore, the sublingual tablet formulation demonstrates an exceptionally short half-life, necessitating that generic versions achieve both rapid drug release kinetics and complete buccal absorption to ensure therapeutic equivalence [12]. These stringent PK requirements underscore the importance of rigorous bioequivalence evaluation and formulation optimization to prevent compromised clinical efficacy due to formulation discrepancies. Consequently, regulatory agencies mandate that generic products demonstrate equivalence to the reference formulation in both rate (C_max, T_max) and extent (AUC) of exposure [13].

Nitroglycerin is a typical highly variable drug with coefficients of variation (CV%) often exceeding 30% [14–16]. This high variability leads to a significant widening of the CI for PK parameters such as AUC and C_max. To satisfy the standard bioequivalence criteria, which stipulate that the 90% CI for the ratio of geometric means (T versus R) must be within the 80–125% range, studies would need to enroll an excessively large number of subjects (e.g., > 48 subjects) [17]. Such requirements impose substantial challenges, including prohibitively high costs, extended study durations, and increased ethical burdens due to unnecessary over-enrollment. To address these challenges, this study adopts the RSABE method—a regulatory-endorsed strategy that dynamically adjusts equivalence limits on the basis of the variability of the reference formulation. By relaxing equivalence boundaries for highly variable drugs (e.g., scaling the limits according to the within-subject standard deviation of the reference formulation when CV% > 30%), RSABE significantly reduces sample size requirements while ensuring controlled clinical risks [18]. This approach overcomes the limitations of traditional standards, achieving a balance between scientific rigor and practical feasibility in the bioequivalence evaluation of highly variable drugs. It provides crucial technical support for optimizing the development of generic drug products.

This study aimed to establish the bioequivalence between a generic version of 0.6 mg sublingual nitroglycerin tablets [test formulation (T), batch no. 38122005, developed by Neoform Biopharmaceutical Co., Ltd. and manufactured by Sinopharm Zhijun (Shenzhen) Pingshan Pharmaceutical Co., Ltd.] in comparison with the reference formulation (R) (Nitrostat^®, 0.6 mg/ tablet, batch no. DN3989, manufactured by Pfizer Pharmaceuticals LLC, USA) in fasting conditions.

Materials and Methods

Study Design

The bioequivalence study of sublingual nitroglycerin tablets was conducted as a single-center, randomized, open-label, single-dose study. Considering that nitroglycerin is a high variability index drug [14–16], the experiment was designed in a two-sequence, four-cycle (2 × 4) replicated crossover design. The study involved the screening of 109 subjects, of which 73 subjects failed the screening owing to failure of physical examination, poor compliance, voluntary withdrawal, etc., and the remaining 36 subjects were enrolled in the study. All 36 HVs recruited eventually completed the experiment. They were randomly assigned to either the T-R-T-R sequence (n = 18) or the R-T-R-T sequence (n = 18). HVs were administered 0.6 mg nitroglycerin sublingual tablets (T/R), with the caveat that subjects fasted for at least 10 h prior to each dosing cycle and took the nitroglycerin sublingual tablets on an empty stomach according to a randomized schedule on the morning of the day of dosing. The specific mode of administration was as follows: the subject remains seated, before administration, ensure that the hands of the investigator are dry, gently pick up the tablet from the inner package, place it on the bottom of the subject’s tongue, and let it dissolve completely without chewing or swallowing; if saliva is produced during this period, try not to swallow until the drug is dissolved completely naturally. Do not drink water before or within 1 h after administration.

Referring to the 2001 US Food and Drug Administration's Guidance for Industry on Statistical Approaches to Establishing Bioequivalence: Since this test was a single administration, and nitroglycerin is non-endogenous, the washout period was designed to be 3 d, which is much greater than the sevenfold half-life [19], taking into account the effects of metabolizing enzymes and ensuring complete metabolism of the drug. The maximum half-life of nitroglycerin in this trial was 0.17 h for the test formulation and 0.26 h for the reference formulation, which are similar to those previously reported in the literature [20] (Fig. 1).

Fig. 1.

The flow chart of the bioequivalence study

The trial design received approval from the Ethics Committee at Hangzhou Red Cross Hospital in Hangzhou City, China (batch no.: 2023-002-001). Ethical approval for this study adhered to the requirements of Good Clinical Practice, the Declaration of Helsinki, and relevant domestic laws and regulations [21]. All subjects provided signed informed consent forms. The trial had a start date of 30 January 2023 and an end date of 9 March 2023. The trial was conducted at Hangzhou Red Cross Hospital and registered at chinadrugtrials.org.cn (CTR20230221, registered 28 January 2023) and chictr.org.cn (ChiCTR2400083302, registered 19 April 2024).

Inclusion Criteria

The inclusion criteria were as follows: (1) age ≥ 18 years old, both male and female; (2) male subjects weighing ≥ 50.0 kg and female subjects weighing ≥ 45.0 kg with a body mass index between 19 and 26 kg/m² (including borderline values); and (3) subjects voluntarily sign a written informed consent form.

Exclusion Criteria

The exclusion criteria were as follows: (1) history of allergy to drugs, food, or other substances, or an allergy to any of the ingredients in this product; (2) any chronic or serious illness; (3) had a needle or blood sickness in history; (4) physical examination with oral mucosal abnormalities confirmed by the investigator; (5) underwent or plan to have surgery within 30 d, such as traumatic dental procedures; (6) had used any kind of drugs within 2 weeks; (7) had a history of drug abuse or positive urine tests in the past year; (8) had a history of heavy smoking or excessive alcohol consumption in the last 3 months; (9) could not avoid using tobacco, alcohol, caffeinated beverages, or strenuous exercise for 48 h during the trial; (10) had taken part in another clinical trial within the last 3 months; (11) individuals who have had unprotected sex within 14 d prior to the first dose of study drug (females), or females who are pregnant or breastfeeding; and (12) subjects deemed unsuitable for inclusion for other reasons.

In phase I clinical trials, eligible subjects are required to stay at the research center for 24 h before and after drug administration. Subjects must fast for at least 10 h and take either test or reference formulations on the basis of random assignment.

Estimation of Sample Size

Considering that nitroglycerin is a highly variable drug [14–16], in accordance with the latest statistical guidelines for bioequivalence studies and technical guidelines for bioequivalence studies of highly variable drugs issued by the National Medical Products Administration, this trial adopted a design of dual formulations, four cycles, two sequences, and complete repeated crossover, involving pharmacokinetic parameters (AUC, C_max). According to the study design, evaluation indices, comparison types, and related literature data, formula (1) was used and the test level α = 0.05, type II error rate β = 0.15, bioequivalence cut-off value δ = ln(0.7695) adjusted by the reference formulation, point estimation θ = ln(0.90), within-subject coefficient of variation (CV_W%) of the reference formulation was equal to 30.00%, σ² = ln(CV_W² + 1), and the number of subjects needed for the formal trial to satisfy the statistical test of efficacy was calculated to be 26, and taking into account a dropout rate of approximately 25%, the final number of subjects needed for the formal trial was determined to be 36.

$$n\ge 2\times \frac{{\left(t_{\left(1,-,\sigma ,,,n,-,2\right)},+,{\text{t}}_{\left(1,-,\mathrm{\beta },,,n,-,2\right)}\right)}^2\times {\sigma }^2}{{\left(\delta ,-,\theta \right)}^2}\times \frac{1}{2},$$ 1

Blood Sampling

During the bioequivalence study, blood samples were collected from Chinese HVs at specific time points to measure the concentration of nitroglycerin. Venous blood samples (approximately 4 mL) were collected pre-dose (0 h) and at 0.02, 0.03, 0.05, 0.07, 0.08, 0.10, 0.12, 0.13, 0.17, 0.20, 0.23, 0.27, 0.33, 0.42, 0.50, 1.00, 1.50, 2.00, 3.00 h post-dose. After the blood samples were collected, they were placed in pre-numbered sodium heparin anticoagulant negative pressure tubes and gently mixed by inversion. The blood samples were centrifuged at 4 °C within 45 min of collection, using a speed of 2000×g for 5 min at −20 °C (or ≤ 0 °C) in a freezer for pre-freezing. The resulting plasma was then transferred to a clinical center or analytical testing center and stored in a −70 °C freezer (or ≤ 0 °C) for further analysis.

Analytical Method

This study employed a validated analytical method using high-performance liquid chromatography with tandem mass spectrometry (HPLC-MS/MS) to detect the blood concentration of the test formulation. The detection process of biological samples conformed to the requirements of the protocol and relevant standard operating procedure (SOPs). The interference at the peak positions of the analyte and internal standard in the blank matrix was in accordance with the regulations, indicating that the HPLC-MS/MS analytical method had high selectivity. The lower limit of quantification for nitroglycerin was 0.0500 ng/mL, demonstrating good precision and accuracy, with a linear range of 0.0500–10.0 ng/mL. For 1,2-dinitroglycerin, the lower limit of quantification was 0.100 ng/mL, also showing good precision and accuracy, with a linear range of 0.100–20.0 ng/mL. Similarly, the lower limit of quantification for 1,3-dinitroglycerin was 0.0500 ng/mL, exhibiting good precision and accuracy, with a linear range of 0.0500–10.0 ng/mL.

Pharmacokinetic and Statistical Analysis

The PK parameter analysis for the bioequivalence study was conducted using Phoenix WinNonlin 8.2 (Certara, USA) with a noncompartmental model. Statistical analysis was performed using the SAS 9.4 Statistical Package (SAS Institute, Cary, North Carolina).

The PK parameters of nitroglycerin sublingual tablets were assessed, including the time to maximum plasma concentration (T_max), the maximum plasma concentration (C_max), the area under the curve from t = 0 to infinity (AUC_0–∞), the area under the curve from t = 0 to the final measurable concentration (AUC_0–t), the elimination rate constant (λ_z), the half-life (t_1/2), the total apparent body clearance (CL/F), and the apparent volume of distribution (V_d/F). These parameters were reported as the arithmetic mean (mean) and standard deviation (SD).

Bioequivalence Analysis and Evaluation Criteria

For nitroglycerin, 1,2-dinitroglycerin, and 1,3-dinitroglycerin, based on bioequivalence studies, we could select either the RSABE method or the average bioequivalence (ABE) method to evaluate the equivalence between the test formulation and the reference formulation, depending on the PK parameters of the reference formulation [18, 22]. It is worth noting that 1,2-dinitroglycerin and 1,3-dinitroglycerin are active metabolites of nitroglycerin and also require bioequivalence verification [14]. However, this part of the data was only used as supporting evidence and not as the basis for determining bioequivalence.

1. Evaluation of the bioequivalence using the RSABE method:

   The RSABE method includes the following steps:

   1. Calculate the intra-subject standard deviation (S_WR) of the reference formulation:
      Based on the intra-subject variability of the reference formulation, evaluate the primary PK parameters using the RSABE method.
      The S_WR of the primary PK parameters, after subjects take the reference formulation twice, can be calculated using the following formula:

      $${\text{S}}_{\text{WR}}^2=\frac{{\sum }_{\text{i}=1}^m{\sum }_{\text{j}=1}^{\text{i}}{({\text{D}}_{\text{ij}}-\overline{{\text{D}}_{\text{i}·}})}^2}{2(n-m)}.$$ 2
      In the formula, i represents the study sequence number; m is the number of sequences, which is 2 in a fully replicated crossover design; j is the subject number within the sequence; n_i denotes the number of subjects in the ith sequence; ${\text{D}}_{\text{ij}}\left({\text{R}}_{\text{ij}1},-,{\text{R}}_{\text{ij}2}\right)$ represents the difference in PK parameters after logarithmic transformation of two administrations of the reference formulation; $\overline{{\text{D}}_{\text{i}·}}=\frac{{\sum }_{\text{j}=1}^{\text{i}}{\text{D}}_{\text{ij}}}{n_i}$; n represents the total number of subjects in the study. The ${\text{S}}_{\text{WR}}$ values for different PK parameters should be calculated separately.
      The following conversion relationship exists between ${\text{S}}_{\text{WR}}$ and CV_W%

      $${\text{CV}}_{\text{W}}\%=\sqrt{e^{{\text{s}}_{\text{WR}}^2}-1}.$$ 3
      If S_WR ≥ 0.294, i.e., CV_W% ≥ 30%, the RSABE method can be used for equivalence evaluation (applicable to C_max, AUC_0–t, and AUC_0–∞, any one or all of them). If S_WR < 0.294, i.e., CV_W% < 30%, the ABE method should be used to evaluate bioequivalence.
   2. Calculate the upper limit of unilateral 95% CI for the following formula:

      $${{(\overline{Y}}_{\text{T}}-{\overline{Y}}_{\text{R}})}^2-\theta {\text{s}}_{\text{WR}}^2.$$ 4
      Howe’s approximation I was used to determine the upper limit of unilateral 95% CI for $({\overline{Y}}_{\text{T}}-{\overline{Y}}_{\text{R}}{)}^2-\theta {\text{s}}_{\text{WR}}^2$. Where ${\overline{Y}}_{\text{T}}$ and ${\overline{Y}}_{\text{R}}$ represent the mean values of the natural log-transformed main PK parameters (C_max, AUC_0–t, and AUC_0–∞) obtained in the bioequivalence study of the test and reference formulations, respectively.

      $$\theta =({\frac{\text{ln}\left(1.25\right)}{{\sigma }_{\text{w}0}})}^2.$$ 5
      In the formula, ${\sigma }_{\text{w}0}$ is the regulatory constant, which is generally taken as ${\sigma }_{\text{w}0}=0.25$.

2. Evaluation of equivalence using the ABE method:

   The primary PK parameters (C_max, AUC_0–t, and AUC_0–∞) of the test and reference formulations were subjected to natural logarithm transformation and analyzed using a mixed-effects linear model with multivariate analysis of variance. Subsequently, bioequivalence was assessed via two one-sided t-tests and the CI method. The geometric least squares mean ratios (T/R) for these parameters, along with their 90% CI, were calculated to evaluate equivalence between formulations.

The rationale for employing the RSABE approach for nitroglycerin is rooted in its PK characteristics and regulatory recommendations. Nitroglycerin exhibits high within-subject variability (CV_W% ≥ 30%) due to its rapid metabolism, extensive first-pass effect, and variable absorption profiles across administrations. These intrinsic properties result in high within-subject variability of systemic exposure, which challenges the fixed 80–125% equivalence margin of conventional ABE criteria, making them statistically over-restrictive for such highly variable drugs. To address this limitation, RSABE is applied when reference formulation variability exceeds the regulatory threshold, whereas standard ABE is retained for parameters with lower variability [22].

Specifically, the following criteria were used:

1. For parameters (C_max, AUC_0–t, and AUC_0–∞) with CV_W% of the reference formulation ≥ 30%, bioequivalence was concluded if: the least squares geometric mean ratio (T/R) fell within 80.00–125.00%, and the upper limit of unilateral 95% CI for ${{(\overline{Y}}_{\text{T}}-{\overline{Y}}_{\text{R}})}^2-{\theta \text{S}}_{\text{WR}}^2$ was ≤ 0.

2. For parameters with CV_W% of the reference formulation < 30%, bioequivalence required that the 90% CI of the geometric mean ratio (T/R) be entirely within 80.00–125.00%.

Bioequivalence between the test and reference formulations was established only if all three parameters (C_max, AUC_0–t, and AUC_0–∞) met their respective criteria.

Safety Assessment

Throughout the study, all subjects were monitored by a clinical investigator. Firstly, all spontaneously reported and directly observed adverse events (AEs) and serious AEs needed to be documented. Secondly, any clinically significant changes in vital signs needed to be recorded. In addition, any clinically significant abnormalities observed during physical examinations, laboratory tests, and electrocardiogram examinations during the trial also need to be documented [Reference range: 90 mmHg ≤ systolic blood pressure < 140 mmHg, 60 mmHg ≤ diastolic blood pressure < 90 mmHg, 60 beats/min ≤ pulse (resting) ≤ 100 beats/min; at the discretion of the study physician]. During the trial, the study personnel were required to conduct clinical safety assessments of the aforementioned aspects.

Results

Participants

In the bioequivalence study, 36 HVs were enrolled and randomly assigned to the T-R-T-R group or R-T-R-T group. All subjects successfully completed the four study cycles. One subject (DX-2301016-K002, group R-T-R-T) experienced co-administration during the trial and was administered a 0.9% physiological saline mouthwash by the investigator during the initial cycle to relieve gingival swelling and pain. The baseline characteristics and demographic information of all subjects are presented in Table 1.

Table 1.

Baseline demographic characteristics

Variable	Mean (SD)
	T-R-T-R (N = 18)	R-T-R-T (N = 18)	ALL (N = 36)
Sex, n (%)
Male	16 (88.9)	14 (77.8)	30 (83.3)
Female	2 (11.1)	4 (22.2)	6 (16.7)
Age (years)	29 (6)	29 (6)	29 (6)
Hight (cm)	169.1 (8.0)	166.3 (7.4)	167.7 (7.8)
Weight (kg)	61.7 (7.3)	60.4 (5.7)	61.1 (6.5)
BMI (kg/m2)	22 (2)	22 (2)	22 (2)

SD standard deviation, BMI body mass index

Pharmacokinetic and Statistical Results

Following a single sublingual administration of 0.6 mg test and reference formulations, the mean semi-logarithmic plasma concentration–time profiles of nitroglycerin, 1,2-dinitroglycerin, and 1,3-dinitroglycerin are depicted in Fig. 2A–C, respectively. The mean PK parameters (AUC_0–t, AUC_0–∞, C_max, T_max, and t_1/2) and CV% are summarized in Table 2. Concurrently, we will separately list the relevant data of T₁, T₂, R₁, and R₂ to distinguish different experimental groups and research stages, so as to directly compare the differences in key parameters (such as C_max and AUC) of each group, verify bioequivalence, and assess the effects of cycle effects, individual differences or formulation, ensuring the reliability, transparency, and statistical compliance of the results.

Fig. 2.

Average semi-logarithmic plasma drug concentration–time curves for A nitroglycerin, B 1,2-dinitroglycerin, and (C) 1,3-dinitroglycerin following a single 0.6mg sublingual dose of either the test or reference formulation. R reference formulation, T test formulation

Table 2.

Plasma PK parameters of nitroglycerin, 1,2-dinitroglycerin and 1,3-dinitroglycerin after a single 0.6 mg nitroglycerin sublingual tablet of a test or a reference formulation in HVs

	Mean ± SD (CV%)
	T1 (N1 = 36)	T2 (N2 = 36)	R1 (N3 = 36)	R2 (N4 = 36)
Nitroglycerin
Cmax (ng/mL)	6.18 ± 3.81 (61.58)	7.26 ± 4.31 (59.38)	6.06 ± 4.01 (66.23)	8.52 ± 4.32 (50.74)
*Tmax (h)	0.08 (0.05, 0.13)	0.08 (0.05, 0.13)	0.08 (0.06, 0.16)	0.08 (0.05, 0.13)
AUC0–t (h × ng/mL)	0.62 ± 0.53 (84.61)	0.72 ± 0.59 (82.02)	0.63 ± 0.45 (72.21)	0.87 ± 0.55 (63.77)
AUC0–∞ (h × ng/mL)	0.63 ± 0.53 (83.96)	0.73 ± 0.60 (81.72)	0.63 ± 0.45 (71.61)	0.88 ± 0.57 (64.18)
t1/2 (h)	0.06 ± 0.02 (41.11)	0.06 ± 0.03 (49.91)	0.07 ± 0.04 (57.50)	0.06 ± 0.03 (49.32)
1,2-Dinitroglycerin
Cmax (ng/mL)	5.81 ± 1.70 (29.21)	6.05 ± 1.71 (28.17)	5.84 ± 1.71 (29.31)	6.52 ± 1.91 (29.24)
*Tmax (h)	0.16 (0.10, 0.26)	0.13 (0.10, 0.23)	0.16 (0.10, 0.50)	0.13 (0.10, 0.50)
AUC0–t (h × ng/mL)	3.45 ± 0.72 (20.81)	3.50 ± 0.77 (21.98)	3.58 ± 0.70 (19.62)	3.74 ± 0.75 (20.05)
AUC0–∞ (h × ng/mL)	3.61 ± 0.74 (20.60)	3.66 ± 0.80 (21.89)	3.73 ± 0.72 (19.37)	3.91 ± 0.78 (20.13)
t1/2 (h)	0.70 ± 0.13 (18.60)	0.71 ± 0.15 (21.03)	0.70 ± 0.14 (19.53)	0.71 ± 0.12 (16.31)
1,3-Dinitroglycerin
Cmax (ng/mL)	1.85 ± 0.73 (39.48)	2.01 ± 0.79 (39.54)	1.94 ± 0.75 (38.81)	2.08 ± 0.71 (34.09)
*Tmax (h)	0.23 (0.12, 0.50)	0.23 (0.11, 0.50)	0.23 (0.10, 0.50)	0.20 (0.11, 0.50)
AUC0–t (h × ng/mL)	1.32 ± 0.56 (42.46)	1.37 ± 0.62 (45.02)	1.38 ± 0.51 (36.54)	1.44 ± 0.54 (37.26)
AUC0–∞ (h × ng/mL)	1.42 ± 0.61 (42.66)	1.47 ± 0.68 (46.00)	1.48 ± 0.55 (37.19)	1.55 ± 0.61 (39.13)
t1/2 (h)	0.66 ± 0.16 (24.14)	0.67 ± 0.19 (28.80)	0.67 ± 0.19 (28.05)	0.71 ± 0.21 (29.25)

*T_max is expressed as the median (minimum, maximum)

This study comprehensively analyzed the PK parameters of nitroglycerin and its metabolites, 1,2-dinitroglycerin and 1,3-dinitroglycerin. The parent drug nitroglycerin exhibited marked variability (C_max CV%: 50.74–66.23%; AUC CV%: 63.77–84.61%), consistent with the characteristics of a highly variable drug. Therefore, for nitroglycerin, in this study, the RSABE method was used to assess the bioequivalence. The t_1/2 and T_max data of nitroglycerin supported its rapid metabolism and absorption kinetics, aligned with the clinical profile of sublingual formulations. Metabolite analysis revealed low variability for 1,2-dinitroglycerin (C_max CV% ≈ 29%; AUC CV% ≈ 20%). For 1,3-dinitroglycerin, the CV% for AUC_0–t and AUC_0–∞ ranged from 36.54 to 46.00%, higher than those of 1,2-dinitroglycerin (CV% ≈ 20%) but substantially lower than the parent drug (CV% > 60%), indicating relatively stable metabolic processes. No significant formulation-related differences were observed in any metabolite parameters, further validating the consistency of the parent drug’s metabolic pathway. These findings conclusively demonstrate the bioequivalence between the test and reference formulations in terms of absorption extent, metabolic pathways, and overall systemic exposure.

Bioequivalence Analysis

The bioequivalence evaluation of the main PK parameters between the test formulation and the reference formulation can be found in Table 3. Since the S_WR of C_max, AUC_0–t, and AUC_0–∞ of nitroglycerin (0.39, 0.34, and 0.34, respectively) exceeded 0.294, and the CV_W% of nitroglycerin (40.53, 35.07, and 34.81, respectively) exceeded 30%, we therefore used the RSABE method in this trial to evaluate its bioequivalence [18]. The least squares geometric mean ratios (T/R) of C_max, AUC_0–t, and AUC_0–∞ for nitroglycerin (94.62%, 89.92%, and 89.44%, respectively) were in the range of 80.00–125.00%, and the upper limit of unilateral 95% CI of C_max, AUC_0–t, and AUC_0–∞ for nitroglycerin (−0.06, −0.03, and −0.03, respectively) were less than 0. Therefore, the two formulations are equivalent in this PK parameter. Since the S_WR of C_max, AUC_0–t, and AUC_0–∞ of 1,2-dinitroglycerin (0.22, 0.11, and 0.10, respectively) and 1,3-dinitroglycerin (0.22, 0.11, and 0.10, respectively) were less than 0.294, and the CV_W% of C_max, AUC_0–t, and AUC_0–∞ of 1,2-dinitroglycerin (22.23, 10.99, and 10.47, respectively) and 1,3-dinitroglycerin (21.90, 10.66, and 10.32, respectively) were less than 30%, this trial used the ABE method to evaluate their bioequivalence. All 90% CI of the least squares geometric mean ratios (T/R) for C_max, AUC_0–t, and AUC_0–∞ of 1,2-dinitroglycerin (89.86–102.53, 91.98–98.06, and 92.31–98.01, respectively) and 1,3-dinitroglycerin (88.32–100.57, 91.72–98.09, and 91.94–97.90, respectively) were within the bioequivalence bounds (80.00–125.00%). In conclusion, the test formulation of nitroglycerin was equivalent to the reference formulation in terms of C_max, AUC_0–t and AUC_0–∞ parameters, indicating bioequivalence of the two formulations.

Table 3.

Evaluation of the bioequivalence of the main PK parameters of nitroglycerin, 1,2-dinitroglycerin, and 1,3-dinitroglycerin after a single 0.6 mg nitroglycerin sublingual tablet of a test or a reference formulation in HVs

Parameter	Least squares geometric means and ratios				S2WR	SWR	CVW% of R	Upper limit of unilateral 95% CI	Methodology
	T (NT)	R (NR)	T/R (%)	90% CI of the T/R ratio (%)
Nitroglycerin
Cmax (ng/mL)	5.42 (36)	5.74 (36)	94.62	80.66–110.99	0.15	0.39	40.53	−0.06	RSABE
AUC0–t (h × ng/mL)	0.52 (36)	0.58 (36)	89.92	78.97–102.40	0.12	0.34	35.07	−0.03	RSABE
AUC0–∞ (h × ng/mL)	0.53 (36)	0.58 (36)	89.44	78.49–101.91	0.11	0.34	34.81	−0.03	RSABE
1,2-Dinitroglycerin
Cmax (ng/mL)	5.62 (36)	5.86 (36)	95.99	89.86–102.53	0.05	0.22	22.23	–	ABE
AUC0–t (h × ng/mL)	3.39 (36)	3.57 (36)	94.97	91.98–98.06	0.01	0.11	10.99	–	ABE
AUC0–∞ (h × ng/mL)	3.54 (36)	3.73 (36)	95.12	92.31–98.01	0.01	0.10	10.47	–	ABE
1,3-Dinitroglycerin
Cmax (ng/mL)	1.76 (36)	1.87 (36)	94.25	88.32–100.57	0.05	0.22	21.90	–	ABE
AUC0–t (h × ng/mL)	1.24 (36)	1.31 (36)	94.85	91.72–98.09	0.01	0.11	10.66	–	ABE
AUC0–∞ (h × ng/mL)	1.34 (36)	1.41 (36)	94.87	91.94–97.90	0.01	0.10	10.32	–	ABE

T test formulation, R reference formulation, CI confidence interval, CV_W% within-subject coefficient of variation

Safety Analysis

Thirty-six HVs were included in the safety analysis after receiving the nitroglycerin sublingual tablets. Among the 36 subjects who took the test formulation, 26 of 36 subjects had 73 AEs (72.20%, 26/36). Moreover, 23 of 36 subjects had 58 adverse reactions (63.90%, 23/36). In the 36 subjects who took the test formulation, the severity of the AEs was grade 1. Among the 36 subjects who took the reference formulation, 25 of 36 subjects had 84 AEs (69.40%, 25/36). In the bioequivalence study, 82 AEs were classified as grade 1 severity, while only 2 were classified as grade 2 severity. In addition, 21 out of 36 subjects experienced a total of 73 adverse reactions (58.30%, 21/36), all of which were categorized as grade 1 severity. No serious AEs or withdrawals due to AEs were reported. The relevant data can be found in Table 4.

Table 4.

Adverse events in the study

SOP/PT	Grade	T (N = 36)		R (N = 36)
		R (%)	NR (%)	R (%)	NR (%)
Laboratory examination
ALT increased	1	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	1.0 (2.8)
LDL increased	1	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	1.0 (2.8)
ERY positive	1	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)	0.0 (0.0)
BLD positive	1	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)	0.0 (0.0)
Body temperature increased	1	0.0 (0.0)	3.0 (8.3)	0.0 (0.0)	3.0 (8.3)
Pulse rate decreased	1	6.0 (16.7)	0.0 (0.0)	5.0 (13.9)	0.0 (0.0)
Pulse rate increased	1	0.0 (0.0)	3.0 (8.3)	0.0 (0.0)	2.0 (5.6)
TG increased	1	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)	0.0 (0.0)
HB increased	1	0.0 (0.0)	0.0 (0.0)	0 (0.0)	1.0 (2.8)
Potassium decreased	1	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)	0.0 (0.0)
BP increased	1	17.0 (47.2)	0.0 (0.0)	17.0 (47.2)	0.0 (0.0)
BP decreased	1	0.0 (0.0)	2.0 (5.6)	0.0 (0.0)	1.0 (2.8)
Various neurological reactions
Hyperalgesia	1	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)	0.0 (0.0)
Head discomfort	1	2.0 (5.6)	0.0 (0.0)	3.0 (8.3)	0.0 (0.0)
Had a headache	1	2.0 (5.6)	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)
Spin	1	5.0 (13.9)	0.0 (0.0)	8.0 (22.2)	0.0 (0.0)
Various musculoskeletal and connective tissue diseases
Back pain	1	0.0 (0.0)	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)
Diseases of the skin and subcutaneous tissue
Perspiring	1	0.0 (0.0)	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)
Frontal sweating	1	0.0 (0.0)	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)
Sweaty hands	1	1.0 (2.8)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)
Systemic diseases and various reactions at the site of administration
Fatigue	1	1.0 (2.8)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)
Chest distress	1	1.0 (2.8)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)
Chest pain	1	0.0 (0.0)	0.0 (0.0)	1.0 (2.8)	0.0 (0.0)
Diseases of the gastrointestinal system
Nausea	1	1.0 (2.8)	0.0 (0.0)	2.0 (5.6)	0.0 (0.0)
Vomiting	1	1.0 (2.8)	0 (0.0)	1.0 (2.8)	0.0 (0.0)
Gum pain	2	0.0 (0.0)	0 (0.0)	0.0 (0.0)	1.0 (2.8)
Sore gums	2	0.0 (0.0)	0 (0.0)	0.0 (0.0)	1.0 (2.8)
Diseases of the heart organs
Tachycardia	1	1.0 (2.8)	0 (0.0)	4.0 (11.1)	0.0 (0.0)
Vascular and lymphatic vessel diseases
Pallor	1	0.0 (0.0)	0 (0.0)	1.0 (2.8)	0.0 (0.0)
Hot flashes on the face	1	1.0 (2.8)	0 (0.0)	1.0 (2.8)	0.0 (0.0)

Grade 1: asymptomatic or mild, there is no treatment; grade 2: moderate, minor, local, or noninvasive treatment is required; grade 3: serious, but not immediately life-threatening, hospitalization or prolonged hospitalization is required; grade 4: life-threatening, urgent treatment is required; grade 5: deaths associated with AEs. NR indicates not relevant (with probably not relevant, definitely not relevant). Adverse medical events that occurred prior to the first dose of the trial drug were not included in the descriptive statistical analysis of adverse events and are listed only

HVs underwent comprehensive clinical laboratory assessments during the screening period and post-dose administration, including hematology, urinalysis, blood biochemistry, coagulation panel, and serum pregnancy testing (female participants only). Clinically observed laboratory abnormalities, as detailed in Table 5, primarily consisted of transient isolated deviations: a single hemoglobin reduction in routine blood test, occult blood positivity with elevated red blood cells in urinalysis, and mild triglyceride/alanine aminotransferase elevations in blood biochemistry. Further causality assessment classified all abnormalities as probably unrelated to the investigational product. Importantly, no severe, persistent, or clinically actionable laboratory abnormalities requiring medical intervention were documented throughout the study.

Table 5.

Statistical summary of laboratory abnormalities

Test category	Abnormal indicator	Occurrences	Related AE and investigational product association	AE outcome
Routine blood	Hemoglobin decreased	1	Probably unrelated	No change
Urinalysis	Urine test positive for hematuria	1	Probably unrelated	Unknown
	Urine red blood cell elevation	1	Probably unrelated	Unknown
Blood lipid	Triglycerides are elevated	1	Probably unrelated	Recovered
	High levels of low-density lipoprotein	1	Probably unrelated	Unknown
Serum electrolytes	Potassium is reduced	1	Probably unrelated	Unknown
Liver/kidney function	Blood glutamic transaminase increased	1	Probably unrelated	Recovered

AE adverse event

Discussion

This study demonstrated the bioequivalence between a generic 0.6 mg nitroglycerin sublingual tablet and the reference drug Nitrostat^® under fasting conditions, employing a replicated crossover design combined with the RSABE approach. The test formulation exhibited favorable tolerability, with no reports of deaths or serious AEs, as detailed in Table 4. Comprehensive safety assessments, including physical examinations, laboratory tests, and electrocardiograms, revealed no significant abnormalities throughout the study period.

In bioequivalence studies of nitroglycerin, both the parent drug (nitroglycerin) and its major active metabolites (e.g., 1,2-dinitroglycerin and 1,3-dinitroglycerin) are typically analyzed. The equivalence conclusion is primarily derived from the parent drug data, and this study employed the RSABE method. For nitroglycerin, the least squares geometric mean ratios (T/R) of C_max, AUC_0–t, and AUC_0–∞ fell within the 80.00–125.00% range, with the upper limits of the unilateral 95% CI for C_max, AUC_0–t, and AUC_0–∞  below 0. These results confirm that the two formulations are bioequivalent in terms of these PK parameters. This trial employed the ABE method to assess metabolite bioequivalence. For 1,2-dinitroglycerin and 1,3-dinitroglycerin, the 90% CIs of the least squares geometric mean ratios (T/R) for C_max, AUC_0–t, and AUC_0–∞ all fell within the bioequivalence range (80.00–125.00%). These findings strengthen the credibility of bioequivalence between the two formulations. The rationale for including active metabolites lies in nitroglycerin’s rapid metabolism. Although metabolite data can indirectly reflect the absorption and metabolism processes of the drug, based on the direct pharmacological activity of the parent drug and the regulatory priority for parent drug data, the conclusions of bioequivalence should still primarily rely on parent drug data. Metabolite analyses remain supplementary. This dual approach balances scientific rigor with regulatory compliance, ensuring that bioequivalence assessments reflect direct pharmacological significance while meeting the statistical reliability requirements needed for approval.

Considering the clinical importance of nitroglycerin and the various advantages of its sublingual formulations, such as rapid absorption and bypassing the first-pass effect, it is essential to develop an innovative sublingual nitroglycerin tablet to meet market demand. Andrade emphasized that demonstrating bioequivalence between generic and branded drugs is a prerequisite for regulatory approval [13]. This study, for the first time, directly quantified the elimination half-life (t_1/2 = 0.06 h) of sublingual nitroglycerin using a highly sensitive detection method. The observed T_max (0.08 h) aligns with literature reports of “T_max < 5 minutes” [8, 20]. Although prior studies did not directly measure the half-life, the undetectable plasma concentrations after 16 min can be logically explained by the half-life data from this study, as concentrations fall below the detection limit after four half-lives. These findings collectively confirm that the core PK characteristics of sublingual nitroglycerin include extremely rapid absorption and exceptionally short elimination time, which underpin its clinically observed fast onset and short duration of action. Notably, in HVs, nitroglycerin’s key PK parameters exhibit high inter-individual variability, regardless of whether administered transdermally [15] or sublingually. To address this, the study employed an innovative 2 × 4 replicated crossover design and the RSABE approach, providing a novel solution for bioequivalence assessment of highly variable drugs. The research further validated the inherent high variability of nitroglycerin’s PK parameters (S_WR > 0.294 for C_max, AUC_0–t, and AUC_0–∞), consistent with variability indices previously reported in similar literature.

Although we recognize the inherent limitations of extrapolating data from HVs to patient populations, such as the potential impact of pathophysiological differences on sublingual absorption efficiency or PK, this study demonstrates through systematic comparative analysis that sublingual nitroglycerin possesses PK characteristics, including rapid absorption and an extremely short elimination half-life. These characteristics exhibit high consistency in HVs and align with key observations previously reported in literature regarding patients, such as onset time and duration of action [8, 20]. This consistency suggests that the PK behavior of nitroglycerin is primarily driven by its physicochemical properties and formulation design, rather than relying solely on the physiological state of the subjects. Furthermore, the 2 × 4 replicated crossover design and RSABE methodology employed in this study enhance the robustness of PK parameter estimation by rigorously controlling high inter-individual variability. These methodological innovations provide a feasible framework for future validation trials in patient populations and lay the groundwork for further clinical research.

This study quantifies the pharmacokinetic characteristics of sublingual nitroglycerin tablets using advanced analytical methods to enhance the accuracy of bioequivalence assessments. At the same time, the RSABE method is employed to evaluate the bioequivalence of nitroglycerin. This research provides a more reliable basis for the bioequivalence assessment of nitroglycerin and high-variability drugs. Given the increasing global burden of cardiovascular diseases, this optimized bioequivalence strategy is expected to facilitate the smooth market entry of cost-effective generic drugs.

Conclusions

This study successfully validated the bioequivalence of the highly variable drug nitroglycerin as sublingual tablets (0.6 mg) in Chinese HVs using the RSABE method, demonstrating that the key PK parameters of the test formulation meet regulatory standards compared with the reference formulation. Its scientific significance lies in overcoming the limitations of conventional ABE approaches, establishing a research paradigm applicable to emergency medications with short half-lives and high variability. Regarding regulatory approval, this study provides empirical evidence for accelerating the assessment of generic drugs. From the perspective of public health, the post-marketing supply of generic drugs is expected to significantly enhance drug affordability through price reduction, and the integration with healthcare policies can expand treatment opportunities for underserved cardiovascular populations, thereby reducing the risk of death from acute events.

Declarations

Funding

This work was supported by the Zhejiang Province Leading Geese Plan (2024C03099) and the Key R&D Program of Hangzhou Science and Technology Bureau (20241203A17).

Conflicts of Interest

The authors affirm that there are no potential conflicts of interest.

Ethics Approval

This study was approved by the Medical Ethics Review Committee of Hangzhou Red Cross Hospital (approval number: 2023-002-001).

Consent to Participate

All participants in this study voluntarily took part after fully understanding the purpose, process, and potential risks. Before the study began, they signed an informed consent form confirming their understanding of the research, its risks and benefits, and their rights, including the right to withdraw at any time. Participants’ information will be kept confidential, and data will be analyzed anonymously to protect their privacy. This study follows ethical guidelines and has been approved by the ethics committee.

Consent for Publication

All authors have consented to the publication of this manuscript. The authors confirm that the manuscript has not been published elsewhere and is not under consideration by any other publication.

Availability of Data and Material

The data that support the findings of this study are available from Neoform Biopharmaceutical Co., Ltd. However, restrictions apply to the availability of these data, as they were used under license for the current study and are not publicly available. Nonetheless, data can be obtained from the corresponding author upon reasonable request and with permission from Neoform Biopharmaceutical Co., Ltd.

Code Availability

Not applicable.

Author Contributions

Qinjiao Fu wrote the manuscript; Ying Wang, Yingying Xu, and Chunqi Huang designed the research; Yuan Yuan, Yu Wang, and Bin Zhu performed the research; Yanzhu Liu and Fang Tian analyzed the data; Xiufeng Xu and Lei Yang contributed new reagents/analytical tools.