The study was conducted ethically in accordance with the principles of the ICMR guidelines (2017) [11] , New Drugs and Clinical Trials Rules 2019 India [12] and adhered to the ethical principles of the Declaration of Helsinki [13] the International Conference on Harmonization Good Clinical Practice Guidelines [14] . The study protocol (N˚096-22) was approved by Aavishkar Ethics Committee, on March 06, 2023 (Version: 00, Dated 06 Dec 2022) and certified by CDSCO/DGHS to ICBio Clinical Research Pvt, Ltd.

This was an open-label, randomized, two-treatment, two-period, two-sequence, single oral dose, and crossover bioequivalence (BE) study under fasting conditions comparing two formulations of Metformin (metformin hydrochloride) 1000 mg tablets extended release (XR), batch N˚18E, date of expiry 04/2023, of Laboratorios Leti S.A.V., as test formulation (T) vs Glugophage® (metformin hydrochloride) 1000 mg tablets XR, batch N˚BR131581, date of expiry 03/2024, of Merck Serono, GmbH, Germany as reference formulation (R).

The subjects were randomized, to one of the two sequences (T - R) or (R - T). The randomization schedule was generated using Statistical Analysis Software (SAS®, version 9.4, Institute. Inc., CARY, USA). One single dose was administered in each period. Subjects who received T product in period I were administered R product in period II and vice versa. Pre-screening period was 21 days. The total duration of the clinical phase of this study was of 12 days from the day of check-in of period I to last blood sample collection of period II (April 27, 2022-May 08, 2022) separated by a washout period of 07 days, considering the terminal half-life for metformin 6.5 h [6] . This BE study met the principles described in bioequivalence EU BE guideline [9] [10] .

The sample size calculation for the study was based on intra-subject coefficient of variation (CV%) obtained of published data for metformin C_max: 11.4%, and ${\text{AUC}}_{0-t}$ 13.63%) [3] [6] . Based on referred literature using the reported 90% confidence intervals, intrasubject CV% was estimated for pharmacokinetics parameters of metformin hydrochloride and the maximum intrasubject CV% with the expected % CV not exceeding 20% and the ratio within 80 and 125% [3] [6] . The study required 22 evaluable subjects to demonstrate BE with a power of ≥90% at 5% level of significance. Thus, with the expected coefficients of variation for C_max and AUC not exceeding 24% and assuming the true ratio falling within 95% to 105% (i.e. a true treatment difference of 5%), the study had at least 22 evaluable subjects to show the bioequivalence with a power of greater than 80% at 5% level of significance. Further, expecting certain dropouts and/ or withdrawals in the study, 04 additional subjects were included [15] - [17] .

All volunteers underwent a screening procedure. A total of 26 healthy, adult male volunteers who met the inclusion and exclusion criteria were enrolled, with a mean age of 33.96 years, mean weight 70.88 Kg, mean height 1.73 cm, and body mass index (BMI) of 23.66 kg/m² ( Table 1 ).

A complete clinical history valid for 6 months before the start of the study; normal laboratory values as determined by medical history and physical examination at the time of screening; normal vital signs and physical examination; creatinine clearance of more than 50 mL/min; negative tests for hepatic transaminases, hepatitis B and C, human immunodeficiency virus, and venereal diseases research laboratory; and normal 12-lead EKG values, normal chest radiography, and negative result in urine drug tests. Urine for drugs of abuse and urine test for alcohol consumption were performed on day of check in of each period. Random blood glucose test was performed in screening. Fasting blood glucose test was conducted prior to check-in of period I. In each period, subject’s blood glucose monitoring was performed before dosing and at 02.00, 04.00, 06.00 and 11.00 h ± (30 minutes) after dosing. Other key inclusion criterion was that subjects must be non-smokers or smokers who had not smoked at least 10 h before the start of the study. They all signed the informed consent.

The exclusion criteria included volunteers incapable of understanding the informed consent, history of diabetes, tuberculosis and systemic hypertension. A history of hypersensitivity to the study medication or to any other medication belonging to the study group or cardiovascular, renal, hepatic, metabolic, gastrointestinal, neurological, endocrine, hematopoietic, psychiatric, or other organic abnormalities; under medication that interferes with the quantification, drugs that can potentially affect the hepatic metabolism of other drugs.

The subjects were admitted to the facility one night before study. All the subjects were fasted for at least 10hours pre-dose and 4hours post-dose. The subjects received standardized meals at 04.00, 08.00, 12.00 and 24.00 hours after dosing in each Period. During housing the meal menu was same in both the periods and drinking water was provided ad libitum (As desired).

The subjects received standard meals at 04.00 (lunch), 08.00 (snacks) and 12.00 (dinner) hours after dosing in each period. All meal plans were identical for each period of study.

A single oral dose (metformin 1000 mg tablets XR) either one T or R, following randomization schedule and was administered with 240 mL ± 2 mL of glucose solution in water, followed by 60 mL of 20% glucose solution in water administered every 15 minutes for up to 04 hours after dosing at ambient temperature to each subject in sitting position [8] - [10] .

A total of 19 × 5 ml of venous blood samples were collected through cannula from each subject during the two periods of the study, withdrawn at pre-dose (00.00 h) and 00.50, 01.00, 01.50, 02.00, 02.50, 03.00, 03.50, 04.00, 05.00, 06.00, 07.00, 08.00, 10.00, 12.00, 16.00, 24.00, 48.00 and 72.00 hours after dosing in each period. While 24.00, 48.00 and 72.00 hours, post dose, blood samples were collected by direct venipuncture. The subjects received standardized meals at 04.00, 08.00, 12.00 and 24.00 hours after dose in each period (800 - 1000 Kcal) and drinking water was provided ad libitum.

Venous blood samples were collected in pre-labelled K₂EDTA (ethylenediaminetetraacetic acid) vacutainers and were centrifuged at 4000 rpm for 10 min at 2˚C - 8˚C within 45 minutes of sample collection. Plasma was separated, labeled and stored at −70˚C ± 15˚C before analysis. Subsequently, the plasma samples were processed, calibration curve of internal standards (ISTD) metformin D6, (Vivian Life Sciences Private Limited, Mumbai, India) and quality control (QC) samples were thawed and vortexed for preparation and analysis. Method of validation of metformin was conducted with calibration 14.040 ng/mL to 4004.046 ng/mL respectively and validation results were reported in Method Validation Report MVR-SOP MV-073-00. The biostudy was performed on Protein Precipitation Method coupled with LC using metformin D6 as internal standard. The interface used was turbo iron spray and positive ions were measured in MRM mode. The samples were extracted by a solid phase extraction chromatography column coupled to a BDS Hypersil C18, 4.6 × 50 mm, 5 μm column. Elution was performed at 25˚C temperature, with a mobile phase of acetonitrile and HPLC water mixture (50:50% v/v). The lower limit of quantification was 10.495 ng/mL whereas the upper limit of quantification was 3053.826 ng/mL. These data were acquired, integrated and quantified on ABSCIEX systems Shimadzu, analyst version 1.7.3 software.

All solutions were prepared using diluent (metanol: water, 50:50 v/v) and stored at 2.0˚C - 8.0˚C. The supernatant identifications were based on multiple reactions monitoring transitions; m/z 60.200 - 130.050 for metformin and m/z 60.200 - 136.100 for the IS-metformin D6. The inter-batch calibration standard (IBCS) for metformin was 3.68% to 5.04%, accuracy 97.61% to 104.50%.

Statistical analysis was conducted on all of the subjects who complete both periods of the study as per protocol, using SAS® (software version 9.4, Institute. Inc., CARY, USA).

The blood concentration-time curve (C-T curve) was obtained by mapping the plasma metformin concentration at each time point. The primary PK variables evaluated were maximum peak concentration (C_max), area under curve from time 0 to last measurable concentration ( ${\text{AUC}}_{0-t}$) and area under the plasma concentration-time curve from time 0 to infinity ( ${\text{AUC}}_{0\to \infty }$) The peak concentration (C_max) was obtained from this graph. The area under the curve (AUC) from time 0 to the time of the last measurable concentration ( ${\text{AUC}}_{0-t}$) was calculated using the linear trapezoidal rule. The AUC from time 0 to infinity ( ${\text{AUC}}_{0\to \propto }$) was obtained using the formula ${\text{AUC}}_{0\to \propto }={\text{AUC}}_{0-t}+C_t/{\lambda }_z$, where C_t is the final measurable concentration and λ_z is the elimination rate constant, which is the slope of the terminal segment of the linear regression curve of the logarithmically transformed drug concentration values vs. Time. Others secondary PK parameters evaluated were: T_max (time to reach C_max) and was obtained directly from the C-T curve_, time required for plasma concentration to decrease by 50% (T_1/2) T_1/2 is the elimination half-life, calculated as 0.693/λ_z. Log-transformed, AUC__%Extrap, constant of elimination (K_el) and the number of points (N_points) of the terminal log-linear phase used to estimate the terminal rate constant. The natural log transformed (i.e., Ln-transformed) values for the pharmacokinetic parameters C_max ${\text{AUC}}_{0-t}$ and ${\text{AUC}}_{0\to \infty }$) was analyzed for statistical difference between test and reference formulations (metformin T & R) with ANOVA by a Generalized Linear Model (GLM) ANOVA using SAS®. Based on these parameters, the 90% confidence intervals (CIs) were constructed for the least square mean differences of log-transformed PK parameters C_max ${\text{AUC}}_{0-t}$ and ${\text{AUC}}_{0\to \infty }$. The metformin formulations regarded as bioequivalent when the 90% (CIs) of the T and R ratio of C_max ${\text{AUC}}_{0-t}$, and ${\text{AUC}}_{0\to \infty }$ ranged from 80% to 125% [8] [9] .

Safety assessments were performed during screening, during the study and the end of the study and the adverse events (AEs) were monitored throughout the study. Vital signs were measured during baseline screening, and at the conclusion of the study. Twelve-lead electrocardiogram was recorded during screening. Random blood glucose test was performed in screening. Fasting blood glucose test (FBGT) was conducted prior to check in of period I. In each period, blood glucose monitoring was performed before dosing and at 03.00 and 06.00 h (± 15 minutes) after dosing. Post study safety evaluation was performed on all the subjects who came for period II of clinical study. Total 25 subject’s post safety samples were collected on last ambulatory visit (72.000 hr) and subject N˚ 19 reported as dropout in period II.

A total of 25 subjects completed the study and were included in the PK and statistics evaluation. A non-compartmental analysis was applied for the estimation of PK parameters C_max, ${\text{AUC}}_{0-t}$, ${\text{AUC}}_{0\to \infty }$, T_max, K_el (h − 1), and T_1/2, of metformin in plasma concentration which are presented in Table 2 .

Data presented as a mean ± standard deviation C_max, maximum concentration, ${\text{AUC}}_{0-t}$, area under the plasma concentration-time curve from time 0 to the last measurable concentration; ${\text{AUC}}_{0\to \infty }$, area under the plasma concentration-time curve from time 0 to infinity, T_max time to reach C_max, K_el elimination constant. T_1/2 time required for plasma concentration to decrease by 50%. Median (range).

The AUC from 0 to last time point with measurable plasma concentrations was computed using linear trapezoidal-rule.

Analysis of variance analysis from Ln C_max, ${\text{AUC}}_{0-t}$ and ${\text{AUC}}_{0\to \infty }$, there were no statistically significant differences between the PK parameters of the metformin formulations (P > 0.05) ( Table 2 and Table 3 ).

Least Square Mean (LSM), Geometric Least Square Mean (GLSM), Ratio, 90% confidence intervals (CI), Intra subject Variability (ISCV) and Power for the Log transformed C_max and ${\text{AUC}}_{0-t}$. Metformin 1000 mg extended release (ER) of Laboratorios Leti S.A.V. as Test formulation and Glugofage® (metformin 1000 mg ER) of Merck Serono as Reference formulation.

The geometric mean ratios for primary PK parameters were: Ln C_max 92.94 (90% IC: 89.08 - 96.97), Ln ${\text{AUC}}_{0-t}$ 92.30 (90% IC: 96.57 - 108.82) and Ln ${\text{AUC}}_{0\to \infty }$ 92.30 (90% IC: 96.57 - 108.82 ( Table 3 ), intervals for test and reference formulations were within the acceptance limits (80.00% - 125.00%) to establish bioequivalence.

The oral dosing for 48 h post-dose is represented on arithmetic and logarithm scales, test and reference formulations as shown in Table 4 ( Figure 1 and Figure 2 ) for metformin 1000 mg XR.

Post study safety evaluation was performed on all subjects who came for period II of the study. Total 25 subjects post safety samples were collected on last ambulatory visit (72.000 hr) in period II and subject N˚ 19 reported as dropout in period II. On evaluation all the safety parameters were found to be normal.

There were no serious adverse events (SAE) and other significant adverse events observed during the study. Hence the test product and reference product were found to be safe and well tolerated upon single dose administration in healthy male adults under fasting conditions.