A Randomized, Two-Treatments, Two-Periods, Crossover, Open label, Laboratory-Blind, Single Dose Bioequivalence Study between Vildagliptin/Metformin 50 mg/1000 mg Film Coated Tablets (Sensityn^®) and Galvusmet^® 50 mg/1000 mg Film Coated Tablets in healthy adults under fed conditions

This was a two-treatment, two-period, crossover, open-label, laboratory-blind, single-dose, bioequivalence study conducted in accordance with the ethical principles that have their origin in the Declaration of Helsinki (Jordan Food and Drug Administration unique trial number: 126/BIO/20). The study protocol was prepared in line with the requirements set in the GCC Guidelines for Bioequivalence (May 2011)²⁰ and the EMA CPMP (CPMP/EWP/QWP/1401/98 Rev. 1/Corr**) Guideline on investigation of bioequivalence.²¹

The registration trial number is ISRCTN18152519. The study was conducted according to GCP, GLP guidelines, as well as JFDA clinical studies Law no. 02 for the year 2011. Dosing was determined according to a predefined randomisation plan prepared as per Triumpharma SOP no.: PHU-1104. Subjects were randomised to receive a single dose of Sensityn^® in one period and Galvusmet^® in the other period. A washout period of 7 days separated the two doses. In sequence 1, subjects received Sensityn^® or Galvusmet^® followed by 7 days of washout period. In sequence 2, subjects were crossed over to either Galvusmet^® or Sensityn^®, depending on what they received in sequence 1.

Each period consisted of 3 days, including the following: -

Day 1: admission of subjects to the clinical site (approximately 16 h before study drug administration)

The subjects were 36 healthy male adults selected from the Jordanian population. For inclusion in the study, subjects had to be physically and mentally healthy, of age 18–50 years, with a body weight of at least 50 kg and body mass index (BMI) range of 18.5–30 kg/m² at screening. The findings of the subject at screening were within the range of clinical acceptability in medical history, physical examination and laboratory results ‘other than RBC indices (MCH, MCV, and MCHC) and haemoglobin’ within ‘normal ranges’ for the laboratory tests performed or abnormalities considered insignificant and justified by the principal/clinical sub-investigator. Results for all red blood cell indices and haemoglobin within the normal range or ±5% of the medical laboratory reference range at screening, a normal electrocardiogram (ECG) at screening, HbA1c within the normal range at screening, kidney function tests (serum creatinine, potassium and sodium) within the medical laboratory reference range, and urea, uric acid, chloride and calcium within the normal reference range were considered clinically acceptable according to the PI/SI judgement at screening. Liver function test (AST, ALT, GGT and total bilirubin) results and alkaline phosphatase, total protein and albumin within the medical laboratory reference ranges were considered clinically acceptable according to the PI/SI judgement at screening. Vital signs within the following ranges: systolic blood pressure: 90–140 mmHg, diastolic blood pressure: 60–90 mmHg, body temperature: 36.1°C–37.2°C, and pulse rate: 60–100 beats per minute at screening and on admission day (before admission) period I. Subjects were excluded based on the identification of significant diseases or clinically abnormal findings during screening and/or admission period I. Furthermore, they were excluded if they had the following conditions: difficulty in swelling; clinically significant history or presence of any clinically significant pathology (e.g. chronic diarrhoea), unresolved gastrointestinal symptoms, or other conditions known to interfere with drug absorption, metabolism or excretion of the drug; known allergy to the drug under investigation (vildagliptin/metformin) or to any ingredient; history of hypersensitivity to heparin; major illness within 1 month preceding screening; acute infection within 1 week before the first dosing; history of acute pancreatitis; glucose level below 70 mg/dl at screening or before the first dosing. Subjects who were receiving cationic medicinal products that are eliminated by renal tubular secretion, glucocorticoids, beta-2-agonists, diuretics or ACE inhibitors were also excluded. Those suffering from acute or chronic disease that can cause tissue hypoxia, such as cardiac or respiratory failure, recent myocardial infarction, shock and those suffering from coronavirus disease 2019 (COVID-19) or having a history of COVID-19 infection and being treated for COVID-19 or having the signs and symptoms of COVID-19 were excluded. Those with a history of or currently having compulsive habit of drinking alcohol or with regular exposure to other substances of abuse, those with a positive test for illicit drugs or alcohol, moderate smokers (more than 10 cigarettes per day), those on a special diet, diabetic as confirmed by HbA1c and those suffering from diabetic ketoacidosis or diabetic pre-coma were excluded. Subjects were also excluded if they participated in a bioavailability/bioequivalence study within 80 days after the first dose, donated (through to the completion of the study) >900 ml of blood in 20 weeks before the study, and had positive human immunodeficiency virus (HIV) I and II and/or hepatitis B and/or hepatitis C.

On day 0 and following an overnight fasting of at least 8 h, a high-fat, high-calorie breakfast was served 30 min before dosing and then the subject received the assigned dose (one tablet of the test product or the reference product) with 240 ± 2 ml of 20% glucose solution, according to the randomisation schedule (Figure 1).

Figure 1.

The main objective of this study was to assess bioequivalence between Sensityn^® 50 mg/1000 mg film-coated tablets (Alpha Pharma Industries, a subsidiary of Cigalah Healthcare LLC, KAEC, Saudi Arabia) and Galvusmet^® 50 mg/1000 mg film-coated tablets (Novartis, Switzerland) in terms of rate and extent of absorption. The assessment was based on a single vildagliptin/metformin 50 mg/1000 mg oral dose (one tablet) administration in each study period.

The pharmacokinetic parameters derived from the plasma concentration–time profiles that were compared were as follows: primary endpoints (C_max, AUC_0-t and AUC_0-∞) and secondary endpoints (T_max [for all subjects], AUC_{_%Extrap_obs}, K_elimination (λz) and T_half for subjects who showed clear elimination).

The safety and tolerability of a single dose administered in healthy human adult subjects under fed conditions were also monitored during the study.

Blood samples were collected in each study period at predose (BK) ‘one sample’ and 0.25, 0.50, 0.75, 1.00, 1.50, 2.00, 2.25, 2.50, 2.75, 3.00, 3.50, 4.00, 5.00, 6.00, 8.00, 10.00, 12.00 and 24.00 h postdosing. A total of 19 PK blood samples (8 ml/sample) were withdrawn in each study period (one predose + 18 postdosing). After collection, the blood samples were kept in wet ice (crushed ice in water) for a maximum period of 1 h and 25 min until being centrifuged. Samples were centrifuged at 3400 rpm for 5 min at room temperature. Each of the pharmacokinetic plasma samples was separated into two aliquots explicitly (Q1 and Q2), Q1 is the primary vial and Q2 is the backup vial, and stored at −20°C. Subjects remained at the clinical facility for 24 h postdosing.

Plasma samples from subjects were analysed for vildagliptin and metformin using a validated bioanalytical LC-MS/MS method. Two analytical methods were developed and validated for the two analytes, vildagliptin and metformin. Vildagliptin was extracted from the plasma samples by a liquid–liquid extraction procedure, and the chromatographic separation was performed with an isocratic mixture of acetonitrile and 0.1% formic acid (7:3) in a C8 column with a total run time of 2.0 min. Metformin was extracted from the plasma samples by protein precipitation procedure and the chromatographic separation was performed with an isocratic mixture of acetonitrile and 0.1% ammonium formate/formic acid buffer solution (1:1) in a silica column with a total run time of 4.0 min.

The lower limit of quantification (LLOQ) of the validated assay for vildagliptin was 0.50 ng/ml and the upper limit of quantification (ULOQ) was 300 ng/ml. LLOQ of the validated assay for metformin was 20.00 ng/ml, and ULOQ was 2500 ng/ml. Concentrations below LLOQ were considered zero during the samples’ statistical and pharmacokinetic analyses.

Pharmacokinetic samples from the pivotal bioequivalence study were reanalysed as part of the incurred samples reproducibility assessment. The result of the incurred sample reanalysis met the acceptance criterion, demonstrating satisfactory reproducibility of the pharmacokinetic assay through the sample analysis period.

From a safety perspective, a COVID-19 questionnaire was filled out and body temperature was measured before the screening assessment began. Screening assessments involved demographic data, full medical history, physical examination, ECG and clinical laboratory tests (full blood count, coagulation tests, clinical chemistry [including HbA1c], serology and urinalysis). Completing the COVID-19 questionnaire and measuring body temperature were again carried out on day -2 of period I and before starting admission assessments. Admission assessments involved physical examination, testing for drugs of abuse through urine, testing for alcohol consumption through saliva and vital signs measurement (body temperature, blood pressure and pulse rate).

Vital signs in sitting position were monitored before dosing and at 30 min, 1, 2, 4, 6, 8, 12 and 24 h postdosing in each period. Glucose blood test was performed at 1 h before dosing, every 30 min during the first 4 h postdosing, and 5, 6, 7, 8, 12 and 24 h postdosing. At the end of period II, all subjects were tested for vital signs in sitting position, and clinical laboratory tests (full blood count, coagulation tests, clinical chemistry and urinalysis), physical examination and ECG were conducted.

Regarding the calculation of the sample size, the guidelines state that the number of subjects must be sufficient to ensure adequate statistical results. Considering an intrasubject coefficient of variation (ISCV) of 26.6% (based on a previous study – Triumpharma study no. VIME.50.1000/273, IntraCV for vildagliptin C_max was 26.6%), geometric mean ratio of 95%–105% and a dropout rate of 10%, 36 subjects were enrolled to achieve at least 80% power, as calculated using SAS.9.4.

Data from the 36 subjects who completed the study was analysed using validated WinNonlin^® version 8.3 for PK and statistical analysis.

The actual times of withdrawal of blood samples were recorded, and the actual times were computed for pharmacokinetics and statistical analysis.

No missing or non-reportable samples were reported for this study. Concentrations below LLOQ were considered zero during the samples’ statistical and pharmacokinetic analysis. Number of samples, mean (arithmetic and geometric), standard deviation (SD), percent coefficient of variation (CV%), standard error of mean (SEM), minimum, median and maximum were calculated for plasma drug concentration of the analysed samples. Median T_max was calculated, while the mean was calculated for all other parameters.

Analysis of variance (ANOVA) analysis (test sequence, subjects nested within sequence, product and period effect) was performed using a significance level of 5% for logarithmic and untransformed data for C_max AUC_0-t and AUC_0-∞ and for untransformed data for T_max, K_elimination (λz) and half-life.

A total of 65 male adult Caucasians selected from the Jordanian population were screened, and 36 subjects aged 18–45 years, with a mean age of 29 years, and four stand-by subjects were found to be eligible for participation as per the selection criteria of the protocol. In period I, a total of 36 subjects with a mean BMI of 24.9 kg/m² (subjects’ randomisation no. 01–36) were dosed in period I and the four stand-by subjects were dismissed after completion of the dosing procedure for the 36 primary subjects. A total of 36 subjects (subjects’ randomisation no. 01–36) were enrolled and dosed in period II and the study was completed. For more details, see Table 1.

Samples from the 36 subjects were analysed in the bioanalytical lab for vildagliptin and metformin. Data from 36 subjects were included in the vildagliptin and metformin PK analyses and in the vildagliptin and metformin statistical analyses for C_max and AUC_0-t.

Data of 36 subjects (subjects’ randomisation no. 01–36) was included in vildagliptin and metformin PK analyses and in vildagliptin and metformin statistical analyses for C_max and AUC_0-t. Data of 33 subjects (subjects’ randomisation no. 01–36, except subjects no. 24, 27 and 31) was included in vildagliptin statistical analysis for AUC_0-∞. Data of 35 subjects (subjects’ randomisation no. 01–36, except subject no. 08) was included in metformin statistical analysis AUC_0-∞.

Subjects no. 24, 27 and 31 were excluded from the AUC_0-∞ statistical analysis for vildagliptin since they had adjusted Rsq values of <0.8 for one of the treatments. Similarly, subject no. 08 was excluded from the AUC_0-∞ statistical analysis for metformin since he had an adjusted Rsq value <0.8 for the treatment.

The P-values for vildagliptin were not significant at >5% for the effects of sequence, treatment and period for C_max, AUC_0-t and AUC_0-∞.

The P-values for metformin were not significant at >5% for the effects of sequence, treatment and period for AUC_0-t and AUC_0-∞. Regarding C_max, the only significant effect was for the period (0.0435).

The summary of the mean and SD of the PK parameters for vildagliptin and metformin is presented in Tables 2 and 3, respectively.

A display of the mean plasma concentrations of vildagliptin and metformin after administration of Sensityn^® and Galvusmet^® in healthy adult subjects using a linear and semilog scales is presented in Figures 2 and 3, respectively.

Figure 2.

Figure 3.

Tables 4 and 5 summarise the statistical comparisons of the pharmacokinetic parameters of vildagliptin and metformin, respectively.

Considering these results, C_max, AUC_0-t and AUC_0-∞ have passed the acceptance limits of 90% confidence interval (CI) of 80.00%–125.00% for vildagliptin and metformin. Consequently, the bioequivalence of vildagliptin/metformin 50 mg/1000 mg film-coated tablets Sensityn^® and Galvusmet^® was demonstrated under fed conditions.

The safety of the test and reference products was assessed on the basis of clinical and laboratory examinations before the administration of the study product and at the end of the study with regular monitoring of the subjects during confinement and recording of adverse events (AEs). During the study, four AEs were reported in three of the study subjects (8.33%). None of these AEs was serious, and there were no AEs that resulted in the death or occurrence of any other significant event of any subject. Three (i.e. nausea, headache and hypoglycaemia) and one (headache) AEs emerged in subjects treated with Sensityn^® and Galvusmet^®, respectively. All subjects with AEs completely recovered and there were no ongoing AEs. In terms of intensity, three AEs (75.0%) were considered mild and one AE (25.0%) was considered moderate. Three (75.0%) AEs (2x headache, 1x nausea) were classified as probably related to the administered treatment and one (25.0%) AE (1x hypoglycaemia) was classified as definitely related to the administered treatment. No subjects were discontinued because of the occurrence of other significant AEs. No clinically significant out-of-range values were observed in the vital signs and physical examination during the study.