Original©2012 Dustri-Verlag Dr. K. Feistle

ISSN 0946-1965

DOI 10.5414/CP201715e-pub: July 4, 2012

ReceivedFebruary 6, 2012;acceptedMarch 21, 2012

Correspondence to Kyun-Seop Bae, MD, PhD 88, Olympic-ro 43-gil, Songpa-gu, Seoul, Korea, 138-736 ksbae@amc.seoul.kr

Key wordsHL-018 – metformin – bioavailability – pharma-cokinetics – pharmaco-dynamics

A single-dose, crossover study comparing the pharmacokinetics and pharmacodynamics of two formulations of metformin in healthy volunteersYook-Hwan Noh1, Hyeong-Seok Lim1,2, Jin Ah Jung4, Seok-Joon Jin1, Mi Jo Kim1, Yo Han Kim1, Hyun-Jung Park3 and Kyun-Seop Bae1,2

1Department of Clinical Pharmacology and Therapeutics, Asan Medical Center, 2College of Medicine, University of Ulsan, 3Clinical Trials Center, Asan Medical Center and 4Department of Clinical Pharmacology and Therapeutics, Samsung Medical Center, Seoul, Korea

Abstract. Objective: To compare phar-macokinetics and pharmacodynamics of two formulations of metformin in healthy male volunteers under fasting conditions. Meth-ods and materials: This was a randomized, 3-treatment, 6-sequence, 3-period, crossover study in healthy Korean volunteers. Subjects received a placebo or a single oral dose of reference formulation (500-mg metformin hydrochloride) or a test formulation (571-mg metformin acetate). The pharmacodynamic profile was assessed according to the blood glucose level. Results: A total of 30 healthy male volunteers was included, although 2 of them withdrew consent before drug ad-ministration. For the test formulation, the mean (SD) pharmacokinetic parameters of Cmax and AUC0–24 were 1,691 (476.7) µg/ml and 11,081 (2,804.5) mg×h/ml, respec-tively. For the reference formulation, the mean (SD) pharmacokinetic parameters of Cmax and AUC0–24 were 1,749 (494.4) mg/ml and 11,814 (3,344.2) mg×h/ml, respectively. The 90% CIs for the test:reference ratio were 86.6 – 103.5 for logarithm-transformed AU-Clast and 87.9 – 107.4 for logarithm-trans-formed Cmax, respectively. In pharmacody-namic analysis, the average blood glucose levels were obtained in the first 4 hours and 5 – 8 hours after drug administration. The 90% CIs for the test:reference ratio of the av-erage blood glucose level were 96.5 – 101.8 for the intervals up to 4 hours during contin-ued fasting and 99.3 – 106.8 for 5 – 8 hours after drug administration, respectively. Con-clusion: The data suggests that the test and reference formulations meet the regulatory criteria for the pharmacokinetic equivalence in fasting healthy Korean male volunteers. The pharmacodynamic profile was similar between the test and reference formulations. Both metformin formulations appeared to be generally well-tolerated.

Introduction

Metformin, an oral antihyperglycaemic agent, is a biguanide developed from galegine, a guanidine derivative contained in Galega officinalis (French lilac; goats rue). Metfor-min is recommended as the first-line treat-ment in all newly diagnosed patients with Type 2 diabetes mellitus, regardless of their age [2, 14].

The pharmacokinetics (PK) of metfor-min have been studied in healthy adults and in patients with Type 2 diabetes [6]. After oral intake, metformin has a bioavailability of 50 – 60%, a half-life (t1/2) of ~ 4 hours, and is not metabolized before being excreted in the urine.

In Korea, there are already more than 30 approved, immediate-release generics, how-ever, all of these agents are hydrochloride salt formulations. Therefore, if one is ad-ministered at the maximum oral daily dose (2,000 mg), a patient takes in more than 440 mg of hydrochloride (A 500-mg tablet of metformin hydrochloride contains ~ 110 mg of hydrochloride). The most common ad-verse reaction to metformin hydrochloride is gastric irritation. However, as HL-018, a new generic tablet formulation, is manufactured with acetate salt and relatively weak acid (hydrochloride salt in 1% aqueous solution – pH 6.68 vs. acetate salt – pH 8.48), it is likely to cause lower gastric irritation and to provide higher solubility in order to be easily absorbed through the gastrointestinal tract.

A generic, 571-mg tablet formulation of metformin was recently developed for the

International Journal of Clinical Pharmacology and Therapeutics, Vol. 50 – No. 8/2012 (605-613)

Noh, Lim, Jung et al. 606

Korean market. Data on the pharmacokinet-ics and comparative bioavailability of this new formulation was required before mar-keting in this region [9]. We also assessed the glucose-lowering effect of HL-018 by mea-suring the plasma fasting glucose level and compared it with the already marketed drug, Glucophage®, in a 500-mg tablet in order to assess the pharmacodynamics of HL-018.

The purpose of the present study is to compare pharmacokinetics and pharmacody-namics of two formulations of metformin in healthy male volunteers under fasting con-ditions. We assessed the tolerability, phar-macokinetics, and pharmacodynamics (PD) of single, oral doses of a 571-mg tablet of metformin acetate (test formulation (HL-018 (HanAll BioPharma Co., Ltd., Seoul, Korea)) and of a 500-mg tablet of metformin hydro-chloride (reference formulation (Trademark: Glucophage® 500 mg (Yuyu Pharma, Inc., Seoul, Korea)) in healthy male volunteers.

Methods and materials

Inclusion and exclusion criteria

Healthy men aged 19 to 50 years, with a body mass index (BMI) between 19 and 25 kg/m2 and an unremarkable medical history; normal laboratory test (hematol-ogy, blood biochemistry, hepatic function, and urinalysis) values; negative test results for HIV/AIDS, hepatitis B, hepatitis C, and syphilis; and who had normal blood pressure were eligible. Baseline laboratory tests were performed within 28 days prior to the study. Subjects who had any disease state before the study, a history of alcohol or drug addiction, or who took any drug two weeks before the beginning of the study, were excluded. Other exclusion criteria considered throughout the study included hypersensitivity to any salt or metformin and 10% outside of the nor-mal blood glucose level before a meal. All laboratory tests (including the plasma fasting glucose level) other than PK analysis, were performed in the Department of Laboratory Medicine, Asan Medical Center, which has been accredited by the Korean Association of Quality Assurance for Clinical Labora-tories and certified by CAP (The College of American Pathologists).

Study design

This randomized, single-dose, 3-treat-ment, 6-sequence, 3-period, crossover study was conducted at the Clinical Trials Center (CTC) of Asan Medical Center (AMC) in Seoul, Korea. Each administration phase was followed by a 7-day (> 5 half-lives of the drug) washout period.[6, 19]. 30 eligible men were randomized to six sequence groups in order of their passing the eligibility tests and using the randomization method before initiation of the study. A table of random numbers generated by R (version 2.10.1, R Development Core Team (2010), R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Austria, URL: http://www.R-project.org) was used to assign volunteers in each same ratio to receive the test or reference formulations or placebo (only water intake).

Test formulation (HL-018 tablet, 571 mg; batch no. 1070006; expiration date, April 2013) and reference formulation (Glu-cophage® tablet, 500 mg; batch no. 090510; expiration date, December 2012) were used in the study.

Volunteers reported to the Clinical Trial Center at Asan Medical Center one day before dosing and were confined until their discharge. After a 10-hour overnight fast, study subjects received a single oral dose of one tablet of the test formulation, a single oral dose of one tablet of the refer-ence formulation, or a placebo (only wa-ter intake with no drug administration). All formulations were administered with 240 ml of water, and fasting continued for an additional 4 hours after administration. Subjects’ mouths were examined by an in-vestigator after ingestion in order to guar-antee swallowing and was then recorded in the written source documents. Because the Korean Food & Drug Administration (KFDA) does not require the testing of food effects, analysis was not performed under fed conditions.

Lunch (200 kcal), snack (bread and milk), and dinner (standardized meal) were served 4, 8 and 10 hours after dose administration. Sub-jects were discharged after all of the sched-uled procedures were finished the next morn-ing, ~ 24 h after dosing. Subjects returned to CTC on Days 7 and 14 (the days before Pe-

Comparison of PK/PD of two formulations of metformin 607

riods 2 and 3). On the following mornings, the next dosing administration period was initiated. All phases were carried out in the same manner as Period 1, with the exception of the administration of a different (alternate) formulation. All dietary, smoking, and drug/herbal product restrictions were maintained throughout the study period. After the last follow-up visit, all subjects were financially compensated for their participation.

The study protocol was reviewed and approved by the Institutional Review Board of Asan Medical Center and was conducted in accordance with the Good Clinical Prac-tice guidelines [16] and the Declaration of Helsinki and its amendments [3]. Written, informed consent was obtained from all subjects before any study-specific proce-dure was undertaken. This clinical trial is registered on-line at www.ClinicalTrials.gov (registration number: NCT01232036).

Sample collection

Serial blood samples for PK study were drawn at 0 hours (prior to metformin admin-istration) and at 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, 10, 12, and 24 h after dosing [3, 13]. Serial blood samples for PD study were also drawn at 0 h (prior to metformin administration) and at 1, 1.5, 2, 2.5, 3, 4, 5, 6, and 8 h after dosing [16].

Approximately 5 ml of blood for PK study and 4 ml of blood for PD study were collected from an indwelling catheter in the antecubital vein of the forearm and were placed in heparinized vacuum tubes. Prior to each sample collection, 1 ml of blood was drawn and discarded. After sampling, the cannula was flushed with 1.5 ml of sa-line solution. Blood samples were placed on ice before centrifugation, and plasma was immediately separated by centrifugation at 3,000 rpm at 4 °C for 10 minutes, transferred to eppendorf tubes, and stored at –70 °C until analysis.

Analytical method

Measurement of plasma metformin

Independent of the salt administered, all pharmacokinetic data reported are based on

plasma concentrations of metformin mea-sured using the validated LC-MS/MS meth-od. All plasma PK samples were handled by the Clinical Trial Center at Asan Medi-cal Center (Seoul, Korea) using validated bioanalytical methods [9]. All procedures of PK analysis were conducted in accor-dance with the standard operating procedure (SOP; number GEM-106) of the Clinical Trials Center at Asan Medical Center, and which were amended in accordance with the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) Q2(R1) [1]. The analytical platform con-sisted of validated, high-performance liquid chromatography (HPLC; Agilent 1200 se-ries, Agilent Technologies, Inc., USA) cou-pled with tandem mass spectrometry (MS/MS; API4000, AB Sciex Inc., Foster City, CA, USA) using Analyst® version 1.4.1 soft-ware (AB Sciex Corporation). Liquid-liquid extraction used acetonitrile from J. T. Back-er, Inc. (Phillipsburg, NJ, USA) and em-ployed a method developed in the analytic laboratory of the Clinical Trial Center of the Asan Medical Center. Phenformin was used as an internal standard for analyses, and was provided by HanAll BioPharma Co. Ltd. A Shiseido MG 3mm column (2.0 × 50 mm, Shiseido, Japan) was used for chromato-graphic separation. The mobile phase con-sisted of a mixture of acetonitrile purchased from J. T. Backer, Inc. and water [5 : 95, v/v] with 0.1% formic acid (pH 3.0) from Fluka (Sigma-Aldrich Co. LLC., Yongin, Korea). The flow rate of the mobile phase was 0.2 ml/min at room temperature, and the injec-tion volume was 10 µl. The tandem mass spectrometry system was operated with electrospray ionization in the positive ion mode. This method had a linear quantifiable range of 5 to 500 ng/ml and an r2 of 0.995 or greater. Intra-day precision, expressed in terms of %CV, ranged from 2.20% to 8.28%, and inter-day precision ranged from 6.92% to 13.43%. Accuracy, expressed as a relative error of measurement, %RE, was –8.32% to 0.90%. The analytes were stable in human plasma following three freeze-thaw cycles, in plasma after storage for 24 h at room tem-perature, in stock solution after storage for 6 h at room temperature, and in HPLC au-tosampler after storage at 4° C for 24 h.

Noh, Lim, Jung et al. 608

Tolerability

Tolerability was assessed based on sub-ject interviews and vital sign monitoring, i.e., changes in body temperature, blood pressure, and heart rate, conducted at rest and measured at predefined, regular inter-vals throughout the study using noninvasive electronic devices. An electrocardiogram examination was performed for volunteer screening. Clinical blood tests were also performed at regular intervals and included hematological analysis (red blood cell count, hemoglobin, hematocrit, platelet count, and white blood cell count), blood chemistry (electrolytes, hepatic enzymes, renal func-tion parameters, fasting blood glucose, albu-min, cholesterol, triglycerides, phosphorus, lactic dehydrogenase, and uric acid), and urinalysis. Volunteers were clinically moni-tored up to seven days after the end of Period 3. The clinical importance of adverse events (AEs) and the likelihood that AEs related to the study drugs, were assessed by investiga-tors throughout the study.

Pharmacokinetic analysis

Pharmacokinetic parameters were calcu-lated using programming plasma data, a sin-gle extravascular dose, and a non-compart-mental model using Phoenix WinNonlin® version 6.1 software (Pharsight Corporation, MO, USA) [11]. The AUC from time zero to the last measurable concentration (AUC0–24), the AUC from Time 0 extrapolated to infin-ity (AUC0–∞), the Vd after non-intravenous administration (Vd/F), the apparent total clearance of the drug from plasma after oral administration (CL/F), and the mean resi-dence time (MRT) were software outputs. The values for Cmax and tmax were obtained directly from the experimental observations of the kinetic profiles of each volunteer. The algorithm used to calculate the t1/2 was con-structed using a linear regression by least squares from the terminal logarithm-normal decay part of the pharmacokinetic profile (elimination phase); the negative slope of the curve is the elimination rate constant (ke); t1/2 = ln(2)/ke.

Pharmacodynamic analysis

We assessed predose blood glucose level as baseline value and two averaged blood glucose levels, one calculated from the first 4 h measurements (1, 1.5, 2, 2.5, 3, and 4 h) after study drug administration (baseline: 0 h) and the other from 5 to 8 h measure-ments (5, 6, and 8 h) after administration (baseline: 4 h). The results were calculated as the mean (SD), and two formulation dif-ference was statistically tested with a 0.05% significance level.

Statistical analysis

The demographic characteristics of the enrolled subjects were summarized using de-scriptive statistics.

As this study was exploratory research to assess the PK/PD and tolerability of met-formin, it was preferable to proceed with the least number of volunteers in order to meet the study objectives. With the assumption of 20% as metformin CV(%) in a 2 × 2 crossover design [6], we calculated the sample size of this study as 19 volunteers using 80% power at a 5% significance level [4, 16]. Consider-ing this result, we then estimated the final sample size of this study as 24 volunteers (four for each sequence). In accordance with the Korean Regulatory Guidelines [9], as 24 volunteers was the minimum sample required (assuming an 80% power to detect a 20% dif-ference), we used a sample size of 30 based on the potential for volunteer drop-out during the study. An ANOVA for a 3 × 6 crossover design was performed on the decimal, loga-rithm-transformed parameters, Cmax, AUC0–t, and AUC0–∞, in order to evaluate fixed effects such as period, sequence, formulation, and carryover. Logarithm-transformed values of these parameters were considered to construct a classic CI at 90%, and with p < 0.05 indi-cating significance. The formulations were considered bioequivalent if the 90% CI of the logarithm-transformed ratios (test/reference) of Cmax (an index of the rate of absorption), AUC0–24h, and AUC0–∞ (indices of the extent of absorption) were within the predefined range of 0.80 – 1.25 [9].

In pharmacodynamic analysis, we consid-ered it to be equivalent between two formu-

Comparison of PK/PD of two formulations of metformin 609

lations if the 90% CI of the logarithm-trans-formed ratios (test/reference) of the average blood glucose level (an index of the blood-glucose-lowering effect) was within the pre-defined conventional range of 0.80 – 1.25.

All statistical analyses were performed using Phoenix WinNonlin® version 6.1 soft-ware (Pharsight Corporation, St. Louis, MO, USA) and R (version 2.10.1; R Foundation for Statistical Computing, Vienna, Austria).

Results

A total of 30 healthy male volunteers (mean (SD) age, 27.8 (5.7) years; height, 172.5 (5.7) cm; weight, 66.3 (6.5) kg; and BMI, 22.3 (1.5) kg/m2), was included. Demo-

graphic data and the sequence of study drug administration are shown in Table 1. The de-mographics did not differ significantly among sequence groups. Two volunteers withdrew before drug administration because of their personal reasons. Safety profiles were as-sessed for all 28 volunteers in whom the study drugs were administered. Pharmacokinetic characteristics were evaluated in the 28 vol-unteers who completed the study.

Tolerability

A total of 23 AEs was reported (7 on the test formulation, 11 on the reference formu-lation and 5 on the placebo) in 10 of the 28 volunteers (35.7%). Among the volunteers, 8 cases of diarrhea, 2 cases of papular rash, and 1 case of nausea were considered likely to be related to the study drug.

The other AEs included dizziness (2 cas-es), abdominal pain (2 cases), nasal stuffi-ness (1 case), sputum (1 case), asthenia (1 case), vasovagal syncope (1 case), papular rash (1 case), eczema (1 case), subcutaneous nodule (1 case), blood creatine phosphoki-nase increased (1 case); these were judged to be unrelated to the study drug considering the onset of administration, time interval of drug administration, and the time course of the AEs (Table 2).

The severity of all the AEs was consid-ered mild except 1 AE which was of moder-ate intensity in the volunteer who had taken placebo treatment. No serious AEs were observed during the study. None of the sub-jects was withdrawn from the study due to AEs. No clinically significant findings were reported on physical examination including vital signs, physical examination, 12-lead ECG, or laboratory tests.

Pharmacokinetic parameters

The mean (SD) plasma concentration-time profiles of the 28 volunteers after sin-gle-dose oral administration of the test and reference formulations, are shown in the Fig-ure. The calculated pharmacokinetic proper-ties of metformin are summarized in Table 3.

For the test formulation, the mean (SD) pharmacokinetic parameters of Cmax and

Table 1. Demographic characteristics and formulation sequence of the healthy Korean volunteers participating in a study assessing the bioequivalence of met-formin hydrochloride 500 mg tablet (R) and metformin acetate 571 mg tablet (T) administered as single oral doses. (P): Placebo (only water intake).

Volunteer Age, y Weight, kg Height, cm BMI, kg/m2 Sequence*1 36 71.4 177 22.8 PTR2 28 59.5 172 20.1 PTR3 28 55.9 160 21.8 PTR4 23 66.4 178 21.0 PTR5 30 72.6 177 23.2 PTR6 26 70.6 174 23.3 TRP7 39 53.8 160 21.0 TRP8 26 66.5 171 22.7 TRP9 26 64.7 176 20.9 TRP

10 32 51.8 164 19.3 TRP11 50 67.1 165 24.7 RPT12 26 71.3 177 22.8 RPT13 28 67.2 166 24.4 RPT14 30 63.1 166 22.9 RPT15 29 63.8 176 20.6 RPT16 23 61.8 169 21.6 RTP17 28 74.8 178 23.6 RTP18 26 70.7 174 23.4 RTP19 25 69.1 173 23.1 RTP20 22 62.4 174 20.6 RTP21 28 74.4 182 22.5 PRT22 28 73.1 180 22.6 PRT23 22 64.1 174 21.2 PRT24 25 57.1 172 19.3 PRT25 27 62.1 173 20.8 PRT26 26 71.3 172 24.1 TPR27 24 78 182 23.6 TPR28 22 63.4 171 21.7 TPR29 24 70.8 171 24.2 TPR30 28 70.6 171 24.1 TPR

BMI = body mass index. *Sequence of formulation administration during the 3-period clinical phase.

Noh, Lim, Jung et al. 610

AUC0–24 were 1,691 (476.7) mg/ml and 11,081 (2,804.5) mg·h/ml, respectively. For the reference formulation, the mean (SD) pharmacokinetic parameters of Cmax and AUC0–24 were 1,749 (494.4) mg/ml and 11,814 (3,344.2) mg×h/ml, respectively.

The 90% CIs for the test:reference ratio were 87.9 – 107.4 for Cmax, 86.6 – 103.5 for AUC0–24 and 86.4 – 102.4 for AUC0–∞, respec-tively. The 90% CI for the logarithm-trans-

formed values of Cmax, AUC0–24, and AUC0–∞ were within the 0.80 – 1.25 interval, and with a statistical power ≥ 0.98 in each tested param-eter. The data is shown in Table 3.

Pharmacodynamic parameters

The calculated fasting blood glucose levels of the 28 volunteers after the single-

Table 2. Adverse events (AEs) per system organ class and treatment (n = 28). Data is presented as number of events (% of volunteers).

Metformin acetate Metformin hydrochloride

Water intake

System Organ Class/AE Test formulation* Reference formulation† PlaceboGastrointestinal disorders 6 (21.4) 4 (14.3) 1 (3.6) Abdominal pain 1 (3.6) 1 (3.6) Diarrhea 4 (14.3) 3 (10.7) 1 (3.6) Nausea 1 (3.6)Investigations 1 (3.6) Blood creatine phosphokinase increased 1 (3.6)Musculoskeletal disorders 1 (3.6) Asthenia 1 (3.6)Nervous system disorder 2 (7.1) Dizziness 2 (7.1)Respiratory, thoracic and mediastinal disorders 2 (3.6)  Nasal stuffiness 1 (3.6) Sputum 1 (3.6)Skin and subcutaneous tissue disorders 1 (3.6) 2 (7.1) 2 (7.1) Rash popular 1 (3.6) 1 (3.6) 1 (3.6) Subcutaneous nodule 1 (3.6) Eczema 1 (3.6)Vascular disorder 1 (3.6) Vasovagal syncope 1 (3.6)Total 7 (21.4) 11 (25.0) 5 (14.3)

*HL-018 (HanAll BioPharma Co., Ltd., Seoul, Korea); †Trademark: Glucophage® 500-mg (Yuyu Pharma, Inc., Seoul, Korea).

Table 3. Pharmacokinetic parameters for metformin and the results of the bioequivalence analysis of metformin after oral administration of single dose of 500-mg metformin hydrochloride tablet (reference formulation; Yuyu pharma Inc., Seoul, Korea) and 571-mg metformin acetate tablet formulation (test formulation; HanAll Biopharma Co., Ltd., Seoul, Korea) in 28 healthy Korean volunteers. Values are ex-pressed as mean (SD).

Parameter Test Reference 90% CI PowerCmax, ng/ml 1,691 (476.7) 1,749 (494.4) 87.9 – 107.4 0.98tmax, h 2.7 (1.0) 2.8 (1.0) 81.9 – 116.5 0.67AUC0–24, ng×h/ml 11,081 (2,804.5) 11,814 (3,344.2) 86.6 – 103.5 0.99AUC0–∞, ng×h/ml 11,403 (2,833.5) 12,213 (3,360.0) 86.4 – 102.4 0.99t1/2, h 4.7 (1.1) 5.2 (1.3) 82.8 – 101.2 0.98Vd/F, l 362.4 (133.7) 332.7 (170.7) 95.8 – 129.0 0.80CL/F, l/h 52.9 (12.2) 44.0 (12.3) 111.5 – 132.2 0.99MRT, h 5.6 (0.6) 5.8 (0.6) 93.5 – 98.4 1.00

CI = confidence interval; Vd/F = apparent volume of distribution after nonintravenous administration; CL/F = apparent total clearance of the drug from plasma after oral administration; MRT = mean residence time.

Comparison of PK/PD of two formulations of metformin 611

dose oral administration of the test formu-lation and the reference formulation are summarized (Table 4). The 90% CIs for the test:reference ratio of the average blood glu-cose level were 96.5 – 101.8 for the intervals up to 4 hours during continued fasting and 99.3 – 106.8 for 5 – 8 h after drug adminis-tration, respectively.

Discussion

In this study, a new salt form of metfor-min acetate formulation showed similar hu-man pharmacokinetic and pharmacodynamic characteristics to those of the currently used metformin hydrochloride formulation. Both formulations were well-tolerated and met the regulatory requirements for pharmacokinetic equivalence.

The value of AUC represents the extent of absorption, and the Cmax and tmax values indicate the rates of absorption [17]. The 90% CIs of the log-transformed ratio of the two metformin formulations were 87.9 – 107.4 for Cmax, 86.6 – 103.5 for AUC0–24, and 86.4 – 102.4 for AUC0–∞, all of which were within the stipulated pharmacokinetic equivalence range of 80 – 125% [5, 9].

Metformin hydrochloride is a hydro-philic drug for which the absolute bioavail-ability of a single 500-mg dose is 55 ± 16% [6]. It has a relatively short elimination t1/2 of 1.7 – 4.5 h [6]. Metformin reportedly causes concomitant gastrointestinal symp-toms such as nausea, diarrhea, and abdomi-nal discomfort [7, 18]. The gastrointestinal AEs of metformin may be related to the hy-drochloride salt of the drug, which suggests that a different salt formulation, i.e., acetate salt form, relatively weak acid, may dimin-ish gastric AEs. However, there was no sta-tistical difference in gastric AEs between the

Figure  1.  Plasma  concentration–time  profiles  of metformin after oral administration of single dose of 500-mg metformin hydrochloride tablet (refer-ence formulation; Yuyu pharma Inc., Seoul, Korea) and 571-mg metformin acetate tablet formulation (test formulation; HanAll Biopharma Co., Ltd., Seoul, Korea) in 28 healthy Korean volunteers under fasting conditions. Values are expressed as mean (SD). (A: linear scale, B: log-linear scale).

Table 4. Pharmacodynamic profile of metformin and comparison of  fasting glucose level of metformin after oral administration of single dose of 500-mg metformin hydrochloride tablet and 571-mg metformin acetate tablet formulation. Values are expressed as mean (SD).

Glucose Test Reference GMR (Test vs. Reference)

90% CI

Pre dose, mg/dl 87.3 (5.6) 87.2 (5.7) 100.0 97.3 – 102.91 – 4 h, mg/dl 87.8 (5.6) 88.5 (5.0) 99.1 96.5 – 101.85 – 8 h, mg/dl 97.1 (7.5) 94.4 (8.2) 103.0 99.3 – 106.8

GMR = Geometric Mean Ratio; CI = confidence interval.

Noh, Lim, Jung et al. 612

test and reference formulations in this study (p = 0.1777, Fisher’s exact test).

We assessed the blood glucose lowering effect of metformin by measuring the fast-ing blood glucose level. In the fasting state, administration of metformin may cause hy-poglycemia in volunteers, although no sig-nificant AEs related to hypoglycemia were expected in this study. It was thought that the dose administered was not a large dose and that the healthy volunteers’ general condi-tion was good enough to accommodate the change of blood glucose level. To assess the pharmacodynamics profile through the blood glucose level, we divided the time period of average glucose level measurement into two intervals starting from study drug adminis-tration time, and based on the tmax of metfor-min. From the published data of a total of 11 healthy study volunteers [19] and our study results, we determined that the peak plasma concentrations of metformin occur ~ 3 h af-ter dosage. Therefore, we analyzed the aver-age glucose level with and without the maxi-mum drug concentration, i.e. the intervals up to 4 h during continued fasting and five to eight hours after drug administration.

As with any clinical trial, the present study had certain limitations that should be considered. First, the drug was administered in the fasting state, and thus may have differ-ent tolerability and pharmacokinetic profiles of the formulations tested. It is recommend-ed that metformin should be taken with food so as to minimize gastrointestinal side effects such as bloating, flatus, and diarrhea. A high-fat meal has been reported to decrease the bioavailability of immediate-release tablets of metformin by ~ 25% [15], although the effect of food is minimal with combination tablets of metformin and other anti-diabetic drugs [8, 10, 12]. The reduced absorption is unlikely to be clinically significant in most patients although, the fasting condition was chosen because the primary objective of the study was to compare the pharmacokinetic properties of the two metformin formula-tions in the fasting state in accordance with regulatory guidelines. Second, because the data was obtained from healthy male sub-jects, the pharmacokinetic characteristics of metformin may differ in patient popula-tions as related to gender and age. However, considering that the pharmacokinetics of the

two drugs were similar in this study, it seems likely that their properties would be similar in target patient populations, although there could possibly be more inter-individual vari-ation. The pharmacokinetic profiles did not differ significantly.

Conclusion

The data suggests that the test and refer-ence formulations met the regulatory criteria for pharmacokinetic equivalence in fasting healthy Korean male volunteers. The phar-macodynamics with respect to the blood glucose lowering effect was also similar be-tween test and reference formulations in the fasting state. Both metformin formulations appeared to be generally well-tolerated.

Acknowledgments

This study was sponsored by HanAll Bi-oPharma Co., Ltd. (Seoul, Korea). The spon-sor had no role in the study development, design, data analysis, or data interpretation. This study was supported by a grant from the Korea Healthcare Technology R&D Proj-ect, Ministry of Health, Welfare & Family Affairs, Republic of Korea (Grant Number: A070001).

All authors agree with the contents of this manuscript, and the corresponding author made the final decision regarding its submis-sion. Drs. Bae, Lim, Jung, Jin, Noh, MJ Kim, and YH Kim contributed to the study design and data analysis. Drs. Bae and Noh contrib-uted to the manuscript writing. Ms. Hyun-Jung Park determined the plasma metformin concentrations. No sources of funding were used to assist in the preparation of the pres-ent manuscript. The authors have indicated that they have no conflict of interest regard-ing the content of this article.

References[1] International Conference on Harmonisation of

Technical Requirements for Registration of Phar-maceuticals for Human Use (ICH) adopts Con-solidated Guideline on Good Clinical Practice in the Conduct of Clinical Trials on Medicinal Prod-ucts for Human Use. Int Dig Health Legis. 1997; 48: 231-234. PubMed

Comparison of PK/PD of two formulations of metformin 613

[2] Global Guideline for Type 2 Diabetes: recommen-dations for standard, comprehensive, and minimal care. Diabet Med. 2006; 23: 579-593.

[3] Caillé G, Lacasse Y, Raymond M, Landriault H, Perrotta M, Picirilli G, Thiffault J, Spénard J. Bio-availability of metformin in tablet form using a new high pressure liquid chromatography assay method. Biopharm Drug Dispos. 1993; 14: 257-263. doi:10.1002/bdd.2510140308 PubMed

[4] Chow SC, Wang H. On sample size calculation in bioequivalence trials. J Pharmacokinet Pharmaco-dyn. 2001; 28: 155-169. doi:10.1023/A:1011503032353 PubMed

[5] Farolfi M, Powers JD, Rescigno A. On the deter-mination of bioequivalence. Pharmacol Res. 1999; 39: 1-4.

[6] Graham GG, Punt J, Arora M, Day RO, Doogue MP, Duong JK, Furlong TJ, Greenfield JR, Gree-nup LC, Kirkpatrick CM, Ray JE, Timmins P, Wil-liams KM. Clinical pharmacokinetics of metfor-min. Clin Pharmacokinet. 2011; 50: 81-98. doi:10.2165/11534750-000000000-00000 PubMed

[7] Gu N, Kim BH, Rhim H, Chung JY, Kim JR, Shin HS, Yoon SH, Cho JY, Shin SG, Jang IJ, Yu KS. Comparison of the bioavailability and tolerability of fixed-dose combination glimepiride/metformin 2/500-mg tablets versus separate tablets: A single-dose, randomized-sequence, open-label, two-peri-od crossover study in healthy Korean volunteers. Clin Ther. 2010; 32: 1408-1418. doi:10.1016/j.clinthera.2010.07.012 PubMed

[8] He YL, Flannery B, Campestrini J, Leon S, Zinny MA, Ligueros-Saylan M, Jarugula V. Effect of food on the pharmacokinetics of a vildagliptin/metformin (50/1000 mg) fixed-dose combination tablet in healthy volunteers. Curr Med Res Opin. 2008; 24: 1703-1709. doi:10.1185/03007990802114070 PubMed

[9] Heinig K, Bucheli F. Fast liquid chromatographic-tandem mass spectrometric (LC-MS-MS) deter-mination of metformin in plasma samples. J Pharm Biomed Anal. 2004; 34: 1005-1011. doi:10.1016/j.jpba.2003.11.017 PubMed

[10] Karim A, Slater M, Bradford D, Schwartz L, Lau-rent A. Oral antidiabetic drugs: effect of food on absorption of pioglitazone and metformin from a fixed-dose combination tablet. J Clin Pharmacol. 2007; 47: 48-55. doi:10.1177/0091270006293756 PubMed

[11]. Malcolm Rowland, Tozer TN. Clinical Pharmaco-kinetics and Pharmacodynamics: Concepts and Applications: Wolters Kluwer/Lippincott Wil-liams & Wilkins; 2011.

[12] Marathe PH, Wen Y, Norton J, Greene DS, Barb-haiya RH, Wilding IR. Effect of altered gastric emptying and gastrointestinal motility on metfor-min absorption. Br J Clin Pharmacol. 2000; 50: 325-332. doi:10.1046/j.1365-2125.2000.00264.x PubMed

[13] Najib N, Idkaidek N, Beshtawi M, Bader M, Ad-mour I, Alam SM, Zaman Q, Dham R. Bioequiva-lence evaluation of two brands of metformin 500 mg tablets (Dialon & Glucophage) – in healthy human volunteers. Biopharm Drug Dispos. 2002; 23: 301-306. doi:10.1002/bdd.326 PubMed

[14] Nathan DM, Buse JB, Davidson MB, Heine RJ, Holman RR, Sherwin R, Zinman B. Management

of hyperglycemia in type 2 diabetes: A consensus algorithm for the initiation and adjustment of therapy: a consensus statement from the Ameri-can Diabetes Association and the European Asso-ciation for the Study of Diabetes. Diabetes Care. 2006; 29: 1963-1972. doi:10.2337/dc06-9912 PubMed

[15] Sambol NC, Brookes LG, Chiang J, Goodman AM, Lin ET, Liu CY, Benet LZ. Food intake and dosage level, but not tablet vs solution dosage form, affect the absorption of metformin HCl in man. Br J Clin Pharmacol. 1996; 42: 510-512. doi:10.1111/j .1365-2125.1996.tb00017.x PubMed

[16] Sambol NC, Chiang J, O’Conner M, Liu CY, Lin ET, Goodman AM, Benet LZ, Karam JH. Pharma-cokinetics and pharmacodynamics of metformin in healthy subjects and patients with noninsulin-dependent diabetes mellitus. J Clin Pharmacol. 1996; 36: 1012-1021. doi:10.1177/009127009603601105 PubMed

[17] Schuirmann DJ. A comparison of the two one-sid-ed tests procedure and the power approach for as-sessing the equivalence of average bioavailability. J Pharmacokinet Biopharm. 1987; 15: 657-680. doi:10.1007/BF01068419 PubMed

[18] Shin KH, Kim SE, Yoon SH, Cho JY, Jang IJ, Shin SG, Yu KS. Pharmacokinetic comparison of a new sustained-release formulation of glimepiride/met-formin 1/500 mg combination tablet and a sus-tained-release formulation of glimepiride/metfor-min 2/500 mg combination tablet in healthy Korean male volunteers: a randomized, 2-se-quence, 2-period, 2-treatment crossover study. Clin Ther. 2011; 33: 1809-1818. doi:10.1016/j.clinthera.2011.10.003 PubMed

[19] Tucker GT, Casey C, Phillips PJ, Connor H, Ward JD, Woods HF. Metformin kinetics in healthy sub-jects and in patients with diabetes mellitus. Br J Clin Pharmacol. 1981; 12: 235-246. PubMed