Assessment of the Pharmacokinetic Interaction Between the Novel DPP-4 Inhibitor Linagliptin Andasulfonylurea,Glyburide,Inhealthysubjects

Drug Metab. Pharmacokinet. 26 (2): 123129 (2011). Copyright © 2011 by the Japanese Society for the Study of Xenobiotics (JSSX) Regular Article Assessment of the Pharmacokinetic Interaction between the Novel DPP-4 Inhibitor Linagliptin andaSulfonylurea,Glyburide,inHealthySubjects

Ulrike GRAEFE-MODY1,*, Peter ROSE2,ArneRING2,KerstinZANDER2, Mario IOVINO2 and Hans-Juergen WOERLE1 1Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim, Germany 2Boehringer Ingelheim Pharma GmbH & Co. KG, Biberach, Germany

Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk

Summary: The aim of this study was to investigate the effect of the dipeptidyl peptidase-4 inhibitor linagliptin on the pharmacokinetics of glyburide (a CYP2C9 and CYP3A4 substrate) and vice versa.This randomized, open-label, three-period, two-way crossover study examined the effects of co-administration of multiple oral doses of linagliptin (5 mg/day © 6 days) and single doses of glyburide (1.75 mg/day © 1day)on the relative bioavailability of either compound in healthy subjects (n = 20, age 1855 years). Coadmin- istration of glyburide did not alter the steady-state pharmacokinetics of linagliptin. Geometric mean ratios

(GMRs) [90% CI] for (linagliptin + glyburide)/linagliptin AUC¸,ss and Cmax,ss were 101.7% [97.7105.8%] and 100.8% [89.0114.3%], respectively. For glyburide, there was a slight reduction in exposure of ³14%

when coadministered with linagliptin (GMRs [90% CI] for (glyburide + linagliptin)/glyburide AUC0¨ and Cmax were 85.7% [79.892.1%] and 86.2% [79.693.3%], respectively). However, this was not seen as clinically relevant due to the absence of a reliable doseresponse relationship and the known large pharmacokinetic interindividual variability of glyburide. These results further support the assumption that linagliptin is not a clinically relevant inhibitor of CYP2C9 or CYP3A4 in vivo. Coadministration of linagliptin and glyburide had no clinically relevant effect on the pharmacokinetics of linagliptin or glyburide. Both agents were well tolerated and can be administered together without the need for dosage adjustments.

Keywords: dipeptidyl peptidase-4 inhibitor; drug interaction; glibenclamide; glyburide; linagliptin; pharmacokinetics; type 2 diabetes

proteolysis of the incretin hormone glucagon-like peptide-1 Introduction ¥ ¤GLP-1¥.3 Linagliptin attains maximum plasma concentra- Type 2 diabetes mellitus ¤T2DM¥ is a progressive disease tions approximately 1.5 h after dosing at a dose of 5 mg.4¥ associated with worsening hyperglycemia, increased periph- The non-speciﬁc protein binding of linagliptin is in the region eral insulin resistance, impaired insulin secretion, and of 70®80%.5¥ In addition, linagliptin binds tightly to DPP-4, reduced pancreatic Ç-cell mass.1¥ Combining two oral which is, however, saturated at low linagliptin concen- 6¥ hypoglycemic agents generally provides greater reductions trations, with an EC50 of 2.82 nM. Due to this tight binding in blood glucose concentrations and/or more sustained to DPP-4, protein binding is concentration dependent and, periods of glycemic control than monotherapy.2¥ Therefore, at very low concentrations, less than 1% of the total combinations of linagliptin and sulfonylureas such as linagliptin is unbound in plasma. Linagliptin therefore shows glyburide appear to be rational. target-mediated, nonlinear disposition kinetics.7,8¥ It has a Linagliptin ¤BI 1356¥ is a novel oral dipeptidyl peptidase-4 long terminal half-life of over 130 h, which is related to the ¤DPP-4¥ inhibitor, which at a clinical dose of 5 mg reduces tight binding to DPP-4 and which is not sensitive to changes blood glucose concentrations by preventing the rapid in absorption or elimination of unbound linagliptin caused by

Received; September 3, 2010, Accepted; October 20, 2010 J-STAGE Advance Published Date: November 12, 2010, doi:10.2133/dmpk.DMPK-10-RG-091 *To whom correspondence should be addressed: Dr. Ulrike GRAEFE-MODY, Therapeutic Area Metabolism, Boehringer Ingelheim Pharma GmbH & Co. KG, Binger Strasse 173, D 55216 Ingelheim, Germany. Tel. +49 (6132) 77-97480, E-mail: [email protected] This study was supported by Boehringer Ingelheim Pharma GmbH & Co. KG.

123 124 Ulrike GRAEFE-MODY, et al. extrinsic factors. The accumulation half-life, which deter- study; if they had an alcohol or drug abuse problem; if they mines the time to attain steady-state conditions, is relatively smoked more than 10 cigarettes, 3 cigars, or 3 pipes per short at 11.4 h for a 5-mg dose.9¥ Thus, linagliptin steady day, or if they could not refrain from smoking for the state is reached by the third day of administration.9¥ duration of the trial. Subjects were also not permitted to Elimination of linagliptin primarily occurs via non-renal take any herbal remedies within 10 days of the start of pathways, and metabolism was shown to play a subordinate dosing and throughout the study. In the case of adverse role with all metabolites being pharmacologically inactive.10¥ events ¤AEs¥ in need of treatment, concomitant therapy was In vitro, linagliptin does not inhibit any CYP enzymes other permitted. In the case of clinical signs of hypoglycemia or a than being a weak to moderate inhibitor of CYP3A4,10¥ blood glucose level below 55 mg/dL ¤g3.1 mmol/L¥, which did not translate into clinically relevant interactions glucose was to be administered stepwise as appropriate. during a sensitive CYP3A4 study in a clinical interaction For minor pain, paracetamol was allowed. Female subjects study.11¥ No indications of biologically relevant changes in of child-bearing age ¤and their male partners¥ were excluded cytochrome P450 activity were observed in rats following unless willing and able to use appropriate barrier contra- repeated once-daily oral administration of 6 or 60 mg/kg ception. A number of restrictions were imposed on the linagliptin for 4 days. In addition, no evidence of enzyme subjects, including avoidance of excessive physical activity induction ¤CYP1A2, 2B6, or 3A4¥ was found in human during the course of the study and abstention from alcoholic hepatocytes ¤unpublished data¥. Therefore, linagliptin is not beverages, caffeine, juices of certain fruits ¤e.g., apples, an inducer of hepatic cytochrome P450. oranges, and grapefruits¥, methylxanthine-containing drinks Glyburide and other sulfonylureas produce their hypo- or foods ¤coffee, tea, cola, energy drinks, chocolate, etc.¥, glycemic effects by stimulating the secretion of insulin from vegetables from the mustard green family ¤e.g., kale, pancreatic Ç-cells.12¥ By closing ATP-sensitive K-channels broccoli, and watercress¥, and charbroiled meats for 24 h in the Ç-cell plasma membrane, a messenger cascade is preceding the first administration of study medication triggered, which promotes insulin release. Glyburide and on the main study days. Citrus fruits, in particular reduces KATP activity by targeting the sulfonylurea receptor grapefruits and Seville oranges and their juices, were not SUR1.13¥ The usual starting dose of glyburide is 2.5 to 5 mg permitted for 5 days before the first administration of study daily, while those patients who may be more sensitive to medication and until after the last sample from each period hypoglycemic drugs should be started at 1.25 mg daily.14¥ was collected. The subjects were not allowed to eat any The major metabolite of glyburide is a 4-trans-hydroxy foods other than those provided by the study center while derivative; a second metabolite, a 3-cis-hydroxy derivative, admitted to the study center. The subjects had to fast for has also been characterized. These metabolites contribute no 10 h prior to medical laboratory blood sampling and drug significant hypoglycemic action, since they are only weakly administration. active. Glyburide is excreted as metabolites in the bile and All subjects gave written informed consent. The protocol urine ¤approximately 50% via each route¥. The major was approved by the Ethikkommittee der Landesärzte- hepatic isozymes responsible for the formation of the kammer Baden-Württemberg, and the study was conducted metabolites of glyburide are CYP3A4, CYP2C9, CYP2C8, in compliance with the guidelines on good clinical practice and CYP2C19.15¥ A recent study has demonstrated that and with ethical standards for human experimentation CYP3A4 is the most important of these enzymes in the established by the Declaration of Helsinki ¤1996 version¥ and metabolism of glyburide in vitro.16¥ in accordance with applicable regulatory requirements. This randomized, open-label, three-period, two-way Study design: This was a randomized, open-label, crossover study was designed to investigate the effect of two-way crossover, multiple-dose study in healthy sub- linagliptin on the pharmacokinetics of glyburide and vice versa. jects of linagliptin ¤1H-purine-2,6-dione,8-ª¤3R¥-3-amino-1- piperidinyl«-7-¤2-butynyl¥-3,7-dihydro-3-methyl-1-ª¤4-meth- Methods ¥ yl-2-quinazolinyl¥methyl«¥;14 Boehringer Ingelheim Pharma Study participants: This study ¤internal reference GmbH & Co. KG; 5 mg/day¥ and glyburide ¤Aventis number 1218.30¥ was carried out on 10 male and 10 female Pharma Deutschland; 1.75 mg/day¥. subjects, aged 18 to 55 years, who were healthy based on a Oral doses of linagliptin were administered alone for 5 complete medical history, vital signs, 12-lead electrocardio- days ¤Days 1®5 of treatment period A¥, linagliptin plus gram ¤ECG¥, and clinical laboratory tests. Subjects were not glyburide were administered for 1 day ¤Day 6 of treatment enrolled if they had any relevant history of renal, hepatic, period B¥, and glyburide alone was administered for 1 day cardiovascular, gastrointestinal, neurologic, metabolic, or ¤Day 1 of treatment period C¥¤Fig. 1¥. Due to the long hormonal disorders; if they had donated blood, participated elimination half-life of linagliptin,4,7,8¥ treatment A was in another clinical trial, or had taken any prescription or immediately followed by treatment B without a washout non-prescription drugs with a long ¤h24 h¥ half-life within at period. Healthy subjects were randomized ¤1:1¥ to receive least 1 month ¤or ten half-lives of the respective drug, one of two treatment sequences: treatment A then B fol- whichever was longer¥ prior to administration or during the lowed by a drug elimination period of at least 35 days before

Treatment sequence AB_C Treatment B Treatment A 1.75 mg glyburide single dose Washout period of Treatment C 5 mg linagliptin q.d. + 5 mg linagliptin q.d. on ≥35 days 1.75 mg glyburide single dose for 5 days Day 6

Randomization

Treatment B Treatment A Treatment C Washout period of 1.75 mg glyburide single dose 5 mg linagliptin q.d. 1.75 mg glyburide single dose ≥7 days + 5 mg linagliptin q.d. on for 5 days Day 6

Treatment sequence C_AB q.d., once daily

Fig. 1. Schematic diagram of trial design and dosing schedule treatment C ¤ABðC¥, or treatment C followed by washout Bioanalytical methods: Plasma concentration of of at least 7 days before treatment A then B ¤CðAB¥. linagliptin was analyzed using a validated high-performance Following an overnight fast, study medication was liquid chromatography/tandem mass spectrometry ¤HPLC- administered with approximately 240 mL of water to MS/MS¥ assay as described previously.17¥ The calibration subjects in the standing position. Medications were curves of undiluted plasma samples were linear over the administered at 0800 and all doses of study medication range of concentrations from 0.10 to 20.0 nmol/L for were witnessed. Subjects were kept under close medical linagliptin. In-study assay validation at nominal concen- surveillance onsite on all days of pharmacokinetic profiling trations of 0.25, 1.0, and 15.0 nmol/L yielded an assay until 25 h after drug administration ¤from the morning of imprecision and inaccuracy of 2.4 to 7.8% and 5.3 to 9.2%, Day 5 to the morning of Day 8 within treatment period AB respectively. and from the morning of Day 1 until the morning of Day 2 Plasma concentrations of glyburide were analyzed by within treatment period C¥. Medical examination was a fully validated HPLC-MS/MS method by Covance performed at screening ¤within 21 days before administration Laboratories Ltd ¤Harrogate, UK¥, using glimepiride as the of any study medication¥ and at the follow-up visit ¤at least 7 internal standard. The assay included sample clean-up by days after the conclusion of treatment¥. protein precipitation. No interference of endogenous Water was allowed ad libitum except for 1 h before and compounds was seen. The calibration curves of undiluted after drug administration. Standardized meals were served at plasma samples were linear over the range of concentrations 1 h, 4 h, 9 h, and 13 h following drug administration on in- from 1.00 to 400 ng/mL using a plasma volume of 25 µL. house days. The observation time after the final drug Imprecision and inaccuracy results for glyburide at nominal administration was at least 8 days. The subjects were not concentrations of 3.0, 160.0, and 320.0 ng/mL were 5.1 to allowed to lie down for 2 h following drug administration, 10.4% and 0.6 to 5.0%, respectively. except for medical examination, carrying out an ECG, or Plasma concentrations of linagliptin are given in nM in assessing vital signs. order to be comparable with previously published data Pharmacokinetic assessments: For quantification of on linagliptin4,7®9¥ ¤1 ng/mL is 2.12 nM¥, whereas plasma the plasma concentrations of the analyte linagliptin, blood concentrations of glyburide are given in ng/mL for easier was taken from a forearm vein into an ethylenediaminetetra- comparison with historical data. acetic acid ¤EDTA¥-anticoagulant blood-drawing tube. For Pharmacokinetic methods: Pharmacokinetic analy- quantification of the plasma concentrations of glyburide, ses were carried out by non-compartmental analysis of the blood samples were taken from a forearm vein into Sarstedt plasma concentration®time data using WinNonlin software S-Monovette plasma lithium-heparin tubes. Blood samples ¤Pharsight Corporation, Mountain View, California¥. The ¤ ¥ were taken at the following time points during treat- maximum concentration Cmax or Cmax,ss and the time to fi ¤ ¥ ment AB for pharmacokinetic pro ling: at 15, 30, and maximum concentration tmax or tmax,ss values were obtained 45 min and at 1, 1.5, 2, 3, 4, 6, 8, and 12 h following by inspection of the plasma concentration data. Actual administration on Day 5 ¤linagliptin alone¥ and Day 6 sampling times were used for the pharmacokinetic analysis. ¤linagliptin and glyburide¥, and at predose or in the morning The apparent terminal rate constant ¤Ð¥ was estimated by on Days 1, 3, 4, 5, 6, and 7. For treatment C, blood regression of the terminal log-linear portion ¤determined by samples were taken for pharmacokinetic profiling at 15 min inspection¥ of the plasma concentration®time profile using ¤ ¥ predose, at 15, 30, and 45 min, and at 1, 1.5, 2, 3, 4, 6, 8, the last three available data points; the half-life t1/2 was and 12 h after administration of glyburide on Day 1, and in calculated as the quotient of ln¤2¥ and Ð. The area under the the morning on Days 2 and 3. The EDTA or lithium-heparin plasma concentration®time curve ¤AUC¥ was calculated anticoagulated blood samples were centrifuged immediately using the linear trapezoidal method for ascending concen- at 2000®4000 ' g for 10 min at 4®8ôC. Plasma was trations and the log trapezoidal method for descending collected in two aliquots ¤each containing at least 0.6 mL concentrations. The steady-state area under the plasma drug ¥ % ô ® Ø¤ ¥ plasma and frozen immediately at or below 20 C. concentration time curve over dosing interval AUCØ,ss

Copyright © 2011 by the Japanese Society for the Study of Xenobiotics (JSSX) 126 Ulrike GRAEFE-MODY, et al. of linagliptin was calculated using the extrapolated or 180 Linagliptin + Glyburide (n = 19) interpolated concentration at time point Ø. AUC ®W of 150 0 Glyburide (n = 19) glyburide was estimated as the sum of the AUC to the last 120 measured concentration and the extrapolated area given by the quotient of the last predicted concentration and Ð. 90 Statistical analysis: Statistical analysis was carried out 60 on the treated set for safety analysis and evaluation of demographics and baseline characteristics, and on the 30 Glyburide plasma conc. [ng/mL] 0 pharmacokinetic analysis set. All 20 subjects were included 0 4 8 12 in the treated set. One subject withdrew after taking Time [hours] ¤ ¥ glyburide alone treatment C ; this subject was excluded Fig. 2. Geometric mean glyburide plasma concentrations after from the pharmacokinetic analysis, resulting in 19 subjects in administration of a single dose of 1.75 mg glyburide given alone the pharmacokinetic set. or in combination with 5 mg linagliptin The relative bioavailability of a single dose of glyburide with and without concurrent linagliptin and the relative Table 1. Geometric mean (% geometric coefficient of bioavailability of linagliptin at steady state with and without variation) non-compartmental pharmacokinetic parameters concurrent glyburide were determined. AUC0®W,Cmax, of glyburide after oral administration of a single dose of 1.75 mg glyburide alone and in combination with 5 mg AUCØ,ss, and Cmax,ss were natural log transformed before fitting the ANOVA model, which included effects accounting linagliptin for úsequence,ûúsubjects within sequences,ûúperiod,û and Glyburide Adjusted GMR ¦ Glyburide ¤ ¥ útreatment.ûúSubjects within sequencesû were considered Parameter Linagliptin 90% CI random, with the other effects considered fixed. For tests on Test ¤n © 19¥ Reference ¤n © 19¥ Test/Reference, % subject, period, and treatment effects, the denominator sum Glyburide of squares was the sum of squares for error. For tests on ª & « ¤ ¥ ¤ ¥ ¤ ® ¥ AUC0®W ng h/mL 299 34.0 348 32.7 85.7 79.8 92.1 sequence effects, the denominator was the sum of squares ª « ¤ ¥ ¤ ¥ ¤ ® ¥ Cmax ng/mL 113 28.2 131 32.2 86.2 79.6 93.3 for the subjects. All other parameters were analyzed ª «a ¤ ® ¥ ¤ ® ¥ tmax h 1.50 1.00 2.02 1.50 1.00 2.02 descriptively. ª « ¤ ¥ ¤ ¥ The difference between the expected log means was t1/2 h 2.09 53.8 2.28 42.7 a ¤ ® ¥ For tmax, the median and range min max is given. estimated by the difference in the corresponding least square fi means ¤point estimate¥. Two-sided 90% confidence intervals GMR, geometric mean ratio; CI, con dence interval. ¤CIs¥ based on the t distribution were computed. These ¤ ¦ quantities underwent back-transformation to the original The adjusted glyburide Cmax GMR glyburide linagliptin/ scale to give the point estimator ¤geometric mean¥ and glyburide¥ was 86.2% ¤90% CI: 79.6®93.3%¥. The adjusted ¤ ® interval estimates for the median intra-subject ratio between glyburide AUC0®W GMR was 85.7% 90% CI: 79.8 the test and reference. The study sample size was not based 92.1%¥. Although the lower limit of the CIs for the mean on a power calculation, but was considered to be adequate to ratios fell just below the usual acceptance range of 80® characterize a potential interaction with sufficient accuracy 125%, the slight reduction in glyburide exposure observed based on previous experience gained in similar studies.11,18¥ ¤approximately 14%¥ when coadministered with linagliptin was not judged to be clinically relevant, and the median t Results max ¤1.5 h¥ was identical for the single and combined regimens. Demographics: Twenty healthy Caucasian subjects, For linagliptin, the geometric mean steady-state plasma aged 19®48 years, were enrolled and 19 of these completed concentration®time profiles ¤Fig. 3¥ and pharmacokinetic all of the treatment and study procedures as outlined in the parameters were similar when administered alone or with fi ¤ ¥ study protocol. Five male and ve female subjects were glyburide Table 2 . Median tmax,ss was 1.52 h when randomized to each sequence ¤ABðCorCðAB¥. One subject linagliptin was given alone and 2.00 h when coadministered randomized to sequence CðAB discontinued the trial with glyburide, but with a similar range of 0.50 to 3.00 h ¤ ¥ prematurely Day 5 of treatment period A because placing for both treatments. The linagliptin AUCØ,ss and Cmax,ss of the venous cannula was not possible. The participantsö GMRs ¤linagliptin ¦ glyburide/linagliptin¥ were 101.7% mean ¤standard deviation¥ age was 35.9 ¤+10.3¥ years, mean ¤90% CI: 97.7®105.8%¥ and 100.8% ¤90% CI: 89.0® weight was 72.7 ¤+10.8¥ kg, and mean body mass index was 114.3%¥, respectively. 23.8 ¤+2.6¥ kg/m2. Safety and tolerability: Linagliptin and glyburide Pharmacokinetic results: The geometric mean were generally well tolerated, whether administered alone glyburide plasma concentration®time profiles with and or coadministered. Only one AE was of severe intensity without linagliptin are shown in Figure 2 and the geometric ¤headache at the screening visit in one subject¥. This AE mean pharmacokinetic parameters are given in Table 1. occurred prior to any study medication and did not require

18 Linagliptin + Glyburide (n = 19) diagnosis. Moreover, despite an expanding armamentarium, = 15 Linagliptin (n 19) around half of T2DM patients still show HbA1c levels in 19¥ 12 excess of 7%. Against this background, there is a pressing

9 need for novel hypoglycemic drugs with distinct mechanisms

6 of action. Glyburide and linagliptin, for example, both increase insulin secretion, but influence pancreatic Ç-cell 3 function through distinct mechanisms. Linagliptin is safe, Linagliptin plasma conc. [nmol/L] 0 well tolerated, and effective as monotherapy or in 0 4 8 12 16 20 24 4,9,11,20¥ Time [hours] combination with metformin. Sulfonylureas remain one of the most commonly prescribed drugs used to manage Fig. 3. Geometric mean linagliptin steady-state plasma concen- 21¥ trations after administration of 5 mg linagliptin monotherapy for T2DM; however, as sulfonylureas can cause severe 5 days or in combination with 1.75 mg glyburide on Day 6 hypoglycemia, appropriate dosing is essential in clinical practice. Therefore, this study assessed the pharmacokinetics of the combination of glyburide and linagliptin in healthy Table 2. Geometric mean (% geometric coefficient of variation) non-compartmental pharmacokinetic parameters volunteers. This multiple-dose study used the expected of linagliptin at steady state following multiple oral admin- therapeutic dose of linagliptin to detect an effect on istration of 5 mg/day linagliptin alone and in combination with glyburide pharmacokinetics. Glyburide has a wide ther- a single dose of 1.75 mg glyburide apeutic window ¤clinical maintenance doses range from 1.25 to 20 mg daily¥,14¥ and in this study, the 1.75-mg dose Linagliptin Adjusted GMR ¦ Linagliptin ¤ ¥ Parameter Glyburide 90% CI was chosen to lower the risk of hypoglycemia. With this Test ¤n © 19¥ Reference ¤n © 19¥ Test/Reference, % dose, mean maximum concentrations of 131 ng/mL were obtained, which is even higher than the 109 ng/mL observed Linagliptin after single-dose administration of 5 mg glyburide in a ª & « ¤ ¥ ¤ ¥ ¤ ® ¥ ¥ AUCØ,ss nmol h/L 168 22.8 166 22.3 101.7 97.7 105.8 comparable study.22 This indicates that the exposure ª « ¤ ¥ ¤ ¥ ¤ ® ¥ Cmax,ss nmol/L 12.9 36.0 12.7 31.9 100.8 89.0 114.3 reached with a dose of 1.75 mg was adequate to detect a ª «a ¤ ® ¥ ¤ ® ¥ tmax,ss h 2.00 0.500 3.00 1.52 0.500 3.00 possible interaction with linagliptin. Despite comparable a ¤ ® ¥ For tmax,ss, the median and range min max is given. maximum concentrations, glyburide plasma concentrations GMR, geometric mean ratio; CI, confidence interval. declined more rapidly in this study than in the study by Karim et al.,22¥ resulting in concentrations falling below the lower limit of quantification of 1 ng/mL beyond 12 h. The treatment intervention; all other AEs were of mild or determined apparent terminal half-lives of 2.1 and 2.3 h moderate intensity. All subjects recovered and study indicated that even increasing the assay sensitivity by a factor medication intake was completed in all cases. One subject of 10 would not have yielded detectable concentrations discontinued the trial prematurely as the investigational staff beyond 24 h, and therefore would not have added significant could not place the venous cannula. information. However, when interpreting the concentra- Six subjects ¤30.0%¥ reported AEs that were considered tion®time profiles, assay sensitivity was not considered to to be drug related. For linagliptin, three subjects ¤15.0%¥ affect the overall study results and interpretation. reported three related AEs ¤headache in two subjects, Administration of a single oral dose of glyburide had sensation of heaviness in one subject¥. For linagliptin ¦ no clinically meaningful effect on the steady-state pharma- glyburide, four subjects ¤21.1%¥ reported five related AEs cokinetics of linagliptin in this study. Glyburide was not ¤hypoglycemia in two subjects, dizziness in two subjects, and expected to significantly alter the pharmacokinetic profile of somnolence in one subject¥. For glyburide, four subjects linagliptin or vice versa. Nevertheless, concomitant admin- ¤20.0%¥ reported four related AEs ¤hypoglycemia in one istration of glyburide and linagliptin reduced the glyburide ¥ subject and headache in three subjects . Two cases of GMRs for Cmax and AUC0®W by approximately 14%. ¤ hypoglycemia one during glyburide monotherapy and one Glyburide median tmax was comparable for both treatments. during combination treatment¥ required treatment with a In view of the absence of a predictable plasma concen- single dose of glucose. Physical examinations, vital signs, 12- tration®response relationship for sulfonylureas in general, lead resting ECGs, and laboratory analyses revealed no and for glyburide in particular,23¥ a 14% reduction in clinically relevant changes. glyburide exposure is considered to have no clinical relevance for the glycemic efficacy or safety of patients Discussion receiving linagliptin and glyburide concomitantly. Although The progressive nature of T2DM means that most patients CIs in the range 80®125% indicate no interaction between eventually require multiple therapies to obtain adequate drugs, the clinical relevance must be judged on an individual glycemic control.2¥ Indeed, around half of T2DM patients drug basis. The FDA Draft Guidance on Drug Interactions require more than one hypoglycemic agent within 3 years of ¤2006¥ states that úwhen the 90% CIs for systemic exposure

Copyright © 2011 by the Japanese Society for the Study of Xenobiotics (JSSX) 128 Ulrike GRAEFE-MODY, et al. ratios fall entirely within the equivalence range of 80®125%, seen in previous clinical studies involving healthy subjects standard Agency practice is to conclude that no clinically and patients with T2DM.4,9®11,18,26¥ In a recently reported significant differences are present. This is, however, a very phase III study of linagliptin as add-on therapy to a conservative standard and a substantial sample would need sulfonylurea in patients with inadequately controlled to be studied to meet it.û24¥ In this study with glyburide, T2DM, linagliptin produced a statistically significant and there is no safety issue in terms of toxicity, as the exposure clinically relevant reduction in HbA1c from baseline to the drug was reduced when coadministered with compared with sulfonylurea plus placebo after 24 weeks of linagliptin. In terms of efficacy, a 14% reduction in exposure treatment.27¥ The study patientsö sulfonylurea drug was was considered not clinically relevant as there is a high administered during the entire trial duration ¤including interindividual variability with this drug.16,23¥ washout and placebo run-in periods¥ in an unchanged dosage Any inhibitory effects of linagliptin on metabolizing and linagliptin was added to this background therapy at a enzymes ¤such as cytochromes¥ or efflux transporters ¤e.g., once-daily dose of 5 mg. The incidence of hypoglycemia P-gp¥ that are involved in the absorption, disposition, or was low and occurred at a similar rate in the two treatment elimination of glyburide would be expected to result in an arms ¤5.6 vs. 4.8% for the combination of sulfonylurea increase of glyburide exposure. Because a minor reduction in and linagliptin or placebo, respectively¥. Therefore, adding glyburide exposure was observed resulting from linagliptin linagliptin to sulfonylurea treatment did not increase the risk co-medication, any relevant inhibitory effects on these of hypoglycemia, and the results of this 24-week trial enzymes or transporters can be excluded. indicate that linagliptin can be used safely as an add-on The data from this study demonstrate that linagliptin therapy in patients with insufficient glycemic control had no relevant effect on the CYP2C9-metabolized drug receiving treatment with a sulfonylurea.27¥ glyburide, and therefore it appears that linagliptin would be In summary, steady-state concentrations of linagliptin unlikely to affect the pharmacokinetics of other drugs have no relevant impact on the pharmacokinetics of primarily metabolized by this pathway. CYP2C9 contributes glyburide. The minor changes in glyburide pharmacokinetics to the metabolism of several widely used agents, including observed in the present study are not expected to non-steroidal anti-inflammatory drugs, oral anticoagulants, meaningfully alter the glycemic efficacy or safety of and oral hypoglycemics.23,25¥ As linagliptin did not alter the glyburide. Linagliptin was well tolerated as monotherapy pharmacokinetics of glyburide, pharmacokinetic interactions and in combination with glyburide, with no drug®drug with other sulfonylureas ¤such as glipizide¥ and other drugs interaction potential with medications metabolized predom- primarily metabolized by CYP2C9 are not expected. A inately by CYP2C9 or CYP3A4. recent study on the contribution of human cytochrome P450 enzymes to glyburide metabolism found that the primary Acknowledgments: Boehringer Ingelheim would like enzyme involved in the metabolism of glyburide in vitro was to thank the subjects and staff who participated in this study. CYP3A4.16¥ When linagliptin was incubated with recombi- This work was supported by Boehringer Ingelheim. Writing nant human P450 enzymes, the only enzyme that was active and editorial assistance was provided by Stephanie Milsom of in metabolizing linagliptin was CYP3A4.10¥ In the present PHASE II International, which was contracted by Boehringer study, if linagliptin were competitively inhibiting the Ingelheim for these services. metabolism of glyburide by CYP3A4 in vivo, an increase in Data from this study have previously been published, in glyburide levels would have been expected, in contrast to part, as Poster No. PI-66 at the 111th Annual Meeting of the the 14% decrease in exposure reported here. A drug®drug American Society for Clinical Pharmacology and Therapeu- interaction study between linagliptin and simvastatin ¤a tics, March 17®20, 2010, Atlanta, GA. CYP3A4 substrate¥ reported no clinically relevant effect on the pharmacokinetics of simvastatin and concluded that References linagliptin coadministration is not expected to exert a 1¥ Kahn, S. E., Zraika, S., Utzschneider, K. M. and Hull, R. L.: The clinically relevant effect on the pharmacokinetics of other beta cell lesion in type 2 diabetes: there has to be a primary CYP3A4 substrates. functional abnormality. Diabetologia, 52: 1003®1012 ¤2009¥. This study involved healthy volunteers, and the hypo- 2¥ Nathan, D. M., Buse, J. B., Davidson, M. B., Ferrannini, E., glycemic effects of linagliptin were not investigated. Holman, R. R., Sherwin, R. and Zinman, B.: Medical management of hyperglycemia in type 2 diabetes: a consensus algorithm for the Nevertheless, no hypoglycemia emerged in this cohort of initiation and adjustment of therapy: a consensus statement of the healthy subjects with normoglycemia during treatment with American Diabetes Association and the European Association for linagliptin alone. The two cases of hypoglycemia occurred the Study of Diabetes. Diabetes Care, 32: 193®203 ¤2009¥. with glyburide monotherapy and combination treatment. 3¥ Thomas, L., Eckhardt, M., Langkopf, E., Tadayyon, M., ¤ ¥ ¤ ¥ This suggests that glyburide was responsible for the Himmelsbach, F. and Mark, M.: R -8- 3-amino-piperidin-1-yl - 7-but-2-ynyl-3-methyl-1-¤4-methyl-quinazolin-2-ylmethyl¥-3,7-di- hypoglycemia. Multiple doses of linagliptin as monotherapy hydro-purine-2,6-dione ¤BI 1356¥, a novel xanthine-based dipep- and with a single dose of glyburide were well tolerated. This tidyl peptidase 4 inhibitor, has a superior potency and longer is in agreement with the excellent tolerability of linagliptin duration of action compared with other dipeptidyl peptidase 4

inhibitors. J. Pharmacol. Exp. Ther., 325: 175®182 ¤2008¥. metabolism. Biopharm. Drug Dispos., 31: 228®242 ¤2010¥. 4¥ Hüttner, S., Graefe-Mody, E. U., Withopf, B., Ring, A. and Dugi, 17¥ Eckhardt, M., Langkopf, E., Mark, M., Tadayyon, M., Thomas, K. A.: Safety, tolerability, pharmacokinetics, and pharmacody- L., Nar, H., Pfrengle, W., Guth, B., Lotz, R., Sieger, P., Fuchs, namics of single oral doses of BI 1356, an inhibitor of dipeptidyl H. and Himmelsbach, F.: 8-¤3-¤R¥-aminopiperidin-1-yl¥-7-but- peptidase 4, in healthy male volunteers. J. Clin. Pharmacol., 48: 2-ynyl-3-methyl-1-¤4-methyl-quinazolin-2-ylmethyl¥-3,7-dihydro- 1171®1178 ¤2008¥. purine-2,6-dione ¤BI 1356¥, a highly potent, selective, long-acting, 5¥ Fuchs, H., Tillement, J.-P., Urien, S., Greischel, A. and Roth, and orally bioavailable DPP-4 inhibitor for the treatment of type 2 W.: Concentration-dependent plasma protein binding of the novel diabetes. J. Med. Chem., 50: 6450®6453 ¤2007¥. dipeptidyl peptidase 4 inhibitor BI 1356 due to saturable binding to 18¥ Graefe-Mody, E. U., Padula, S., Ring, A., Withopf, B. and Dugi, its target in plasma of mice, rats and humans. J. Pharm. Pharmacol., K. A.: Evaluation of the potential for steady-state pharmacokinetic 61:55®62 ¤2009¥. and pharmacodynamic interactions between the DPP-4 inhibitor 6¥ Retlich, S., Duval, V., Ring, A., Staab, A., Hüttner, S., Jungnik, linagliptin and metformin in healthy subjects. Curr. Med. Res. Opin., A., Jaehde, U., Dugi, K. A. and Graefe-Mody, U.: Pharmacoki- 25: 1963®1972 ¤2009¥. netics and pharmacodynamics of single rising intravenous doses 19¥ Kendall, D. M., Cuddihy, R. M. and Bergenstal, R. M.: Clinical ¤0.5 mg®10 mg¥ and determination of absolute bioavailability of the application of incretin-based therapy: therapeutic potential, patient dipeptidyl peptidase-4 inhibitor linagliptin ¤BI 1356¥ in healthy male selection and clinical use. Eur. J. Intern. Med., 20: S329®S339 subjects. Clin. Pharmacokinet., 49: 829®840 ¤2010¥. ¤2009¥. 7¥ Retlich, S., Duval, V., Graefe-Mody, U., Jaehde, U. and Staab, A.: 20¥ Forst, T., Uhlig-Laske, B., Ring, A., Graefe-Mody, U., Friedrich, Impact of target-mediated drug disposition on linagliptin pharma- C., Herbach, K., Woerle, H.-J. and Dugi, K. A.: Linagliptin ¤BI cokinetics and DPP-4 inhibition in type 2 diabetic patients. J. Clin. 1356¥, a potent and selective DPP-4 inhibitor, is safe and efficacious Pharmacol., 50: 873®885 ¤2010¥. in combination with metformin in patients with inadequately 8¥ Retlich, S., Withopf, B., Greischel, A., Staab, A., Jaehde, U. and controlled type 2 diabetes. Diabet. Med., 27: 1409®1419 ¤2010¥. Fuchs, H.: Binding to dipeptidyl peptidase-4 determines the 21¥ Tzoulaki, I., Molokhia, M., Curcin, V., Little, M. P., Millett, disposition of linagliptin ¤BI 1356¥®investigations in DPP-4 C. J., Ng, A., Hughes, R. I., Khunti, K., Wilkins, M. R., Majeed, deficient and wildtype rats. Biopharm. Drug Dispos., 30: 422®436 A. and Elliott, P.: Risk of cardiovascular disease and all cause ¤2009¥. mortality among patients with type 2 diabetes prescribed oral 9¥ Heise, T., Graefe-Mody, E. U., Hüttner, S., Ring, A., antidiabetes drugs: retrospective cohort study using UK general Trommeshauser, D. and Dugi, K. A.: Pharmacokinetics, pharma- practice research database. BMJ, 339: b4731 ¤2009¥. codynamics and tolerability of multiple oral doses of linagliptin, a 22¥ Karim, A., Laurent, A., Munsaka, M., Wann, E., Fleck, P. and dipeptidyl peptidase-4 inhibitor in male type 2 diabetes patients. Mekki, Q.: Coadministration of pioglitazone or glyburide and Diabetes Obes. Metab., 11: 786®794 ¤2009¥. alogliptin: pharmacokinetic drug interaction assessment in healthy 10¥ Blech, S., Ludwig-Schwellinger, E., Graefe-Mody, E. U., Withopf, participants. J. Clin. Pharmacol., 49: 1210®1219 ¤2009¥. B. and Wagner, K.: The metabolism and disposition of the oral 23¥ Kirchheiner, J., Brockmoller, J., Meineke, I., Bauer, S., Rohde, dipeptidyl peptidase-4 inhibitor, linagliptin, in humans. Drug Metab. W., Meisel, C. and Roots, I.: Impact of CYP2C9 amino acid Dispos., 38: 667®678 ¤2010¥. polymorphisms on glyburide kinetics and on the insulin and glucose 11¥ Graefe-Mody, U., Huettner, S., Stähle, H., Ring, A. and response in healthy volunteers. Clin. Pharmacol. Ther., 71: 286®296 Dugi, K. A.: Effect of linagliptin ¤BI 1356¥ on the steady-state ¤2002¥. pharmacokinetics of simvastatin. Int. J. Clin. Pharmacol. Ther., 48: 24¥ FDA Draft Guidance on Drug Interactions. US FDA Center for 367®374 ¤2010¥. Biologics Evaluation and Research ¤CBER¥, Sept 2006. Available 12¥ Ashcroft, F. M.: Mechanisms of the glycaemic effects of at: http://www.fda.gov/downloads/Drugs/GuidanceCompliance sulfonylureas. Horm. Metab. Res., 28: 456®463 ¤1996¥. RegulatoryInformation/Guidances/UCM072119.pdf ¤accessed 13 13¥ Vila-Carriles, W. H., Zhao, G. and Bryan, J.: Defining a binding Oct 2010¥. pocket for sulfonylureas in ATP-sensitive potassium channels. 25¥ Rettie, A. E. and Jones, J. P.: Clinical and toxicological relevance FASEB J., 21:18®25 ¤2007¥. of CYP2C9: drug-drug interactions and pharmacogenetics. Annu. 14¥ Sanofi-aventis US LLC. DIAßETA¬ ¤Glyburide USP tablets¥ Rev. Pharmacol. Toxicol., 45: 477®494 ¤2005¥. Prescribing information, July 2009. Available at: http://products. 26¥ Deacon, C. F. and Holst, J. J.: Linagliptin, a xanthine-based sanofi-aventis.us/diabeta/diabeta.pdf ¤accessed 27 Aug 2010¥. dipeptidyl peptidase-4 inhibitor with an unusual profile for the 15¥ Zharikova, O. L., Fokina, V. M., Nanovskaya, T. N., Hill, R. A., treatment of type 2 diabetes. Expert Opin. Investig. Drugs, 19: 133® Mattison, D. R., Hankins, G. D. and Ahmed, M. S.: Identification 140 ¤2010¥. of the major human hepatic and placental enzymes responsible for 27¥ Lewin, A. J., Arvay, L., Liu, D., Patel, S. and Woerle, H.-J.: the biotransformation of glyburide. Biochem. Pharmacol., 78: 1483® Safety and efficacy of linagliptin as add-on therapy to a 1490 ¤2009¥. sulphonylurea in inadequately controlled type 2 diabetes. Presented 16¥ Zhou, L., Naraharisetti, S. B., Liu, L., Wang, H., Lin, Y. S., at the European Association for the Study of Diabetes 46th Annual Isoherranen, N., Unadkat, J. D., Hebert, M. F. and Mao, Q.: Meeting, Stockholm, Sweden. 20®24 September 2010, poster no. Contributions of human cytochrome P450 enzymes to glyburide 821.