Journal of Clinical Pharmacy and Therapeutics (2006) 31, 469-476
`
`ORIGINAL ARTICLE
`
`Pharmacokinetic and pharmacodynamic modelling of the
`effects of glimepiride on insulin secretion and glucose
`lowering in healthy humans
`H.-Y. Yun* MS, H.-C. Park* MS, W. Kangft PhD and K.-I. Kwon* PhD
`*College of Pharmacy, Chungnam National University, Daejeon, Korea and +Department of pharmacy,
`Catholic University of Daegu, Daegu, Korea
`
`SUMMARY
`
`glimepiride and its insulin secretion and hypo-
`glycaemic effects in healthy volunteers.
`
`INTRODUCTION
`
`Glimepiride is an oral sulfonylurea antihyper-
`glycaemic agent. We used pharmacokinetic—
`Keywords: glimepiride, pharmacodynamic, phar-
`macokinetic, pharmacokinetic-pharmacodynamic
`pharmacodynamic (PK-PD) modelling to analyse
`the relationship between plasma glimepiride
`modelling
`
`concentration, insulin secretion and_glucose
`lowering to determine the effects of the drug in
`healthy volunteers. A single 2-mg oral dose of
`1-[[p-[2-@-ethyl-4-methyl-2-oxo-3-
`Glimepiride,
`glimepiride was administered to six healthy vol-
`pyrroline-1-carboxamido)ethyl]|phenyl]sulfonyl]-
`unteers. The control group received a placebo. All
`3-(trans-4-methylcyclohexyl) urea, is an oral sulfo-
`subjects consumed 12g of sugar immediately
`nylurea antihyperglycaemic agent that contains a
`after drug administration in order to standardize
`sulfonylurea nucleus and a cyclohexyl ring (1).
`the initial plasma glucose levels. Serial blood
`Glimepiride may be given oncedaily.It has a long-
`sampling was performed for 9 h after oral dosing.
`lasting effect without markedly increasing plasma
`Plasma glimepiride, insulin and glucose levels
`insulin compared with other sulfonylureas (2).
`were determined by validated methods (LC/MS/
`Glimepiride’s majorsite of action is thoughtto be a
`MSassay, hexokinase method and radioimmu-
`membrane receptor on pancreatic B-cells, whereit
`noassay respectively). Time courses of plasma
`acts via ATP-regulated potassium (Karp) channels
`glimepiride concentration, insulin secretion, and
`to cause membrane depolarization and insulin
`glucose lowering effects were analysed by means
`release (3). The association rate of glimepiride is
`of PK-PD modelling with the ADAPT II pro-
`2:5- to 3-fold that of glibenclamide, its dissociation
`gram. The time course of the plasma concentra-
`rate is 8- to 9-fold that of glibenclamide, and its
`tions followed a two-compartmental model with a
`in vitro binding affinity for rat B-cell tumour and
`lag time. The glimepiride concentration peaked at
`insulinomacells is 2:5- to 3-fold lower than that of
`191:5 ng/mL at approximately 4 h after adminis-
`tration. The maximal increase in insulin secretion
`glibenclamide (4). Sulfonylureas interact with dif-
`was 9:98 mIU/L and the maximal decrease in
`ferent sites on the pancreatic f-cell membrane.
`Glimepiride binds to a 65-kDa protein, whereas
`plasma glucose was 19:33 mg/dL. Both peak
`glibenclamide binds to a 140-kDa protein. These
`effects occurred at approximately 2°5 h after drug
`proteins are believed to both be part of the same
`intake. The glucose disappearance model was
`sulfonylurea receptor, as each agent inhibits the
`used to analyse glimepiride’s insulin secretion
`other’s binding to its protein target (5).
`and glucose lowering effects. The PK-PD model
`In single-dose studies with healthy volunteers,
`described well the relationship between plasma
`the peak glimepiride plasma concentration (Cmax)
`and the area under the plasma concentration—time
`curve (AUC) were generally dose-proportional. In
`12 healthy volunteers, the Cymax rose linearly from
`
`Received 24 March 2005, Accepted 20 June 2006
`Correspondence: Kwang-Il Kwon, College of Pharmacy, Chung-
`nam National University, Daejeon 305-764, Korea. Tel.: 82 42 821
`5937; fax: 82 42 823 6781; e-mail: kwon@cnu.ac.kr
`
`© 2006 Blackwell Publishing Ltd
`
`469
`
`MPI EXHIBIT 1046 PAGE1
`
`MPI EXHIBIT 1046 PAGE 1
`
`MPI EXHIBIT 1046 PAGE 1
`
`DR. REDDY’S LABORATORIES, INC.
`IPR2024-00009
`Ex. 1046, p. 1 of 7
`
`

`

`470
`
`=~=4#H.-Y. Yun et al.
`
`103-2 to 550-8 ug/L as the dose increased from 1 to
`8 mg, whereas the AUC increased from 339 to
`2634 pg h/L. The glimepiride Cmax occurred at 0:7—
`2:8 h (Tmax) after the single-dose administration to
`healthy volunteers (6). Glimepiride plasma protein
`binding was 99:-4% (7), and the volumeof distri-
`bution (Vq) was 8:8 L (8).
`From dose-ranging studies in patients with
`type 2 diabetes mellitus, glimepiride appears to
`reduce fasting and postprandial blood glucose
`levels,
`as well
`as glycosylated haemoglobin.
`These effects are dose-dependent over a range of
`1-4 mg daily. For patients receiving the maxi-
`mum daily dose (8 mg), the average reduction in
`glycosylated haemoglobin is 2% in absolute
`units. Age, gender, weight, and race do not affect
`glimepiride’s efficacy (9).
`The main objective of this study was to examine
`the relationship between plasma glimepiride con-
`centration and its insulin secretion and glucose
`lowering effects (i.e. the increase in blood insulin
`and decrease in blood glucose) after
`its oral
`administration to healthy volunteers. This should
`permit prediction of the time course of glimepi-
`ride’s therapeutic and side effect profiles after oral
`dosing. The relationship between the pharmacoki-
`netics of glimepiride andits insulin secretion and
`glucose lowering effects has not yet been analysed.
`Our goal was to assess the usefulness of pharma-
`cokinetic-pharmacodynamic (PK-PD) modelling
`in describing this relationship.
`
`METHODS
`
`Subjects
`
`Six healthy male subjects with a mean age of
`25 years (range = 22-25 years) and a mean weight
`of 69:67 kg (range = 55-83 kg)
`took part
`in this
`study. All subjects underwent a thoroughhistory, a
`complete physical examination, and a battery of
`routine
`laboratory tests
`(haematology,
`serum
`chemistry and urinalysis). None had taken any
`drugs knownto interfere with the study for at least
`10 days beforehand. The exclusion criteria inclu-
`ded health problems, drug or alcohol abuse, and
`abnormalities in laboratory screening tests. All
`subjects were told the full details of the study and
`gave written informed consent. The study was
`approved by the local ethics committee.
`
`Study design
`
`All subjects fasted for at least 10 h before taking
`their study medication. At time zero, an intraven-
`ous cannula was inserted into the forearm vein and
`control blood samples were collected.
`First period, six subjects received a single-oral
`dose Amaryl 2 mg and second period, same sub-
`jects received a placebo. There was a 6-day wash-
`out period between the periods. After baseline
`sampling, the test group took glimepiride (Ama-
`ryl® 2 mg tablet; Handok/Aventis Pharma Co. Ltd,
`Seoul, Korea) with 240 mL of water. The control
`group took a placebo with 240 mL of water. All
`subjects consumed 12 g of sugar cubes immedi-
`ately after drug administration in order to prevent
`hypoglycaemia and maintain standard initial
`plasma glucose level. All subjects were given a
`standardized meal 4h after drug administration.
`They were not allowed to remain supine or to sleep
`until 4 h after drug administration.
`insulin and
`Samples for plasma glimepiride,
`glucose determinations were taken at 0-5, 1, 1:5, 2,
`2:5, 3, 4,5, 7 and 9 h after drug administration. The
`samples were collected in heparinized tubes,
`immediately centrifuged (10 min at 1650 g), and
`stored at —80 °C for later analysis.
`
`Plasma assay
`
`Plasma glimepiride was assayed by the reported
`LC/MS/MS method, with a slight modification (1).
`Briefly, 50 wL of internal standard (glibenclamide,
`500 ng/mL) and 0:5 mL of 1 mM NaOH were added
`to 0:5 mL of plasma, followed by a 10-min liquid—
`liquid extraction with 5 mL of ethyl ether : ethyl
`acetate (1 : 1, v/v). The organic layer was separated
`and evaporated to dryness at ambient temperature
`in a Speed-Vac (Savant, Holbrook, NY, USA). The
`residue was reconstituted in 100 “L of acetonitrile
`by vortexing for 15 s; then 5 wL of this solution was
`injected onto the column. The mobile phase was a
`mixture of 01% formic acid buffer : acetonitrile
`
`(20: 80, v/v), and the column was eluted at
`0:2 mL/min with an HP 1100 series pump(Agilent,
`Wilmington, DE, USA). The turbo-ion spray inter-
`face was operated in positive ion mode at 5500 V
`and 350 °C. Using flow injection of a mixtureof all
`analyses, the operating conditions were optimized
`to: nebulizing gas flow, 1:04 L/min; auxiliary gas
`
`© 2006 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 31, 469-476
`
`MPI EXHIBIT 1046 PAGE 2
`
`MPI EXHIBIT 1046 PAGE 2
`
`MPI EXHIBIT 1046 PAGE 2
`
`DR. REDDY’S LABORATORIES, INC.
`IPR2024-00009
`Ex. 1046, p. 2 of 7
`
`

`

`PK/PD modeling of glimepiride
`
`471
`
`kinetic model
`to reflect
`the oral administration
`flow, 40 L/min; curtain gas flow, 1-44 L/min; ori-
`fice voltage, 80 V; ring voltage, 400 V and collision
`and flip-flop kinetics. Model development was an
`gas (nitrogen) pressure, 3:58 x 10° Torr. Quantita-
`iterative process with regard to both the under-
`tion was performed by multiple reaction monitor-
`lying data set and the selected model structure.
`Models were constructed as a series of differen-
`ing of the protonated precursor ion and the related
`product
`ion for glimepiride, using the internal
`tial equations that were solved numerically and
`fitted to the data with the ADAPT II software
`standard method with the peak area ratio. The mass
`transitions used for glimepiride and the internal
`(Biomedical Simulation Resource, Los Angeles,
`standard
`were m/z
`4910 — 3524
`and
`CA, USA)
`(11). The fitting with individual data
`
`4940 — 3694 (20 eV_collisionrespectively
`
`was performed by means of maximum likelihood
`energy, 200 ms dwell time). Quadruples Q1 and Q3
`estimation. The following information was used
`wereset on unit resolution. The analytical data were
`to evaluate the goodness of fit and the quality of
`processed using Analyst Software (version 1:2;
`the parameter estimates: coefficient of variation of
`POETsoftware corporation, London, UK).
`parameter estimates, parameter correlation mat-
`rix, sums of squares of residuals, visual exam-
`For all plasma samples, glucose concentrations
`ination of the distribution of residuals and the
`were determined enzymatically by the hexokinase
`method, and insulin concentrations were deter-
`mined by radioimmunoassay (10).
`
`Akaike Information Criterion (12). The differential
`equations
`that described the changes
`in the
`amounts of glimepiride in the compartments after
`oral administration are given by Eqs (1) to (3):
`
`Pharmacokinetic/pharmacodynamic model and
`data analysis
`
`dx
`Vinaxl2
`(EN x
`1
`a Ge xm
`(1)
`Pharmacokinetics (plasma_glimepiride) and
`
`
`pharmacodynamics (plasma insulin and glucose)
`were modelled sequentially. We developed a
`parsimonious compartmental model that reflects
`the rate of change of glucose as the difference
`between the net hepatic glucose balances. As
`shownin Fig. 1, a two-compartment model with
`a lag time, nonlinear absorption and elimination
`was selected as the most appropriate pharmaco-
`
`2
`
`(2)
`
`(3)
`
`— Ko3
`
`Vimaxl2
`dx2
`es
`dt
`Kmi2 To)
`pense
`— = (SY x ey +h x x
`Vinax2e
`X X — (=) x x
`23
`2 (Gt 2
`dxpo Kas x 32 ~ Kan X83
`
`Fig. 1. The model selected to des-
`cribe the effects of glimepiride on
`insulin secretion and glucose
`lowering in healthy volunteers:1,
`absorption lag time; Vinax and Kyi,
`Michaelis-Menten transport
`parameters; Kj, first-order rate
`constant; K;,, zero-order rate
`constant for insulin production;
`Kour the first-order constant for
`insulin loss.
`
`
`
`2. Central
`
`3. Peripheral
`
`compartment
`compartment
`
`
`
`
`
`4. Insulin
`compartment
`i
`
` 5. Liver
`
`compartment
`
`© 2006 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 31, 469-476
`
`MPI EXHIBIT 1046 PAGE 3
`
`MPI EXHIBIT 1046 PAGE 3
`
`MPI EXHIBIT 1046 PAGE 3
`
`DR. REDDY’S LABORATORIES, INC.
`IPR2024-00009
`Ex. 1046, p. 3 of 7
`
`

`

`472 H.-Y. Yun et al.
`
`First-order rate constants describing intercom-
`partmental transport are denoted by Kj, and the
`active transport with Michaelis-Menten type kin-
`etics is characterized by the apparent maximal
`transport rates Vmax; and the apparent Michaelis
`constants Kyyjj
`(13). After the parameters of the
`pharmacokinetic model were fixed,
`the model
`served as an input function for the pharmacody-
`namic models (14).
`An insulin-dependent glucose disappearance
`model was used to analyse the PK-PDrelationship.
`(15) In our study, the response variables measured
`were the insulin plasma concentration (milli-inter-
`national units per litre, mIU/L) and the glucose
`plasma concentration (mg/mL). The system of
`differential equations shown below describes the
`model:
`
`dt
`
`x4 = Kin x 51 (t) _ Kout xR
`dxaKes X X65 —Kee x So(t) X X53 — KeeS(t) x X5
`dxFr = Kge x S(t) x x5 — KesXxe6
`_
`Smax (x2/V2)
`sil) = Th Scsgy (x2/V2)
`
`So(t) = 14 x4
`
`Kj
`Ro = ZK.
`
`(4)
`
`(5)
`(6)
`
`”)
`
`(8)
`
`(9)
`
`where Ki, represents the zero-order constant for
`production of the insulin response and Ko. defines
`the first-order rate constant for loss of the insulin
`response. In our study, a model
`that assumed
`glimepiride simulated K;, was considered to be
`closest to the pharmacological action of the drug.
`However, both modelling approaches, with glim-
`epiride inhibiting Kou: or stimulating Kj,, were
`investigated to compare the performance of the
`model and the physiologic relevance of
`the
`parameters obtained.
`In this model, the rate of change of plasma glu-
`cose is the difference between the net hepatic glu-
`cose balance and the disappearance of glucose into
`peripheral tissues only (16). We have shown that
`the hepatic glucose balance varies according to the
`relationship shownin Eq. (5). To explain glucose
`disappearance,it is assumed that insulin acts from
`a remote compartment, that insulin increases the
`
`Fig. 2. Plasma glimepiride concentration after a single 2-
`mg oral dose to healthy volunteers (mean + SEM, n = 6).
`Data points are observed values; the solid line is the
`result of maximum likelihood fitting with the ADAPT II
`program.
`
`© 2006 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 31, 469-476
`
`MPI EXHIBIT 1046 PAGE 4
`
`mobility of glucose across cell membranes, and that
`glucose mobility potentiates glucose disappear-
`ance. Wheninsulin increases, it stimulates Ks, and
`Keo according to the stimulation function given in
`Eq.(8).
`
`RESULTS
`
`Pharmacokinetic analysis
`
`The mean plasma concentration-vs.-time curve
`after oral administration of 2 mg of glimepirideis
`shownin Fig. 2, where the solid line represents the
`best fit of the pharmacokinetic model to the meas-
`ured concentration, based on the means of the
`individual parameter estimates. Based on the
`maximum likelihood criterion and visual inspec-
`tion of the fits, a two-compartment model with a
`lag time, nonlinear absorption and elimination was
`chosen to describe the data. The estimated phar-
`macokinetic parameters are listed in Table 1: The
`Michaelis-Menten type absorption rate into the
`central compartment Vimaxi2 and the apparent
`Michaelis constant K,,12 equal 1042:58 ng/h and
`70:31 ng respectively. The Michaelis-Menten kin-
`etic analysis of the data indicated the existence of a
`second carrier-mediated transport process and of
`an interaction between sulfonylurea and highly
`protein-bound drugs that govern elimination from
`the
`central
`compartment
`(Vmax= 0-14 ng/h,
`Kye = 0-006 ng) (2). The terminal elimination half-
`
`2 1000,
`“Sp
`&oO
`3a
`z
`“eb
`‘*s
`
`1004
`
`;
`
`10
`
`0
`
`7
`2
`
`r
`4
`Time (h)
`
`7
`6
`
`!
`
`+
`8
`
`
`
`o& So
`
`yg
`
`2wo
`
`a &S
`
`a
`
`MPI EXHIBIT 1046 PAGE 4
`
`MPI EXHIBIT 1046 PAGE 4
`
`DR. REDDY’S LABORATORIES, INC.
`IPR2024-00009
`Ex. 1046, p. 4 of 7
`
`

`

`Table 1. Pharmacokinetic parameters for glimepiride
`after a single 2-mg oral dose in healthy volunteers
`(mean + SEM, 1 = 6)
`
`Value
`Parameter
`
`
`PK/PD modeling of glimepiride
`
`473
`
`1021-43 + 199-55
`191-52 + 29:80
`
`2:58 + 052
`2:26 + 0:33
`7:82 + 1:22
`2:55 + 0:37
`
`Glucose lowering effect
`
`secretion was
`insulin
`the maximal
`group,
`4643 mIU/L at 5h, and secretion returned to
`baseline at 9h.
`In the test group, glimepiride
`caused a further significant
`increase (maximum
`increase of 9:98 mIU/L)
`in plasma insulin at
`between 1 and 5hafter its administration, and the
`Model independent parameter
`insulin concentration returned to baseline by 9 h.
`AUC (ng h/mL)
`Glimepiride produced a statistically significant
`Cmax (ng/mL)
`increasein insulin secretion, relative to placebo, for
`Tmax (h)
`3h (from 1 to 4 h after administration) (P < 0-01).
`CL(inf) /F (L/h)
`Vz(terminal) /F (L)
`ti2
`Model dependent parameter
`70:31 + 30-51
`Kmiz (ng)
`Figure 4 showsthe glucose profiles after drug and
`1042:58 + 161-82
`Vmaxcaz (ng/h)
`placebo administration. The decline in plasma
`0-006 + 0-0004
`Kaze (ng)
`glucose was induced after a lag of approximately
`0-14 + 0-02
`Vmaxze(ng/h)
`15h, and the decrease reached its maximum
`0-21 + 0-11
`Ko3 (/h)
`
`(2100 mg/dL decrease plasma_glucose)in
`
`Kap (/h)
`0-72 + 0:36
`approximately 2 h after dosing. Compared with the
`Tiag (h)
`0-34 + 0-04
`placebo group, glimepiride produced a statistically
`significant decrease in glucoselevels for a period of
`administration
`35h,
`from 15 to 5h after
`(P < 0:01).
`
`805
`
`604
`
`404
`
`204
`
`
`
`2=g =
`
`§ =
`
`&&o
`
`goS
`
`&2
`
`wo
`
`a 5
`
`Ss
`
`* 04
`:
`:
`2
`4
`6
`8
`
`Time (h)
`
`Fig. 3. Plasma insulin concentration after a single oral
`dose of placebo or 2 mg glimepiride to healthy volun-
`teers (mean + SEM,n = 6). Open circles, plasma insulin
`concentration after placebo. Closed circles, plasma insu-
`lin concentration after 2 mg glimepiride.
`
`life (t1/2p) was 2:548 + 0-901 h and the CLiota/F
`was 2-262 + 0814 L/h. Our pharmacokinetic pro-
`file for glimepiride is similar to that found in other
`studies involving healthy volunteers and the same
`glimepiride dose (10).
`
`Insulin secretion effect
`
`Plasma insulin profiles after drug and placebo
`administration are shown in Fig. 3. In the control
`
`Pharmacokinetic_pharmacodynamic modelling of
`insulin secretion and glucose lowering effects
`
`The corresponding pharmacodynamic parameter
`estimates for plasma insulin and glucose are shown
`in Table 2. Figures 5 and 6 show the plasmainsu-
`lin profile (post-drug level minus placebo level;
`
`
`
`
`
`Plasmaconcentrationofglucose(mg/dL)
`
`160 -
`
`===OQho£OooOoOiii
`
`80 -
`
`60 -
`
`40 -
`
`
`
`20 +
`OQ
`ho
`oO
`4
`6
`8
`
`Time(h)
`
`Fig. 4. Plasma glucose concentration after a single oral
`dose of placebo or 2 mg glimepiride to healthy volun-
`teers (mean + SEM, n = 6). Opencircles, plasma glucose
`concentration after placebo. Closed circles, plasma glu-
`cose concentration after 2 mg glimepiride.
`
`© 2006 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 31, 469-476
`
`MPI EXHIBIT 1046 PAGE 5
`
`MPI EXHIBIT 1046 PAGE 5
`
`MPI EXHIBIT 1046 PAGE 5
`
`DR. REDDY’S LABORATORIES, INC.
`IPR2024-00009
`Ex. 1046, p. 5 of 7
`
`

`

`474 H.-Y. Yun et al.
`
`Table 2. Pharmacodynamic parameters estimated by
`individual fitting of maximum likelihood method with
`ADAPTII software (mean + SEM, 1 = 6)
`
`Parameter
`Value
`
`
`Estimated parameter
`Kin (mIU h/L)
`RO (mIU/L)
`SC50 (ng/mL)
`Sax
`Ke (/h)
`Kse (/h)
`Kes (/h)
`Secondary parameter
`Kout (]Kin/RO) (/h)
`
`4-05 + 1:01
`4-54 + 1-08
`78°89 + 33-66
`21-17 + 10-20
`0-012 + 0-005
`0-007 + 0-002
`0:37 + 0:20
`
`0-89
`
`
`
`|
`
`|
`
`d
`
`a 20,
`
`== 2
`
`3Qa
`
`£= 10
`
`15!
`
`|
`2
`
`54
`
`ol
`0
`
`8B
`
`g
`
`a 2
`
`|
`
`,
`4
`
`J
`
`|
`6
`
`|
`8
`
`This study analysed the effects of a single oral dose
`of glimepiride on insulin secretion and glucose
`lowering as a function of glimepiride concentra-
`tion. This is the first attempt
`to apply PK-PD
`modelling to glimepiride’s effects on these para-
`meters. Our results may help elucidate the rela-
`
`
`
`tionship plasma_concentrationbetween of
`Fig. 5. Time course of insulin secretion after a single
`glimepiride and its physiological effects.
`2-mg oral dose of glimepiride to healthy volunteers
`Glimepiride absorption kinetics have previously
`(mean + SEM, n= 6). Curve, pharmacodynamic fit
`to
`been studied in vitro and in vivo. Glimepiride dis-
`measure increases relative to placebo (circles).
`solves very rapidly in vitro (>80% after 15 min). In
`contrast,
`in vivo dissolution is delayed (>80%in
`approximately 4h) because of
`the low, pH-
`dependentsolubility of the drug (17). Poor aqueous
`solubility and slow dissolution are factors in non-
`linear absorption kinetics (18). Thus, first-order
`absorption kinetics were unsuccessful
`in repre-
`senting the behaviour of glimepiride, as evidenced
`by the poor correlation between observed andcal-
`culated data.
`Depending on the knowledge of a drug’s
`mechanism of action, different types of indirect
`response models may be used. In our study, we
`investigated pharmacokinetic and pharmacody-
`namic modelling approaches to link glucose dis-
`appearance and glimepiride levels. Given that
`
`
`
`
`
`
`
` 30- Glucosedisappearancerelativetoplacebo(mg//dL)
`
`Time (h)
`
`Fig. 6. Time course of glucose disappearance after a
`single 2-mg oral dose of glimepiride to healthy volun-
`teers (mean + SEM, n = 6). Curve, Pharmacodynamicfit
`to measure decreases relative to placebo (circles).
`
`consistent with the known physiology of glucose
`metabolism.
`
`DISCUSSION
`
`Time(h)
`
`mIU/L) and the plasma glucose profile (placebo
`level minus post-drug level; mg/dL), respectively,
`after a single 2-mg dose of glimepiride. The solid
`lines in Figs 5 and 6 representthe best fit of the PK—
`PD model. Pharmacokinetic and pharmacody-
`namic parameters for insulin secretion and glucose
`lowering effects after glimepiride dosing were
`estimated for each subject. Glimepiride plasma
`concentrations could be linked to the observed
`insulin secretion effects by means of an indirect-
`response model, and the change in plasma glucose
`was explained by a glucose disappearance model.
`Weselected this model because it
`is acceptable
`from the identification point of view and is broadly
`
`© 2006 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 31, 469-476
`
`MPI EXHIBIT 1046 PAGE 6
`
`MPI EXHIBIT 1046 PAGE 6
`
`MPI EXHIBIT 1046 PAGE 6
`
`DR. REDDY’S LABORATORIES, INC.
`IPR2024-00009
`Ex. 1046, p. 6 of 7
`
`

`

`glimepiride stimulates insulin secretion, a model
`that assumes the stimulation of Kj, may better
`describe the drug’s pharmacological action.
`A nonlinear relationship between plasma glim-
`epiride levels and the stimulation of insulin secre-
`tion was observed. We therefore assume that
`insulin was secreted in accordance with drug-
`receptor binding via a sigmoid E,,,, model (4). The
`adaptation of other (e.g. linear and exponential)
`models was also explored, but only the sigmoid
`Emax model
`fit
`the observed plasma insulin
`dynamics. We explain the direct response between
`plasma insulin and the stimulation of glucose dis-
`appearance by assuming that the unit of plasma
`insulin Gin mIU/L) reflects the glucose lowering
`effect directly. Similar multi-compartments have
`been demonstrated by Ole et al. (14).
`Our model might more closely simulate insulin
`and glucose regulation if other factors, such as the
`stimulation of Ko, by plasma glucose concentra-
`tion, were considered. However, the insulin secre-
`tion
`and
`glucose
`lowering
`effects
`of
`oral
`glimepiride fitted sufficiently well, as shown in
`Figs 5 and 6. This indicated that our present model
`appropriately describes the PK-PDrelationships of
`oral glimepiride. As the onset of glimepiride’s
`pharmacological effect
`is
`fundamentally inde-
`pendent of dose and route, this quantified model
`for the insulin-glucose regulation system could
`describe plasma levels after another different dose
`or route of administration.
`In conclusion,
`the use of the glucose disap-
`pearance model in the context of PK-PD model-
`ling
`showed
`that
`the
`two measures
`of
`pharmacodynamic response (insulin secretion and
`glucose lowering effects) are valid and clinically
`relevant ways to investigate the effects of glim-
`epiride. By using an indirect response model, we
`could describe the stimulatory effect of glimepi-
`ride on insulin secretion. The glucose disappear-
`ance model, regulated by insulin, also fit well
`with the relation between glimepiride-induced
`changes in insulin secretion and glucose reducing
`effects.
`The established PK-PD model may be used to
`predict the plasma level of glimepiride, glucose
`and insulin under new conditions, for example
`following multiple dosing, and increasing or
`decreasing dose. This study involved healthy vol-
`unteers. By definition in clinical practice glime-
`
`PK/PD modeling of glimepiride
`
`475
`
`pride is going to be used in patients with diabetes.
`To overcome the limitation of this PK/PD model,
`we need more pharmacokinetics and pharmaco-
`dynamics data from such patients.
`
`ACKNOWLEDGEMENTS
`
`This study was supported by the contract, ‘PK/PD
`model development of glimepiride 2 mg after
`administering a single oral dose to healthy Korean
`volunteers’,
`from SAM-NAM Pharm Co., Ltd,
`Chungnam, Korea.
`
`REFERENCES
`
`1. Salem I, Idress J, Al Tamimi JI (2004) Determination
`of glimepiride in human plasma by liquid chroma-
`tography-electrospray
`ionization
`tandem mass
`spectrometry. Journal of Chromatography B, 79, 103-
`109.
`
`2. Langtry HD, Balfour JA (1998) Glimepiride - a re-
`view ofits use in the managementof type 2 diabetes
`mellitus. Drugs, 55, 563-584.
`3. Kramer W, Muller G, Geisen K (1996) Characteriza-
`tion of the molecular mode of action of the sulfo-
`
`and
`
`f-cells. Hormone
`at
`nylurea, glimepiride,
`Metabolic Research, 28, 464-468.
`4. Schwanstecher M, Manner K, Panten U (1994) Inhi-
`bition of K* channels and stimulation of insulin
`secretion by the sulfonylurea, glimepiride, in relation
`to its membrane binding in pancreatic islets. Phar-
`macology, 49, 105-111.
`5. Malerczyk V, Babian M, Korn A (1994) Doselinearity
`assessment of glimepiride (Amaryl®)
`tablets
`in
`healthy volunteers. Drug Metabolism and Drug Inter-
`actions, 11, 341-357.
`6. Profozic V, Mrzljak V, Rosenkranz B, Lehr KH,
`Waldhausl W (1991) Pharmacokinetics of glimepi-
`ride in kidney disease. Diabetes, 40, 343-347.
`7. Rosekranz B (1996) Pharmacokinetic basis for the
`safety of glimepiride in risk groups of NIDDM pa-
`tients. Hormone and Metabolic Research, 28, 434-439.
`8. Campbell RK (1998) Glimepiride: role of a new sul-
`fonylurea in the treatment of type 2 diabetes melli-
`tus. Annals of Pharmacotherapy, 32, 1044-1052.
`9. McEvoy GK, (2002) AHFS drug information. Wiscon-
`sin: American Society of Health-Systems Pharma-
`cists Inc., 3025-3028.
`10. Badian M, Korn A, Lehr K-H, Malerczyk V, Wald-
`hausal W (1996) Pharmacokinetics and pharmaco-
`dynamics of the hydroxyl metabolite of glimepiride
`(Amaryl®) after intravenous administration. Drug
`Metabolism and Drug Interactions, 13, 69-85.
`
`© 2006 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 31, 469-476
`
`MPI EXHIBIT 1046 PAGE 7
`
`MPI EXHIBIT 1046 PAGE 7
`
`MPI EXHIBIT 1046 PAGE 7
`
`DR. REDDY’S LABORATORIES, INC.
`IPR2024-00009
`Ex. 1046, p. 7 of 7
`
`