Title of study CV181010: Placebo-controlled, ascending multiple-dose study to evaluate
`the safety, pharmacokinetics and pharmacodynamics of higher doses of saxagliptin
`(EMS-477118) in healthy subjects.
`'
`
`Study period: 25—Aug-2003 - 01-May-2004
`
`Method: This was a placebo-controlled, randomized, double-blind, sequential, multiple
`ascending dose study. 50 subjects were randomized to receive either 40, 100, 150, 200,
`300, or 400 mg saxagliptin or matching placebo. Forty (40) subjects received saxagliptin
`(10 subjects at the 40 mg dose level and 6 subjects per every other dose level) and 10
`subjects received placebo. All doses were administered 1 hour prior to breakfast.
`
`Bioanalytical:
`
`Play/22a ayyayfiryaraga’ptza' The standard curves were well fitted by a l/x—weighted
`quadratic equation over the concentration range of 5.00 to 1000 ng/mL. Values for the
`between-run and within-run precision for analytical quality control samples were no
`greater than 6.4% coefficient of variation (CV), with deviations from the nominal
`concentrations of no more than :|:3.5%.
`
`Play/72a assay/0r 57/55/0672- The standard curves were well fitted by a 1/x-weighted
`quadratic equation over the concentration range of 10.0 to 2000 ng/mL. Values for the
`between-run and within-run precision for analytical quality control samples were no
`greater than 6.3% CV, with deviations from the nominal concentrations of no more than
`13.7%.
`
`Results:
`
`The mean plasma —time concentration profiles are shown in the Figure and the summary
`of PK parameters are shown in the Table 1. As seen, the PK profiles on Day 1 and Day
`14 was similar. There is no evidence of accumulation following once daily dosing for 2
`weeks. There is no evidence of saxagliptin inhibiting or inducing its own metabolism
`following daily oral doses of 40 to 400 mg for 2 weeks. Across the dose groups on Days
`1 and 14, the mean amounts of a saxagliptin dose excreted into the urine (unchanged
`saxagliptin) ranged between 18 and 29%. In general, saxagliptin trough concentrations
`suggested that saxagliptin was at steady-state by Day 4.
`
`128
`
`

`

`Figure 1: Mean (+ SD) Plasma Concentration-Time Profiles for Saxagliptin on Days
`1 and 14
`
`10:00
`
`
`
`
`
`
`
`
`
`Saxwliumplasmamnunn’aaon(12¢le 8
`
`Day 1
`+40 mg Saagm'n In - 10)
`+103 mg 33:99:69 [n a 6]
`+15le m9 Smack-tin [n = 6]
`{PM me Saragfpcin [n : 5]
`+393 mg Sanctum [n I 6]
`+400 mg Gaugiptin [n I 6]
`
`
`
`
`
`
`
`
`0451235203
`Tlfl‘é posubse 01)
`‘
`
`Day 14
`
`aaiccnlfllialmmL)
`8155477118dasma
`
`0481216324
`Time pas-cm (h)
`
`ii]
`
`(37181010
`
`
`
`Table 1: Summary Statistics forSaxagliptinPK Parameters
`
`
`
`my u
`DayI
`mm for40 ll
`pmfor» in:
`“Wflfiw‘lm Hmmmam'
`220 (40)
`224 (33)
`
`535 on609 (25)
`
`+87 (24)on (19)
`1207 (n)
`m (:3
`ms (20)
`mo (31)
`m (29)
`too (29)
`2545 (II)
`2512 (9)
`1m 05)
`I»: u»
`use 05)
`mo (20)
`6559 an
`1.05 (33)
`1.00 (15)
`1.05 (15)
`0.99 (19)
`0.99 m)
`oas to50.2.00)
`-0 (050,290
`125 (01,2.
`
`1.00 (9.75.100)
`ms 959.100)
`150 (0512.00)
`:50 w50. mo
`:50 (1.00. L50)
`
`2.:9 (oin)
`‘
`
`:03 (1.29)
`2.69 (OSI)
`3.5: (1.25)
`
`9-10 h} .10 in:
`“’6 I" all m1" ““5
`
`259177)
`13’ ‘5‘)
`1:9 (54)
`199 (69)
`191 (53)
`213130
`
`10-18 {or 40 In;
`11-6{or all Minx dons .
`:5 no)
`-- (8)
`if, (3)
`1'3 (5)
`
`:30 (3:)
`241 (56)
`\95 (37)
`159 :1
`
`_
`Can: ml.
`mom“ 312:: (c v. a)
`
`.
`,
`-
`lull
`- «\LC AY-
`5
`; mgulfid 2-.)
`
`-
`-
`.
`LC AL
`gag; sari-1&5: 9-.)
`
`Smmplln PK
`Parana-m
`
`MuMSD.)
`
`Mn:(9.D.)
`
`Dose Proportionality: Daseprqoor/z’wzaflg/ was eytz‘ma/edmizzg lflepower model {7 =
`0:. Dose” where Y, a and ,6 correspond to the PK parameter (AUC or Cmax),
`
`129
`
`

`

`prqoor/zbnd/z'gl constant and an exponent; rerecfl'Veé/j, using data/font [fie multiple
`dose .y/zzajz. fine 90% Conr Me slope ,3 contains 1, the relationship between dose and
`the PKparameters is considered to be dose proportional.
`
`Linear regressions of log[Cmax] on log(dose) and of log[AUC] on log(dose) were
`estimated for saxagliptin and EMS-510849, using the power model described by Gough
`et al. A slope of 1 would indicate perfect dose proportionality. Point estimates and 90%
`confidence intervals
`for
`the dose-proportionality parameter
`(slope of the linear
`regression) were calculated and are shown below:
`
`Cmax Day 1: 1.00 [0.84 — 1.17]
`
`Cmax Day 14: 0.95 [0.755 — 1.14]
`AUC Day 1: 1.06 [0.93 — 1.20]
`AUC Day 14: 1.03 [0.87— 1.18]
`
`V As the 90% CI for the slope B contains 1, the relationship between dose and the PK
`parameters is considered to be dose proportional for saxagliptin.
`
`ag§§3§
`
`
`Saxlai'pflnwcmu(nuhImL)‘
`
`
`Sanglipuncrnax(no/17L)
`
`o
`
`50
`
`100
`
`150
`
`250
`zoo
`DWBUW)
`
`300
`
`350
`
`«do
`
`a
`
`so
`
`100
`
`150
`
`250
`200
`00590“)?
`
`300
`
`350
`
`400
`
`BMS-510849: The mean plasma — time concentration profiles are shown in the Figure
`and the summary of PK parameters are shown in the Table below.
`
`Figure: Mean (+ SD) Plasma Concentration-Time Profiles for BMS-510849 on Days
`1 and 14 Following Once— Daily Oral Doses of Saxagliptin .
`
`130
`
`

`

`IOCOD
`
`Day1
`HWSWWIW‘ l“l
`—o—‘-00mq Snafipminttj
`+150ma sawhtn=fil
`43—23mm SmWnin-E]
`+300ms Sawhin‘fl
`«3—40me $¥W'fl5fl*fl
`
`
`
`
`
`
`'8
`
` 10
`
`o 4-812162024
`um- pmbdmt in)
`
`Day 14
`
`109:0
`
`cal-cannula)mu 8
`BMS~510849stun-n
`
`
`la
`
`o
`
`4
`
`3 0182024
`11m- pom-dam (h)
`
`
`
`
`
`EMS-510443pawncameramanmm
`
`Both Cmax and AUC(TAU) of EMS-510849 appeared to increase proportionally with
`
`saxagliptin doses up to 300 mg but appeared to increase less than proportionally at the
`400 mg saxagliptin dose. EMS-510849 trough concentrations suggested that BMS-
`510849 was at steady-state by Day 4. Mean EMS-510849 urinary recoveries was 21 and
`33% of the saxagliptin dose over a dose interval.
`
`Table: Summary Statistics for BMS-510849 PK Parameters
`mu.
`Dore
`
`Day!
`mu
`I
`
`
`[P10 m- 40 mg)
`(0-10 m- .L0 mg)
`l
`
`
`(”'5 I" all “a" ‘9’”) (n-G (01' all other than) ' l
`
`
`331 (33)
`314 (40)
`Om: (ax-hat.)
`
`
`Gum-ck Meanwx'fl‘o)
`"15 (12)
`919 (I4)
`
`1202 (10)
`1550 (10)
`
`
`
`1501 (24)
`my (24)
`
`
`3022 (30)
`2433 (so)
`
`"400 13
`‘6—9
`is
`
`
`
`os (30)
`
`m9 (:5)
`: autumn-war.)
`
`
`momkmtcyfih)
`609‘: (15)
`53-“ (H)
`m4 (:2)
`7902 (13)
`
`
`
`
`9410 (1:) mo (2:)
`
`
`
` .u fix AL'CflAL’)
`
`60mm: Mon (C37. 9'.)
`0.94 (l l)
`
`
`0,94 (5)
`0.92 (0)
`UH (u)
`
`2.20 (39)
`
`
`3.60 (I?)
`2.68 {17)
`Malt: we m: AUC(TAU)“
`
`
`
`
`600mm: Mu: (CV. $0)
`294 (31)
`2.?6 Q!)
`
`2.12 (31)
`2.02 (30)
`
`
`
`129 (41)
`1.17 (35)
`
`
`L”:
`'-
`
`1.50 (1.50, 3.00)
`1.50 (1.50, 3.00)
`1.75 (1.00,- 100)
`
`
`2.00 (1.50, 3.00)
`1.00 (150, 3.00)
`
`2.00 (1.00, 3.00)
`
`2.00 (1.50, 3.00)
`
`2.50 (1.50.100)
`2.00 (1.50= 5.00)
`
`
`s
`
`
`
`v 00 00)
`
`
`
`.n
`
`131
`
`

`

`
`I
`
`
`
`
`i
`Bus-slow in:
`5
`Parana-«r
`
`
`{
`
`5mm“
`Don
`
`
`
`pm:
`Day)
`(Ia-lo [nub 13:)
`(Ii-lo for 40 mg)
`(“.5 m m m," an“) i ‘
`(p6 m .n on»: dam)
`3.71 (1.05)
`2.96 (0.29)
`5.25 (1.35)
`4.—--._ (on)
`
`5.2; (1.96)
`4.19 (054)
`
`5.93 (1.35)
`493 (0.14)
`
`
`4.27 (0.52)
`7.243 (103)
`{4-1
`0.5
`
`
`Men(5.3.)
`
`
`
`28 (10)
`
`21
`10
`
`110 (36)
`
`
`as (2:)
`10° (2?)
`
`s: (30)
`111 (42)
`
`
`$9 (32)
`m (:5)
`
`
`‘1 n=5 for 200 mg
`b molar ratio = (Metabolite AUC/Parent AUC)*(455.55/487.55)
`
`Plasma DPP-4 Activity: Plasma DPP-4 activity remained constant over the 24 hour
`observation period in the subjects who received placebo. For subjects who received
`saxagliptin, DPP-4 inhibition peaked, on average, between 0.75 and 4 hours after closing
`on both Day 1 and Day 14. Plasma DPP—4 inhibition on Days 1 and 14 appeared to be
`dose-dependent both in terms of the maximum inhibition and the amount remaining
`inhibited at the end of the dose interval (24 h) fi'om 40 to 150 mg QD saxagliptin. Dosing
`with saxagliptin at 100, 150, 200, 300 and 400 mg resulted in larger inhibition of plasma
`DPP-4 activity than dosing with saxagliptin 40 mg, no clear difference was observed
`between the 150 mg — 400 mg doses. For all doses, plasma DPP-4 activity was inhibited
`by at least 74% at 24 hours after a single dose and following two weeks of daily dosing.
`The peak inhibition of plasma DPP-IV activity on Days 1 and 14 was between 1 and 2 h
`post-dose which tended to coincide with the Tmax values for saxagliptin and EMS-
`510849.
`
`Figure: Plot of Mean Percent Changes from Baseline for Plasma DPP—4 Activity on
`Day 1 (top) and Day 14 (bottom):
`Day!
`
`
`
`Menu19changehumBaez-law
`
`~31:
`-20 1
`~40
`
`‘szLW’:
`-1-so
`
`
`
`
`do}
`.4.go 4
`
`no:
`
`Tlrn. SIM. DOS. (0)
`.-.-~ 3867*.
`s—n—n 13338
`
`PLOT m Elfii$
`
`5-.» 150 mg
`
`132
`
`

`

`Day 1.:
`
`mi
`01WV“...—
`«0“-m—
`-35..
`iii.50 -
`~70
`~90
`ID
`
`-
`
`.
`
`“
`
`I
`
`
`
`Mars“inClmuuauumUuwblm
`
`m1
`
`D
`
`4
`
`a
`
`1:
`Time arm. Dan in)
`o-;-‘ §83ngg
`Q—fl-d 153%:
`
`15
`
`2a
`
`:4
`
`,
`
`PLOT -.- gazex
`
`an... 159 m9
`
`In general, dosing with saxagliptin (in the dose ranges of 40 to
`Plasma GLP-l Levels:
`200 mg) produced an increase in mean changes from baseline (although not dose-
`dependent effect) for postprandial AUC(0-3h) plasma active GLP-l over those observed
`for subjects on placebo for all meals.
`
`Other PD parameters: No apparent dose—dependent effects were observed for glycemic
`indices or lipid parameters studied in subjects who received saxagliptin in this study. A
`small saxagliptin dose-dependent decrease was observed in AUC(0-3h) values for plasma
`glucose after breakfast on Day 8 and 13 compared to baseline. No saxagliptin dose,
`saxagliptin treatment, or time effect was observed in AUC(0-3h) for serum insulin, serum
`C-peptide, or the HOMA insulin resistance index were observed on Days 8 or 13
`compared to baseline.
`
`[Voter The lack of a clear saxagliptin dose, saxagliptin treatment, or time efi’ect on these
`indices of glycemic control despite robust plasma DPP-IV activity inhibition and
`increases in active GLP—I plasma concentrations may have been due to a number of
`factors such as, but not restricted to: the limited length of the study (2 weeks) may not
`have been long enough to observe an eflect with this mechanism;
`the study was
`conducted in healthy subjects who did not have poor glycemic control; confinement of the
`subject to the clinical facility on a controlled diet; or high within-subject and between-
`subject variability in the analytes measured.
`Comments:
`
`0
`
`1 subject (CV181010-1-30; 25 yr white male) was discontinued from the study
`due to an AB of mild rash (possible related to study drug) after receiving 200 mg
`saxagliptin for 9 days.
`
`0 Most common AE was pain in extremities (n=3). There were no dose-related
`trends.
`-
`
`0 As the 90% CI for the slope B contains 1, the relationship between dose and the
`PK parameters is considered to be dose proportional for saxagliptin.
`
`0 Within each dose group, the molar AUC ratio of BMS-S 10849 to saxagliptin did
`not appear to be different between Days 1 and 14. The apparent terminal-phase
`half-life values for both saxagliptin and BMS-510849 appeared to be slightly
`higher (2-3 h more) on Day 14 compared to Day 1. Thus, saxagliptin does not
`
`133
`
`

`

`seem to display time dependent PK when administered once daily to healthy
`subjects for 2 weeks at doses up to 400 mg, suggesting saxagliptin does not
`induce or inhibit its own metabolism over time.
`
`the mean EMS—510849 plasma
`o Across the dose groups on Days 1 and 14,
`exposures on a molar basis were between 1.7 and 3.0-fold higher than parent
`saxagliptin.
`
`Title of study CV181002: Placebo-controlled, ascending multiple-dose study to evaluate
`the safety,
`
`Pharmacokinetics and pharmacodynamics of Saxagliptin in diabetic subjects
`
`Study period: 25-Feb-2002 - 07-Jun-2002
`
`Method: This was a placebo—controlled, randomized, double-blind, sequential, ascending
`multiple dose designed study. The original protocol stated that subjects were to be
`assigned to 4 sequential panels (5, 15, 30 or 50 mg EMS-477118 or matched placebo).
`Based on the preliminary PD data from Study CV181001, the protocol was amended to
`evaluate lower doses (2.5 mg and 1 mg). However, the 1 mg dose was never evaluated.
`The 2.5 mg dose was administered as solution (1 mg/mL) while the other doses were
`administered as capsules. Subjects taking antidiabetic medications at screening were to
`have a Z 7 day washout period. Subjects taking sulfonylureas were to have a 14 day
`washout period. Drug was administered ~ 5 min afier breakfast.
`
`The primary objective of this study was to assess the safety and tolerability of EMS-
`477118 following oral doses (2.5, 5, 15, 30 and 50 mg) administered once daily for 14
`days in subjects with Type 2 diabetes mellitus (T2DM). Effects of saxagliptin on DPP-4
`inhibition, glucagon-like peptide-1 (GLP-l), plasma glucose, serum insulin, homeostasis
`assessment model
`(HOMA),
`lipids
`(LDL, VLDL, HDL triglycerides and total
`cholesterol) and serum C-peptide was also assessed.
`
`The PK profile of saxagliptin appeared to be similar for Days 1, 7 and 14 and there was
`no accumulation. The apparent terminal elimination half life appeared to be similar up to
`50 mg dose in T2DM patients.
`
`Figure: Mean (plus SD) plasma concentration-time profiles for BMS-477118 on days
`1, 7, and 14 following daily dosing administration of 2.5 to 50 mg doses of BMS—
`477118 to type 2 diabetic subjects in study CV181002
`
`134
`
`

`

`my 1
`
`Day 7
`
`Day It
`
`ICDD
`
`103
`
`10
`
`l
`
`
`
`04§l2ll$24
`Tmpan-dbuonanHm)
`
`04812132113
`Tmpon-doumnay7m)
`
`(OW
`
`‘00
`
`1|)
`
`1
`
`'—o—25mg
`+5m:
`$15M
`+30mg
`—D-5°"l9
`
`111:0
`
`E
`‘9
`5'
`ft m)
`
`10 x8
`
`g g
`
`E‘1
`Qas
`
`1
`
`35312102024.
`TMMMMMDIY‘IM
`
`Table 1: Summary Statistics for EMS-4771 18 Pharmacokinetic Parameters
`3315471110
`rnm'fiums:
`Dost
`Pannier
`: mungn- )
`.
`600mm: Mm
`‘ u;
`(ct-12'.)
`
`Dav D (11:6)
`1_
`23
`
`E
`3
`5
`i|
`i
`
`5
`l
`I
`5i
`
`ilI I
`
`AUC(O—D
`(as-Mal)
`620:2:ka
`(cs-'33)
`
`,
`‘
`
`
`
`1.503(035300)
`2'
`.00.
`_
`12% $35,333;
`$.00
`(100.430)
`:50 (1.00.300)
`3.3;
`2.21“
`1-‘6
`2.;5
`2.17
`
`.
`.
`.
`
`.
`..
`
`150 (035,100)
`.
`,
`.
`2.00 (150.000) 5
`.
`1.15 (1.00.100)|
`(1.50,;00)
`2.00_ (1.00, 3.00)
`.
`(2.00900)
`“5°59” i.5o"(l.50,3.00)_
`3.9-
`.
`233‘
`(024)
`2.55
`(0.35)
`255
`(035)
`a ,1
`
`.
`.
`.
`.
`
`:
`
`140‘
`190‘
`
`(50)
`(57)
`
`149‘
`163’
`124
`
`(36)
`(69)
`
`Figure: Saxagliptin AUCO-t and Cmas on Day 14 in TZDM patients
`
`135
`
`

`

`'
`
`o Cmax(nglmL)
`
`I ALICO-t(ng.hlmL)
`
`511151510349.- The mean plasma concentration — time profiles for the metabolite
`following various doses of saxaglitpin is shown below. As seen, there appeared to be no
`difference in the profiles on Days 1, 7, and 14. The molar ratio of metabolitezparent was
`similar on Days 1, 7, and 14 within each dose. The apparent terminal half-life was also
`not changed on Days 7 and 14 as compared to Day 1. The mean exposure of BMS-
`510849 was 4-7 fold higher than the parent in TZDM patients.
`
`Figure: Mean (plus SD) Plasma Concentration-Time Profiles for BMS-510849 on
`Days 1, 7, and 14 Following Daily Dosing Administration of 2.5 to 50 mg Doses of
`EMS-477118 to Type II Diabetic Subjects in Study CV181002
`Day ‘I
`Day 7
`Day 1‘
`
`
`
`
`
`
`
`EMS-510849plasmacone-anion(637le
`
`Km
`
`E
`
`a8
`
`04812932024
`Timgost-msqmflaylm)
`
`04812162024
`Timepogl-deuenDny7m)
`
`10900
`
`1500
`
`100
`
`0491:2‘02024
`Timpoudononnay 140i)
`
`Plasma DPP-4 activity: Mean percent changes from baseline for plasma DPP-4 activity
`values are plotted over 24 h since dosing on Days 1 and 14 by treatment. Dosing with
`saxagliptin appeared to have-a dose-dependent and time-dependent effect on plasma
`DPP-IV activity. As expected, DPP-IV inhibition was negligible for subjects receiving
`placebo. For subjects receiving saxagliptin, DPP-IV inhibition peaked, on- average,
`between 1.5 and 6 hours after dosing. The amount remaining inhibited at the end of the
`dose interval (24 h) was 37% and 65% at the proposed clinical dose of 2.5 mg and 5 mg
`respectively.
`
`136
`
`

`

`Figure: Plot of Mean Percent Changes from Baseline for Plasma DPP-IV Activity on
`Day 1 (Top) and Day 14 (bottom)
`Day 1
`
`"m m‘-‘ Tint-333°
`
`322;:
`Day 14
`
`,3? l
`
`:2; gm,
`
`"omlamina “m. flint. 05:. (h)
`Mom"51-change
`
`.1fl-D ‘nrnc 5m:- Dnzt (h)
`
`
`Mun"«3Chum"llulllautumn”
`
`0
`
`u‘u
`
`"°" Ci“: 7'33? °
`
`:33 in'rfi'a'
`
`323:3 2mg
`
`Plasma active GLP-l concentrations: Dosing with saxagliptin did not appear to have a
`dose-dependent or time-dependent effect on plasma active GLP—l concentrations. For
`subjects receiving saxagliptin, plasma active GLP-l concentrations generally peaked, on
`average, at 6 hours after dosing. Exceptions were the 50 mg dose-group which produced
`plasma active GLP-l concentrations which peaked, on average, at 1 hour after dosing on
`Days 1 and 14, and the 2.5 mg dose-group which produced plasma active GLP-l
`concentrations which peaked, on average, at 45 minutes after dosing on Day 1.
`
`Other PD parameters: Dosing with saxagliptin did not appear to have a dose dependent
`or time-"dependent (day of dosing) effect on glucose over 24 hours from the time of
`dosing. Dosing with saxagliptin did not appear to have a dose-dependent or time-
`dependent effect on HOMA, glucose AUCO-4 values, serum insulin or C-peptide over 4
`hours fi'om the time of meal.
`
`Bioanalytical: Saragép/z‘zzx Plasma samples were assayed using a MS/MS method. The
`values for within-run and between run precision (%CV) for quality control samples were
`< 4.8%. The standard curve was linear in the 5 — 1000 ng/mL range. For urine samples
`the assay was linear in the range of 25-5000 ng/mL. Values for the between-run and
`within-run precision for analytical quality control samples were no greater than 3.6%
`(%C.V.). The metabolite standard curve and QC data also indicate that the plasma and
`urine assay method was adequate.
`
`C'flllllliellb'.’
`
`0 T2DM patients were enrolled who had established diagnosis for < 10 years and
`who were treated with either diet or drugs (insulin, sulfonylureas or acarbose
`only). HbAlc was in the range of 6.5 -9.5%.
`
`137
`
`

`

`0 Dose-proportionality was examined using the power model for saxagliptin AUC
`and Cmax. Dose proportionality was demonstrated on all days. The results slope
`(90%CI) for Ln Dose Vs. Ln (AUC/Cmax) are as follows:
`
`0 Cmax Day 14: 0.976 (0.79 — 1.15)
`
`0000
`
`AUC Day 14: 1.13 (0.87 — 1.39)
`
`Cmax Day 1: 0.98 (0.71 —- 1.25)
`
`AUC Day 1: 1.14 (0.9 ~1.39)
`
`Cmax Day 7: 0.99 (0.82 - 1.16)
`
`o AUC Day 7: 1.15 (0.86 — 1.43)
`
`I The between subject variability in the molar ratio (metabolite: parent) was
`relatively large, ranging from 18 to 47% CV. I” who studies suggest CYP3A is
`the primary enzyme involved in EMS-477118 biotransformation to BMS—S 10849,
`and the between subject variability in the molar ratio probably reflects the
`between subject variability in CYP3A activity.
`
`Protocol No.: CV 1 81022
`
`Study Date: 06-Oct-2004 — 25-Oct-2004
`
`Title: A study to assess the effects of ketoconazole and EMS-477118 on lymphocyte
`count in healthy subjects
`
`This was an open-label, randomized, three-sequence study in 36 healthy subjects. After
`10-hour fast, subjects were randomized to one of the following 3 sequences:
`'1=5 mg saxagliptin on Days 1 and 9
`
`2=20 mg saxagliptin on Days 1 and 9
`3=20 mg saxagliptin on Day 1, followed by 200 mg ketoconazole q12h on Days 3-8,
`followed by 200 mg ketoconazole q12h + 20 mg saxagliptin on Day 9
`
`The subjects were maintained in a fasted state for 4 h post-dose after saxagliptin
`administration. Ketoconazole was given following breakfast and dinner on Days 3-8. PK
`of saxagliptin and EMS-510849 was measured after Day 1 and Day 9 up to 48 h post-
`dose and trough samples were measured for ketoconazole on Days 7, 8, and 9 pre-dose.
`
`To determine the effects of the treatments administered on Day 9 on the changes in
`absolute lymphocyte counts (ALC), an analysis of variance was performed on Day 9
`%AALC, where
`
`%AALC = [(lowest recorded ALC on Day 9/ALC at corresponding time on Day 8) -1] X
`100%
`
`138
`
`

`

`The mean (SD) plasma saxagliptin concentration versus time profile following a single
`dose of saxagliptin on Days 1 and 9 and with 200 mg ketoconazole q 12 h is shown
`below. When 20 mg saxagliptin was co-administered with 200 mgketoconazole quh,
`the geometric means for Cmax, AUCinf, and AUC(O‘T) of saxagliptin increased by
`144%, 267%, and 279%, respectively, relative to those observed following administration
`of 20 mg saxagliptin alone.
`
`Correspondingly, a decrease in EMS-510849 plasma level was observed when 20 mg
`saxagliptin was co-administered with 200 mg ketoconazole q 12 h. Subjects CV181022-
`1-18, CV181022-1-19 and CV181022-1-22 did not receive a second dose of saxagliptin
`on Day 9 and were therefore not included in the summary statistics. Additionally, eight
`(8) of the eleven (11) subjects who were administered ketoconazole on Day 9 did not
`have sufficient quantifiable EMS-510819 plasma (LLQ = 10 ng/mL) levels to calculate
`the PK parameters and were not included in the summary statistics. Statistical analyses
`were not performed on Cmax, AUC(INF), and AUC(O-T) of BMS—510849, since only 3
`subjects have pharmacokinetic data of EMS-510849 on both Days 1 and 9 (see figure
`below).
`
`*Dzyl-Sflllmlfi‘f-(PIE)
`«mp-squmrum
`
`1cm +Ehy1-3Jmm3ucalfl)
`*myi‘mmxmupu)
`
`§
`
`22
`
`
`
`n.“awn-mt...own.)
`-ugipduplan-n“xv-mud":(Iv-l.)
`
`
`
`uxufirihtH...
`
`‘01!l!l4$)ld+§fll§flll
`11:09::qu
`Thy-mt!)
`
`1(00
`
`
`
`
`+Dryh1az’rmwu
`(Fill
`*Drfi- 10:; 1mm-
`noalmmwahq Db
`will
`
`100
`
`10
`
`
`
`94512162021
`“ammo;
`
`
`
`uualpmIn“.«mcmmum(an-l.)
`
`
`
`
`
`§ l
`
`1000
`
`-
`1"
`a
`.
`g m
`..
`E
`a
`.-
`g
`-_
`é-
`)2
`
`xo
`
`*Dayl-loaqnnm
`(rm
`
`*m,_.oqm _
`:oéqzmmqm
`
`(3-11)
`
`
`
`OIIGSWDMN
`no mm (I)
`
`9:4d310l21i16
`nu pum- (b)
`
`in
`
`- W *Nl-fiwmflififlrm woo
`.
`;'
`w—DcyD-Sflmfiphwlu) g
`3
`Z
`g
`3
`g
`E
`:3 m
`.
`2
`5
`a
`.
`.2
`2
`in
`_1
`3.
`3
`:1.
`a
`3
`
`no
`
`01463IOIII4X9
`Tn- yoq—dm a)
`
`Table 2: Results of Statistical Analyses on Saxagliptin Cmax, AUC(INF) and AUC(O-T)
`1
`Phamatokinéfit
`Gcomuk Means
`Ratio of Goomtrk Means
`
`j
`3
`Pan-amm-
`D 1
`bar 9
`(Day 9: Day 1)
`
`
`”
`-
`Pointisflmm ms or 5
`
`
`WW) —-1
`
`AUC(ZNF) (ng‘mL-h) ~
`371
`1360
`3.67
`(3.30.4133) i
`
`349
`
`
`? AUC(o-r) (nymph)
`1322
`3.79
`(339, 4.24)
`Days: 1 = 20 mg saxagliptin
`
`
`
`
`139
`
`

`

`9 = 20 mg saxagliptin + 200 mg ketoconazole q12h
`
`Absolute lymphocyte count:
`
`Table below shows the mean i SD values for absolute lymphocyte count by sequence,
`study day, and time point.
`
`Following administration of a single 5 or 20 mg dose of saxagliptin one week earlier,
`administration of a second single dose of 5 or 20 mg saxagliptin did not result in a Z 30
`% decrease in absolute lymphocyte counts. Following administration of a single 20 mg
`dose of saxagliptin one week earlier, co~administration of a second single dose of 20 mg
`saxagliptin and 200 mg ketoconazole q12h dosed to steady-state resulted in a decrease (-
`30.6%) in absolute lymphocyte counts. The levels returned to baseline levels within 72 h
`in all sequences. The sponsor claims that this change was not associated with any clinical
`signs or symptoms.
`
`'Abébiuré'iymé'necfio com Shh: Sb (x 10' inn.)
`
`'
`
`
`_—-fl
`
`Saleem
`
`
`
`
`
`2.22
`:055
`2.18
`i 0.54
`2.06
`t 036
`
`2.04»
`1 0.43
`2.00
`2 0.53
`1.95
`A 057
`
`2.45
`g 0.59
`2.29
`1 0.60
`2.6»
`2 0.40
`
`2.72
`g 0.75
`2.5!
`z 0.59
`2.49
`t 0.55
`
`2.16
`:t 0.45
`2.03
`t 0.44
`1.8?
`* 0.3l
`
`1.07
`t 0.32
`1.83
`t 059
`1.86
`t 0.55
`
`232
`t 0.45
`2.02
`:t 053
`1.32
`2 0.62
`
`2.4-3
`i 0.55
`2.20
`:t: 0.68
`1.87
`t 0.32
`
`1.82
`1: 0.31
`l 05
`to.“
`1.35
`:E 0:52
`
`1.59
`1 0.48
`LED
`2 0.50
`1.67
`i 0.31
`
`2.13
`i 0.43
`l.99
`t 0.54
`1.28
`i 0.49
`
`
`
`
`
`The present study was powered to detect a decrease of ALC of greater than 30%
`
`(%AALC). If the point estimate and/or the 90% confidence interval for %AALC on Day
`9 was less than -30%, then a decrease in ALC would be concluded. A decrease in ALC
`
`was only observed in Sequence 3 (200mg ketoconazole q12h co-administered with 20 mg
`
`saxagliptin). In sequence 3 (200mg ketoccnazole q12h co—administered with 20 mg
`saxagliptin) both
`
`the point estimate of %AALC (—3 0.6%) and the lower 90% confidence bound (-36.9%) on
`Day 9 fell below the pre-specified -30% %AALC level. The
`
`Table 12.8.48: Results of Statistical Analyses 'of Day 9 %AALC
`'
`’
`‘
`‘
`P0111! Estimate
`90% c1"
`
`
`=
`
`
`
`
`
`(328.5. -15.9)
`
`
`(56.9, -24.3)
`
`
`€1.19th
`
`Comments:
`
`140
`
`

`

`o A twice daily dosing regimen (q12 h) of 200 mg ketoconazole was selected to
`ensure adequate tissue and plasma concentrations of the inhibitor over 24 hours.
`Considering saxagliptin’s‘ half—life (2-3 h), this dosing regimen is acceptable.
`
`0 Ketoconazole was administered for 6 days prior to assessing interaction with
`saxagliptin. Trough plasma levels of ketoconazole was measured to validate that
`ketoconazole was at steady-state.
`
`0 Sponsor’s cutoff for a significant decrease in ALC was at 30%. The effect was
`concluded to be less than 30% if the 90% CI for %AALC was entirely above -
`30%.
`
`o The clinical significance of the detectable and transient decrease in ALC in
`Sequence 3 is unkown. However, the magnitude of the decrease upon the re-
`challenge of the
`subject with 20 mg saxagliptin co-administered with
`ketoconazole after a single 20 mg dose of saxagliptin 8 days earlier appeared to be
`less than that observed in Studies CV181005 and CV181017 where some subjects
`were considered to be lymphopenic and had flu-like syndrome.
`
`Protocol No.: CV181031
`
`Study Date: 26-Jun-2005 —— 20—Aug-2005
`Title: Effects on lymphocyte count and the potential for cyanide formation following
`single and multiple doses of saxagliptin in healthy subjects
`
`This study simultaneously further investigated both the lymphocyte changes upon
`multiple daily dosing with saxagliptin and also the role of the interrupted closing in the
`lymphocyte changes in the same individuals. Due to the presence of a cyano group in the
`structure of saxagliptin and its active metabolite, EMS-510849, and the preclinical
`findings in rats, there is a possibility of human exposure to free cyanide (CN—) via its
`metabolic cleavage from saxagliptin. Therefore, information about CN-exposure at the
`highest Phase 3 dose (10 mg) and a supratherapeutic dose (40 mg) was also determined
`by the sponsor in this study. Plasma samples for whole blood cyanide and plasma
`thiocyanate (SCN) concentrations were collected throughout
`the study. Urine was
`collected for the measurement of total SCN excretion.
`
`This was a double-blind, multiple-dose, randomized, parallel group, placebo-controlled
`study in 48 healthy subjects. Flu syndrome had occurred in previous saxagliptin studies,
`possibly related to dose or intermittent dosing. No flu-like syndrome was observed in the
`present study. The only treatment Sequence in which absolute lymphocyte count
`decreases were observed was on Day 23 in subjects administered placebo on Day 1, 40
`mg QD from Day 2 through Day 15, placebo from Day 16 through Day 22 and a single
`dose of 40 mg saxagliptin on Day 23 (Sequence 2). The decreases in absolute lymphocyte
`count in subjects in Sequence 2 were transient and were most pronounced 24 h following
`the 40 mg saxagliptin dose on Day 23 (mean decrease of 22 :l: 20% at 24 h post-dose on
`Day 23).
`
`Maximum percent changes in absolute lymphocyte counts (%AALC) were determined
`following single and multiple doses of saxagliptin, where
`
`141
`
`

`

`DayX‘E'V :1c_(A:m*“m°:MA1C°nEYx nylons;
`Caxtonespmdingnmeonmyi
`and X32, 0:10, 6215, or 23
`
`IawestmcadedALCanDavY
`.
`“HA! _ _—____-_
`.nCatomzonmyl
`my“ ‘
`c (
`:nd'1'=4, or 5, 019, «11,0212, or 33., or 1?, or 18, «19, «:25, o: 26,0117.
`
`4”:sz
`
`Dosing with saxagliptin did not appear to have a dose-dependent, time-dependent or
`sequence-dependent effect on plasma or urine concentrations of thiocyanate.
`
`Mean Plasma Thiocyanate Concentrations Versus Time Profiles Following
`Interrupted Administration of 10 and 40 mg Saxagliptin
`We 1 (F13)
`squat. 2 (2-12)
`"0" square 3 01-12)
`squat-I 4 (F12)
`
`QPPE'I¢!~'P'P‘.‘~."!"OWQVIOV'OV'OVIO
`l‘lmmmiwymncwin-manual(mknmmnhul.)
`
`
`
`o
`
`5
`
`‘io
`
`15
`5m:- 4,}.
`
`20
`
`:5
`
`Sequence l=Single dose 40 mg saxagliptin on Day 2, Placebo on Days 3-9, 40 mg saxagliptin on Days 9-23
`Sequence 2=40 mg saxagliptin on Days 2-15, Placebo on Days 16-22, Single dose 40 mg saxagliptin on Day 23
`Sequence 3=10 mg saxagliptin on Days 2-15, Placebo on Days 16-22, Single dose 10 mg saxagliptin on Day 23
`Sequence 4=Placebo on Days 2-23
`
`Mean (S.D.) Statistics for Amounts of Thiocyanate Recovered in Urine (ug)
`
`V
`Treatment Group
`
`.
`
`..
`
`.
`
`.
`
`.. ”anal-nae .
`
`
`
`
`
`‘
`-_. Dario
`
`'S'NM‘MU "m
`'
`'
`
`
`m_ 14:12:31) tum—mama]
`
`
`
`:W
`W...._.$..%*(h‘12
`.Wm 13mm
`
`
`
`
`
`.
`r.
`
`
`
`In vitro Studies
`
`142
`
`

`

`Study 930025627: Protein binding
`
`Serum protein binding was determined by equilibrium dialysis method. The
`concentration of parent and metabolite in the serum protein binding assay was 100
`ng/mL. Following completion of dialysis, concentrations of BMS-477ll8 and BMS-
`510849 were obtained by analyzing samples fi'om the serum and buffer compartments of
`the dialysis cell. From each dialysis cell, the free fraction of EMS-477118 and BMS-
`510849 were calculated as follows:
`'
`
`% free = 100 x (Cbuffer)/(Cserum)
`
`is the concentration of EMS-477118 or EMS-510849 in the buffer
`Where Cbuffer
`compartment, and Cserum is concentration of EMS-4771 18 or EMS-510849 in the serum
`compartment of the dialysis cell. The serum protein binding for saxagliptin and
`metabolite EMS-510849 was very low in all species tested (Table 1).
`
`Table 1: Free Fraction (%) of BMS-477118 and BMS-510849 in Mouse, Rat, Dog,
`Monkey and Human Serum (Mean 5: SD, n=3)
`
`
`
`
`—_—-“
`
`
`
`Permeability Studies:
`
`Study 930023241 Title: Evaluation of EMS-477118 Membrane Permeability via dual
`pH PAMPA (Parallel Artificial Membrane Permeability Assay)
`
`The 1'12 V1270 permeability of EMS-477118 was evaluated using the dual pH Parallel
`Artificial Membrane Permeability Assay (PAMPA). Dual pH PAMPA consists of a
`specially formulated lecithin-based lipid combination referred to here
`as
`the
`gastrointestinal tract (GIT) lipid. The GIT lipid is used to form a membrane in a sandwich
`plate assembly. The permeability coefficient (Pc) values generated in the dual pH
`PAMPA assay are reported in nm/sec and reflect the ability of a compound to passively
`cross a lipid membrane. The P0 values (mean i SD) of BMS—477l l 8 were 4 :1: 1 nm/sec
`and 59 i 10 nm/sec at pH 5.5 and 7.4, respectively. These results demonstrate pH
`dependent intrinsic membrane permeability. EMS-477118 was found to be significantly
`more permeable at the higher pH tested. Compounds are classified as highly permeable
`when Pc values are greater than 100 nm/sec and are predicted to be highly absorbed
`(>75%) in humans if the compound is primarily absorbed via a transeellular route and is
`not a substrate for active transport mechanisms. These results suggest that EMS-477118
`is likely to be minimally absorbed in humans via a transcellular absorption route. It is
`
`143
`
`

`

`to consider Caco-2' data and other models that
`important
`mechanisms when refining 1'): Viva activity predictions. I
`Study report 300861090: Assessment of pgp mediated transport in LLC-PKl cell
`monolayers
`
`include active transport
`
`Time- and concentration dependence of saxagliptin transport was evaluated. The test
`article was assayed in the A to B and B to A directions in both the P-gp transfected
`(22L1) and the vector carrying (CLD) cell line at three concentrations (3, 10, 100 uM).
`The positive controls used were mannitol, propranolol and digoxin for low, high
`permeability marker and P-glycoprotein substrate respectively. EMS-477118 transport
`was linear over the time course (left) and concentration range (right) tested.
`Mp WMTanned
`P49 rm Tmsmn
`O Sled”!!! 3 W!
`0 ”Gunilla-mu
`I IW~|77113 1: “M
`I EmdmflEE'nln
`A sworn: EEC IM
`A llfidflfliflfimln
`
`3
`
`i
`
`ES g
`
`Pre-experimental trans—epithelial electrical resistance (TEER), post—experimental lucifer
`yellow A to B flux values, as well as the digoxin polarization ratio (9.4 fold) were
`consistent with a properly functioning LLC—PKl monolayer model. EMS-477118 was
`subject to active B—A transport with polarization ratios in 22L1 (P-gp expressing) cells
`approximately 3—fold higher than in CLD (control) cells.
`
`Study report 300877516: Assessment of pgp mediated transport in LLC—PKl cell
`monolayers
`
`~3-PK1 cells expressing human P-gp cDNA
`Porcine kidney-derived, BD C
`line (LLC-PKI cells containing the vector
`(designated as 22L1) and the control cell
`without human P-gp cDNA, designated as CLD) were used. Monolayer integrity was
`evaluated by pre-experimental transepithelial electrical resistance (TEBR) measurements
`and post-experimental lucifer yellow A-to-B flux determinations for each'cell monolayer.
`Time- and concentration dependence of test article transport was evaluated. Saxagliptin
`was assayed in the A—to-B and B-to-A directions in both the P-gp transfected (22L1) and
`the vector carrying (CLD) cell line at three concentrations (3, 10, 100 uM). Samples from
`the receiver chambers were taken at three time points (60, 90, 120 min) and replaced by
`an equal volume of receiver solution. The positive controls used were mannitol,
`propranolol and digoxin for low, high permeability marker and P—glycoprotein substrate
`respectively. The digoxin polarization ratio in 22L1 was 9.6.
`
`144
`
`“0‘
`
`

`

`EMS-510849 was not subject to active B~to-A transport with polarization ratios in both
`the 22L1 (P-gp expressing) and CLD (control) cells ranging from 0.8 to 1.1 at the
`con