CENTER FOR DRUG EVALUATION AND
`
`RESEARCH
`
`'
`
`APPLICA TION NUMBER:
`
`22-350
`
`PHARMACOLOGY REVIEW; S)
`
`

`

`Executive CAC
`
`Date of Meeting: March 3, 2009
`
`Committee: David Jacobson—Kram, Ph.D., 0ND 10, Chair
`Abby Jacobs, Ph.D., 0ND IO, Member
`Paul Brown, Ph.D., 0ND IO, Member
`Barry Rosloff, Ph.D., DPP, Alternate Member
`Todd Bourcier, Ph.D., DMEP, Team Leader/Presenting Reviewer
`
`NDA # 22-350
`
`Drug Name: Saxagliptin (Onglyza)
`Sponsor: Bristol Myers Squibb
`
`.
`Background
`Saxagliptin is a dipeptidyl peptidase-4 inhibitor currently under Agency review as a
`blood-glucose lowering therapy for type 2 diabetes. The ECAC concurred with 2-yr
`carcinogenicity study protocols in rats and mice in 2003, and the Division received final
`study reports in the NDA submitted in June 2008.
`
`Mouse Carcinogenicity Study
`CD-1 mice were dosed 50, 250, and 600mg/kg saxagliptin in a vehicle of 0.5%
`methylcellulose for 2 years. Dual controls consisted of vehicle alone. Reduced survival
`prompted termination of high dose males at week 90 and of all remaining male groups at
`week 100. Female groups survived to scheduled termination at week 104. The Division '
`and ECAC were notified prior to termination of dose groups. There was no evidence of
`drug-related increases in the incidence of neoplastic lesions in any dose group. Exposure
`margins relative to AUC'at the 5mg clinical dose ranged from ~20 to 1000x (low to high
`dose) for saxagliptin and ~15 to 300x for the active metabolite BMS-510849.
`
`Rat Carcinogenicity Study
`Sprague Dawley rats were dosed 25, 75, 150, and. 300mg/kg saxagliptin in a vehicle of
`0.5% methylcellulose for 2 years. Dual controls consisted of vehicle alone. Reduced
`survival prompted termination of high dose males at week 68; this dose group was
`excluded from statistical analysis of tumor incidence. All remaining male groups were
`terminated at week 99 when survival in the dual controls collectively reached ~ 22
`animals. Female groups survived to scheduled termination at week 104. The Division and
`ECAC were notified prior to termination of dose groups. There was no evidence of drug-
`related increases in the incidence of neoplastic lesions in any dose group. Exposure
`margins relative to AUC at the 5mg clinical dose ranged from ~50 to 2200x (low to high
`dose) for saxagliptin and ~3 to 68x for the active metabolite BMS-510849.
`
`

`

`Executive CAC Recommendations and Conclusions:
`
`Mouse:
`
`0 The Committee found that the study was adequate, noting prior Exec CAC
`concurrence with the doses.
`
`0 The Committee agreed that the study was negative for drug related tumor
`findings.
`
`Rat:
`
`o The Committee found that the study was adequate, noting prior Exec CAC
`concurrence with the doses.
`
`0 The Committee agreed that the study was negative for drug related tumor
`findings.
`
`David Jacobson-Kram, Ph.D.
`Chair, Executive CAC
`
`cc:\
`
`/Division File, DMEP
`/Todd Bourcier, DMEP
`/Fred Alavi, DMEP
`/Julie Marchick, DMEP
`/ASeifried, 0ND IO
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`-.---------------—-—------.----------------.-----------------------——------—__-----—---—----------——-_--_------------
`
`David Jacobson—Kram
`3/5/2009 08:30:07 AM
`
`

`

`Tertiary Pharmacology Review
`
`By:
`-
`
`' Paul C. Brown, Ph.D., ODE Associate Director for Pharmacology and Toxicology
`0ND IO
`'
`
`NDA: 22—350
`
`_
`
`Submission date: June 30, 2008
`Drug: saxagliptin (OnglyZa)
`Sponsor: Bristol Myers Squibb
`Indication: type 2 diabetes
`
`' Reviewing Division: Division of Metabolism and Endocrine Products
`
`Comments: The pharm/tox reviewer and supervisor found the nonclinical information
`submitted for saxagliptin to be sufficient to support approval for the indication as
`proposed above;
`
`Developmental and Reproductive toxicity:
`Saxagliptin was tested in rats and rabbits in oral embryofetal studies and no indication of
`teratogenicity was observed at doses of up to 500 mg/kg in rats and 200 mg/kg in rabbits.
`Saxagliptin was also tested in rats in combination with metformin. This study had three
`groups: one vehicle and two that received 200 mg/kg metformin and either 5 or 25 mg/kg
`saxagliptin. The group receiving 25 mg/kg saxagliptin and 200 mg/kg metformin
`exhibited malformations such as craniorachischisis, absentrenal papilla, cleft palate and
`absent or shortened digits of forepaws/hindpaws. Since neither saxagliptin nor metformin
`has shown teratOgenicity alone, it appears that the combination might be responsible for
`this signal, if it is a real signal. These two drugs may be used together in the patients;
`therefore, in order to better understand the teratogenic potential, the reviewer and
`supervisor recommend that the applicant conduct additional embryofetal studies in rats
`and rabbits studying the combination of saxagliptin with metformin as post marketing
`requirements. It was also recommended that these studies include saxagliptin and
`metformin alone arms to aid in interpretation.
`
`i
`
`The reviewer and supervisor have concluded that pregnancy category B is acceptable for
`the labeling of the product under consideration in this NDA, which contains saxagliptin
`as the sole active ingredient. Some incomplete ossification was observed in the rat
`reproductive toxicity studies but this occurred at exposures that were much higher than
`expected clinically.
`'
`
`,
`Carcinogenicity:
`No drug—related tumors were observed in carcinogenicity studies in rats or mice. These
`studies were considered adequate by the executive carcinogenicity assessment committee.
`
`_
`Other issues:
`Saxagliptin produced cutaneous lesions in cynomolgus monkeys and erosive lesions in
`the paws of dogs. These lesions were only observed at exposures that were significantly
`higher than those expected in humans and no such lesions were observed in the clinical
`studies. SUch lesions have also been described in studies with other compounds of the
`
`

`

`class. Consequently, the reviewer and supervisor recommended including a description of
`these findings in the labeling.
`'
`
`Other nonclinical issues were adequately deScribed and discussed in the primary and
`secondary reviews and will not be further discussed here as they have no further impact
`on approval or labeling.
`
`‘ Conclusions:
`
`I concur with the Division pharm/tox conclusion that the nonclinical data support
`approval of this NDA. I agree that it is appropriate to collect additional information on
`the teratogenic potential of the combination of saxagliptin and metformin and that these
`may be postmarketing requirements. I agree that pregnancy category B is acceptable and
`that the labeling may contain a brief description of the cutaneous lesions noted in the
`animal studies.
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`
`Paul Brown
`
`7/23/2009 12:45:57 PM
`PHARMACOLOGIST
`
`

`

`MEMORANDUM
`
`DEPARTMENT OF HEALTH & HUMAN SERVICES
`Public Health
`Service
`
`Food and Drug Administration
`
`_____________________———————————————-
`
`Division of Metabolism and Endocrinology Products
`Center for Drug Evaluation and Research
`
`Date: June 23, 09
`From: Fred Alavi, Ph.D.
`
`Subject: Rat and rabbits embryofetal development study protocols
`
`Reference is made to the sponsor’s rat and rabbit embryofetal development study protocols
`(Protocol DN09018 and DN09020) submitted on June 17, 2009 (sequence # 0036). The
`protocolswere submitted in part to fulfill the Post Marketing Requirement (PMR) for the
`approval of saxagliptin (Onglyza®). It should be noted that both studies have been initiated and
`dosing has commenced or completed.
`
`Protocol DN09018 dated April 27, 2009 titled: Saxagliptin (MES-477118) and
`0
`' Metformin (EMS—207 150): Oral combination Study of Embryo—Fetal Development in
`Rats
`‘
`.
`0 Protocol DN09020 dated May 21, 2009 titled: Saxagliptin (BMS—4771 18) and
`Metformin (BMS-207150): Oral CombinatiOn Study of Embryo-fetal Development in
`Rabbits
`
`The proposed studies were requested by the agency in March 25 of 2009 to determine whether
`malformations found in the rat saxagliptin/metformin combination (25/200 mg/kg) embryofetal
`development study were not coincidental and due to drug-drug pharmacodynamic interaction.
`Since the study also lacked separate arms for saxagliptin and metformin alone, the study was
`considered inadequate to support the sponsors claim that metformin was responsible for the
`malformations. The malformations identified in 2 fetuses from 1 litter at 25 /200 mg/kg of
`saxagliptin/metformin werezcraniochischisis, a type of neural tube defect (incomplete closure of
`the skull and spinal column) and absence of renal papillae.
`
`Rat Emb ofetal Develo ment Stud Protocol:
`In the proposed rat study, 4 groups of pregnant SD rats (30/group) will be treated orally with the
`vehicle (water and 1.25% Avicel), saxagliptin (25 mg/kg), metformin (600 mg/kg) or
`combination of saxagliptin/metformin (25/600 mg/kg) from gestation Day 6 (GD6) through
`GD15. Standard clinical chemistry parameters including glucose will be measured from blood
`samples collected via jugular vein at 2 and 24 hrs from the first 15 rats in each group on GD 13.
`
`

`

`The plasma chemistry analysis will not include analysis of Vit B12 or folate. The toxicokinetic
`analysis will be carried out on GD 15 on the second 15 rats in- each group. Rats will be
`’ necropsied on GD 21 (May 19-22, 2009). The report is expected to be finalized by Oct 14,
`2009.
`
`In the original study pregnant rats were treated with two saxagliptin dose combination levels
`while metformin dose was kept the same (5/200 and 25 /200 mg/kg of saxagliptin/metformin).
`There were no separate saxagliptin and metformin treatment groups alone. The agency had
`asked the sponsor to repeat the study but to also include separate arms for saxagliptin and
`metformin alone. Based on the protocol, the sponsor appears to have added separate arms for
`saxagliptin and metformin but the dose of metformin is raised to 600 mg/kg from 200mg/kg in
`the original study. ,
`'
`Although the study overall appears to be fine, the increase in metformin dose is likely to pose
`additional problems: a) metformin dose of 600 mg/kg may cause maternal toxicity thus
`confounding interpretation of any positive fetal findings, b) the higher dose of metformin may
`cause fetal toxicity and change reproductive parameters that excludes comparison to the lower
`metformin dose in the original. study, 0) the study does not include the metformin dose that
`produced craniochischisis in .the original study, d) the protocol does not include not include
`plasma analysis of glucose, Vit B12 and folate. The sponsor may have chosen 600 mg/kg of
`metformin to show that metformin was responsible for the malformations, but by doing So they
`have also introduced additional complexities to the study. The new study would only be
`interpretable if there is no maternal or excessive fetal toxicity, ifreproductive parameters are
`comparable to the original rat study, and if no fetal abnormalities are observed. An outcome
`other than this would not adequately address our concern described in the PMR. The reviewer
`recommends repeating the study at dose levels of saxagliptin/metformin (25/200 mg/kg of
`saxagliptin/metformin) that caused malformation in the original rat study. The study also should
`include separate arms for saxagliptin (25 mg/kg) and metformin (200 mg/kg) alone.
`
`M S
`
`axagliptin/Metformin (or Respective Vehicle) Number of .
`
`Dose level
`Concentration Dose Volume
`Females
`(mg/kg/day)‘
`(mg/mL)
`(mL/kg/day)
`ASSlgned
`
`Group
`No.
`
`I
`
`Identification
`
`1
`2
`3
`4
`
`Vehicle controla
`Saxagliptinb
`Metformin‘
`Saxagliptin + Metformind
`
`0/0
`25/0
`0/600
`25/600
`
`0/0
`25/0
`0/150
`25/150
`
`'
`
`1
`
`1/4
`1/4
`1/4
`1/4
`
`30
`30
`‘30
`30
`
`a — 1.25% Avicel will be dosed first and immediately followed by deionized water.
`b - Deionized water will be administered first and immediately followed by saxagliptin formulation
`c - 1.25% Avicel will be administered first and immediately followed by metformin formulation.
`(1 - Saxagliptin formulation will be administered first and immediately followed by metformin formulation.
`
`,
`
`Rabbit Embryofetal Development Study Protocol:
`In the proposed rabbit study, 4 groups ofpregnant New Zealand White Hra rabbits (30/group)
`will be treated orally with the vehicle (water and 1.25% Avicel), saxagliptin (40 mg/kg),
`metformin (50 mg/kg) or combination of saxagliptin/metformin (40/50 mg/kg) from gestation
`
`

`

`Day 7 (GD7) through GD19. Standard clinical chemistry parameters including glucose will be
`measured from blood samples collected via the marginal ear vein at 1, 4 and 24 hrs from 5
`rabbits in each group on GD 18. The plasma chemistry analysis will not include analysis of Vit
`B12 or folate. The toxicokinetic analysis will-be carried out on GD 1.9 on designated animals for
`TK (5 /group). Rabbits will be necropsied on GD 29. The report is expected to be finalized by
`Novof 2009.
`‘
`
`The dose selection for saxagliptin was based on studies in the saxagliptin NDA. The metformin
`dose was based on 13—day dose ranging study where pregnant rabbits were treated With 25, 50,
`100 and 150 mg/kg from GD 7 through GD19. Metformin was maternally toxic at 100 and 150
`mg/kg to the point than none of the pregnant rabbits survived to the scheduled'C-section. In
`contrast, 50 mg/kg metformin dose was not maternally toxic, nor did it produce any gross
`external changes in fetuses. Based on that, the sponsor chose 50 mg/kg/d for-the rabbit
`combination Study. The reviewer concurs with the doses the sponsor has selected for
`saxagliptin/metformin combination embryofetal development study in rabbits.
`
`Experimental Design
`
`Metformin Daily Dose
`Number of
`Saxaghptin a y
`ose
`~
`Group
`Number
`Dose
`Volume
`Cone.
`Dose
`Volume
`Cone.
`Female Rabbits
`
`(mg/kg)
`(mL/kg)
`(mg/mL)
`(ms/kg)
`(“IL/kg)
`(mymL)
`
`1
`
`2
`
`3
`
`_ 0
`
`0
`
`40
`
`4
`
`4
`
`4
`
`o
`
`0
`
`10
`
`O
`
`50
`
`0
`
`2
`
`2
`
`2
`
`0
`
`25
`
`0
`
`’
`
`30
`
`30
`
`30
`
`30
`25
`2
`50
`10
`4
`40
`4
`
`
`

`

`External recommendationi
`
`We have reviewed protocols #DNO9018 and #DN09_020 for the rat and rabbit embryofetal
`, development studies with the saxagliptin/metformin combination and have the following
`comments:
`'
`
`The intent of repeating the rat embryofetal study was to assess the reproducibility of the neural
`tube malformation observed with the metfonnin/saxagliptin combination along with an
`evaluation of each component alone. The proposed 600mg/kg dose of metforrnin selected for the
`rat study is 3-fold greater than the metformin dose evaluated in the original rat study. Increasing
`the metforrnin dose to 600mg/kg may confound interpretation of the repeat study due to several
`factors, including unexpected maternal/fetal toxicity when combined with saxagliptin and
`alteration of C-section data that differs fi‘om the original study. Not including the dose
`combination associated with the neural tube malformation in the original study is also a
`deficiency of the current protocol. It is reasonable that a negative’ study outcome would be
`interpretable and acceptable; however, a confounded outcome will not adequately address the
`PMRissue. Therefore, we request that you submit a new study protocol that incorporates the
`following elements, should additional studies be required:
`
`0
`
`Include a 25/200 mg/kg saxagliptin/metformin combination group plus separate arms for
`saxagliptin and metforrnin in the rat embryofetal study. Additional combination groups at
`doses that bracket the 25/200mg/kg group are acceptable. Signs of maternal toxicity at
`the highest dose combination group is desirable. Doses of the separate saxagliptin and
`metforrnin arms should'equal those in the highest combination dose group;
`
`0 Evaluate at least two combination dose levels in the rabbit study that enable identification
`of a NOAEL and a maternally toxic dose. Separate arms for metformin and saxagliptin
`should also be incorporated.
`
`0 Monitor blood glucose, folate and vitamin B12 in both rat and rabbit studies to determine
`the potential role of metformin in malformations observed in rats.
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`
`Fred Alavi
`
`6/25/2009 01:24:47 PM
`PHARMACOLOGIST
`
`Brief review of the embryofetal development study protocols. Most
`of the information was conveyed to the sponsor
`V(Dr. Smith, BMS) by Tcon on Wed June
`24, 2009. We reserved the right to reject
`the studies since they had already started the
`studies witho
`
`Reveiw of rat and rabbit embryofetal development study protocols—
`part of the PMR
`
`Todd Bourcier
`6/25/2009 02:25:11 PM
`PHARMACOLOGIST
`'I concur.
`
`

`

`gJMCEr.
`
`as"
`
`“tr
`
`a
`
`If / DEPARTMENT OF HEALTH & HUMAN SERVICES
`,2
`C
`afi
`
`- Food and Drug Administration
`
`Memorandum
`
`'
`
`SUPERVISOR MEMO
`
`
`
`
`
`
`
`
`
`
`.
`.
`Onglyza (saxagliptin, DPP4 inhibitor)
`
`Drug/Indlcatlon Type2 diabetes
`
`
`Bristol Myers Squib
`
`Bristol Myers Squibb is seeking marketing approval for saxagliptin, proposed trade name
`Onglyza®, as a treatment option for Type 2 diabetes. Saxagliptin is a member of the DPP4
`inhibitor class of compounds whose primary mode of action consists of extending the half-life of
`. the incretin GLP-l, thereby enhancing glucose-induced insulin release from pancreatic beta cells.
`C
`_ 3 only Januvia® (sitagliptin) is
`appfiroved for US and global markets while Galvus® is approved for non—US markets. - ST
`
`.
`
`>
`
`\
`
`MM
`
`Dr. Fred Alavi, the primary nonclinical reviewer, concludes that the pharmacology and
`toxicology data support approval of saxagliptin (5 mg q.d.) with a post-marketing requirement to
`conduct embryofetal development studies in rats and rabbits administered saxagliptin and
`metforrnin in combination. I concur with Dr. Alavi ’3 assessment.
`
`Dr. Alavi’s decision is based on reasonable exposure margins between animal toxicities and
`clinical exposure at the proposed maximum dose of 5mg/day. Toxicity in rats, dogs, and
`monkeys was largely dose-dependent, with no adverse effects observed at low multiples (S 8
`fold) of clinical exposure. Adverse effects that were relatively minimal and reversible including
`splenic lymphoid proliferation and multi—organ lymphoid/monocytic infiltration in all species
`and non-necrotizing cutaneous lesions in monkeys were observed at 20 to 30-fold multiples of
`clinical exposure. Severe toxicity that exceeded tolerability in all test species occurred at higher
`multiples of clinical exposure. Saxagliptin did not alter tumorigenesis in rats and mice after
`lifetime (2 year) exposure to exceedingly high doses of saxagliptin and its active metabolite,
`EMS-510849 (>800 and 300 fold higher, respectively, than clinical exposure). Findings from the
`reproductive toxicology studies with saxagliptin alone did not reveal relevant toxicities without
`also producing signs of maternal toxicity. A Moderate increase in drug exposure at relevant
`clinical doses in susceptible patients (i.e. renal insufficiency) is unlikely to reproduce toxicities
`noted in animals and presents negligible risk to humans.
`
`Dr. Alavi recommends a non—clinical post-marketing requirement to conduct additional
`embryofetal development studies based on findings of teratogenicity in rats co—administered
`saxagliptin and metformin. I agree with his recommendation. If approved, saxagliptin would be
`commonly prescribed as an add-on therapy to metformin; therefore, findings of teratogenicity
`
`

`

`with the saxagliptin/metformin combination are concerning and are relevant to the monotherapy
`NDA. Craniochischisis identified in the saxagliptin/metformin combination study is a rare and
`serious neural tube malformation involving the incomplete closure of the skull and spinal cord.
`The relevance of this finding should not be dismissed, as yet, based on exposure margins alone.
`Alternative explanations for the apparent teratogenic interaction of saxagliptin and metfonnin
`were confounded by the lack of separate arms for saxagliptin and metformin alone in the study
`design. Arguments that metformin altered vitamin B12 and folate levels which caused the neural
`tube malformation were unconvincing because metformin exposure was equal in the two
`combination dose groups but teratogenicity occurred only in the group with the higher dose of
`saxagliptin. Also, the original embryofetal studies with metfonnin did not report
`craniochischisis. Dr. Alavi’s and my recommendation are consistent with advice received from
`the Associate Director and Director of Pharmacology/Toxicology and the Reproductive
`Toxicology Subcommittee at FDA. Special contraindications in the monotherapy label are not
`recommended at this time, but will be reconsidered afier review of the additional studies.
`
`One issue of some concern during development of saxagliptin identified by Dr. Alavi was the
`finding of cutaneous lesions in cynomolgus monkeys. Since first reported to the FDA nearly five
`years ago, similar toxicity has been encountered in monkey studies with several other DPP4
`inhibitors in development. Past clinical experience suggest that the findings in monkeys are
`relevant to human risk, and therefore should not be dismissed. Mechanistic data from several
`sponsors variably implicate off-target inhibition of DPPs 8 or 9, peripheral .vasoconstriction, or
`ischemic vascular injury underlying the lesions, but as Dr. Alavi rightly states, the etiology
`remains unresolved. The lesions observed with saxagliptin are remarkably consistent with the
`class regarding the peripheral location, time-dependent emergence, and dose-related severity of
`the lesions. Saxagliptin at clinical exposure and low multiples thereof did not produce lesions in
`monkeys, and a further 20-fold increase in exposure was required to produce minimal, self-
`resolving lesions. Of note, saxagliptin also induced minimal erosive lesions of the paws in dogs
`after 12 months exposure at doses 2 35 times clinical exposure. This is unlike the case with
`vildagliptin, currently approved in non—US markets, whereby an equivalent 20-fold multiple of
`clinical exposure resulted in lesions of such severity in monkeys that amputation and humane
`sacrifice was required. The currently marketed DPP4 inhibitor sitagliptin does not induce
`cutaneous lesions in monkeys, which has been independently verified by several sponsors
`including the sponsor of saxagliptin. It is important to note that the cutaneous lesions described
`herein are separate from hypersensitivity reactions that have been reported from post-market
`experience with sitagliptin and vildagliptin. I agree with Dr. Alavi’s assessment that the risk to
`human subjects is minimal based on the exposure margins to the no-effect and lowest—effect
`doses in monkeys. This assessment is further supported by the lack of adverse cutaneous events
`in human subjects given up to 400mg saxagliptin in phase I studies or given 5 to 10mg
`saxagliptin in phase 3 clinical studies.
`
`Dr. Alavi discusses potential. immunotoxicity issues with saxagliptin, concluding that the risk of
`severe adverse effects is minimal but that the risk of more subtle immune system-related AEs
`remains to be seen in the post-approval period. Preclinical signs of potential adverse effects on
`immunity generally occurred at high doses (2 20x) of saxagliptin, notably splenic lymphoid
`hyperplasia and multiorgan lymphoid or monocytic aggregates in rats and dogs. Saxagliptin at
`high exposure also partially inhibited the humoral response to KLH, but this effect likely reflects
`inhibition of DPP8/9 activity which is known to suppress lymphocyte proliferation. A small but
`consistent 5-10% reduction in circulating lymphocytes was reported in human subjects given
`10mg saxagliptin, though without any clear clinical correlate. There are instances of reduced
`
`

`

`lymphocytes in monkeys, dogs, and rats given high doses of saxagliptin, but the finding was not
`consistent across studies and did not follow a dose- or time-dependency; thus, the animal studies
`do not provide much insight to this observation in humans. I agree with Dr. Alavi that there
`appears to be minimal to no risk of severe immunotoxicity with saxagliptin. I also agree that the
`risk of more subtle changes in immunity cannot be excluded from the available preclinical data,
`particularly given the reported role of DPP4 in deactivating numerous chemokines and its role as
`CD26 in lymphocyte responses to antigen.
`
`Another issue of some interest discussed by Dr. Alavi was the finding of severe lesions in
`various parts of the brain in male rats administered high doses of saxagliptin. The sponsor
`undertook a number of mechanistic studies which convincingly demonstrated that the brain
`lesions were secondary to release of cyanide from the saxagliptin structure (which contains a
`cyano group) by CYPZCl 1, an androgen—regulated metabolizing enzyme. As detailed in Dr.
`Alavi’s review, the studies demonstrated that expression of CYP2C11 is highest in male rats, that
`methods to reduce or inhibit CYP2C11 also reduced development of brain lesions, and that
`notably high levels of cyanide were present in blood from male but not female rats, nor from any
`other species including dogs, monkeys, and humans. The molecular structure of several DPP4
`inhibitors includes a cyano group, but the release of cyanide via a metabolic pathway has only
`been documented with saxagliptin. Studies comparing the metabolism of two cyano-containing
`DPP4 inhibitors, saxagliptin and vildagliptin, demonstrated cyanide release from the former but
`not the latter. Our preclinical experience with saxagliptin ensures that the potential release of
`cyanide from other cyano-containing DPP4 inhibitors in development would be detected in the
`course of regulatory toxicology studies.
`
`Non-clinical labeling issues to be resolved prior to an ‘approval’ action include refining the post-
`marketing requirement and revising language in sections on Pregnancy, Nursing Mothers, and
`Animal Toxicology.
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`
`Todd Bourcier
`6/1/2009 03:34 :08 PM
`PHARMACOLOGIST
`
`

`

`CARCINOGENICITY ASSESSMENT COMMITTEE (CAC/CAC-EC) REPORT
`AND
`
`FDA-CDER RODENT CARCINOGENICITY DATABASE FACTSHEET
`
`P/T REVIEWER(s): Fred Alavi, PhD.
`DATE:
`Feb 23, 2009
`
`NBA:
`DRUG CODE#:
`CAS#:
`
`22-350 (IND 63,634)
`BMS-477118
`945667-22-1
`
`DIVISION(s):
`DRUG NAME(s):
`
`DMEP
`Saxagliptin (ONGLYZA®)
`
`Bristol Meyer and Squib Pharmaceuticals (BMS)
`SPONSOR:
`(-
`D
`LABORATORY:
`CARCINOGENICITY STUDY REPORT DATE:
`June 2008
`
`b 4)
`(
`
`THERAPEUTIC CATEGORY: Type 2 Diabetes
`PHARMACOLOGICAL/CHEMICAL CLASSIFICATION:
`
`Dipeptidylpeptidase IV (DPIV) inhibitor
`
`MUTAGENIC/GENOTOXIC (y/n/equivocal/na; assay): No
`Saxagliptin was not genotoxic in Ames assay, in—vz'vo rat micronucleus assay, in-vivo/z'n-vitro
`peripheral blood lymphocyte assay (1 month rat study) and oral DNA repair assay in male
`rats with BMS-477l 18 containing ( I .degradant ‘
`3
`However, in an earlier
`clastogenicity assay using primary human lymphocyte assay saxagliptin containing, ( )
`degradant ("
`_
`, saxagliptin was found to be clastogenic. The clastogenicity ’was
`attributed to 7C
`3 ,degradant impurities. The impurities including C
`D
`which were as 'high as C 1' in preclinical and {w -5 in clinical studies were later reduced
`to less than C
`in commercial preparations (Process D). Overall saxagliptin was
`considered nongenotoxic.
`
`11(4)
`
`Alavi-
`
`1
`
`

`

`MOUSE CARCINOGENICITY STUDY:
`
`MOUSE STUDY DURATION (weeks): 104 Weeks (DN03100)
`STUDY STARTING DATE:
`Dec 30, 2003
`STUDY ENDING DATE:
`Jan 16, 2006
`MOUSE STRAIN:
`Crl:CD-l®(ICR)BR mice
`ROUTE:
`Oral gavage, 5 ml/kg, acidified water
`
`DOSING COMMENTS:
`
`The carcinogenicity protocol was submitted to the agency (IND 63,634) and reviewed by Dr.
`Colerangle. The sponsor had proposed to use 0, 0, 50, 150 and 500 mg/kg/d for both sexes
`with two identical vehicle controls (acidified water). The dose selection was based on MTD
`determined by relatively mild toxicity seen at 2 1000 mg/kg/d in the 3-month mouse dose—
`ranging study (30, 100, 300, 600, 1000 and 1500 mkd). The toxicology reviewer agreed with
`the sponsor’s doses selection, however, the ECAC recommended 0, 0, 50, 250 and 600
`mg/kg/d for both male and female mice due to mild toxicity of saxagliptin in the dose~
`ranging study.
`
`NUMBER OF MICE:
`
`60 per sex
`- Control-l (Cl):
`60 per sex
`— Control-2 (C2):
`60 per sex
`- Low Dose (LD):
`- Middle Dose (MD): 60 per sex
`- High Dose-l (HDl): 60 per sex
`
`MOUSE DOSE LEVELS* (mg/kg/day):
`- Low Dose:
`50
`- Middle Dose:
`250
`
`- High Dose-1:
`
`600
`
`BASIS FOR DOSES SELECTED: MTD, determined by relatively mild toxicity in mice
`
`PRIOR FDA DOSE CONCURRENCE: Yes, Nov 4, 2003
`
`MOUSE CARCINOGENICITY: Negative
`There were no statistical significant differences in the incidence of tumors between control
`and saxagliptin treated groups. There was no statistically significant difference in survival
`rate in female mice. However, saxagliptin at 250 and 600 mg/kg/d significantly reduced
`survival rate in male mice. The analysis of mouse survival rate and tumor incidences was
`also carried out by Dr. Rahman which was in agreement with the sponsor’s conclusion.
`
`MOUSE TUMOR FINDINGS: There were no statistically significant drug-related tumor
`findings in mice.
`
`Alavi-
`
`2
`
`

`

`MOUSE STUDY COMMENTS:
`
`Mice were give saxagliptin doses of 50, 250 and 600 mg/kg/day for 2 years with 2 identical
`vehicle control (acidified water). The two controls were combined and compared to the
`treated mice. Study protocol was reviewed by the Division and approved by eCAC. The
`study design and statistical method used in the study appeared to be adequate and were
`similar to statistical methods used by FDA statistician.
`
`Oral administration of 250 and 600 mg/kg/d to male mice produced significant increase in
`mortality leading to early termination of the MD and HD males at week 90 and all remaining
`males at week 100 when the survival rate was S 25%. Final survival in male mice were 51%,
`38%, 36%, 25% and 25% in Control 1, Control 2, 50, 250, and 600 mg/kg/day groups,
`respectively. There was no significant drug—related increase in mortality in female mice and
`reached scheduled necropsy. Final survival rate in female mice were 22%, 28%, 33%, 27%,
`and 27% in Control 1, Control 2, 50, 250, and 600 mg/kg/day groups, respectively.
`
`Statistical analysis provided by the sponsor found no statistically significant difference in the
`incidence of tumors between control and saxagliptin treated male or female mice. There
`were no notable differences in gross and histopath findings between control and saxagliptin
`treated mice. The analysis carried Out by the FDA statistician, Dr. Raman was in agreement
`with the sponsor analysis. There were no saxagliptin related increases in incidence of tumors
`in CD mice. Saxagliptin exposure in male mice at 50, 250 and 600 was 20, 428 and .870 fold
`the clinical dose of 5 mg based on AUC, respectively. Saxagliptin exposure in female mice
`was 32, 376 and 1165 fold the clinical dose of 5 mg based on AUC, respectively. The
`exposure to active metabolite, EMS—510849 at parent doses of 50, 250 and 600 mg/kg/day in
`male and female mice are shown in table below. EMS-510849 exposure in mice and humans
`were overestimated by 20% and 6.7%, respectively. Slightly lower safety margin for EMS-
`510849 is unlikely to change safety profile of saxagliptin or its active metabolite.
`
`
`
`Safe marins:
`Species
`
`_-
`.
`104-WeekMouse Cami
`3de
`
`
`
`
`
`
`-
`. nose" .fSaXagliptin. ,-_ ..
`
`.
`:mg/kg/d _.AUC_,?ng'.h_/rnl
`.
`.
`,
`-
`
`
`M1605: F1615
`
`
`
`-BMS-510849
`' safetyz'margins based .on AUC
`,.
`
`'
`- Animal/Hurrian)‘
`AUC: nah/m1
`
`
`
`
`
`,
`-
`U
`-
`...Sfaxaglipfinh_
`
`
`
`
`
`
`
`
`
`BMS-510849 *
`
`“___—
`'438
`
`
`
`
`
`*Saxagliptin is metabolized in all species primarily to an active metabolite, EMS-510849. This
`metabolite is half as potent as the parent but more selective to DPPIV. The initial HPLC analysis
`used for AUC calculations apparently overestimated EMS-510849 exposure due to in adequate
`resolution of BMS-S 10849 peak from two minor metabolites, thus all submitted AUC values for
`BMS—510849 in mice, rats, pregnant rabbits, dogs, Cyno monkeys and humans were overestimated by
`20%, 42.7%, 11.1%, 36.2%, 15.1% and 6.8%, respectively. Therefore the corrected safety margins
`for EMS-510849 are lower than safety margins shown in the table above. The safety margins for
`EMS-510849 in the table are from the original data and not corrected for lower EMS-510849
`exposures.
`
`Alavi-
`
`3
`
`

`

`RAT CARCINOGENICITY STUDY:
`
`RAT STUDY DURATION: 104 weeks (DN05004)
`STUDY STARTING DATE: Jan 10, 2005
`STUDY ENDING DATE:
`Jan 30, 2007
`RAT STRAIN:
`Hsd: Sprague Dawley
`ROUTE: Oral gavage 5m1/kg
`
`DOSING COMMENTS: The ca