CENTER FOR DRUG EVALUATION AND
`
`RESEARCH
`
`APPLICA TION NUMBER:
`
`22-341
`
`PHARMACOLOGY REVIEW! S)
`
`

`

`Tertiary Pharmacology/Toxicology Review
`
`Paul C. Brown, Ph.D., ODE Associate Director for Pharmacology and
`By:
`Toxicology
`OND IO
`
`NDA: 22-341
`Submission date: May 23, 2008
`Drug: Iiraglutide
`Sponsor: Novo Nordisk Inc.
`Indication: treatment of type 2 diabetes mellitus
`
`Reviewing Division: Division of Metabolism and Endocrinology Products
`
`»
`Background comments:
`Liraglutide is a recombinant human GLP-1 analog. It is Iipidated and,
`consequently, is resistant to peptidase degradation thus it has a prolonged half-
`life compared to unmodified GLP-1. The pharmacology/toxicology reviewer and
`supervisor recommend that Iiraglutide not be approved for the treatment of type 2
`diabetes mellitus. The primary concern is that Iiraglutide produced increased
`thyroid C-cell tumors in rats and mice.
`
`-
`
`Carcinogenicity:
`Nonclinical studies of Iiraglutide included assessment of carcinogenic potential in
`two-year rat and two year mouse studies. These studies were conducted by
`subcutaneous injection which is the route of administration in humans. The
`protocols for these studies were reviewed by the executive carcinogenicity
`assessment committee. The high doses were selected based on the maximum
`tolerated dose in male rats and on achieving an AUC in female rats and male
`and female mice of greater than 25 fold the human AUC. Male and female
`Sprague Dawley rats received doses of 0 (vehicle), 0.075, 0.25, or 0.75
`mg/kg/day. Male and female CD-1 mice received doses of 0, 0.03, 0.2, 1, and 3
`mg/kg/day.
`
`Liraglutide significantly increased the incidence of thyroid C-cell adenomas in
`male and female rats at 20.25 mg/kg, C-cell carcinoma in males at 0.75 mg/kg,
`and combined C-cell adenomas and carcinomas in males and females at 20.25
`mg/kg.‘The doses of 0.075, 0.25 and 0.75 produced systemic exposures in rats
`of approximately 0.5, 2 and 8 fold the human AUC.
`.
`
`Liraglutide significantly increased the incidence of thyroid c-cell adenomas at 2 1
`mg/kg in male and female mice and C—cell adenomas and carcinomas
`(combined) at 2 1 mg/kg in females. The doses of 0.03, 0.2, 1 and 3 mg/kg
`produced systemic exposures in mice of approximately 0.2, 2, 10 and 45 fold the
`human AUC.
`
`The executive carcinogenicity assessment committee reviewed these studies
`and although there was some doubt that AUCs of greater than 25 fold the human
`
`1
`
`

`

`AUC were achieved, the studies were found to be acceptable since the doses
`were clearly high enough to elicit a carcinogenic response.
`
`The occurrence of the same tumors in both sexes of both species at relevant
`human exposures raises the possibility that patients may be at increased risk for
`this tumor. The applicant conducted a number of nonclinical studies to explore
`the mechanism of tumor formation and its relevance to humans. The applicant’s
`initial theory was that liraglutide bound to GLP-1 receptors on thyroid C-cells and
`increased the synthesis and secretion of calcitonin from these cells. As a result of
`this activation, the applicant proposed that C-cells proliferated and eventually,
`with persistent activation, progressed to adenomas and carcinomas. The
`applicant believed that this mechanism was not relevant to primates including
`humans. The pharm/tox reviewer concluded that the data submitted by the
`applicant did not adequately establish this mechanism nor did the data
`demonstrate that such a mechanism would be irrelevant to humans. The
`executive carcinogenicity assessment committee agreed that the applicant had
`not shown convincingly that the tumor findings were irrelevant to humans.
`
`The applicant also noted that, regardless of mechanism, no hyperplasia or
`tumors were observed in monkey studies of up to 87 weeks duration. However,
`the monkey data for liraglutide is not particularly reassuring. Although the data on
`mechanism provided by the applicant is not definitive, the highly specific nature
`of the tumor induction and the lack of genotoxicity suggest that the mechanism is
`most likely related to an exaggerated pharmacologic effect. The only systemic
`tumors that were increased in the studies were thyroid C-cell tumors and it is
`reasonable to expect at least some binding and activation of thyroid C-cells by
`GLP-1 receptor agonists based on the collective data. The applicant provided
`data that showed a lack of GLP-1 receptors in the thyroids' of monkeys. However,
`published information has shown that the GLP-1 receptor can be‘detected in
`human thyroids and in human medullary thyroid tumors. If a compound was
`causing an increase in tumors in multiple tissues by some general mechanism
`then a lack of tumor findings in the monkey might be somewhat reassuring;
`however, liraglutide is producing a very specific signal which suggests that a very
`specific mechanism is the cause. The monkey data would be reassuring if it were
`clear that the monkey accurately represented the human with regard to the
`suspected mechanism. Since the currently available data suggests that at least a
`subset of humans have GLP-1 receptors in the thyroid while it is not clear that
`monkeys do, it does not appear that additional monkey studies would be
`informative unless it can be shown that they express GLP-1 receptors in their
`thyroids similarly to the subset of humans that express the receptor in the thyroid.
`
`The concern about the tumorigenic potential of liraglutide would be diminished if
`it was clearly demonstrated that it produced tumors by a mechanism that was not
`relevant to humans. The concern of using liraglutide for a particular duration
`might also be diminished if it could be shown that there was a duration of
`treatment during and after which the probability of tumorigenesis was low.
`
`

`

`Liraglutide is a nongenotoxic carcinogen therefore it is possible that early
`preneoplastic events induced by liraglutide could be reversible. Studies of up to
`26 weeks in rats did not show preneoplastic effects in the thyroid; however,
`thyroid C—cell hyperplasia was noted in mouse studies of 4 to 13 week duration.
`The reversibility of the hyperplasia was assessed in mice treated for 9 weeks.
`The hyperplasia was mostly although not completely reversed after a 15 week
`recovery period. Whether any of the lingering thyroid effects would have
`progressed to adenoma or carcinoma is unknown.
`'
`
`Conclusions:
`The thyroid C-cell tumors observed in the mouse and rat raises the concern that
`patients taking liraglutide may be at increased risk for this tumor. Relevance to
`humans is not firmly established but it appears possible that at least a segment
`of the population could be at increased risk. Additional nonclinical studies may
`help better define the risk. An understanding of the mechanism by which the
`tumors are induced and of its relevance to humans could assist in determining
`risk. In addition, demonstration that tumors did not develop in a responsive
`animal species after subchronic exposure followed by long term observation may
`help determine whether the risk of tumorigenesis might be relatively low for
`patients treated for a limited duration even if the mechanism of tumor formation is
`relevant to humans.
`
`The pharm/tox reviewer and supervisor concluded that liraglutide should not be
`approved unless the sponsor can demonstrate that the tumors are not of human
`relevance.
`
`_
`Options for approval:
`I agree that the tumor findings are significant enough to warrant further
`evaluation of the risk. One option would be to require this evaluation prior to
`approving the drug. Once additional studies were completed, the relevance and
`risk could be reassessed. This would prevent any risk to patients until the drug
`was later approved, if it was approved at all.
`
`If clinical benefit is considered great enough, then another possible option would
`be to approve the drug with postmarketing commitments to further assess the
`relevance and risk. Once additional studies were completed, the relevance and
`risk could be reassessed and further regulatory action considered, if any. This
`would potentially place some patients at an increased risk of tumor development.
`The exact number of patients at risk and level of risk would be unknown. It may
`be possible that the risk would be relatively low for the period of time during
`which further studies were being conducted even for those patients expressing
`GLP—1 receptors in the thyroid given that liraglutide is a nongenotoxic carcinogen
`in animals. In addition, limiting the maximum duration that patients should be
`treated with liraglutide and excluding patients with pre-existing thyroid disease
`could further reduce the risk. Enhanced clinical monitoring for thyroid tumor
`development until the issue of human risk is further resolved might also permit
`
`

`

`liraglutide use while minimizing the number of patients that develop malignant
`tumors. However, it is acknowledged that it would be challenging to incorporate,
`as a part of routine therapy, adequate and clinically acceptable monitoring for
`preneoplastic thyroid effects and thyroid tumors.
`
`Possible additional nonclinical studies:
`A number of nonclinical studies might be conducted to address the relevance
`and risk of the thyroid C-cell tumors and the division has been considering these
`possibilities. A plan would need to be developed with the applicant. The primary
`areas to consider are:
`1. Additional studies that establish a mechanism of action and the relevance of
`the mechanism to humans.
`2. Studies that establish that the thyroid GLP-1 receptor expression in the thyroid
`is an appropriate model of human thyroid GLP-1 receptor expression and tumor
`development for those humans that do express GLP-1 receptor in the thyroid. A
`lack of tumor development in monkeys expressing GLP—1 receptor in the thyroid
`could then be used to support a treatment duration of relatively low risk in
`humans.
`
`3. Studies that establish a duration of treatment which does not result in tumor
`development during or after the period of treatment. That is, that the
`preneoplastic events do not progress in the absence of continued treatment. This
`might be demonstrated by treating mice for various durations and following
`lifetime tumor development.
`
`

`

`Application
`Type/Number
`
`Submission
`Type/Number
`
`Submitter Name
`
`Product Name
`
`NBA-22341
`
`ORlG-1
`
`NOVO NORDISK
`INC
`
`VICTOZA (LIRAGLUTIDE)
`,
`
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`
`PAUL C BROWN
`
`01/22/2010
`
`

`

`DEPARTMENT OF HEALTH & HUMAN SERVICES
`Food and Drug Administration
`
`Date: July 13 , 2009
`From: Karen Davis-Bruno PhD; Pharmacology Supervisor; DMEP
`To: NDA 22-341 Victoza (liraglutide)
`Re: Supervisor’s Memo Pharmacology/Toxicology Review NDA 22-341; A. Parola; PhD
`
`Memnramlum
`
`Glucagon-like peptide-l (GLP-1) is a circulating endogenous peptide of 30 amino acid
`sequence (7-36 amide) that is secreted from epithelial L—cells in the distal small intestine
`and colon in response to food. GLP-l improves glycemic control by stimulating glucose-
`dependent insulin secretion, increases insulin synthesis, inhibits glucagon secretion,
`slows gastric emptying and acid secretion and decreases food consumption through the
`GLP-1 receptor (GLP-IR). This receptor is widely distributed throughout the body (e.g.
`alpha, beta, delta cells of the pancreas, peripheral and central nervous system, heart,
`kidney, type II pneumocytes, parietal cells). Various signal transduction pathways have
`been associated with GLP—IR depending on the tissue. Systemic activity of GLP—1 is
`limited because it is readily degraded (tl/2<2 minutes) by a pervasive endopeptidase
`(membrane, soluble forms) known as dipeptidylpeptidase-4 (DPP-4) and subsequent
`renal clearance. Liraglutide is a GLP-l analogue that is lipidated so that it is resistant to
`metabolism by DPP-4 (ti/2 =hours). Its chemical structure tends to promote self—
`association following subcutaneous injection slowing systemic absorption which in
`conjunction with high protein binding tends to further increase its elimination half-life
`particularly limiting its renal excretion.
`
`Liraglutide has been evaluated in pharmacology, acute and chronic, genetic, and
`reproductive toxicology studies and carcinogenicity studies consistent with a
`development program for a new molecular entity for chronic use. A series of
`mechanistic studies were performed to explore liraglutide-induced thyroid C-cell tumors
`observed in rodent carcinogenicity studies.
`
`Clinical Relevance of Rodent Thyroid C-cell Tumors
`Liraglutide was negative in a series of genetic toxicity evaluations. Liraglutide has
`demonstrated dose—related, carcinogenic potential in both genders with multiple rodent
`species (rat, mouse) at multiple tissue sites (thyroid C-cell, skin) following life-time
`treatment with liraglutide at clinically relevant systemic exposures.
`In rats, thyroid C—cell
`adenomas and carcinomas occur at low multiples of the proposed clinical exposure. In
`' mice thyroid C-cell adenomas occur at 10-times (males, females) and carcinomas 45-
`times (females) the clinical exposure. The earliest appearance of thyroid C-cell
`carcinoma occurred after 15 months treatment in an early decedent mouse from the
`carcinogenicity study. Mechanistic studies using adolescent (age 2 months) or adult (age
`8 months) rats exhibit tumors after 7 months of treatment. Thyroid C—cell
`hyperproliferative changes (hyperplasia, adenoma, carcinoma) are rare findings (<1%) in
`mouse carcinogenicity studies. Diffuse and focal hyperplastic responses as well as
`adenomas are common findings in aging rats, however malignant C-cell carcinoma is a
`rare finding (<1%) in a rat carcinogenicity study.
`
`

`

`Mechanistic studies were performed to establish a mode of action for the rodent tumors
`based on liraglutide—induced GLP-1 receptor mediated calcitonin synthesis and secretion.
`The thyroid C-cell hyperplasia with progression to tumors was attributed to an increased
`release of calcitonin in the thyroid. These mechanistic studies have been extensively
`reviewed by DMEP, CDER Executive Carcinogenicity Assessment Committee (ECAC)
`and discussed at an April 2009 Advisory Committee as well as a CDER Regulatory
`briefing June 2009, convened to evaluate the safety and efficacy of liraglutide. The
`outcome of these series of reviews is that the mechanistic studies were inadequate to
`support the sponsor’s proposed mechanism of action and that there was insufficient
`evidence to demonstrate that the C-cell tumors were rodent specific and not therefore
`relevant to humans. NovoNordisk has acknowledged that a mechanism has not yet been
`established to explain the thyroid tumorigenicity.
`
`Recent data under review from other GLP—1 receptor agonists with longer half-lives
`(t1/2) than endogenous GLP-l peptide as well as sustained release formulation of short—
`acting GLP-l analogues (i.e. exenatide), suggest that persistent receptor activation is
`associated with thyroid C-cell hyperproliferation in rodents. Monkeys have not shown
`proliferative thyroid C-cell lesions following liraglutide treatment up to 20 months at >60
`times human exposure. Caution needs to be exercised in interpreting the relevance of
`monkey toxicities studies to address a potential human risk for carcinogenicity. These
`monkey studies were not powered nor designed to evaluate carcinogenicity. The duration
`of treatment in these'primate studies was only 5% of a monkey’s lifespan whereas in the
`rodent carcinogenicity studies many more animals were evaluated following lifetime
`exposure which more closely mimics the intended therapeutic use of liraglutide as anti-
`diabetic agent. Furthermore, liraglutide was immunogenic in monkeys but not in mice or
`rats. Anti—liraglutide antibodies were shown which cross react with GLP—l after 12
`months of liraglutide treatment and the neutralizing effects of these anti-drug antibodies
`were not assessed with regard to liraglutide exposure in these monkeys.
`
`Dr. Parola’s Pharmacology/Toxicology review of the available nonclinical data
`recommends not approving this marketing application.
`I agree with his recommendation
`based on the clear drug-related carcinogenicity signal observed in rodent life-time
`bioassays at relevant therapeutic exposures. He concludes that the mechanistic data
`presented regarding thyroid C-cell tumors were inconclusive in substantiating these
`findings as rodent specific. Therefore by inference these findings are potentially
`clinically relevant as thereis an absence of supporting evidence otherwise. He has
`recommended that NovoNordisk perform further mechanistic studies to establish their
`claim that these findings are rodent specific. Suggestions for further studies have been
`discussed with NovoNordisk at the end of review meeting held June 2009. These
`suggestions are included in Dr. Parola’s review under section I.B. Recommendations for
`nonclinical studies. The nonclinical deficiency lies in the inability to dismiss the rodent
`carcinogenicity findings as rodent specific and therefore of human relevance until proven
`otherwise by supportive studies of the sponsor’s design and at their discretion. The
`sponsor suggests that clinical monitoring for calcitonin is an appropriate biomarker for
`hyperproliferative thyroid C-cell lesions. However calcitonin has not been established as
`
`

`

`an adequate predictive biomarker for these lesions. Human medullary thyroid carcinoma
`(MTC) is mediated by activating rearranged during transfection (RET) mutations. There
`may be a correlative increase in plasma calcitonin in these patients, however the elevation
`is not considered predictive of the MTC diagnosis. It is unclear if drug—induced
`medullary carcinoma follows the same mechanisms as RET associated MTC.
`Fibrosarcomas, Local Tolerance and Impurities Concerns
`Liraglutide is administered by daily subcutaneous injection. Dose—related malignant
`fibrosarcomas were observed in male mice at the injection site as well as in the dorsal
`skin and subcutis. The incidence of injection site fibrosarcomas in high dose male mice
`exceeded concurrent and historical controls and reached statistical significance at 45-
`times clinical systemic exposure by trend test and not pair-wise comparisons. These
`tumors were not considered drug related by ECAC. Inflammation was not noted at the
`injection sites in mice above concurrent control incidence. Injection site reactions were
`described as subacute inflammation and fat necrosis in pigs and monkeys._ In primates
`thickening at the injection site with inflammation, necrosis and fibrosis were noted with
`continued dosing. This has relevance in that Dr. Parola’s Pharmacology/Toxicology
`review indicates that the mouse carcinogenicity study was performed using a liraglutide
`dosing solution that was 10-times less concentrated than the clinical formulation. Other
`repeat dose toxiCology studies used dilute dosing solutions of liraglutide relative to the
`clinical concentration suggesting that the local toxicity of liraglutide may not have been
`thoroughly assessed in the nonclinical program. Furthermore, different drug lots were
`used in the various nonclinical studies which may have diverse impurity profiles.
`NovoNordisk groups the impurities in liraglutide and specifications as a group, rather
`than on the individual impurity. Since the sponsor has structurally identified these
`impurities (see pg. 15 Table 1 of review) it isn’t clear why the specification was
`established for groups of impurities.
`
`Subchronic animal studies qualify the general toxicity profile up to the proposed drug
`product specifications. However impurities were not qualified in stand alone genetic
`toxicology studies (in vitro) and the carcinogenicity studies establish a positive drug-
`related thyroid C-cell tumorigenic response in both species at low clinical exposures. A
`statistically significant incidence by trend analysis for skin/subcutis fibrosarcomas are
`seen in male mice at 45-times clinical exposure only. This might reflect the
`concentration of impurity or this may be coincidental. The sponsor has identified the
`grouped impurities components as 7 ."K——-—-
`. These are unlikely to have a
`
`structural alert for genotoxicity.
`'
`_
`_
`are generally exempt from
`genotoxicity testing according to ICHS6 guidelines on biologics. Liraglutide is a
`lipidated peptide synthesized by acylating a recombinant peptide. Subcutaneously
`injected anti-diabetic agents (i.e. insulin) are commonly associated with local injection
`site reactions under clinical use. While an extensive evaluation of injection site reactions
`may not have been assessed in the nonclinical development program the clinical
`formulation has been evaluated in the clinical trials without significant concern and
`therefore additional nonclinical testing of local toxicity and impurities genotoxicity may
`not be necessary. The fibrosarcoma incidence and thyroid C—cell tumors are described in
`the labeling comments.
`
`b(4
`)
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`
`Karen Davis—Bruno
`
`7/13/2009 12:31:29 PM
`PHARMACOLOGI ST
`
`P/T sup memo
`
`

`

`
`
`DEPARTMENT OF HEALTH AND HUMAN SERVICES
`PUBLIC HEALTH SERVICE
`FOOD AND DRUG ADMINISTRATION
`CENTER FOR DRUG EVALUATION AND RESEARCH
`
`PHARMACOLOGY/TOXICOLOGY REVIEW AND EVALUATION
`
`NDA NUMBER:
`
`SERIAL NUMBER:
`
`22,341
`
`000
`
`DATE RECEIVED BY CENTER:
`
`5/23/2008
`
`PRODUCT:
`
`Victoza® (liraglutide for injection)
`
`INTENDED CLINICAL POPULATION:
`
`Adults with Type 2 Diabetes Mellitus
`
`SPONSOR:
`
`Novo Nordisk Inc., Princeton, NJ
`
`DOCUMENTS REVIEWED:
`
`Electronic submission
`
`REVIEW DIVISION:
`
`Division
`
`of Metabolic
`
`and Endocrine
`
`Products (HFD-510)
`
`PHARM/TOX REVIEWER:
`
`_
`
`Anthony Parola, PhD
`
`PHARM/TOX SUPERVISOR:
`
`Karen Davis-Bruno, PhD
`
`DIVISION DIRECTOR:
`
`PROJECT MANAGER:
`
`Mary Parks, MD
`
`John Bishai, PhD
`
`Date of review submission to Division File System (DFS): July 10, 2009
`
`

`

`TABLE OF CONTENTS
`
`EXECUTIVE SUMMARY ......................................................................................................................... 3
`
`2.6 PHARMACOLOGY/TOXICOLOGY REVIEW ............................................................................ 10
`
`2.6.1
`
`INTRODUCTION AND DRUG HISTORY .............................................................................. 10
`
`2.6.2
`
`PHARMACOLOGY .................................................................................................................... 20
`
`2.6.2.1
`2.6.2.2
`2.6.2.3
`2.6.2.4
`2.6.2.5
`
`Brief summary ......................................................................................................... 20
`Primary pharmacodynamics .................................................................................... 23
`Secondary pharmacodynamics ................................................................................ 43
`Safety pharmacology ............................................................................................... 43
`Pharmacodynamic drug interactions ........................................................................ 49
`
`2.6.4
`
`PHARMACOKINETICS/TOXICOKINETICS ........................................................................ 54
`
`Brief summary ......................................................................................................... 54
`2.6.4.1
`2.6.4.2 Methods of Analysis ..........................................._..................................................... 56
`2.6.4.3
`Absorption ............................................................................................................... 59
`2.6.4.4
`Distribution .............................................................................................................. 64
`2.6.4.5 Metabolism .............................................................................................................. 73
`2.6.4.6
`Excretion .................................................................................................................. 80
`
`Pharmacokinetic drug interactions ........................................................................... 83
`2.6.4.7
`2.6.4.10 Tables and figures to include comparative TK summary ........................................ 84
`
`2.6.6 TOXICOLOGY ............................................................................................................................ 87
`
`2.6.6.1
`2.6.6.2
`2.6.6.3
`2.6.6.4
`2.6.6.5
`2.6.6.6
`2.6.6.7
`
`2.6.6.8
`
`Overall toxicology summary ................................................................................... 87
`Single-dose toxicity ............................................................................................... 100
`Repeat—dose toxicity .............................................................................................. 101
`Genetic toxicology ................................................................................................. 164
`Carcinogenicity: ..................................................................................................... 180
`Reproductive and developmental toxicology ........................................................ 188
`Local tolerance ...................................................................................................... 225
`
`Special toxicology studies ..................................................................................... 230
`
`2.6.7 TOXICOLOGY TABULATED SUMMARY .......................................................................... 247
`
`OVERALL CONCLUSIONS AND RECOMMENDATIONS ............................................................ 285
`
`APPENDICIES ........................................................................................................................................ 298
`Appendix A: Mouse Carcinogenicity Study Review .............................................................. 298
`Appendix B. Rat Carcinogenicity Study Review .................................................................... 348
`Appendix C: Mechanistic Studies of Liraglutide—Induced Rodent C-cell Tumors ................. 379
`Appendix D: Rat Embryofetal Historical Control Data from CRL ......................................... 508
`Appendix D: Rabbit Embryofetal Historical Control Data from CRL ................................... 512
`
`

`

`EXECUTIVE SUMMAR Y
`
`Recommendations
`
`A. Recommendation on approvability: Based on the nonclinical data, this application
`is not approvable because there is insufficient nonclinical information about
`liraglutide to determine if it is safe for chronic use. In 2-year lifetime exposure
`carcinogenicity studies, liraglutide caused thyroid C-cell tumors in mice and rats at
`clinically relevant exposures. The human relevance of liraglutide-induced rodent C-
`cell tumors is unknown, and mechanistic studies performed by the applicant did not
`mitigate this risk.
`
`Recommendation for nonclinical studies:
`
`The proposed mode—of-action for liraglutide-induced rodent thyroid C—cell tumors
`based on drug-induced calcitonin secretion driving C-cell hyperplasia and tumor
`formation was not supported by mechanistic studies. The applicant should determine
`a mode-of-action for liraglutide-induced rodent C—cell tumors and evaluate the
`human'relevance of rodent C-cell tumors based on this mode—of-action. These studies
`
`may include evaluating:
`a.
`the effect of liraglutide on REarranged during Transfection (RET)
`protooncogene signaling in normal and focal hyperplastic and/or neoplastic
`thyroid C—cells in mice and/or rats. Consider determining if liraglutide alters
`phosphorylation of tyrosine residues in RET important for C-cell
`proliferation / transformation, such as Y1062. In humans, RET mutations
`constitutively activating its tyrosine kinase activity are a common cause of
`spontaneous and inherited medullary thyroid carcinoma.
`b. GLP—1 receptor expression in normal, focal hyperplastic, and neoplastic C-
`cells in liraglutide treated mice and/or rats. A published study showed 100%
`of rats, but only 60% of mice had detectable GLP-1 receptor in their thyroid,
`but liraglutide induces C-cell focal hyperplasia and tumors in both rats and
`mice. Whether or not a thyroid GLP—1 receptor is required for liraglutide’s
`proliferative effects on C-cells in mice or rats is unknown.
`0. GLP-1 receptor dependence of liraglutide—induced thyroid C—cell focal
`hyperplasia and/or neoplasia in mice and/or rats. Consider determining if
`liraglutide-induced C-cell focal hyperplasia occurs in mice or rats treated
`with a GLP-1 receptor antagonist or in GLP-1 receptor knockout mice.
`
`2.
`
`L1)
`
`Provide evidence that local toxicity after repeat subcutaneous injection with
`liraglutide was adequately assessed in nonclinical studies. In chronic repeat-dose
`toxicity studies, liraglutide caused irreversible injection site reactions in monkeys
`using drug formulations that were at least 3-times more dilute than the clinical
`formulation. Liraglutide caused fibrosarcomas in the dorsal skin and subcutis in high
`dose male mice in a 2-year subcutaneous dose carcinogenicity study and these tumors
`were attributed to local toxicity due to high drug concentration at or near the injection
`site. The concentration of liraglutide in the high dose drug formulation in the mouse
`study was 0.6 mg/mL, lO—times more dilute than the clinical formulation (6 mg/mL).
`
`Some liraglutide impurities were not qualified in genetic toxicity studies. Evaluate
`the in vitro genetic toxicity of liraglutide impurities at impurity levels consistent with
`drug substance and drug product acceptance criteria.
`
`

`

`C. Recommendations on labeling
`
`8.1 Pregnancy
`Pregnancy Category C
`Liraglutide has been shown to cause abnormalities in fetal rats at maternal
`systemic exposures 0.8 times the human exposure resulting from the maximum
`recommended human dose (MRHD) of 1.8 mg/day based on AUC. Liraglutide has been
`shown to cause reduced fetal growth, fetal abnormalities and it increased total major
`abnormalities in fetal rabbits at maternal systemic exposures 0.2 times the human
`exposure at the MRHD. There are no adequate and well-controlled studies in pregnant
`women. Liraglutide should be used during pregnancy only if the potential benefit
`justifies the potential risk to the fetus.
`
`Female rats given so. doses of 0.1, 0.25 and 1.0 mg/kg liraglutide once a day
`beginning 2 weeks—before mating to gestation day 17 had estimated systemic exposures
`0.8, 3, and 11 times the human exposure at the MRHD of 1.8 mg/day based on AUC. The
`number of early embryonic deaths slightly increased in the 1 mg/kg high dose group.
`Fetal abnormalities in kidneys and blood vessels, irregular ossification in the skull, and a
`more complete state of ossification in bones occurred at all doses. Mottled liver and
`minimally kinked ribs occurred at the highest dose. The incidence of malformations,
`including misshapen oropharynx or narrowed larynx at 0.1 mg /kg and umbilical hernia at
`0.1 and 0.25 mg/kg, exceeded concurrent and historical controls.
`
`Pregnant rabbits given 3.0 doses of 0.01, 0.025 and 0.05 mg/kg liraglutide from
`gestation day 6 through day 18 had maternal systemic exposure less than 1 times the
`human exposure at the MRHD of 1.8 mg/day based on AUC, at all doses. Liraglutide
`decreased fetal weight and dose—dependently increased total major fetal abnormalities at
`all doses. The incidence of malformations exceeded concurrent and historical controls at
`
`0.01 mg/kg (kidneys, scapula), 3 0.01 mg/kg (eyes, forelimbs), 0.025 mg/kg (brain, tail
`and sacral vertebrae, major blood vessels and heart, umbilicus), 3 0.25 mg/kg (sternum),
`and at 0.05 mg/kg (parietal bones, major blood vessels). Irregular ossification and/or
`skeletal anomolies occurred in the skull and jaw, vertebrae and ribs, sternum, pelvis, and
`scapula. Visceral anomalies occurred in blood vessels, lung, liver, esophagus, tail, and
`gall bladder.
`
`In pregnant rats given doses of 0.1, 0.25 and 1.0 mg/kg liraglutide from gestation
`day 6 through lactation day 24 (weaning), liraglutide delayed parturition to gestation day
`22 in the majority of treated rats, and decreased maternal food consumption and body
`weight gain during gestation, but not during lactation, at all doses. F1 generation body
`weight decreased at all dose levels from postpartum day 7 to week 16 in males and from
`day 7 to week 10 in females. Bloody scabs and agitated behavior occurred in F1
`generation males descended from 1 mg/kg liraglutide treated rats. Body weight of F2
`generation rats descended from liraglutide-treated females trended lower than controls
`from birth to postpartum day 14, but differences never reached statistical significance.
`
`8.3 Nursing mothers
`It is not