`
`CENTER FOR DRUG EVALUATION AND RESEARCH
`
`APPROVAL PACKAGE FOR:
`
`APPLICATION NUMBER
`
`21 -476
`
`Medical Review(s)
`
`

`

`
`
`MEMORANDUM
`
`. DATE:
`
`December 6, 2004
`
`FROM:
`
`Director
`
`Division of Neuropharmacological Drug Products/HFD-120
`
`TO:
`
`File, NDA 21-476
`
`SUBJECT: Recommendation for Approval Action for NDA 21-476, for the use
`of Lunesta (eszopiclone) in the treatment of patients with insomnia
`
`NDA 21—476, for the use of Estorra (eszopiclone; new proposed tradename
`Lunesta) in the treatment of patients with insomnia, was submitted by Sepracor,
`Inc., on 1/30/03. Although the review team recommended that the sponsor be
`sent a Not Approvable letter on initial review (although efficacy had been
`demonstrated, there were concerns about safety based on evidence in animals
`that the drug was carcinogenic), Dr. Robert Temple, Director, Office of Drug
`Evaluation l, concluded that the application should be considered Approvable
`(see his memo of 3/4/04). For this reason, the Agency issued an Approvable
`letter on 2/27/04.
`In that letter, numerous requests were made:
`
`1) additional analyses of human tumor data were requested
`2) additional analyses of adverse events listed as "Infection" and "accidental
`injury“ were requested
`3) additional analyses of the effectiveness data were requested, based on our
`concerns that the high rate of "unpleasant taste“ in the controlled trials could
`have invalidated the treatment blind
`
`4) additional analyses of orthostatic vital sign and EKG data were requested
`5) additional analyses of adverse events related to memory impairment and
`psychomotor impairment were requested _
`6) additional analyses of withdrawal phenomena and rebound insomnia were
`requested
`7) the sponsor was requested to adopt specific dissolution specifications
`8) the sponsor was asked to produce and make available a 1 mg tablet strength
`(a dose shown to be effectiveIn elderly patients)
`9) multiple CMC deficiencies were noted
`10) the sponsor was requested to supply a new tradename, because their
`proposed name, ESTORRA, was found to bear an unacceptable simiiarity to
`the marketed drug ESTRACE.
`
`The sponsor responded to the Approvable letter in a submission dated 6/14/04.
`The response has been reviewed by Dr. Karen Brugge, medical officer, Dr.
`Andre Jackson, Office of Clinical Pharmacology and Biopharmaceutics, Dr.
`Gurpreet Gill-Sangha, chemist, Dr. Aisar Atrakchi, pharmacologist, Dr. Michael
`
`

`

`
`
`Klein, Controlled Substance Staff, Dr. James L. McVey, Microbiology, Dr. Jinhee
`L. Jahng, Division of Medication Errors and Technical Support, and Dr. Paul
`Andreason, Psychiatric Drugs Team Leader. The review team recommends that
`the application be approved.
`
`I will very briefly review the sponsor's responses to the major questions posed in
`our Approvabie letter, and offer the division's recommendations for action on the
`NDA.
`
`Human Tumor Data
`
`In the initial application, it was unclear how many human tumors were treatment
`emergent.
`In particular, a number of events appeared to have been neopiasia in
`a 6-month placebo controlled trial (Study 049). Given our primary concern about
`carcinogenicity, we had asked the sponsor for clarification and analyses of these
`data.
`
`Upon re-analyses of these data, the sponsor noted 3 basal cell carcinomas (each
`of which were diagnosed about 150-170 days after treatment initiation, with some
`evidence that in two of these patients there was a lesion pre-existing prior to
`initiation of treatment) and 7 benign neoplastic events (2 uterine leimyomas, one
`each of cervical dysplasia, nevus, actinic keratosis, lipoma, and GI polyp) in the
`drug group and no tumors in the placebo group (see Tables 14 through 1-? in
`Dr. Brugge's review, pages 14-17). There was a total of 427 patient-yrs of
`exposure to eszopiclone and 67 person-years of exposure to placebo (although
`Dr. Brugge concludes that exposure should not be expressed in person-yrs for
`data in a placebo controlled trial, I disagree; this is an entirely appropriate, and
`preferred way to compare exposures, especially in the context of controlled
`data).
`
`There were two more malignancies (one basal cell, one ductal carcinoma in situ
`of the breast) in Open label exposure, and 8 more benign neoplasms in open
`label exposure (3 of which were uterine leiomyomas). Dr. Brugge notes the
`occurrence of 9 cases of fibrocystic disease, although only 3 were in controlled
`trials (this represents 3/373 women treated with eszopiclone vs. 0/125 women
`receiving placebo).
`
`Accidental Injury
`
`The sponsor performed a detailed search of terms that could reasonably be
`considered to fall under the term “accidental injury". As Dr. Brugge notes, the
`largest difference in incidence was in the 6 month controlled trial, in which 10.1%
`of drug-treated patients experienced an injury, compared to 6.2% of placebo
`patients. There was no material difference in the rate of "falls" between these
`two groups in this study (0.8% vs 0.5%, drug and placebo rates, respectively).
`the trials in elderly patients, 1.4% of eszopiclone-treated patients and 0.5% of
`
`In
`
`
`
`
`
`

`

`
`
`placebo-treated patients experienced "falls". There was no evidence that these
`events were related to hypotension.
`
`Infection
`
`The sponsor categorized infections into bacteriai, viral, or fungal; Dr. Brugge
`describes the details of this categorization, but, in brief, this categorization was
`based on the verbatim term (e.g., "flu syndrome" as a verbatim term was
`considered a viral infection).
`In some cases, listed medication used to treat the
`infection was the basis for the categorization (e.g., if antibiotics were prescribed,
`the infection was considered bacterial).
`
`In the 6-month study, 39% of eszopiclone and 28% of placebo treated patients
`had infections (in the 6 week study, 15% and 23% of placebo and eszopiclone
`patients, respectively, had infections). Table 2A—6 (Dr. Brugge's review, page 31)
`displays the incidence of viral infections in the 6 month study; as can be seen,
`only Pharyngitis and Infection were frequent and about twice the incidence in the
`drug compared to the placebo treated patients.
`In this 6 month study, the
`incidence of bacterial infection was 15% and 12% in the drug and placebo
`groups, respectively, with incidences of fungal infections of 3% and 1% in the
`drug and placebo groups, respectively.
`In the 6 week study, the incidence of
`viral infections was 19% and 13% in the drug and placebo groups, respectively.
`lncidences of bacterial and fungal infections were low (0-4% range, minimally
`greater on drug vs placebo).
`
`Effectiveness
`
`As noted above, we were concerned that the frequent occurrence of an
`unpleasant taste in the drug treated patients could have broken the blind.
`However, as Dr. Brugge describes, the sponsor has re-anaiyzed the data from
`multiple trials (transient insomnia, the 6 week trial, one of the two 2 week elderly
`trials, and the 6 month trial), in which they included only those patients who did
`not experience an unpleasant taste. The re-analyses revealed statistically
`significant drug-placebo differences.
`
`Vital signs and EKG
`
`There were no important differences between drug and placebo treated patients
`on the percentage of patients meeting outlier criteria for EKG intervals.
`
`Regarding vital signs, the sponsor evaluated orthostatic vital signs in a single
`dose study in healthy volunteers, and in two seven day studies, one in elderly
`subjects, and one in younger adults.
`
`in younger adults, episodes of orthostatic hypotension occurred in 2H2 subjects
`at Day 7 at 3 mg, and 0/12 at 6 mg. A total of 5/123 subject receiving a dose of 3
`
`
`
`

`

`
`
`mg (combined Phase 1 studies of 1-7 days duration) had an episode of clinically
`significant decreased systolic blood pressure (< 90 mm Hg and > 20 mm Hg
`decrease from baseline) at 60 minutes post-dose; 2/52 subjects met this criterion
`at 60 minutes post-dose at 2 mg. Mean changes from baseline peaked at -5.3
`mm Hg at 3 mg systolic and -2.8 mm Hg diastolic (with small increases in the
`placebo patients). There were no episodes of syncope in the database.
`
`In elderly subjects in a 7 day study, small numbers of subjects experienced
`orthostatic hypotension between 30-90 minutes post-dose at doses of 3-5 mg,
`although the incidence was not consistently greater than in the placebo group
`(see, for example, Table 6A-3, Dr. Brugge's review, page 39-40).
`In this study,
`mean change in systolic blood pressure from baseline was maximal at -20 mm
`Hg at 3 mg (although placebo mean was -9 mm Hg) at 90 minutes post close.
`The maximum recommended dose in the elderly is 2mg, a dose at which there
`were no important drug~placebo differences in vital signs.
`
`In the 6 month controlled trial, eszopiclone treated patients had a 10% incidence
`of dizziness (no information related to blood pressure changes, if any, coincident
`with this adverse event) compared to 3% in the placebo group.
`In the elderly
`controlled studies, the rates were 6% and 2%, drug and placebo, respectively.
`
`Cognitive and Psychomotor Effects
`
`The sponsor evaluated cognitive, psychomotor, and memory function in two
`crossover studies evaluating 2 mg, 3 mg, and placebo, one in patients with
`chronic insomnia, one in healthy volunteers. The effects were measured 9.5 and
`12.5 hours post—dose (next morning) with a battery of 20 tests. Rare individual
`tests showed some decrements on drug compared to placebo, with no
`discernible pattern of dose relatedness or consistent abnormality on any specific
`test (though of those tests that were abnormal, memory was the faculty most
`commonly affected).
`
`In the 6 month controlled trial, 1.3% of drug treated and XX % of placebo treated
`patients reported memory impairment.
`In a 6 week non-elderly study, 3% of
`patients treated with 3 mg reported confusion, compared to 0% for the 2 mg and
`piacebo groups.
`In one 2 week study in elderly patients, 2.5 % of patients
`treated with 2 mg reported confusion, compared to 0% in the 1 mg and placebo
`groups (in a second similar study, there were no reports of confusion).
`
`Withdrawal effects (Anxiety and Rebound Insomnia)
`
`Low rates of anxiety were reported in the controlled trials (see draft labeling for a
`description).
`
`Rebound insomnia (defined as a worsening of insomnia compared to baseline
`
`
`
`
`
`

`

`
`
`after treatment discontinuation) was evaiuated in a 6 week study (2 mg, 3 mg,
`placebo) on the first 2 days after treatment discontinuation.
`
`Significant rebound was noted on the first post-treatment day on sleep latency
`and wakenings after sleep onset in the patients previousiy treated with 2 mg.
`These parameters resolved by the second post-treatment night, and no
`significant rebound on these measures occurred in the (previously treated) 3 mg
`group, although in this dose group, sleep efficiency was reduced on the first post-
`dosing night.
`
`Dissolution Specifications
`
`The sponsor has adopted the Agency's proposed dissolution specifications.
`
`One mg tablet
`
`As noted, the Agency requested that the sponsor produce a 1 mg tablet, because
`this dose was effective in elderly patients. The sponsor has done this, and the
`chemist and the OCPB reviewer consider the relevant data for this tablet
`
`acceptable.
`
`CMC
`
`The CMC deficiencies described in the Approvable letter have been resolved.
`
`New Name
`
`The Sponsor has proposed a new tradename, Lunesta, that has been found
`acceptable by both DMETS and DDMAC.
`
`Labefing
`
`We have attached a version of draft labeling which has been discussed with the
`sponsor, and on which we have agreed. The major sections that are new (in
`comparison to the draft sent with the Approvable letter) are a section in the
`Clinical Trials section related to safety concerns for sedative/hypnotic drugs and
`the Adverse Events Tables. The former is a standard sub-section in labeling for
`these drugs, but the Sponsor had not included such a section in their originally
`proposed labeling; we had asked them to do so in the Approvable label. This
`sub-section now consists of two parts: a Cognitive, Memory, Sedative, and
`Psychomotor Effects section, and a Withdrawal Emergent Anxiety and insomnia
`section.
`
`Regarding the Adverse Event Tables, in our Approvable letter, we had asked for
`three tables: one which compared ADR incidences on drug to placebo in all
`controlled trials, one which included ADR incidences in all controlled trials but
`
`

`

`
`
`broken down by age (elderly data separately from young adult data), and one in
`which those ADRs that were dose related (both ages combined) were presented.
`
`The firm has submitted two tables (each containing the data from the most
`relevant controlled trials in each age group), and added text to describe those
`that are dose related in each case (there are few of these). We have found this
`acceptable.
`
`COMMENTS
`
`The sponsor has responded adequately to all of the questions posed in our
`Approvable letter. Not unexpectedly, the data do not suggest a signal for the
`occurrence of malignancies (at least not in the controlied trials). None of the
`other re—analyses of the safety issues raised in the Approvable letter have
`identified issues that would preclude Approval of the application at this time. and
`we believe the attached draft labeling accurately describes the data.
`
`For these reasons, then, we recommend that the application be approved, and
`that the attached Approval letter with appended labeling be issued.
`
`Russell Katz, MD.
`
`
`
`
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`
`
`
`
`/s/
`
`Russell Katz
`12/15/04 12:51:56 PM
`MEDICAL OFFICER
`
`
`
`

`

`
`
`REVIEW AND EVALUATION OF CLINICAL DATA
`
`NDA:
`
`Sponsor:
`
`Drug
`Established Name:
`Chemical Name:
`.
`
`Code Name:
`
`Formulation:
`Indication:
`
`21—476
`
`Sepracor Inc
`
`Eszoplicone
`(+)-(SS)—6—(chIoropyridine—2—yi)—7—ox0-6,7-
`dihydro—SH—pyrrolo[3,4ablpyrazin—5—yl 4~
`methylpiperazine- l -carboxylate
`NA
`
`2 and 3 mg oral tablets
`Chronic and Transient Insomnia
`
`Dates of Submission:
`
`Letter Date: 6/14/04
`
`Materials Reviewed:
`
`Submission EDR Date: 6/ 17/04
`Response to February 25, 2004 Approvable
`Letter
`
`Clinical Reviewer:
`
`Karen L. Brugge, MD.
`
`Review Completion Date:
`
`10/18/04
`
`I. Background.
`
`The purpose of the submission is to assist the Team Leader and Division Director of the
`Division of Neuropharmacological Drug Products in the regulatory processing of this
`NDA.
`
`The current submission is a response to 3 2/25/04 Approvable Letter. This reviewer,
`recommended in the Clinical review of the original NDA submission, that the NDA not
`be given an approvable status. However, the submission was given an approvable action
`at the Agency Level. Therefore this review focuses on individual clinical items raised in
`the 2/25/04 Approvable Letter provides reviewer comments and recommendations to
`each of these items,
`
`The Structure of this Review. The sponsor itemized each bulleted item in the
`Approvable Letter, as well as some bracketed comments in labeling that was attached to
`the Approvable letter and categorized each comment as follows:
`0
`“Clinical" Comments: 1,2 A and B, 3, 6A—C (these items were reSponses to
`specific safety or related information), 10—14 (these items related to the Safety
`Update information), 15 (on postmarketing experience), 16 (English translations
`to foreign approved labeling)
`“Clinical Labelin " Comments: 7 A and B, 8, 9 A—G (these items are additionai
`safety related items). These items appeared either as comments in the Approvable
`
`0
`
`NDA 21-476 Response to the 2/25/04 Approvable Letter
`
`1
`
`
`
`

`

`
`
`letter (Items 7-8) or as bracketed comments in labeling attached to the Approvable
`Letter (Items 9, A-G).
`
`Additional Comments of the Approvable Letter that are itemized by the sponsor
`as follows: Comments 4 (Controlled Substance Staff related topic), 5 (CMC
`topic), 17 (OCPB item), 18 (CMC item), 19 (on nomenclature for the drug, a
`DMET item)
`
`Comment 20 regarding the Pediatric Research Equity Act (PREA)
`Comment 21 (Promotional Materials and Advertising information).
`
`This review has the sections that provide the sponsor’s response, reviewer comments,
`conclusions and recommendations (including pertinent labeling recommendations)
`regarding each of the above itemized comments in the Approvable letter, as itemized by
`the sponsor in their response submission. The sections were organized with the effort to
`group itemized comments that are related as described in the following:
`Section II on Specific Safety Concerns in the Approvable Letter: Comments 1, 2
`A and B, 3, 6A-C, 10—16 that are regarding Specific safety concerns
`(corresponding to bulleted comments in the Approvable letter, as previous
`described).
`
`Section III on Drug Class Safety Concerns (comments in the Approvable Letter
`or in bracketed comments in attached labeling) and Other Bracketed Labeling
`Comments in Labeling (attached to the Approvable Letter to which the sponsor
`responded as itemized cements in the response submission):
`0 Clinical comments (7 and 8) related to safety issues (such as memory,
`cognitive and psychomotor function effects) for the drug class including
`bracketed labeling comment itemized by the sponsor as 9 A (attached to the
`Approved Letter) regarding the section on “Studies Pertinent to Safety
`Concerns for Sedative/Hypnotic Drugs.”
`0 Other bracketed comments in labeling attached to the Approvable letter
`(portions of Comment 9A and 93-0).
`Section IV on Updated Safety-Related Information: Comments 10—16 (in the
`Approvable Letter) that pertain to updated safety-related information.
`Section V on Comments Related to Other Specialties: Additional Comments:
`Items 4—5, 17, 18, and 19 that focus on CMC, OCPB or DMET topics
`Section VI Pediatric Research: Comment 20 on PREA.
`
`Section VII. Promotional Materials: Comment 21 on Promotional/Advertising
`information.
`
`Section VIII. Overall Conclusions and Additional Key Labeling
`Recommendations, Not Addressed in Previous Sections:
`this section addresses
`issues not addressed elsewhere in the review.
`
`The above itemized comments are henceforth referred to as items in this review (rather
`than as comments).
`
`The sponsor’s response (not italicized), and reviewer’s comments and recommendations
`(italicized subseCtions) are provided for each item of each section.
`
`NDA 21—476 Response to the 2/25/04 Approvable Letter
`
`2
`
`
`
`

`

`The final section of this review (Section VI) addresses additional key labeling issues.
`
`II. Clinical Items in the Approvable Letter
`
`The Sponsor’s Response to Each Clinical Item in the 2425/04 Approvable Letter
`Each Item below corresponds to each item in the response submission. See the above
`section for details on the structure of this review. Each Item is followed by a copy of the
`comments of the Approvable Letter to which the given item refers. Non—itealicezed
`sections describe the sponsor’s response. Italicized sections reflect reviewer comments,
`conclusions andfor recommendations the given item.
`
`Clinical Item 1.. Adverse Events of Neoplasia
`
`Item 1 in the Approvable Letter.
`
`Please clarify the actual numbers of reports of neoplasm in study 190—049. There appears
`to be a disproportionate number of reports of adverse events of neoplasia in the eszopiclone
`group in this long-term double blind study of eszopiclone in patients with chronic
`insomnia. Depending on the tables we consult there are somewhere between 16 and 24
`reports ot‘neoplasia in the 593 eszopiclone treated patients and Oil 95 reports in the placebo
`group. We recognize from the verbatim terms that many of these reports may have been
`improperly coded; however, in the absence of the patient data or a clearer. explanation, we
`cannot make that assumption. Though we are interested in an explanation of all of these
`cases, we are particularly curious about three cases:
`
`a.
`
`this patient seems to be progressing
`Subject 0450024- by your description,
`steadily in a work up for disseminated cancer and then appears lost to foliow-up
`after she drops out of the study. This case was not reported as a serious adverse
`event even though the reason for her discontinuation is coded as "oeoplasia".
`
`Continued on the next page...
`
`NDA 21—476 Response to the 2/25/04 Approvable Letter
`
`3
`
`

`

`
`
`b.
`
`0.
`
`Subject 0406001- dropped out of the study for an adverse event coded as Breasr
`Neoplasm. The summary reports that she experienced a “lump” in her lefi breast
`after approximately W2 months of double-blind treatment.
`It was considered
`benign, presumably based on ultrasound and mammography that were conducted,
`but the results were not described. The subsequent course of her breast lump over
`time was not described yet study drug was discontinued upon discovery of the
`lump.
`
`Subject 0421004- a 62 year old female with no medical problems at screening
`who reported a “nodule in throat“ after approximately 5 months of double-blind
`treatment. This nodule was described as resolving 10 days after cessation of
`treatment. The narrative provides no other information and she appears lost to
`follow—up.
`
`Once all of the cases have been adequately examined, comparative incidence rates for the
`occurrence of neoplasia need to be calculated. Of the potential comparisons that you may
`make on the occurrence rates for neoplasia, one should be based on patient-years exposure
`to drug or placebo.
`if patients were last to follow—up before a definitive diagnosis of the
`problem was made, then these cases should be counted as neoplasia in at least one analysis.
`it will also be important to examine the timing of the observations of limplasia, as the
`plausibility of such an event as drug—related could be affected (e.g.. a finding at 2-4 weeks
`would not be plausibly drug—related but one at 6 months might be).
`
`We can not say that these cases represent a persuasive signal ofdrug-induced neoplasia, but
`the numericalimbalance of the reports of neoplasia and case histories that these numbers
`represent need to be thoroughly examined prior to considering eszopiclone for approval,
`especially given the pro—clinical findings of mammary and lung tumors with zopiclone and
`the finding of clastogenicity ofeszopiclonc and S-desmethyl-zopiclone.
`
`Smnsor’s Response to Item 1.
`
`A. Narrative Descriptions of 3 Subiects at Interest (Subjects 0450024, 0406001, and
`0421004) Requested in the Approvable Letter under Item 1 and Reviewer Comments.
`Subjects 0406001 and 0421008 are described in subsections below on “Breast Events
`. " and on ”Other Neoplasias.. ., " respectively. Upon a review of the narratives of these
`subjects, the former subject most likely hadfibrocystic disease (examined by
`mammography). The latter subject most likely had throat or esophageal related
`complication (complained of a “knot or lump" in the throat that worsened with
`swallowing, diagnosed as “globus”) due to gastric-esophageal reflux (the symptoms were
`accompanied by this condition and were reported approximately 2 1/2 weeks after starting
`lipitor).
`
`The third subject (subject 0450024) had multiple tumors revealed by CT scan or other
`imaging techniques (bilateral breasts, liver. pulmonary and a renal cyst). The etiology of
`these tumors and ascertaining a potential role of the study drug is more complicated.
`
`NDA 21-476 Response to the 2/25/04 Approvable Letter
`
`4
`
`
`
`
`
`

`

`
`
`
`
`The breast and renal tumors were reported as cystic and breastfindings were found at
`baseline and were bilateral. These observations suggest non-neoplastic and non-drug—
`related pre-existing conditions. Yet, the status of the breast findings during treatment in
`the study is not adequately documented to determine if a potential contributory role of the
`study existed in the progression of the breast-related pre-existing condition. A potential
`role of the study drug is a potentially serious concern, given other clinical observations
`for a greater incidence ofpotentially related events in longer term trials (breast pain.
`engorgement and dysmennorhea, among others) and given the preclinical mammary
`gland tumorfindings (refer to the Pharmacology Toxicology review of the original NBA).
`Yet, the breast condition is likely to be fibrocystic disease on the basis of the information
`provided. However a definitive diagnoses remains unclear.
`
`The pulmonary and liver tumors found on CT scan of the above subject (subject 0450024)
`may be neoplasia and the role of the study drug in at least as a contributoryfactor to
`these conditions remains unclear. Yet, the subject is reportedly stable based on self-
`report to the sponsor when they contacted the subject in March of 2004 (the subject
`withdrew from the study approximately 2 and halfyears ago). However, a completed
`diagnostic work-up for any of the tumors cannot be found in the narratives (e.g. hepatitis
`screen. biopsy, among others).
`In conclusion, information remains inadequate to verify
`that this subject does not have neOplasia and that the study drug did not play a role in
`events during the study. See below for a more detailed description of this subject.
`
`A More Detailed Description of Subject 0450024 with Multiple Nodules with Multi—
`organ Involvement (renal, breast, lung and liver). The third subject of interest has a
`complicated history and presentation. Furthermore, the etiology and/or diagnosis of the
`nodules remain unclear. Therefore, this subject is described in more detail in the
`following (refer to the above synopsisfor a brief summary of this subject). Refer to the
`review of the original NDA for background information on this subject. The following
`summarizes any new and/or relevant information found in an updated narrative provided
`in the current submission.
`
`The subject had abdominal and CT scans on Day 176 of ESZ treatmentfor
`evaluation of abdominal pain by her gynecologist (the narrative in the original NDA
`indicated the subject was 43 year old, although other information appears similar to
`information provided in the current submission, but in less detail and without more
`recently obtained infonnation). Thesefindings resulted in early withdrawal from the
`study on November 9, 2001.
`Upon review of the narrative that included some updated information (upon
`contacting the subject on 3/15/2004) this subject was found to be stable since almost 3
`years ago when the multi—organ “nodules” were first revealed by CT. The subject
`underwent spine surgery in 2002 andfollow-up chest CT scans revealed a "stable
`pulmonary nodule, " except a “slight interval increase in size” that “may be due to slight
`difference in patient positioning " (the sponsor provided these phrases, as quotes) on
`1/16/02. A subsequent chest CT on
`-\
`there was “no change in the sized ” of the
`nodule. No adenopathy was found on CT and no other nodules or masses are described.
`The subject is a 28 year 1 ‘72 packs/day smoker who stopped in 2000. The etiology of the
`pulmonary nodule remains unclear to this reviewer.
`
`NDA 21-476 Response to the 2125l04 Approvable Letter
`
`5
`
`.. _-‘. A-.._..k
`
`

`

`
`
`Regarding the liver hypo~dense nodulesfound in the initial abdominal CT in
`subject 0450024, the sponsor only indicates that this subject had elevated ALT at
`screening and study Visit 5 (42 and 5] Ufl, respectively) with no history of alcohol or
`drug abuse. No other information is provided regarding the liver lesions and regarding
`the' renal cyst (also found on the initial abdominal CT). No comment on the status of
`these lesions over time or comment about any follow-up diagnostic tests could be found
`in the narrative. A general comment was provided, based on the subject ’s self report
`{upon a recent contact with the subject), that the subject was stable, asymptomatic, well,
`and not aware of requiring anyfurtherfollow-up evaluations (last radiographicfollow-
`up was in 2002, 2 years before the subject was last contacted). This reviewer is unclear
`as to the etiology of any of the nodulesfound in this subject.
`Subject 0450024 also had “small fairly well defined nodular densities in both
`. breasts ” and a normal chest x-ray at screening. At baseline the diagnosis was (as quoted
`by the sponsor): “BitRads Category 2-Benign Finding: Benign nodular densities in both
`breast. ” A follow-up ultrasound revealed no identifiable mass. This subject is likely to
`have fibrocystic breast disease at baseline. The outcome of this condition during and
`after the study is not described in the narrative, other than the general statement thatthe
`subject was stable, as previously described.
`In conclusion, Subject 0450024 remains stable regarding the pulmonary nodule.
`It is not clear if the liver nodules are neoplastic or neoplastic metastesic tumors from the
`pulmonary nodule. Yet as above, this subject has been “stable. " The definitive
`diagnoses of the nodules in any ofthe subject's organs remains unclear. However, the
`renal and breast lesions are reportedly cystic in nature and are likely to be benign non-
`neoplastic events. Yet, the outcome of either of these lesions and their diagnosis upon
`follow-up remains unclear. Despite these caveats the patient who was recently
`contacted, reported being well and asymptomatic and she was unaware ofplans for
`additional scans (it is not clear if this part of the narrative is only referring to chest scans
`and the pulmonary nodule). Typically if the subject had a malignancy such as in the lung
`and liver and possibly in the breast, she would be having signs and symptoms and
`metastases afterfour years from the time the nodules were first found. However,
`consideration must be given to the possibility that she may have neoplasia that may be
`malignant (e. g. pulmonary).
`
`B. A Search and Review of Adverse Events reported as Neoplasia in the Chronic
`Insomnia Trial (Study 190-0491
`The sponsor provided a listing of adverse events (AE5) found in the database search of
`ABS reported as neoplasia or related events in the longterm Chronic Insomnia trial, Study
`190—049. The sponsor had these events classified on the basis of the likelihood that a
`given event was malignant, benign or non-neoplastic in nature. The methods for this
`search and for classifying the events is described later.
`
`Before showing the sponsor’s results of their classification system, a synopsis is first
`provided and recommendations. The synopsis summarizes neoplasias enumerated by this
`reviewer on the basis of a review of narratives of the AEs provided by the sponsor (the
`Sponsor provided narratives for ABS captured by their search of ABS of neoplasias in the
`
`NDA 21—476 Response to the 2/25/04 Approvable Letter
`
`6
`
`
`
`

`

`
`
`long term Chronic Insomnia Trial, Study 190—049). Refer to the review of the original
`NDA for additional background information.
`
`After the synopsis and reviewer recommendations (italicized subsections), the following .
`subsections provide more detailed information:
`
`A subsection of the sponsor’s enumeration of neoplasia events based on
`0
`their classification method is provided (non-italicized subsection),
`-
`Followed by a more detailed description of the reviewer’s methods for
`classifying AEs of neoplasia and reviewer observations based on a review of the
`narrative information. These subsections include a description of gynecological
`events.
`
`All sections reflecting reviewer’s comments and observations are italicized.
`
`Reviewer Synopsis of Sponsor’s Response to Item 1: Exposure, Adverse Events of
`NeoglasiaI and Reviewer Conclusions
`Summary of Background and ofReviewer’s Findings Upon Review of Narratives
`Provided in the Current Submission. Refer to the original review of the NDA regarding
`concernsfor a signalfor adverse events of “neoplasia " in the longterm trial, Study 190-
`049. The Approvable Letter requested mOre detailed information and clarification of the
`adverse events reported as “neoplasia” and other events that may be ne0plastic in nature
`in this Chronic Insomnia Phase III trial (refer to the above copy of the relevant section of
`the Approvable Letter). The sponsor was also asked in the Approvable letter to
`determine the incidence of events ofneoplasia in this trial.
`
`Study 190-049 was a longterm Chronic Insomnia Phase III trial. This trial included
`almost exclusively non-elderly adults that received 6-months of double-blind placebo or
`E32 (3 mg daily; 593 eszopiclone and 195 placebo HT safety subjects) followed by open
`label ESZ (3 mg daily) for 6~months (471 ITT safety subjects).
`
`A review of information in the current submission yielded multiple events of neoplasia
`involving skin and other organ systems (e.g. breast) in subjects of the longterm Chronic
`Insomnia trial, Study 190—049. Given the complexity in classifying these subjects,
`limitations with the information on these adverse events reported as events of neoplasia,
`fairly stringent eligibility criteria with respect to ruling out pre-existing neoplasia, as
`well as other potential