CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`
`APPLICATION NUMBER:
`
`202155Orig1s000
`
`
`OTHER REVIEW(S)
`
`

`

`SEALD Director Sign-Off Review of the End-of—Cycle Prescribing
`Information: Outstanding Format Deficiencies
`
`
`
`
`\
`\
`
`
`Product Title
`ELIQUIS (apixaban) tablets for oral use
`Applicant
`I Bristol-Myers Squibb
`
`Application]Supplement Number
`I NDA 202155
`
`Type of Application
`Original
`Indication(3)
`To reduce the risk of stroke and systemic embolism in patients
`with nonvalvular atrial fibrillation.
`Established Pharmacologic Classl factor Xa inhibitor anticoagulant
`
`
`Office/Division ODE I/DCRP
`
`
`Division Project Manager
`Alison Blaus
`
`I Date FDA Received Application I September 17, 2012
`
`Goal Date
`March 17, 2013
`
`
`
`
`
`December 28, 2012
`Date PI Received by SEALD
`SEALD Review Date
`December 28, 2012
`
`SEALD Labeling Reviewer
`Elizabeth Donohoe
`
`I SEALD Division Director ‘ Laurie Burke
`P1 = prescribing information
`1 The established pharmacologic class (EPC) that appears in the final draft PI.
`
`This Study Endpoints and Labeling Development (SEALD) Director Sign-Off review of the end-0f-
`cycle, draft prescribing information (PI) for critical format elements reveals outstanding labeling
`format deficiencies that must be corrected before the final PI is approved. After these outstanding
`labeling format deficiencies are corrected, the SEALD Director will have no objection to the
`approval of this PI.
`
`The critical format elements include labeling regulation (21 CFR 201.56 and 201.57), labeling
`guidance, and best labeling practices (see list below). This review does not include every
`regulation or guidance that pertains to PI format.
`
`Guide to the Selected Requirements of Prescribing Information [SRPH Checklist: For each SRPI
`item, one of the following 3 response options is selected:
`
`NO: The PI does not meet the requirement for this item (deficiency).
`
`YES: The PI meets the requirement for this item (not a deficiency).
`0 N/A (not applicable): This item does not apply to the specific PI under review.
`
`Reference ID: 3237536
`
`Page 1 of 8
`
`

`

`Selected Requirements of Prescribing Information
`
`
`
`
`Highlights (HL)
`GENERAL FORMAT
`1. Highlights (HL) must be in two-column format, with ½ inch margins on all sides and in a
`minimum of 8-point font.
`Comment:
`2. The length of HL must be less than or equal to one-half page (the HL Boxed Warning does not
`count against the one-half page requirement) unless a waiver has been is granted in a previous
`submission (i.e., the application being reviewed is an efficacy supplement).
`Instructions to complete this item: If the length of the HL is less than or equal to one-half page
`then select “YES” in the drop-down menu because this item meets the requirement. However, if
`HL is longer than one-half page:
` For the Filing Period (for RPMs)
` For efficacy supplements: If a waiver was previously granted, select “YES” in the drop-
`down menu because this item meets the requirement.
` For NDAs/BLAs and PLR conversions: Select “NO” in the drop-down menu because this
`item does not meet the requirement (deficiency). The RPM notifies the Cross-Discipline
`Team Leader (CDTL) of the excessive HL length and the CDTL determines if this
`deficiency is included in the 74-day or advice letter to the applicant.
` For the End-of Cycle Period (for SEALD reviewers)
` The SEALD reviewer documents (based on information received from the RPM) that a
`waiver has been previously granted or will be granted by the review division in the
`approval letter.
`Comment:
`3. All headings in HL must be presented in the center of a horizontal line, in UPPER-CASE letters
`and bolded.
`Comment:
`4. White space must be present before each major heading in HL.
`Comment:
`5. Each summarized statement in HL must reference the section(s) or subsection(s) of the Full
`Prescribing Information (FPI) that contains more detailed information. The preferred format is
`the numerical identifier in parenthesis [e.g., (1.1)] at the end of each information summary (e.g.
`end of each bullet).
`Comment: Under D&A, the second bullet should reference 2.2; in DI the references should be
`7.1 under the first bullet and 7.2 under the second bullet.
`6. Section headings are presented in the following order in HL:
`Section
`Required/Optional
`Required
` Highlights Heading
`Required
` Highlights Limitation Statement
`Required
` Product Title
`Required
` Initial U.S. Approval
`Required if a Boxed Warning is in the FPI
` Boxed Warning
`
`YES
`
`YES
`
`YES
`
`YES
`
`NO
`
`YES
`
`
`
` Page 2 of 8
`
`Reference ID: 3237536
`
`

`

`
`
`
`
`Selected Requirements of Prescribing Information
`
`Required for only certain changes to PI*
` Recent Major Changes
`Required
` Indications and Usage
`Required
` Dosage and Administration
`Required
` Dosage Forms and Strengths
`Required (if no contraindications must state “None.”)
` Contraindications
`Not required by regulation, but should be present
` Warnings and Precautions
`Required
` Adverse Reactions
`Optional
` Drug Interactions
`Optional
` Use in Specific Populations
` Patient Counseling Information Statement Required
`Required
` Revision Date
`* RMC only applies to the Boxed Warning, Indications and Usage, Dosage and Administration, Contraindications,
`and Warnings and Precautions sections.
`Comment:
`7. A horizontal line must separate HL and Table of Contents (TOC).
`Comment:
`
`
`HIGHLIGHTS DETAILS
`Highlights Heading
`8. At the beginning of HL, the following heading must be bolded and appear in all UPPER CASE
`letters: “HIGHLIGHTS OF PRESCRIBING INFORMATION”.
`Comment:
`
`
`Highlights Limitation Statement
`9. The bolded HL Limitation Statement must be on the line immediately beneath the HL heading
`and must state: “These highlights do not include all the information needed to use (insert
`name of drug product in UPPER CASE) safely and effectively. See full prescribing
`information for (insert name of drug product in UPPER CASE).”
`Comment:
`
`Product Title
`10. Product title in HL must be bolded.
`Comment:
`
`Initial U.S. Approval
`11. Initial U.S. Approval in HL must be placed immediately beneath the product title, bolded, and
`include the verbatim statement “Initial U.S. Approval:” followed by the 4-digit year.
`Comment:
`
`Boxed Warning
`12. All text must be bolded.
`Comment:
`13. Must have a centered heading in UPPER-CASE, containing the word “WARNING” (even if
`more than one Warning, the term, “WARNING” and not “WARNINGS” should be used) and
`other words to identify the subject of the Warning (e.g., “WARNING: SERIOUS
`INFECTIONS”).
`
`YES
`
`YES
`
`YES
`
`YES
`
`YES
`
`YES
`
`YES
`
`
`
`Reference ID: 3237536
`
`
`
`Page 3 of 8
`
`

`

`YES
`
`YES
`
`YES
`
`N/A
`
`N/A
`
`N/A
`
`N/A
`
`YES
`
`
`
`
`
`Selected Requirements of Prescribing Information
`
`Comment:
`14. Must always have the verbatim statement “See full prescribing information for complete boxed
`warning.” in italics and centered immediately beneath the heading.
`Comment: Consider adding one space between this statement and the wording of the BW in HL
`to improve readability.
`15. Must be limited in length to 20 lines (this does not include the heading and statement “See full
`prescribing information for complete boxed warning.”)
`Comment:
`16. Use sentence case for summary (combination of uppercase and lowercase letters typical of that
`used in a sentence).
`Comment:
`
`
`Recent Major Changes (RMC)
`17. Pertains to only the following five sections of the FPI: Boxed Warning, Indications and Usage,
`Dosage and Administration, Contraindications, and Warnings and Precautions.
`Comment:
`18. Must be listed in the same order in HL as they appear in FPI.
`Comment:
`19. Includes heading(s) and, if appropriate, subheading(s) of labeling section(s) affected by the
`recent major change, together with each section’s identifying number and date (month/year
`format) on which the change was incorporated in the PI (supplement approval date). For
`example, “Dosage and Administration, Coronary Stenting (2.2) --- 3/2012”.
`Comment:
`20. Must list changes for at least one year after the supplement is approved and must be removed at
`the first printing subsequent to one year (e.g., no listing should be one year older than revision
`date).
`Comment:
`
`Indications and Usage
`21. If a product belongs to an established pharmacologic class, the following statement is required in
`the Indications and Usage section of HL: “(Product) is a (name of established pharmacologic
`class) indicated for (indication)”.
`Comment: The PI contains the correct format. Note, eLIST (http://elist/prpllr/public/query/)
`and the Xarelto PI states the EPC for rivaroxaban as "factor Xa inhibitor" . We recommend that
`the EPC in the I&U statement in the Highlights for rivaroxaban and apixaban be consistent if
`appropriate.
`
`N/A
`
`Dosage Forms and Strengths
`22. For a product that has several dosage forms, bulleted subheadings (e.g., capsules, tablets,
`injection, suspension) or tabular presentations of information is used.
`Comment:
`
`
`
`Reference ID: 3237536
`
`
`
`Page 4 of 8
`
`

`

`YES
`
`YES
`
`YES
`
`YES
`
`
`
`Selected Requirements of Prescribing Information
`
`
`Contraindications
`23. All contraindications listed in the FPI must also be listed in HL or must include the statement
`“None” if no contraindications are known.
`Comment:
`24. Each contraindication is bulleted when there is more than one contraindication.
`Comment:
`
`Adverse Reactions
`25. For drug products other than vaccines, the verbatim bolded statement must be present: “To
`report SUSPECTED ADVERSE REACTIONS, contact (insert name of manufacturer) at
`(insert manufacturer’s U.S. phone number) or FDA at 1-800-FDA-1088 or
`www.fda.gov/medwatch”.
`Comment: The applicant underlined the www.fda.gov/medwatch website; it should not be
`underlined.
`
`Patient Counseling Information Statement
`26. Must include one of the following three bolded verbatim statements (without quotation marks):
`
`
`
`
`
`
`If a product does not have FDA-approved patient labeling:
` “See 17 for PATIENT COUNSELING INFORMATION”
`If a product has FDA-approved patient labeling:
` “See 17 for PATIENT COUNSELING INFORMATION and FDA-approved patient labeling.”
` “See 17 for PATIENT COUNSELING INFORMATION and Medication Guide.”
` Comment:
`
`
`
`YES
`
`YES
`
`YES
`
`YES
`
`Revision Date
`27. Bolded revision date (i.e., “Revised: MM/YYYY or Month Year”) must be at the end of HL.
`Comment:
` Contents: Table of Contents (TOC)
`
`
`
`
`GENERAL FORMAT
`28. A horizontal line must separate TOC from the FPI.
`Comment:
`29. The following bolded heading in all UPPER CASE letters must appear at the beginning of TOC:
`“FULL PRESCRIBING INFORMATION: CONTENTS”.
`Comment:
`30. The section headings and subheadings (including title of the Boxed Warning) in the TOC must
`match the headings and subheadings in the FPI.
`Comment:
`
`
`
`Reference ID: 3237536
`
`
`
`Page 5 of 8
`
`

`

`
`
`Selected Requirements of Prescribing Information
`
`
`31. The same title for the Boxed Warning that appears in the HL and FPI must also appear at the
`beginning of the TOC in UPPER-CASE letters and bolded.
`Comment:
`32. All section headings must be bolded and in UPPER CASE.
`Comment:
`33. All subsection headings must be indented, not bolded, and in title case.
`Comment:
`34. When a section or subsection is omitted, the numbering does not change.
`Comment:
`35. If a section or subsection from 201.56(d)(1) is omitted from the FPI and TOC, the heading
`“FULL PRESCRIBING INFORMATION: CONTENTS” must be followed by an asterisk
`and the following statement must appear at the end of TOC: “*Sections or subsections omitted
`from the Full Prescribing Information are not listed.”
`Comment:
`
`
`
`Full Prescribing Information (FPI)
`GENERAL FORMAT
`36. The following heading must appear at the beginning of the FPI in UPPER CASE and bolded:
`“FULL PRESCRIBING INFORMATION”.
`Comment:
`37. All section and subsection headings and numbers must be bolded.
`Comment:
`
`38. The bolded section and subsection headings must be named and numbered in accordance with
`21 CFR 201.56(d)(1) as noted below. If a section/subsection is omitted, the numbering does not
`change.
`
`YES
`YES
`
`YES
`
`YES
`
`YES
`
`YES
`
`YES
`
`YES
`
`Boxed Warning
`1 INDICATIONS AND USAGE
`2 DOSAGE AND ADMINISTRATION
`3 DOSAGE FORMS AND STRENGTHS
`4 CONTRAINDICATIONS
`5 WARNINGS AND PRECAUTIONS
`6 ADVERSE REACTIONS
`7 DRUG INTERACTIONS
`8 USE IN SPECIFIC POPULATIONS
`8.1 Pregnancy
`8.2 Labor and Delivery
`8.3 Nursing Mothers
`8.4 Pediatric Use
`8.5 Geriatric Use
`9 DRUG ABUSE AND DEPENDENCE
`9.1 Controlled Substance
`9.2 Abuse
`
`
`
`
`
`Reference ID: 3237536
`
`
`
`Page 6 of 8
`
`

`

`
`
`
`
`Selected Requirements of Prescribing Information
`
`9.3 Dependence
`10 OVERDOSAGE
`11 DESCRIPTION
`12 CLINICAL PHARMACOLOGY
`12.1 Mechanism of Action
`12.2 Pharmacodynamics
`12.3 Pharmacokinetics
`12.4 Microbiology (by guidance)
`12.5 Pharmacogenomics (by guidance)
`13 NONCLINICAL TOXICOLOGY
`13.1 Carcinogenesis, Mutagenesis, Impairment of Fertility
`13.2 Animal Toxicology and/or Pharmacology
`14 CLINICAL STUDIES
`15 REFERENCES
`16 HOW SUPPLIED/STORAGE AND HANDLING
`17 PATIENT COUNSELING INFORMATION
`Comment:
`
`YES
`
`YES
`
`N/A
`
`NO
`
`YES
`
`YES
`
`
`39. FDA-approved patient labeling (e.g., Medication Guide, Patient Information, or Instructions for
`Use) must not be included as a subsection under Section 17 (Patient Counseling Information).
`All patient labeling must appear at the end of the PI upon approval.
`Comment:
`40. The preferred presentation for cross-references in the FPI is the section heading (not subsection
`heading) followed by the numerical identifier in italics. For example, “[see Warnings and
`Precautions (5.2)]”.
`Comment: A possible incorrect cross reference was noted in the BW and Section 5.1 W&P:
`instead of Dosage and Administration (2.4) in BW and Dosage and Administration (2.3), we
`recommend Dosage and Administration (2.4, 2.5) for both sections. Recommend also that
`under Drug Interactions Studies subheader in Section 12.3, that the word "also" be removed
`from the first cross reference.
`41. If RMCs are listed in HL, the corresponding new or modified text in the FPI sections or
`subsections must be marked with a vertical line on the left edge.
`Comment:
`FULL PRESCRIBING INFORMATION DETAILS
`Boxed Warning
`42. All text is bolded.
`Comment: The summary text is not bolded.
`43. Must have a heading in UPPER-CASE, containing the word “WARNING” (even if more than
`one Warning, the term, “WARNING” and not “WARNINGS” should be used) and other words
`to identify the subject of the Warning (e.g., “WARNING: SERIOUS INFECTIONS”).
`Comment:
`44. Use sentence case (combination of uppercase and lowercase letters typical of that used in a
`sentence) for the information in the Boxed Warning.
`Comment:
`Contraindications
`
`
`
`
`
`Reference ID: 3237536
`
`
`
`Page 7 of 8
`
`

`

`
`
`Selected Requirements of Prescribing Information
`
`
`45. If no Contraindications are known, this section must state “None”.
`Comment:
`Adverse Reactions
`46. When clinical trials adverse reactions data is included (typically in the “Clinical Trials
`Experience” subsection of Adverse Reactions), the following verbatim statement or appropriate
`modification should precede the presentation of adverse reactions:
`
`“Because clinical trials are conducted under widely varying conditions, adverse reaction rates
`observed in the clinical trials of a drug cannot be directly compared to rates in the clinical
`trials of another drug and may not reflect the rates observed in clinical practice.”
`Comment:
`47. When postmarketing adverse reaction data is included (typically in the “Postmarketing
`Experience” subsection of Adverse Reactions), the following verbatim statement or appropriate
`modification should precede the presentation of adverse reactions:
`“The following adverse reactions have been identified during post-approval use of (insert drug
`name). Because these reactions are reported voluntarily from a population of uncertain size, it
`is not always possible to reliably estimate their frequency or establish a causal relationship to
`drug exposure.”
`Comment:
`Patient Counseling Information
`48. Must reference any FDA-approved patient labeling, include the type of patient labeling, and use
`one of the following statements at the beginning of Section 17:
` “See FDA-approved patient labeling (Medication Guide)”
` “See FDA-approved patient labeling (Medication Guide and Instructions for Use)”
` “See FDA-approved patient labeling (Patient Information)"
` “See FDA-approved patient labeling (Instructions for Use)"
` “See FDA-approved patient labeling (Patient Information and Instructions for Use)”
`Comment:
`
`
`
`
`
`
`
`
`
`
`
`N/A
`
`YES
`
`N/A
`
`YES
`
`
`
`
`
`Reference ID: 3237536
`
`
`
`Page 8 of 8
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`ERIC R BRODSKY
`12/28/2012
`Eric Brodsky, signing for Elizabeth Donohoe, SEALD labeling reviewer
`
`LAURIE B BURKE
`12/28/2012
`
`Reference ID: 3237536
`
`

`

`
`
`DIVISION OF CARDIOVASCULAR AND RENAL PRODUCTS
`Divisional Memo
`
`
`NDA:
`202155 apixaban (Eliquis) for prevention of stroke and
`systemic embolism in patients with non-valvular atrial fibrillation
`Sponsor:
`BMS
`Review date: 27 December 2012
`
`Reviewer:
`N. Stockbridge, M.D., Ph.D., HFD-110
`Distribution: NDA 202155
`This memo conveys the Division’s recommendation to issue an Approval letter for
`apixaban.
`I reference Dr. Grant’s CDTL/Divisional Memo of 22 June 2012 for a description of
`most aspects of the original submission and of issues that led to a Complete Response
`letter, also dated 22 June 2012. In this memo, I summarize aspects of the clinical
`review (Beasley and Rose; 22 May 2012) not covered by Dr. Grant, the clinical review in
`response to the sponsor’s resubmission (Beasley and Rose; 10 December 2012), an
`unsolicited review by Dr. Marciniak (revised 17 December 2012), and a commentary on
`the Marciniak review (Beasley and Rose; 21 December 2012).
`Support for a claim to prevent stroke and systemic embolism in patients with atrial
`fibrillation comes from two studies, AVERROES and ARISTOTLE.
`AVERROES was a randomized, double-blind comparison of apixaban and aspirin in
`patients with AF, some additional risk factor, and the perceived need to avoid warfarin.
`The primary analysis was a test of superiority in reducing stroke or systemic embolism.
`The trial was stopped early (enrollment complete, but before the targeted number of
`events were observed) for overwhelming benefit, but the sponsor was dissuaded from
`seeking a claim based on this alone, as warfarin is superior to aspirin in this setting,
`rationale for avoiding warfarin was questionable in many cases, and ARISTOTLE was
`soon to complete.
`The vast majority of reviewer attention was then on ARISTOTLE, a randomized, double-
`blind comparison of apixaban (5 mg BID in most subjects) and warfarin (titrated to INR
`2-3). Its primary analysis was non-inferiority for prevention of stroke and systemic
`embolism, using the margin of 1.38 previously accepted by FDA.
`The Complete Response letter (22 June 2012) listed 6 issues. Four of these related to
`difficulty ascertaining who got incorrect study drug dispensed, a problem of both
`process (no subject-specific kits) and documentation (most importantly, failure to
`collect bottle labels centrally). The problem of mixing up who received what drug had
`obvious implications with regard to any non-inferiority analysis, but it also potentially
`led to worse outcomes among incorrectly dosed warfarin subjects (because of resulting
`dosing changes introduced to deal with anomalous INR readings, the effects of which
`would last even after subjects returned to assigned treatment).
`As Drs. Beasley and Rose describe, the sponsor provided analyses of labels from a
`sample of about 1/3 of the 450,000 bottles used, placing a reliable upper bound on the
`magnitude of the dosing problem. Reasonable analyses show that the impact of this was
`not enough to account for the superiority of apixaban over warfarin on either stroke or
`bleeding.
`
`C:\Users\STOCKBRIDGEN\Documents\NDA\N202155 Apixaban\ApixabanAFDivMemo.doc Last saved
`
`1
`16:07 Thursday, December 27, 2012
`
`Reference ID: 3237140
`
`

`

`Divisional memo
`Eliquis (apixaban)
`
`
`
`NDA 202155
` (stroke prevention in AF)
`
`0.45 (0.32-0.62)
`0.66 (0.53-0.83)
`0.79 (0.62-1.02)
`
`Two unrelated issues were also raised in the CR letter. One had to do with apparent
`discrepancies between the medication error dataset and underlying case report forms.
`This issue was resolved by fixing some data entry errors, with no material impact on the
`resulting analyses. The final issue had to do with multiple entries for the same adverse
`event in the sponsor’s adverse event dataset. While this problem was incompletely
`resolved, its residual impact is minor.
`Key results from AVERROES are summarized in the table below (events per 1000
`patient-years; abstracted from pages 110-112 of the original clinical review):
`
`Apixaban
`Aspirin
`HR (95% CI)
`N=2807
`N=2791
`16.2
`36.3
`Stroke/SE
`42.1
`63.5
`Stroke/SE/MI/Vasc death
`35.5
`44.2
`All-cause death
`26.5
` Vascular
`30.3
`8.5
` Non-vascular
`13.9
`Although the early study termination seems to have been appropriate, one should
`expect that early termination will tend to overestimate the treatment effect size.
`Key results from ARISTOTLE are shown in the table below:
`
`Apixaban
`Warfarin
`N=9120
`N=9081
`12.7
`18.0
`Stroke/SE1
`11.9
`15.1
` Stroke
`8.4
`8.2
` Ischemic
`2.4
`4.7
` Hemorrhagic
`0.7
`1.2
` Ischemichemorrhagic
`0.8
`1.3
` Uncertain
`0.9
`1.0
` Systemic embolism
`35.2
`39.4
`All-cause death
`18.0
` Cardiovascular
`20.2
`2.2
` Stroke
`3.8
`4.4
` Heart failure
`5.4
`7.4
` Sudden
`7.6
` Other2
`4.0
`3.4
` Non-cardiovascular
`11.4
`12.2
` Respiratory failure
`1.1
`2.1
` Malignancy
`3.5
`3.9
` Trauma
`0.4
`0.8
` Bleeding
`0.9
`1.0
` Other
`1.6
`1.5
` Infection
`3.9
`3.1
` Unknown
`5.8
`6.9
`Results on the primary analysis establish superiority of apixaban to warfarin (non-
`inferiority is not an issue).
`
`HR (95% CI)
`
`0.79 (0.66-0.95)
`
`0.89 (0.80-1.00)
`0.89 (0.76-1.04)
`
`
`
`
`0.93 (0.77-1.13)
`
`
`1 Subordinate events shown at subjects with events at any time, not necessarily as the first event.
`2 Systemic embolism, MI, other CV, unobserved
`C:\Users\STOCKBRIDGEN\Documents\NDA\N202155 Apixaban\ApixabanAFDivMemo.docLast saved
`2
`16:07 Thursday, December 27, 2012
`
`
`
`Reference ID: 3237140
`
`

`

`Divisional memo
`Eliquis (apixaban)
`
`
`
`NDA 202155
` (stroke prevention in AF)
`
`Results on mortality are more difficult to interpret. The nominal p-value according to
`the pre-specified analysis plan is 0.0465, so “significant”. I address first what ought to
`be the easier question, whether this represents a treatment effect on mortality, before
`addressing whether this should be considered superiority over warfarin.
`If warfarin had a mortality effect, than one might conclude, even with a p-value for
`superiority to warfarin >0.05, that apixaban was superior to placebo. Dr. Marciniak
`addressed this issue in the 6 studies of warfarin vs. placebo that formed the basis for
`FDA’s determination of a non-inferiority margin for warfarin. In one of these studies
`(BAATAF), warfarin reduced mortality by >50%, highly statistically significant, but Dr.
`Marciniak’s random-effects meta-analysis of all six studies has p=0.192 for mortality.
`However, this result is at odds with a published meta-analysis of the same 6 studies
`(Hart et al., 19993), which showed a 26% decrease in the risk of all-cause mortality
`(95% CI of 3-43%), a difference upon which Dr. Marciniak does not comment.
`Whether one might consider apixaban superior to warfarin in reducing all-cause
`mortality with a p≈0.05, one could more reliably conclude that even p≈0.05 is indicative
`of a benefit of apixaban over placebo. Nor is apixaban the only anticoagulant with at
`least a favorable lean on mortality compared with warfarin; dabigatran has quite similar
`data, and Drs. Beasley and Rose suggest that these results might be considered
`mutually supportive of a mortality effect, much as we used data from losartan and
`irbesartan to support one another’s effects on diabetic nephropathy.
`Credibility to the mortality benefit is aided by the main driver of the benefit, a reduction
`in mortality attributed to stroke, consistent with the benefit on overall stroke and
`hemorrhagic stroke4. Other contributors to the apparent benefit on stroke are less easy
`to interpret—reductions in heart failure and respiratory failure.
`There is no question that the mortality findings need display in the label. I would favor
`wording in section 14 that says that various anticoagulants appear to reduce mortality
`in AF, and that these particular data are not compelling evidence of superiority in this
`regard to warfarin.
`Dr. Marciniak takes a more dichotomous view of the mortality findings, making much of
`the fact that many quite reasonable sensitivity analyses—count errors, date errors,
`censoring issues, etc.—yield p-values >0.05. Were the comparison between apixaban
`and warfarin the only information available, these issues would be more important than
`I consider them to be.
`Dr. Marciniak raises several other issues warranting response.
`He is rightfully concerned about adequacy of follow-up, particularly for vital status, but
`he counts anyone with data missing for the closeout date the same whether the gap in
`knowledge is a day or a year. This exaggerates the missing information.
`Dr. Marciniak reminds us that prasugrel caused more bleeding than did clopidogrel in
`TRITON, and was (by Dr. Marciniak’s counts) associated with more cancers reported.
`In APPRAISE-2 (ACS study), apixaban plus antiplatelet therapy was associated with
`more bleeding than was the antiplatelet therapy alone, and it was (by Dr. Marciniak’s
`counts) associated with more reported cancers. In ARISTOTLE, apixaban was
`associated with less bleeding and fewer cancers. In none of these studies was there an
`effect on deaths from cancer. It seems most likely to me that bleeding leads to more
`
`3 Hart RG et al. 1999. Antithrombotic therapy to prevent stroke in patients with atrial fibrillation: a meta-
`analysis. Ann Int Med 131:492-501.
`4 Beasley and Rose (21 December) note that a similar trend for superiotity of apixaban over warfarin for all-
`cause mortality persists post-treatment. They show that is largely attributable to deaths in the few days after
`discontinuing for strokes.
`C:\Users\STOCKBRIDGEN\Documents\NDA\N202155 Apixaban\ApixabanAFDivMemo.docLast saved
`3
`16:07 Thursday, December 27, 2012
`
`
`
`Reference ID: 3237140
`
`

`

`Divisional memo
`Eliquis (apixaban)
`
`
`
`NDA 202155
` (stroke prevention in AF)
`
`vigilance, whether it relates to the affected organ or not, and bleeding thus leads to
`earlier detection of cancers, an effect that persists for as long as you take these drugs.
`This sounds like a benefit to me.
`The review team and I recommend approval of Eliquis to prevent stroke in patients with
`non-valvular atrial fibrillation.
`
`C:\Users\STOCKBRIDGEN\Documents\NDA\N202155 Apixaban\ApixabanAFDivMemo.docLast saved
`4
`16:07 Thursday, December 27, 2012
`
`
`
`Reference ID: 3237140
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`NORMAN L STOCKBRIDGE
`12/27/2012
`
`Reference ID: 3237140
`
`

`

`FOOD AND DRUG ADMINISTRATION
`Center for Drug Evaluation and Research
`Office of Prescription Drug Promotion
`Division of Professional Drug Promotion
`Division of Consumer Drug Promotion
`****Pre-decisional Agency Information****
`
`Memorandum
`
`December 10, 2012
`
`Alison Blaus
`Regulatory Project Manager
`Division of Cardio-Renal Products (DCRP)
`
`Emily Baker, PharmD
`Regulatory Review Officer
`Division of Professional Drug Promotion (DPDP)
`Office of Prescription Drug Promotion (OPDP)
`
`
`
`Zarna Patel, PharmD
`Regulatory Review Officer
`Division or Consumer Drug Promotion (DCDP)
`Office of Prescription Drug Promotion (OPDP)
`
`
`
`
`
`
`
`
`Eliquis (apixaban)
`NDA 202155
`
`
`
`
`Date:
`
`
`To:
`
`
`
`
`
`
`
`From:
`
`
`
`
`
`
`
`
`
`
`
`
`Subject:
`
`
`
`
`
`OPDP has reviewed the proposed Package Insert (PI) and Medication Guide submitted for consult on
`October 2, 2012, for Eliquis (apixaban). Our comments are based on the proposed labeling at the
`following EDR location: \\CDSESUB1\EVSPROD\NDA202155\0004.
`
`DPDP reviewed the proposed PI and our comments are provided directly on the attached proposed
`PI.
`
`DCDP also reviewed the comments on the proposed Medication Guide from the Division of Medical
`Policy Programs (DMPP) dated December 7, 2012. We agree with DMPP’s comments and have the
`following additional comments. Our comments on the proposed Medication Guide are provided
`directly on the version sent to DCRP from DMPP.
`
`Thank you for the opportunity to comment on these proposed materials.
`
`If you have any questions on the comments for the PI, please contact Emily Baker at 301.796.7524 or
`emily.baker@fda.hhs.gov.
`
`If you have any questions on the comments for the Medication Guide, please contact Zarna Patel at
`301.796.3822 or zarna.patel@fda.hhs.gov.
`
`1
`
`Reference ID: 3228233
`
`24 Page(s) of Draft Labeling have been Withheld in Full as b4 (CCI/TS) immediately following this page
`
`

`

`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`
`EMILY K BAKER
`
`12/10/2012
`
`ZARNA PATEL
`
`1 2/1 0/2012
`
`Reference ID: 3228233
`
`

`

`Department of Health and Human Services
`Public Health Service
`Food and Drug Administration
`Center for Drug Evaluation and Research
`Office of Medical Policy Initiatives
`Division of Medical Policy Programs
`
`PATIENT LABELING REVIEW
`
`December 7, 2012
`
`Norman Stockbridge, MD, PhD
`Director
`Division of Cardiovascular and Renal Products (DCRP)
`
`LaShawn Griffiths, MSHS-PH, BSN, RN
`Associate Director for Patient Labeling
`Division of Medical Policy Programs (DMPP)
`Barbara Fuller, RN, MSN, CWOCN
`Team Leader, Patient Labeling
`Division of Medical Policy Programs (DMPP)
`
`Sharon R. Mills, BSN, RN, CCRP
`Senior Patient Labeling Reviewer
`Division of Medical Policy Programs (DMPP)
`
`DMPP Review of Patient Labeling: Medication Guide (MG)
`
`
`ELIQUIS (apixaban)
`
`
`Date:
`
`To:
`
`Through:
`
`From:
`
`Subject:
`
`
`Drug Name (established
`name):
`
`Dosage Form and Route:
`
`tablets
`
`Application
`Type/Number:
`
`NDA 202-155
`
`Applicant:
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 3227673
`
`Bristol-Myers Squibb Co. Pharmaceutical Research Institute
`
`
`
`

`

`1
`
` 2
`
`INTRODUCTION
`On September 17, 2012, Bristol-Myers Squibb Co. Pharmaceutical Research Institute
`re-submitted for the Agency’s review, their Original New Drug Application (NDA)
`202-155 for ELIQUIS (apixaban) tablets, in response to a Complete Response Letter
`issued on June 22, 2012. The proposed indication for ELIQUIS (apixaban) tablets is
`to reduce the risk of stroke and systemic embolism in patients with nonvalvular atrial
`fibrillation. On October 2, 2012, the Division of Cardiovascular and Renal Products
`(DCRP) requested that the Division of Medical Policy Programs (DMPP) review the
`Applicant’s proposed Medication Guide (MG) for ELIQUIS (apixaban) tablets.
`This review is written in response to a request by DCRP for DMPP to review the
`Applicant’s proposed MG for ELIQUIS (apixaban) table