1

 

                DEPARTMENT OF HEALTH AND HUMAN SERVICES

 

                      FOOD AND DRUG ADMINISTRATION

 

                CENTER FOR DRUG EVALUATION AND RESEARCH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

                          JOINT MEETING OF THE

 

           CDER PSYCHOPHARMACOLOGIC DRUGS ADVISORY COMMITTEE

 

                                AND THE

 

                    FDA PEDIATRIC ADVISORY COMMITTEE

 

 

 

 

 

 

 

 

 

 

 

                       Monday, September 13, 2004

 

                               8:00 a.m.

 

 

 

                          Holiday Inn Bethsda

                         8120 Wisconsin Avenue

                           Bethesda, Maryland

                                                                 2

 

                              PARTICIPANTS

 

      PSYCHOPHARMACOLOGIC DRUGS ADVISORY

        COMMITTEE MEMBERS:

 

        Wayne K. Goodman, M.D., Chair

        Jean E. Bronstein, R.N., M.S.

          (Consumer Representative)

         James J. McGough, M.D.

         Philip S. Wang, M.D., M.P.H., Dr.P.H.

         Lauren Marangell, M.D.

         Dilip J. Mehta, M.D., Ph.D.

           (Industry Representative)

         Delbert G. Robinson, M.D.

         Daniel S. Pine, M.D.

         Barbara G. Wells, Pharm.D.

         Bruce G. Pollock, M.D., Ph.D.

 

      PEDIATRIC ADVISORY COMMITTEE MEMBERS:

 

         P. Joan Chesney, M.D., Chair

         Deborah L. Dokken, M.P.A.

         Michael E. Fant, M.D., Ph.D.

         Richard L. Gorman, M.D.

         Robert M. Nelson, M.D., Ph.D.

         Thomas B. Newman, M.D., M.P.H.

         Judith R. O'Fallon, Ph.D.

         Victor M. Santana, M.D.

 

      CONSULTANTS AND GUESTS (Voting):

 

         Norman Fost, M.D., M.P.H.

         Charles E. Irwin, Mr., M.D.

         Laurel K. Leslie, M.D., F.A.A.P.

         Steven Ebert, Pharm.D. (Consumer Representative)

         James M. Perrin, M.D.

         Cynthia R. Pfeffer, M.D.

         Gail W. Griffith

           (Patient Representative, Voting)

         Robert D. Gibbons, Ph.D.

         Tana A. Grady-Weliky, M.D.

         Richard P. Malone, M.D.

         Irene E. Ortiz, M.D.

         Matthew V. Rudorfer, M.D.

                                                                 3

 

                        PARTICIPANTS (Continued)

 

      GUEST SPEAKERS AND GUESTS (Non-Voting):

 

         Kelly Posner, Ph.D.

         John March, M.D., M.P.H.

         Samuel Maldonado, M.D., M.P.H.

           (Industry Representative

         Barbara Stanley, Ph.D.

         Madelyn Gould, Ph.D., M.P.H.

 

      FDA PARTICIPANTS:

 

         Robert Temple, M.D.

         Russell G. Katz, M.D.

         Thomas Laughren, M.D.

         M. Dianne Murphy, M.D.

         Anne Trontell, M.D., M.P.H

         Anuja M. Patel, M.P.H., Executive Secretary

                                                                 4

 

                            C O N T E N T S

 

      Call to Order and Opening Remarks,

         Wayne Goodman, M.D.                                     6

 

      Introduction of Committee                                  9

 

      Conflict of Interest Statement,

         Anuja Patel, M.P.H.                                    20

 

      Overview of Issues:

         Dianne Murphy, M.D., Director, Office of

           Pediatric Therapeutics,

           Office of the Commissioner                           25

 

         Russell Katz, M.D., Director, Division of

           Neuropharmacological Drug Products,

             DNDP, CDER                                         34

 

      Regulatory History and Background,

         Thomas Laughren, M.D., Team Leader, DNDP, CDER         46

 

      Recent Observational Studies of Antidepressants

        and Suicidal Behavior,

         Diane Wysowski, Ph.D., Division of Drug Risk

           Evaluation, Office of Drug Safety, CDER              62

 

      Brief Report on TADS Trial,

         John March, M.D., M.P.H., Duke University              74

 

      Committe Discussion on TADS Trial                         93

 

      Characteristics of Pediatric Antidepressant Trials,

         Greg Dubitsky, M.D., Medical Officer,

           DNDP, CDER                                          107

 

      Classification of Suicidality Events,

         Kelly Posner, Ph.D., Columbia University              117

 

      OCTAP Appraisal of Columbia Classification

        Methodology,

         Solomon Iyasu, M.D., M.P.H., Team Leader,

           Office of Counter-Terrorism and Pediatric

             Drug Development                                  140

                                                                 5

 

                      C O N T E N T S (Continued)

 

      Results of the Analysis of Suicidality in Pediatric

        Trials of Newer Antidepressants,

         Tarek Hammad, M.D., Ph.D., M.Sc., M.S., Senior

           Medical Reviewer, DNDP, CDER                        152

 

      Comparison Between Original ODS and DNDP Analyses

        of Pediatric Suicidality Data Sets,

         Andrew Mosholder, M.D., M.P.H., Division of

           Drug Risk Evaluation, ODS, CDER                     200

 

      Citalopram and Escitalopram Product Safety Data,

         Jeffrey Jonas, M.D., Forest Laboratories, Inc.        219

 

      Sertraline Use in Pediatric Population: A

      Risk/Benefit Discussion,

        Steven J. Romano, M.D., Pfizer, Inc.                   232

 

      Wyeth Pharmaceuticals, Joseph S. Camardo, M.D.           247

 

      Open Public Hearing                                      255

 

      Summary by the Committee Chair                           444

 

                                                                 6

 

                         P R O C E E D I N G S

 

                   Call to Order and Opening Remarks

 

                DR. GOODMAN:  I wish to welcome you to

 

      this two-day joint session of the

 

      Psychopharmacologic Drugs Advisory Committee and

 

      the Pediatric Advisory Committee, being held on

 

      September 13th and 14th here, at the Holiday Inn in

 

      Bethesda, Maryland.

 

                I am Wayne Goodman, Professor of

 

      Psychiatry at the University of Florida, today

 

      wearing my hat as chair of the advisory committee.

 

      As you settle in, please take this opportunity to

 

      put into silent mode your cell phones and any other

 

      devices that emit sounds in the audible range of

 

      human beings.

 

                Some of you may be surprised not to see

 

      Matt Rudorfer in this seat but we arm-wrestled for

 

      the position and he won.

 

                [Laughter]

 

                In all seriousness, his term has ended but

 

      we are fortunate to see him return as a voting

 

      consultant to the committee.

 

                                                                 7

 

                I have some official language to read to

 

      you.  All committee members and consultants have

 

      been provided with copies of the background

 

      materials, from both the sponsors and the FDA, and

 

      with copies of letters from the public that we

 

      received by the August 23rd deadline.  The

 

      background materials have been posted on the FDA

 

      website.  Copies of all these materials are

 

      available for viewing at the FDA desk outside this

 

      room.

 

                We have a very large table, a full house

 

      and important topic today so I would like to start

 

      with a few rules of order.  Please speak directly

 

      into the mike when called on.  We will be keeping

 

      track of individuals at the table who wish to speak

 

      and we will call upon them in order.

 

                FDA relies on the advisory committee to

 

      provide the best possible scientific advice

 

      available to assist us in the discussion of complex

 

      topics.  We understand that issues raised during

 

      the meeting may well lead to conversations over

 

      breaks or during lunch.  However, one of the

 

                                                                 8

 

      benefits of an advisory committee meeting is that

 

      the discussions take place in an open and public

 

      forum.  To that end, we request that members of the

 

      committee not engage in off-record conversations on

 

      today's topic during the breaks and lunch.

 

                Whenever there is an important topic to be

 

      discussed there are a variety of opinions.  One of

 

      our goals today and tomorrow is for the meeting to

 

      be conducted in a fair and open way where every

 

      participant is listened to carefully and treated

 

      with dignity, courtesy and respect.  Anyone whose

 

      behavior is disruptive to the meeting will be asked

 

      to leave.  We are confident that everyone here is

 

      sensitive to these issues so understand that these

 

      comments are as a gentle reminder.

 

                We look forward to a productive and

 

      interesting meeting.  This is an unusual meeting in

 

      that we have two advisory committees represented

 

      here, Psychopharmacological Drugs Advisory

 

      Committee, chaired by myself, and the Pediatric

 

      Advisory Committee, chaired by Joan Chesney, to my

 

      left.  We will now go around the table and have the

 

                                                                 9

 

      committee introduce themselves, starting on my

 

      right.  Please indicate your expertise and

 

      affiliation.  We will start in that corner, over

 

      there.

 

                             Introductions

 

                DR. TEMPLE:  Bob Temple.  I am the Office

 

      Director, ODE I.

 

                DR. KATZ:  Russ Katz, Division Director,

 

      Division of Neuropharmacological Drug Products,

 

      FDA.

 

                DR. LAUGHREN:  Tom Laughren, phychopharm.

 

      team leader, in the Neuropharmacological Division.

 

                DR. MURPHY:  Dianne Murphy, Office

 

      Director, Office of Pediatric Therapeutics.

 

                DR. TRONTELL:  Anne Trontell, Deputy

 

      Director, Office of Drug Safety.

 

                DR. FANT:  I am Michael Fant, University

 

      of Texas Health Science Center in Houston.  My

 

      expertise is neonatology and biochemistry.

 

                DR. PFEFFER:  Cynthia Pfeffer.  I am a

 

      child psychiatrist at Weill Medical College of

 

      Cornell University, and I have expertise in

 

                                                                10

 

      depression suicidal behavior in children and

 

      adolescents.

 

                DR. FOST:  Norm Fost, University of

 

      Wisconsin, Professor of Pediatrics, Director of the

 

      Bioethics Program and Chair of the IRB.

 

                DR. ORTIZ:  Irene Ortiz, University of New

 

      Mexico, Albuquerque VA.  My expertise is in

 

      depression in the elderly.

 

                DR. MALONE:  Richard Malone, Drexel

 

      University College of Medicine, and my area is

 

      child psychiatry.

 

                DR. NELSON:  Robert Nelson, Children's

 

      Hospital of Philadelphia and the University of

 

      Pennsylvania.  My expertise is in pediatric

 

      critical care medicine and ethics.

 

                DR. PERRIN:  Jim Perrin, Professor of

 

      Pediatrics, Harvard Medical School and Head of the

 

      Division of General Pediatrics at the Mass. General

 

      Hospital.  I have shortened my expertise as being

 

      in general pediatrics.

 

                DR. GRADY-WELIKY:  Tana Grady-Weliky,

 

      Associate Professor of Psychiatry at the University

 

                                                                11

 

      of Rochester School of Medicine and Dentistry.  My

 

      expertise is in mood disorders and women across the

 

      reproductive life cycle and medical education.

 

                DR. EBERT:  Steven Ebert, Department of

 

      Pharmacy of Meriter Hospital and School of

 

      Pharmacy, University of Wisconsin, Madison.

 

                DR. GIBBONS:  Robert Gibbons, Professor of

 

      Statistics and Professor of Psychiatry and Director

 

      of the Center for Health Statistics at the

 

      University of Illinois, Chicago.  I only do math!

 

                DR. PINE:  Danny Pine, child and

 

      adolescent psychiatrist, National Institute of

 

      Mental Health intramural research program.  I am a

 

      clinical child psychiatrist.

 

                MS. BRONSTEIN:  Jean Bronstein,

 

      psychiatric nurse, Stanford University Hospital,

 

      the consumer representative.

 

                DR. RUDORFER:  Matthew Rudorfer, National

 

      Institute of Mental Health.  My areas of expertise

 

      are mood disorders and psychopharmacology.

 

                MS. PATEL:  Anuja Patel, Advisors and

 

      Consultants Staff, Executive Secretary for the

 

                                                                12

 

      Psychopharmacologic Drugs Advisory Committee.

 

                DR. CHESNEY:  Joan Chesney, the University

 

      of Tennessee, in Memphis, and Professor of

 

      Pediatrics, and my specialty is infectious

 

      diseases.

 

                DR. MCGOUGH:  Jim McGough, Professor of

 

      Psychiatry, UCLA.  My area is child and adolescent

 

      psychopharmacology.

 

                MS. GRIFFITH:  My name is Gail Griffith

 

      and I serve as the patient rep. on this committee,

 

      and I would just like to take this opportunity to

 

      say why I am here.  First, I am not a medical

 

      professional; I am a consumer.  I have suffered

 

      from major depression since I was a teen.  Second,

 

      I have a son who suffers from major depression and

 

      three years ago, at age 17, after he was diagnosed

 

      and placed on a regimen of antidepressants he

 

      attempted suicide by overdosing intentionally on

 

      all his medications.  He nearly died.  So, I know

 

      this illness.  I know what it does to adolescents.

 

                For the record, I would simply like to

 

      state that I have no professional ties to any

 

                                                                13

 

      advocacy group or any patient constituency.  I also

 

      wish to affirm that I have no ties to any

 

      pharmaceutical company, nor do I hold any

 

      investments in pharmaceutical manufacturers.  My

 

      sole responsibility is to ensure that the interests

 

      of concerned parents and families are represented

 

      at this meeting.

 

                DR. MARANGELL:  Lauren Marangell, Baylor

 

      College of Medicine.  I specialize in adult

 

      interventions in mood disorders, both unipolar and

 

      bipolar.

 

                DR. ROBINSON:  I am Delbert Robinson.  I

 

      am from the Albert Einstein College of Medicine, in

 

      New York, and I specialize in psychotic disorders

 

      and anxiety disorders.

 

                DR. LESLIE:  Laurel Leslie.  I am a

 

      behavioral developmental pediatrician at Children's

 

      Hospital, San Diego and my area of expertise is in

 

      children's mental health services research.

 

                DR. IRWIN:  Charles Irwin.  I am a

 

      professor of pediatrics at the University of

 

      California, San Francisco.  I am in charge of the

 

                                                                14

 

      Division of Adolescent Medicine at the University,

 

      which is a multi-disciplinary program that cares

 

      for adolescents and trains large numbers of

 

      individuals caring for teenagers, and my research

 

      is in the area of risk-taking during adolescence.

 

                MS. DOKKEN:  I am Deborah Dokken.  I

 

      reside in the Washington, D.C. Metro area.  I do

 

      not have a specific institutional affiliation, and

 

      I have for several years been involved in parent

 

      and family advocacy and health care.

 

                DR. NEWMAN:  I am Thomas Newman.  I am a

 

      professor of epidemiology and biostatistics in

 

      pediatrics at the University of California, San

 

      Francisco, and a general pediatrician.

 

                DR. WELLS:  I am Barbara Wells.  I am a

 

      professor and Dean of the School of Pharmacy at the

 

      University of Mississippi.  My expertise is in

 

      psychiatric pharmacotherapy.

 

                DR. POLLOCK:  I am Bruce Pollock.  I am a

 

      professor of psychiatry, pharmacology and

 

      pharmaceutical sciences at the University of

 

      Pittsburgh.  I head the Division of Geriatric

 

                                                                15

 

      Psychiatry at the university.

 

                DR. O'FALLON:  Judith O'Fallon, Emeritus

 

      Professor of Biostatistics from the Mayo Clinic,

 

      with 30 years of experience particularly in cancer

 

      clinical trials but clinical trials methods.

 

                DR. SANTANA:  Good morning.  I am Victor

 

      Santana.  I am a pediatric hematologist/oncologist

 

      at St. Jude's Children's Research Hospital in

 

      Memphis, Tennessee.

 

                DR. WANG:  I am Philip Wang, Harvard

 

      Medical School.  I am a psychiatrist and

 

      epidemiologist and those are my areas of expertise.

 

                DR. GORMAN:  Richard Gorman, a practicing

 

      pediatrician for 20 years in the Baltimore suburbs,

 

      Chair of the American Academy's Committee on Drugs,

 

      and representing the American Academy of Pediatrics

 

      at this table.

 

                DR. MALDONADO:  Sam Maldonado.  I work at

 

      pediatric drug development at Johnson & Johnson.  I

 

      am one of the industry representatives to this

 

      committee.

 

                DR. MEHTA:  Dilip Mehta, retired industry

 

                                                                16

 

      executive and industry representative on the

 

      Psychopharmacologic Drugs Advisory Committee.

 

                DR. GOODMAN:  Thank you, all, for being

 

      with us these two days.  Our session today is the

 

      second of two planned advisory committee meetings,

 

      convened to address recent concerns about reports

 

      of suicidal ideation and behavior developing in

 

      some children and adolescents during treatment of

 

      depression with a selective serotonin reuptake

 

      inhibitor, an SSRI, or other newer generation

 

      antidepressants.  Our goal is to gather information

 

      from a variety of sources and perspectives to help

 

      us understand this complex situation, and

 

      ultimately to offer the best possible

 

      recommendations to the FDA.

 

                I would like to thank the many groups,

 

      individuals and families that submitted written

 

      statements in advance of the meeting, many of which

 

      were quite informative as well as moving.  A major

 

      portion of today's meeting will be devoted to a

 

      four-hour open public hearing during which dozens

 

      of people from around, and even beyond, the country

 

                                                                17

 

      will have the opportunity to present their own

 

      personal or professional experiences and ideas

 

      about the relative risks and benefits of

 

      antidepressant medication in children and

 

      adolescents.  Although the necessary consideration

 

      of the clock will permit only a short time at the

 

      microphone for each speaker, I can assure you that

 

      the committee welcomes and values input from all

 

      viewpoints and feels it is essential to our work

 

      that all voices be heard.

 

                The committee's task is more difficult

 

      than usual.  Our review is not confined to whether

 

      one agent is safe and effective based upon the

 

      corresponding clinical trials submitted to the FDA.

 

      We are faced, instead, with assessing efficacy and

 

      safety for nine drugs that represent more than one

 

      chemical class of antidepressants, all of which are

 

      already available on the market.

 

                Although the cornerstone of the data under

 

      examination is derived from randomized clinical

 

      trials submitted to the FDA this time, following a

 

      reclassification of the adverse events, we find

 

                                                                18

 

      ourselves turning to information from a wide

 

      variety of sources, in particular to inform

 

      ourselves about the drugs' possible benefits in

 

      this population.  However, once we open our minds

 

      to consideration of data originating outside

 

      randomized clinical trials we rest upon a slippery

 

      slope in which variations in interpretation are

 

      introduced according to the weighting each member

 

      places on the merits of the source.

 

                For me, the difficulty in assessing the

 

      balance between benefit and risk is multiplied by

 

      the nature of the adverse events under scrutiny.

 

      Psychiatrists grapple, for the most part, with

 

      illnesses that produce significant morbidity and

 

      more rarely mortality except from suicide.  Nothing

 

      in my experience is more tragic than the loss of a

 

      child to suicide.  To think that I might prescribe

 

      an agent that contributed to that outcome is

 

      unbearable.  Equally unbearable is to think that I

 

      did not do enough to prevent it.  This is the

 

      essence of the dilemma before us.

 

                We may not have all the data we would

 

                                                                19

 

      like, especially to assess long-term benefit.  We

 

      can make recommendations about what research should

 

      be conducted, but we will be faced at the

 

      conclusion of business tomorrow to make

 

      recommendations based upon what we know at this

 

      cross-section in time.  In deliberating on the

 

      safety of antidepressant treatment in children, let

 

      us not forget the toxicity of the underlying

 

      disease.  Major depression remains an

 

      under-diagnosed, under-studied and under-treated

 

      serious disorder among many thousands of our

 

      nation's youth, leading to considerable suffering,

 

      disability and heartbreak in many families.

 

                I believe that all of us in this room

 

      share the desire to alleviate the suffering from

 

      this disorder through the successful use of

 

      interventions that are made available to all those

 

      who need them.  Despite the daunting task before

 

      us, I remain hopeful that with a fair and

 

      open-minded review of the evidence this advisory

 

      committee will constructively address the issues

 

      and ensure that interventions for this serious

 

                                                                20

 

      disorder meet high standards for both effectiveness

 

      and safety.

 

                Now I will ask Anuja Patel, executive

 

      secretary for the advisory committee, to review

 

      some of the ground rules for this committee and the

 

      public hearing.

 

                     Conflict of Interest Statement

 

                MS. PATEL:  Good morning.  Before I

 

      continue, I would like to notify you of a

 

      correction on the roster attached to the agenda.

 

      The following consultants, Dr. Robert Gibbons, Dr.

 

      Matthew Rudorfer, Dr. Richard Malone, Dr. Tana

 

      Grady-Weliky and Dr. Irene Ortiz will be added to

 

      the roster.  Amended copies of the roster will be

 

      available later this morning at the information

 

      desk outside this ballroom.

 

                As you know, we have a very full open

 

      public hearing today, and in the interest of both

 

      fairness and efficiency we are running it by some

 

      strict rules.  To make transitions between speakers

 

      more efficient, all speakers will be using the

 

      microphone and podium in front of the audience. 

 

                                                                21

 

      Each speaker has been given their number in the

 

      order of presentations and when the person ahead of

 

      you is speaking, we ask that you move to the nearby

 

      next speaker chair.  Individual presenters and

 

      families have been allotted three minutes for their

 

      presentations.  The one consolidated presentation

 

      has been given five minutes.  We will be using a

 

      timer and speakers who run over their time will

 

      find that the microphone is no longer working.  We

 

      apologize for the need for the strict rules, but we

 

      wanted to be fair and to give as many people as

 

      possible an opportunity to participate.

 

                The public may submit comments after this

 

      meeting directly to the FDA's Division of Dockets

 

      Management.  Instructions for submitting electronic

 

      and written statements are available at the

 

      registration desk outside this room.  The docket

 

      will remain open until July 29, 2005.  Thank you

 

      for your cooperation.

 

                I would like to read the meeting statement

 

      into the record now.  The following announcement

 

      addresses the issue of conflict of interest and is

 

                                                                22

 

      made a part of the record to preclude even the

 

      appearance of such at this meeting.  The topics to

 

      be discussed today are issues of broad

 

      applicability.  Unlike issues before a committee in

 

      which a particular company's product is discussed,

 

      issues of broader applicability involve many

 

      industrial sponsors and products.

 

                All special government employees and

 

      invited guests have been screened for their

 

      financial interests as they may apply to the

 

      general topics at hand.  The Food and Drug

 

      Administration has granted particular matters of

 

      general applicability waivers under 18 USC

 

      208(b)(3) to the following special government

 

      employees, which permits them to participate fully

 

      in today's discussion and vote:  Jean Bronstein,

 

      Dr. Joan Chesney, Dr. Wayne Goodman, Dr. Lauren

 

      Marangell, Dr. James McGough, Dr. James Perrin, Dr.

 

      Bruce Pollock.  In addition, Dr. Philip Wang has

 

      been granted a limited waiver that permits him to

 

      participate in the committee's discussions.  He is,

 

      however, excluded from voting.

 

                                                                23

 

                A copy of the waiver statements may be

 

      obtained by submitting a written request to the

 

      agency's Freedom of Infection Office, Room 12A-30

 

      of the Parklawn Building.

 

                In addition, Dr. Judith O'Fallon and Dr.

 

      Victor Santana have financial interest under 5 CFR,

 

      Part II, Sec. 40.202 that are covered by a

 

      regulatory waiver under 18 USC 208(b)(2).

 

                Because general topics impact so many

 

      entities, it is not practical to recite all

 

      potential conflicts of interest as they may apply

 

      to each member, consultant and guest speaker.  FDA

 

      acknowledges that there may be potential conflicts

 

      of interest but, because of the general nature of

 

      the discussion before the committee, these

 

      potential conflicts are mitigated.

 

                With respect to FDA's invited industry

 

      representatives, we would like to disclose that Dr.

 

      Dilip Mehta and Dr. Samuel Maldonado are

 

      participating in this meeting as industry

 

      representatives, acting on behalf of regulated

 

      industry.  Dr. Mehta is retired from Pfizer and Dr.

 

                                                                24

 

      Maldonado is employed by Johnson & Johnson.

 

                With respect to all other participants, we

 

      ask in the interest of fairness that they address

 

      any current or previous financial involvement with

 

      any firm whose product they may wish to comment

 

      upon.  Thank you.

 

                DR. GOODMAN:  We will now proceed with a

 

      series of formal presentations that will bring us

 

      to 11:45 a.m. and then a 15-minute discussion

 

      before lunch.  In the interest of time, I would

 

      like to ask my fellow committee members to restrict

 

      their questions after each presentation to issues

 

      of clarification only.  There will be time, 15

 

      minutes, for some discussion between 11:45 and

 

      12:00 and tomorrow there will be a great deal of

 

      time for discussion and consideration of the

 

      questions before us.  So, please, if you have

 

      questions about clarification, you can ask them

 

      after each presentation but restrict it to those

 

      kinds of issues.

 

                With that, I would like to introduce Dr.

 

      Dianne Murphy, of the FDA, who will be followed by

 

                                                                25

 

      Dr. Russell Katz, also of the FDA.

 

                           Overview of Issues

 

                DR. MURPHY:  Good morning and welcome to

 

      this very important discussion.  Before we begin

 

      today's important deliberations, I would ask us to

 

      step back and see the broader context in which this

 

      meeting is occurring.  I am going to spend a few

 

      minutes trying to describe that for you.

 

                There are four points I hope you take from

 

      this short presentation.  One is that the majority

 

      of medicines given to children in this country are

 

      prescribed off-label and have not been studied in

 

      all the pediatric populations in which they are

 

      used.

 

                Second, because of new legislation and

 

      regulations since 1998, FDA has seen an increase in

 

      products that are used in children being studied in

 

      children.

 

                Third, for the first 100 products,

 

      involving over 200 studies conducted as a result of

 

      the new legislation, FDA has found that

 

      approximately one-fourth of the time there was a

 

                                                                26

 

      need to change the dose, a new pediatric-specific

 

      adverse event was described or the product was not

 

      found to be efficacious despite the fact that it

 

      was efficacious in adults.

 

                Fourth, part of the reason we are here

 

      today is because we are finally studying the

 

      therapies that are being given to children.

 

      Children deserve the same level of evidence that is

 

      required for adults to determine that their use by

 

      them is safe, effective and properly dosed.  They

 

      are a heterogeneous group who undergo rapid

 

      metabolic, hormonal, physiologic, development and

 

      growth changes in comparison to us, adults, who are

 

      rather static and tend only to deteriorate.

 

                Over the last two decades FDA has actively

 

      supported, along with the American Academy of

 

      Pediatrics and many other groups, the efforts to

 

      encourage development of information and

 

      appropriate use of therapies in the pediatric

 

      population.

 

                Very quickly, and this is important to

 

      understand, again, the context in which some of

 

                                                                27

 

      this information has been brought to you, in the

 

      last decade we have made tremendous progress.  In

 

      1994 the agency published an approach that it hoped

 

      would help foster and encourage development of

 

      therapies that we be used in children.  Congress

 

      passed legislation in 1997 which is referred to as

 

      the exclusivity or the incentive to develop studies

 

      on products that are being used in children.

 

                In 1998 the FDA published the Pediatric

 

      Rule, which was an effort to say that if a sponsor

 

      is going to develop a product in adults and that

 

      same disease occurs in children, or condition, that

 

      product in most circumstances and certain

 

      conditions would be required to be studied.

 

                We are going to go more into the 2001

 

      adoption by FDA of Subpart D, Pediatric Ethics

 

      Regulations, and I wanted to bring up the Best

 

      Pharmaceuticals for Children Act, which you will

 

      hear referred to as BPCA, because it renewed the

 

      congressional legislation of '97 and is important

 

      in that, again, it is the renewal of the incentive

 

      to study products that are being used in children.

 

                                                                28

 

                Another congressional legislative

 

      activity, the Pediatric Research Equity Act, in

 

      essence confirmed FDA's authority to require

 

      studies in children in certain circumstances.

 

                In the last decade, particularly really

 

      since 1997, the FDA has issued over 290 written

 

      requests to sponsors asking them to study products

 

      in children because these products are being used

 

      in children.  We have had submitted to us over 110

 

      products, involving over 220 studies in children,

 

      and have now more than 76 new labels that have new

 

      pediatric information from these studies.

 

                The major depressive disorders were

 

      included in the written requests that were issued,

 

      and written requests were issued for the products

 

      you see listed here, Prozac, Zoloft, Remeron, Paxil

 

      Effexor, Celexa and Serzone.  Those studies were

 

      all conducted under this program or in response to

 

      this program.

 

                This is a list of some of the programs and

 

      activities that are in place at FDA to help ensure

 

      the quality and ethical conduct of studies and the

 

                                                                29

 

      approach to pediatrics.  This is really focusing

 

      mostly on the drugs component of this.  But there

 

      is, at the Commissioner's level, an Office of

 

      Pediatric Therapeutics.  This was enacted by the

 

      Best Pharmaceuticals for Children Act in 2002 and

 

      first staff were hired last year.  We now have in

 

      place an ethicist whose focus is pediatric ethics.

 

      You will hear a little bit more about the Pediatric

 

      Advisory Committee and Subpart D referrals in a

 

      minute.  You just heard about the exclusivity

 

      process which has been important in making sure

 

      that trials do get conducted.  I will spend a few

 

      moments at the end talking about disclosure

 

      requirements that are unique to pediatric studies

 

      that are conducted under exclusivity.

 

                This is another meeting, actually in a

 

      long series of meetings that have occurred to

 

      ensure the scientific and ethical quality of

 

      activities involving studies that are being

 

      conducted in children.  Since 1999, the Pediatric

 

      Advisory Subcommittee has had, including today's

 

      meeting, over ten meetings that have addressed over

 

                                                                30

 

      ten scientific issues, three ethical issues and, in

 

      addition, starting last year, began having specific

 

      safety reviews of products that have been approved,

 

      again under the exclusivity provisions, so that all

 

      adverse events that occurred in the year after

 

      product was granted exclusivity were reviewed.

 

      Again, this is just to inform you of the ongoing

 

      pediatric activities that are occurring at the FDA,

 

      some of them.

 

                The new advisory committee, I should say

 

      full Pediatric Advisory Committee is meeting for

 

      the first time today.  It was  chartered this year

 

      and is mandated to include patient and family

 

      organizations, and its mandated responsibilities

 

      include safety, labeling disputes and Subpart D

 

      referrals and general pediatric issues.

 

                The first Subpart D ethics panel met this

 

      past Friday and will report to this committee on

 

      Wednesday.  I will tell you a little bit more about

 

      that.

 

                It is important to understand that Subpart

 

      D, which is part of the Common Rule, was those

 

                                                                31

 

      extra protections for children applied to only

 

      federally funded activities until recently.  In

 

      2000, the Children's Health Act required FDA

 

      regulated products to be in compliance with

 

      additional protections for children that are

 

      embodied in the Subpart D of the Common Rule.

 

                Subpart D is fundamentally a referral

 

      process.  There is much more to it but it is a

 

      process for IRBs when they are unsure they can

 

      approve or under which regulation they should

 

      approve a study involving children.  The Pediatric

 

      Ethics Subcommittee reports to the Pediatric

 

      Advisory Committee and this is a public process.

 

                The disclosure of the studies that are

 

      conducted in children is distinct for studies that

 

      are conducted in children under the exclusivity

 

      provisions of BPCA.  I mention this because it is

 

      unusual in the FDA if a product is not approved

 

      that those studies would be disclosed.  However, we

 

      now have, again under BPCA, a requirement that

 

      within 180 days of submission of a pediatric study

 

      a public summary of the medical and clinical

 

                                                                32

 

      pharmacology reviews will be posted.  There are now

 

      41 pediatric summaries posted at this website.

 

      Basically, you can go to the FDA page and get there

 

      by going to the Center for Drugs or pediatric

 

      summary\summary review.

 

                The summaries of Effexor, Paxil, Serzone,

 

      Celexa, Zoloft and Remeron are available on the

 

      pediatric summary review site.  As you know, and

 

      will hear, Prozac is the only antidepressant that

 

      is approved for use in children, and it is posted

 

      up on FDA's site for approved applications.  That

 

      URL is provided for you here and in your handouts.

 

                The new pediatric data has taught us that

 

      our knowledge of pediatric therapeutics is in its

 

      infancy; that we must study children if we are to

 

      understand pediatric-specific adverse events and

 

      reactions or if a product is going to work in

 

      children.  The pharmacokinetics in children has

 

      proven to be more variable than anticipated.  The

 

      submitted studies that we are receiving are

 

      teaching us that we need to know more about

 

      pediatric endpoints, pediatric trial designs and

 

                                                                33

 

      how to conduct these trials, and that we will need

 

      to change some of our trials as we move forward.

 

      Ethical issues require reassessment from a

 

      pediatric perspective.  Therefore, at this point no

 

      longer shall each child be an experiment of one in

 

      which not much knowledge is gained.

 

                As we move forward, it will require our

 

      careful attention if we are to discover why

 

      children are behaving differently.  If children are

 

      to be appropriately treated, we will need to know

 

      more than how to correct those things or describe

 

      adverse events.  We are going to need answers to

 

      such fundamental questions as to why children react

 

      differently, what are the metabolic, physiologic

 

      events that are occurring that necessitate

 

      different dosing, or why is there a therapy that

 

      works in adults and does not work in children.  Our

 

      public policy must be more knowledge to replace our

 

      ignorance.  Thank you, and we look forward to your

 

      discussion.

 

                DR. MARANGELL:  Dr. Murphy, a quick

 

      question, when the FDA requests a study can you

 

                                                                34

 

      specify methodology and assessments that you would

 

      like to see included?

 

                DR. MURPHY:  When FDA requests a study we

 

      do put in that written request the trial design,

 

      the number of patients, the adverse events--you

 

      know, under exclusivity all of that does go into

 

      the written request.

 

                DR. GOODMAN:  Thank you, Dr. Murphy.  Now

 

      Dr. Katz?

 

                           Overview of Issues

 

                DR. KATZ:  Thank you, Dr. Goodman, and

 

      good morning.  I would like to welcome you to this

 

      joint meeting of the Psychopharmacologic Drugs

 

      Advisory Committee and the Pediatric Advisory

 

      Committee.

 

                As you know, we are here to present to you

 

      and to ask for your guidance in interpreting the

 

      results of our analyses of the relationship between

 

      antidepressant drug use and suicidal behavior in

 

      controlled trials in pediatric patients.  This

 

      meeting is in follow-up to the meeting of these two

 

      committees held in February of this year.  At that

 

                                                                35

 

      meeting, as you recall, we presented to you and

 

      obtained your endorsement of our plans to perform

 

      these analyses.

 

                At this point I would like to very briefly

 

      recap how we arrived to this point.  As you know,

 

      we first became aware of a possible relationship

 

      between antidepressant drug use and suicidal

 

      behavior in pediatric patients in May of 2003 when,

 

      in response to our request for further

 

      clarification of their data, GlaxoSmithKline, the

 

      manufacturer of Paxil, submitted data to us that

 

      suggested such a link for that drug.  As a result

 

      of this submission, the agency issued a public

 

      statement recommending that this drug not be used

 

      in pediatric patients with depression, and

 

      independently we asked all other manufacturers of

 

      antidepressant drugs to resubmit the relevant data

 

      from controlled studies with their drugs in

 

      pediatric patients.

 

                Based on our review of this data, we

 

      issued a statement informing prescribers that there

 

      was a potential relationship between all of these

 

                                                                36

 

      drugs and suicidal behavior, and that these drugs

 

      should be used with caution in these patients.

 

                However, at that time we also noticed that

 

      the data submitted to us from the various companies

 

      was not reported to us in a form that would permit

 

      definitive analyses.  Specifically, each company

 

      classified various behaviors as being

 

      suicide-related adverse events in their own

 

      idiosyncratic manner.  This led to questions about

 

      whether or not these events were, in fact,

 

      suicide-related and, in addition, prevented

 

      meaningful comparisons between drugs in this class.

 

                For this reason, we decided that an

 

      independent assessment of these possible events by

 

      experts in suicidology would be the most

 

      appropriate way to definitively answer the question

 

      of whether or not any, all or none of these drugs

 

      increased the risk of suicidal behavior in

 

      pediatric patients.  Let me just add that by

 

      definitive analyses I mean analyses that make the

 

      best possible use of the available data.

 

                It was at this time that we brought the

 

                                                                37

 

      issue before you.  At that meeting we presented you

 

      with our plans to submit blinded narrative

 

      descriptions of possible suicide-related events to

 

      a group of independent experts, to be coordinated

 

      by Columbia University, whose task it would be to

 

      classify these events as being suicide-related or

 

      not.  Although no formal vote was taken, this

 

      committee fully endorsed this effort and agreed

 

      that the data in hand at that time did not permit

 

      definitive analyses to be done.

 

                The committee also recommended, based in

 

      part on the data in hand but also, I believe

 

      importantly, on the basis of testimony from members

 

      of the public who had suffered the tragedy of loved

 

      ones who had committed suicide while taking these

 

      drugs, that the agency should ask sponsors of these

 

      products to warn prescribers that patients being

 

      treated with these drugs, especially at the

 

      beginning of treatment, should be closely watched

 

      for the emergence of signs and symptoms that might

 

      suggest a worsening in their clinical state.

 

                Since that February meeting a number of

 

                                                                38

 

      important things have happened.  Based on your

 

      advice, the agency drafted, and all of the sponsors

 

      of these drugs have adopted, language in product

 

      labeling warning about the possibility of

 

      significant behavioral changes at the outset of

 

      treatment with these drugs in both pediatric and

 

      adult patients, and the prescribing community and

 

      the public have been informed of these changes.

 

                Critically, this warning made clear that

 

      the possibility of worsening and a possible

 

      increased risk of suicidal behavior at the outset

 

      of treatment could not necessarily be attributed to

 

      the drugs because the data did not permit such a

 

      definitive conclusion.  Nonetheless, it was

 

      considered appropriate and prudent to inform

 

      prescribers and patients and their families that

 

      changes in behavior could occur with the onset of

 

      treatment.

 

                Also, the Columbia group has completed

 

      their task of reclassifying the potential cases of

 

      suicidal behavior and, importantly, we have

 

      completed our reanalyses of these data.  As

 

                                                                39

 

      promised at our February meeting, we are now ready

 

      to present to you the results of these analyses.

 

                At this point I would just like to give

 

      you a brief overview of the agenda for today's and

 

      tomorrow's session.  First Dr. Tom Laughren, of the

 

      Neuropharmacology Division, will provide you with a

 

      more detailed account of the regulatory history and

 

      events that have brought us here this morning.  He

 

      will be followed by Dr. Diane Wysowski, of the

 

      agency's Office of Drug Safety, who will briefly

 

      present the results of some recently published

 

      epidemiologic studies relevant to this question.

 

      We have provided the committees with copies of

 

      these published materials.  Although, of course, we

 

      consider our reanalyses of the controlled data to

 

      be the primary source of data on which your

 

      discussions and recommendations will be based, we

 

      thought it important to present at least briefly

 

      the available relevant epidemiologic data.

 

                Next, Dr. John March, of Duke University,

 

      will present a brief report of the Treatment for

 

      Adolescent Depression Study, or TADS trial, a

 

                                                                40

 

      recently completed trial that evaluated the effects

 

      of fluoxetine in adolescents with depression.  As

 

      you know, fluoxetine is the only drug approved in

 

      the United States for the treatment of depression

 

      in pediatric patients and, as you will see, these

 

      data make an important contribution to our overall

 

      assessment of the problem before us.

 

                Dr. Greg Dubitsky, again of the

 

      Neuropharmacology Division, will then present an

 

      overview of the design of the pediatric trials from

 

      which the data for our analyses were derived.  This

 

      exploration is important because similarities and

 

      differences in design elements among these trials

 

      can have important implications for whether or not

 

      these data can be examined as a whole, or whether

 

      they must be considered separately.

 

                Then, Dr. Kelly Posner, of Columbia

 

      University and the primary person responsible for

 

      coordinating the blinded reclassification effort,

 

      will present to you the methodology her group used

 

      to produce what we now consider to be the

 

      definitive data on which we have based our

 

                                                                41

 

      reanalyses.

 

                Because the reclassification of these

 

      clinical events was the critical activity on which

 

      all subsequent analyses and decisions will have

 

      been based, and because by its nature it involved

 

      subjective assessments of the primary data, we felt

 

      strongly that it was appropriate to ensure that the

 

      methodology used by the Columbia group could

 

      reliably and reproducibly yield similar results

 

      when applied by an independent group.  For this

 

      reason, agency scientists performed such an

 

      independent reclassification of a percentage of

 

      these cases, utilizing the Columbia classification

 

      schema, and Dr. Solomon Iyasu, of the agency's

 

      pediatric group, will present the results of this

 

      independent audit of the Columbia process.

 

                At that point, Dr. Tarek Hammad, of the

 

      neuropharmacology group, will then present the most

 

      critical results of his extensive reanalyses of the

 

      data as reclassified by the group of outside

 

      experts.  These analyses look at the data for

 

      individual drugs, as well as across all drugs, and

 

                                                                42

 

      will provide the data on which the committee

 

      subsequent discussions will be based.

 

                Finally, the last formal agency

 

      presentation will be given by Dr. Andrew Mosholder,

 

      of the Office of Drug Safety.  Dr. Mosholder's name

 

      is undoubtedly familiar to you.  Dr. Mosholder was

 

      the agency reviewer who had, prior to the February

 

      advisory committee meeting, performed analyses on

 

      the cases as submitted, that is, the

 

      non-reclassified cases, and had concluded that

 

      these drugs did, in fact, increase the risk of

 

      suicide-related behaviors in this population.  As

 

      you know, Dr. Mosholder did not present his

 

      conclusions at the February meeting, although the

 

      data on which his analyses were based were

 

      presented and we noted at that time that some in

 

      the agency had already reached a definitive

 

      conclusion on this question.

 

                There has been since that meeting

 

      considerable public discussion and controversy

 

      related to the fact that Dr. Mosholder was not

 

      given the opportunity to present his conclusions at

 

                                                                43

 

      that meeting.  The reasons for our decision at that

 

      time were straightforward.  As I have discussed

 

      today and as we have discussed publicly on numerous

 

      occasions, we had decided that at the time of the

 

      February meeting the data had not been submitted in

 

      a form in which we could reliably agree that the

 

      events described as representing suicide-related

 

      behavior did, in fact, represent such behavior.

 

      We, therefore, felt that conclusions reached on the

 

      basis of analyses that relied on these descriptions

 

      could no, in turn, be considered completely

 

      reliable.

 

                We felt, and still feel, that presenting

 

      conclusions based on potentially unreliable

 

      analyses could have led to errors in either

 

      direction, that is, resulted in a conclusion that

 

      the drugs were dangerous when they really were not,

 

      or resulted in a conclusion that the drugs were

 

      safe when they were not.  A mistake of either kind

 

      could have, in our view, disastrous consequences.

 

      For this reason it was, and remains, our view that

 

      these decisions must be based on the most reliable

 

                                                                44

 

      analyses possible.  Now that the definitive

 

      analyses have been done, however, Dr. Mosholder

 

      will present his own analyses and conclusions, with

 

      particular attention to a comparison between his

 

      results and Dr. Hammad's.

 

                Following lunch we will hear brief

 

      comments from several of the pharmaceutical

 

      companies who have antidepressant drug products on

 

      the market, and the day will end with the open

 

      public hearing.  A total of 73 members of the

 

      public have signed up to make statements.  As you

 

      have heard and as in the February meeting, we will

 

      again need to limit the statements from the public,

 

      this time to three minutes per individual.  We

 

      recognize that this is not much time and we

 

      apologize for the limit but it would be impossible

 

      for all those who wish to make statements to do so

 

      without imposing this limitation.  We appreciate

 

      your understanding on this point and, as you have

 

      heard, anyone who wishes may submit written

 

      testimony to the docket.

 

                Tomorrow the committee will discuss the

 

                                                                45

 

      data you will have heard.  We, of course, look

 

      forward to this discussion and in particular to

 

      your answers to the questions we have brought to

 

      you and which we have provided in your background

 

      packages.  We are, in brief, interested in your

 

      views on our approach to the reclassification

 

      effort and, critically, whether you believe that

 

      the analyses establish that one or more of the

 

      drugs studied increases the risk of suicidality in

 

      pediatric patients.  Importantly, if you do

 

      conclude that there is a signal for suicidality,

 

      whether for one or more of these drugs, we need to

 

      know what additional regulatory action, if any, you

 

      believe should be taken.

 

                The results of our reanalyses are complex

 

      and their interpretation is not immediately

 

      obvious.  They raise difficult questions, not only

 

      about the fundamental meaning of the results of the

 

      analyses for each drug, but also about the

 

      comparability of the various treatments and,

 

      therefore, whether it is appropriate to consider

 

      the drugs as a class for which any conclusion

 

                                                                46

 

      reached should globally apply or whether the drugs

 

      must be considered individually.  Further, the

 

      question of any further regulatory action is also a

 

      thorny one and must take into account the

 

      consideration of the lack of available

 

      effectiveness data for all of the drugs, except

 

      fluoxetine, although the absence of this

 

      effectiveness data is not easily interpreted

 

      either.

 

                Because of the complex nature of the

 

      evidence and because of the extraordinary

 

      importance for the public health of the decisions

 

      that we need to make, we are turning to you, the

 

      experts, for guidance on these matters.  We thank

 

      you in advance for your efforts.

 

                DR. GOODMAN:  Thank you, Dr. Katz.  I

 

      would like to invite Dr. Tom Laughren to come to

 

      the podium.

 

                   Regulatory History and Background

 

                DR. LAUGHREN:  Thank you, and I would also

 

      like to welcome everyone to the meeting today.  I

 

      am going to begin by briefly giving an overview of

 

                                                                47

 

      events leading up to today's meeting.  I am then

 

      going to talk about the key elements in the

 

      division's exploration and analysis of the

 

      pediatrics suicidality data.  I will then spend a

 

      little time talking about our March 22nd public

 

      health advisory and the subsequent labeling changes

 

      that have now been implemented.  Then I am going to

 

      spend a little time talking about the effectiveness

 

      data.  I did this at the last meeting; I will do

 

      this again because I think it is important to have

 

      these data in mind since they are an important part

 

      of the context of this discussion about pediatric

 

      suicidality.  Then I am going to quickly go over

 

      the questions and the issues for which we are going

 

      to be seeking feedback tomorrow.  I think it is

 

      important that you have these questions in mind as

 

      you hear the talks this morning.

 

                This slide lists a number of the people at

 

      FDA who have been involved in looking at these

 

      data.  As you can see, these people come from

 

      various sections of the agency.  It is a long list,

 

      and really the point of this slide is that we take

 

                                                                48

 

      this matter very seriously and we have invested a

 

      lot of effort into trying to understand these data.

 

                You heard earlier about the two laws,

 

      FDAMA and BPCA, that give FDA authority to grant

 

      additional market exclusivity for companies which

 

      do pediatric studies.  The point of this slide is

 

      that most of the data that we are dealing with this

 

      morning come from these types of studies, in other

 

      words, studies that were done to obtain additional

 

      marketing exclusivity.  However, we have also

 

      included in our analysis data from a ninth

 

      antidepressant drug, Wellbutrin, that was not

 

      studied for exclusivity.  That was one study in

 

      ADHD.  We are also including in our analysis data

 

      from the TADS trial that you will hear about in

 

      more detail later in the morning from John March,

 

      from Duke.

 

                As Dr. Katz pointed out, this issue first

 

      came to our attention based on a review of the

 

      Paxil supplement.  In that review, the reviewer

 

      noticed that events suggestive of possible

 

      suicidality were subsumed, along with other

 

                                                                49

 

      behavioral events, under the preferred term

 

      "emotional lability."  This led FDA to issue a

 

      request to the sponsor, GSK, to explain this coding

 

      practice.  Ultimately, that resulted in a report to

 

      FDA, in May of last year, on pediatric suicidality

 

      with Paxil.  As Dr. Katz pointed out, that report

 

      did suggest a signal of increased suicidality in

 

      association with drug use, particularly in one of

 

      three depression trials in that program.

 

                What I am going to do in this slide is

 

      very quickly run through subsequent events that led

 

      us up to the February advisory committee meeting.

 

      So, in June of last year we issued a public

 

      statement cautioning about the use of Paxil in

 

      pediatric patients with depression.  In July we

 

      issued requests to sponsors of eight other

 

      antidepressant products to ask them to give us the

 

      same kind of summary data that GSK had provided for

 

      Paxil.

 

                In September of last year we held an

 

      internal regulatory briefing at FDA.  The purpose

 

      of this was to brief upper management about this

 

                                                                50

 

      signal.  The two points that I took away from that

 

      meeting from the standpoint of the division's work

 

      were, number one, there was general agreement that

 

      it would be important to try and classify these

 

      events since many of them were not clearly related

 

      to suicidality and we felt it would be very

 

      important to do a rational classification.

 

      Secondly, there was sentiment that we ought to try

 

      and obtain patient-level data information, beyond

 

      the summary information, in order to try and

 

      explain differences among trials and between

 

      programs.

 

                In September and October we began to get

 

      responses to our July requests.  Also, in October

 

      we issued requests to sponsors for the

 

      patient-level data sets that I mentioned earlier.

 

      Also in October, we decided to go outside of FDA to

 

      get a classification of these cases accomplished.

 

      Then, again as Dr. Katz mentioned, in October we

 

      issued a second public health advisory, this time

 

      extending the cautionary language to all current

 

      generation antidepressants.  Finally, in November

 

                                                                51

 

      and December, having looked at the responses to the

 

      July request for summary data, it occurred to us

 

      that we may not have obtained all of the relevant

 

      events and so we sent additional requests to have a

 

      broader search for events that we would then try

 

      and get classified.

 

                That brings us up to the February advisory

 

      committee that we held.  At that meeting you

 

      advised us to basically continue with our analysis

 

      of the data but, in the meantime, to go ahead and

 

      make some labeling changes.  In March of this year

 

      we issued a public health advisory announcing the

 

      changes that we had requested.  In the meantime,

 

      the classification of the cases was ongoing by the

 

      Columbia group.  Those were completed in June of

 

      this year.  Then, in August of this year we

 

      completed our analysis of the pediatric suicidality

 

      data.

 

                In this slide what I am doing is basically

 

      summarizing what I think is the major contribution

 

      of the division to this effort.  Again, we went to

 

      a lot of effort to try to ensure completeness of

 

                                                                52

 

      case findings, that we had a complete set of events

 

      to have classified.  We then worked with Columbia

 

      University to have these events classified.

 

                As an aside, I would like to mention that

 

      this effort, conducted by Kelly Posner and her

 

      group at Columbia and the very exceptional group of

 

      outside experts that they assembled to do this,

 

      represents a very substantial effort that has not

 

      only helped us to understand these data but I think

 

      will have implications for the field in terms of

 

      developing a standard approach to classifying these

 

      kinds of data, and also will lead to guidance

 

      document that, hopefully, will improve

 

      ascertainment for suicidality, which was a very

 

      significant problem in these trials.

 

                Finally, the third effort that we were

 

      involved in was, again, in obtaining the

 

      patient-level data sets that allowed us to try and

 

      explore for confounding and effect modification, in

 

      essence, to try to explain some of the striking

 

      differences we were seeing in the signal across

 

      trials within programs and across programs.

 

                                                                53

 

                As mentioned, at the February advisory

 

      committee you advised us to go ahead and strengthen

 

      labeling, in particular for monitoring for

 

      suicidality, while we were completing our analysis.

 

      We did this and we announced the request that we

 

      were making in a March 22nd public health advisory.

 

                The changes in labeling that we requested

 

      have now all been implemented for the ten drugs of

 

      interest.  I would add here that our plan is to

 

      extend the standard language to all

 

      antidepressants, not just the current generation

 

      and, in fact, that has already been done for some

 

      of these drugs.  We are waiting to do it for the

 

      others until we work out the final standard

 

      language, which will be based on advice we get from

 

      you at this meeting.

 

                What I want to do in this slide is to very

 

      quickly go over the labeling changes that have been

 

      implemented now.  This slide focuses on the advice

 

      for clinicians who are using antidepressants for

 

      treating any condition really, whether in adult of

 

      pediatric patients.  So, the advice is as follows,

 

                                                                54

 

      first of all, we are asking clinicians to closely

 

      observe patients who are being treated with

 

      antidepressants for clinical worsening and for the

 

      emergence of suicidality, especially at the

 

      beginning of therapy but also at times of dose

 

      change.

 

                Secondly, we are asking clinicians to

 

      consider changing the therapeutic regimen in

 

      patients whose depression is either persistently

 

      worse or whose emergent suicidality is severe,

 

      abrupt in onset, or was not part of the patient's

 

      presenting symptoms.

 

                Finally, we are also asking clinicians to

 

      observe for the emergence of other symptoms as

 

      well, for example, anxiety, agitation, panic

 

      attacks, insomnia, irritability, hostility,

 

      impulsivity, and so forth.  The idea here is that

 

      there is a belief, not really solidly empirically

 

      established but a belief that many of these events

 

      may represent precursors to emerging suicidality.

 

      So, we are also asking clinicians to be alert to

 

      these symptoms.

 

                                                                55

 

                This slide focuses on advice for families

 

      and caregivers that also is included in labeling.

 

      We are asking those folks to also be alert to the

 

      emergence of these same symptoms and to report

 

      those symptoms to healthcare providers if they

 

      emerge.

 

                Now I want to turn to briefly describing

 

      the efficacy data for the 15 short-term trials that

 

      we looked at in our review of these pediatric

 

      supplements.  I am going to be focusing on primary

 

      outcomes in those trials.  I also want to spend a

 

      little time talking about the difficulty in

 

      interpreting negative findings in this setting and

 

      again I want to note that although I am not going

 

      to be talking about the TADS efficacy data, you

 

      will be hearing about the TADS efficacy data from

 

      John March a little bit later in the morning.

 

                This is kind of a busy slide but basically

 

      each row in this table represents a different

 

      trial.  Again, there was a total of 15 trials.

 

      This is color-coded so you can separate the

 

      different programs.  There were seven programs. 

 

                                                                56

 

      Paroxetine had three trials.  The rest all had two

 

      trials.  The column to look at is the far column

 

      where I have summarized the results on the primary

 

      endpoint.

 

                Basically I have characterized the results

 

      as follows:  Where the p value on drug versus

 

      placebo on the primary endpoint was less than 0.05,

 

      I am calling it positive.  As you can see, that

 

      applies to the two fluoxetine trials and one of the

 

      citalopram trials.  If the p value fell between

 

      0.05 and 0.1 I am characterizing it as a trend.

 

      That applies to one of the sertraline trials and

 

      one of the nefazodone trials.  If the p value was

 

      greater than 0.1 on that primary endpoint I am

 

      characterizing it as negative.  That applies to all

 

      the remaining trials.  So, basically what you have

 

      here is three out of 15 trials meeting FDA's

 

      standard for being positive.

 

                The other point I want to make on this

 

      slide is that this represents FDA's view and I

 

      think it is a reasonable standard, however, it is

 

      not the only standard.  To illustrate that, I want

 

                                                                57

 

      to talk about two published papers, one for study

 

      329, the paroxetine trial, a paper that was

 

      published by Keller in 2001.  That paper

 

      characterized that trial as a positive study, the

 

      argument being that although it failed on the

 

      primary endpoint it succeeded on all the secondary

 

      endpoints.  So, the authors of that paper

 

      considered it a positive trial and many in the

 

      community also considered that a positive study.

 

                Secondly, there was a paper published on

 

      the two sertraline trials by Wagner et al., in

 

      2003, that was based on a pooling of the two

 

      trials.  Individually those trials did not make it

 

      but if you pooled them you got a significant p

 

      value.  Again, many in the community view that as

 

      evidence of effectiveness of sertraline in

 

      pediatric depression.  This does not meet FDA's

 

      standard but the point is that different folks have

 

      different views of the same data.

 

                Now I want to talk a little bit about the

 

      problem of interpreting negative findings in this

 

      setting.  First I want to turn to adult depression

 

                                                                58

 

      trials for drugs that we believe work.  Llooking at

 

      trials that on face should work, about half the

 

      time those trials fail. If that failure rate can be

 

      extrapolated to the pediatric population, the

 

      expectation in two study programs and most of these

 

      programs were two study programs--the expectation

 

      is that three out of four times you would fail to

 

      get two positive studies.  So, perhaps it shouldn't

 

      have come as such a surprise that many of these

 

      programs failed.  On the other hand, the fact that

 

      the overall success rate, again according to FDA's

 

      standard, is only 20 percent success is clearly a

 

      concern.

 

                Other factors to think about in looking at

 

      negative trials in this setting is, first of all,

 

      the history of antidepressant trials in pediatric

 

      depression.  If you go back to the tricyclic era,

 

      there were 12 trials comparing tricyclics with

 

      placebo in this population.  All of them failed.

 

      There are many interpretations of that.  One, of

 

      course, is that these drugs simply don't work in

 

      that population.

 

                                                                59

 

                Another might be that there is even

 

      greater heterogeneity in this population of

 

      patients captured under the diagnostic criteria for

 

      major depression than we see in adults.  That would

 

      work against getting positive trials.

 

                Another factor to think about is the

 

      somewhat unusual regulatory context in which these

 

      studies were done.  Ordinarily, when companies do

 

      studies they only benefit if they get a positive

 

      trial.  In this setting they would win in terms of

 

      getting exclusivity whether the trial succeeded or

 

      failed.  I don't know whether or not that was a

 

      factor in the conduct of these trials but it is

 

      another thing to think about.

 

                Finally, at the time that we issued

 

      written requests for these programs we were not

 

      routinely asking for phase 2 dose-finding studies

 

      as we are now.  That, again, maybe a factor.  It is

 

      possible that the dose was not right in some of

 

      these trials.

 

                In any case, the bottom line in terms of

 

      efficacy is that I think there are plausible

 

                                                                60

 

      reasons for failure to find efficacy other than the

 

      obvious one that maybe the drugs don't work.  On

 

      the other hand, a very important point I believe is

 

      that even though most of these programs have failed

 

      to meet FDA's standard for approval, this is not

 

      the same thing as saying that we have proof that

 

      the drugs have no benefit.  The drugs may have

 

      benefit that has simply not yet been demonstrated.

 

      On the other hand, the failure to demonstrate

 

      benefit clearly is a concern, especially when we

 

      have a risk, as we have now seen, of emerging

 

      suicidality.  So, the burden is clearly on those

 

      who believe that these drugs do have benefit to

 

      show that benefit.  Tomorrow I am going to talk a

 

      little bit about some possible designs for looking

 

      at longer-term benefits with these drugs.

 

                Now what I would like to do is quickly

 

      move through the questions that we are going to be

 

      asking you to discuss and comment on tomorrow.

 

      Again, we think it is important that you have these

 

      in mind as you hear the presentations this morning.

 

                First of all, we are going to ask you to

 

                                                                61

 

      comment on our approach to classifying the possible

 

      cases of suicidality and our subsequent analyses of

 

      the resulting data for the now 24 trials--again,

 

      the additional trial is the TADS trial.

 

                The question then would be do the

 

      suicidality data from these trials support the

 

      conclusion that any or all of these drugs increase

 

      the risk of suicidality in pediatric patients?  If

 

      the answer to that question is yes, to which of

 

      these nine drugs does that risk apply?  In other

 

      words, is this a class effect of all

 

      antidepressants?  Does it apply to certain

 

      subclasses within this broader class or only to

 

      specific drugs?

 

                If you believe there is a class risk or a

 

      risk that applies only to certain drugs, how should

 

      this information be reflected in the labeling for

 

      each of these products?  What, if any, additional

 

      regulatory actions do you think we need to take?

 

                Finally, again we would like you to

 

      consider what additional research might be needed

 

      to further delineate the risks and the benefits of

 

                                                                62

 

      these drugs in patients with pediatric depression?

 

      Thank you very much.

 

                DR. GOODMAN:  Thank you, Tom.  I imagine

 

      people have questions but I want to try to catch up

 

      this morning to make sure that we have time for all

 

      the presentations.  So, I will ask you to hold your

 

      questions and I would like to invite the next

 

      speaker, Dr. Diane Wysowski, who will be looking at

 

      data from different sources other than clinical

 

      trials.

 

            Recent Observational Studies of Antidepressants

 

                         and Suicidal Behavior

 

                DR. WYSOWSKI:  Good morning.  In this

 

      presentation I will be reviewing recent studies of

 

      antidepressants and suicidal behavior and briefly

 

      discuss their methods, results and limitations.

 

                I reviewed two types of studies,

 

      ecological and patient-level controlled,

 

      observational studies.  Ecological studies show

 

      increasing antidepressant use and simultaneous

 

      decreasing suicide rates.  However, such

 

      correlations do not necessarily imply causality. 

 

                                                                63

 

      Findings of ecological studies can be merely

 

      coincidental.  Numerous factors such as changes in

 

      risk factors, social and economic changes, more

 

      available counseling, changes in gun access and

 

      choice of a less lethal means of suicide in

 

      children and adolescents may coincide with

 

      decreases in the suicide rate in children and

 

      adolescents.

 

                Ecological studies don't show which factor

 

      or factors are responsible for an observed trend.

 

      Furthermore, an increased relative risk of suicide

 

      with antidepressants in children and adolescents

 

      may coexist with a decreased suicide rate.  To

 

      better examine causality, we turned to

 

      patient-level controlled studies, such as

 

      observational studies, in clinical trials.

 

                For the rest of this presentation I will

 

      be focusing on two patient-level controlled,

 

      observational studies.  The first is the Jick study

 

      that was published this July in The Journal of the

 

      American Medical Association.  It is a matched case

 

      control design based on patient prescriptions and

 

                                                                64

 

      diagnoses obtained from the United Kingdom's GPRD,

 

      the General Practice Research Database, for the

 

      period 1993-1999.  The GPRD is a database of

 

      medical records from general practitioners of more

 

      than three million patients in the United Kingdom.

 

                For this study subjects were 10 through 69

 

      years of age.  Exposures studied were the most

 

      widely used antidepressants in the U.K.,

 

      amitriptyline, fluoxetine, paroxetine and

 

      dothiepin.  Dothiepin was chosen as the reference

 

      category.  From data on these antidepressants

 

      users, the investigators identified 555 cases of

 

      nonfatal suicidal behavior, defined as ideation or

 

      attempts.  They identified 17 cases of suicide.

 

      From the base group of antidepressant users, the

 

      investigators matched the cases with more than 2000

 

      controls who did not develop suicidal behavior.

 

      The researchers then compared the suicidal cases to

 

      the non-suicidal controls for initiation of each

 

      antidepressant.

 

                Controlling for age, sex, calendar time

 

      and time from first antidepressant prescription to

 

                                                                65

 

      onset of suicidal behavior, the range of relative

 

      risk for nonfatal suicidal behavior was 0.83 to

 

      1.29 for the antidepressants compared to dothiepin.

 

      None of these risks were statistically significant.

 

      Paroxetine, with a relative risk of 1.29 and a 95

 

      percent confidence interval of 0.97 to 1.7, had

 

      borderline statistical significance.

 

                Similar results were obtained for those

 

      10-19 years old.  No statistically significant

 

      association was found between each antidepressant

 

      and completed suicide.  No statistically

 

      significant association was found between stopping

 

      an antidepressant and nonfatal suicidal behavior.

 

                The relative risk for nonfatal suicidal

 

      behavior and suicide were highest for patients

 

      first prescribed an antidepressant within 1-9 days,

 

      versus 90 days or more, before the suicidal

 

      behavior of the case in the same time period for

 

      the control.  For nonfatal suicidal behavior the

 

      relative risk for antidepressant use within 1-9

 

      days was 4, with a 95 percent confidence interval

 

      of 2.89 to 5.74.  For suicide the relative risk for

 

                                                                66

 

      antidepressant use within 1-9 days was 38, with a

 

      wide confidence interval of 6.2 to 231.

 

                Reviewing the limitations of this study,

 

      the results are only as good as the GPRD data.

 

      There are concerns about possible missing data,

 

      possible incomplete ascertainment and

 

      misclassification of patients, and possible

 

      uncontrolled biases among the antidepressant drugs,

 

      such as selection by severity of depression.  There

 

      were no interviews of cases and controls so

 

      medication compliance is not systematically known.

 

      There was no unexposed group, and the

 

      antidepressant risks are only in reference to the

 

      dothiepin group.

 

                FDA asked Dr. Jick and colleagues to

 

      reanalyze their results with amitriptyline as the

 

      reference category.  They kindly responded to our

 

      request, and asked that their interpretation of the

 

      results be presented verbatim to the committee.  If

 

      the committee wishes to see these supplemental

 

      analyses I will be glad to present them in the

 

      question and answer period.

 

                                                                67

 

                Other limitations of the study include the

 

      fact that suicidal ideation is a more subjective,

 

      softer diagnosis than suicide attempts and the

 

      risks were not examined separately.  This was a

 

      study of mostly adults and there is limited

 

      information on children and adolescents.

 

                Finally, the investigators excluded

 

      patients with a history of 11 other

 

      neuropsychiatric diagnoses, calling into question

 

      the representativeness of the patients compared

 

      with those in clinical practice.

 

                Another patient-level controlled study

 

      examined the relationship between antidepressants

 

      and the risk of suicide attempt by adolescents with

 

      major depressive disorder diagnoses.  The study was

 

      done by investigators at the University of Colorado

 

      School of Pharmacy and Medicine.  Robert Valuck was

 

      the principal investigator.  It was presented as a

 

      poster at the International Society of

 

      Pharmacoeconomics and Outcomes Research meeting,

 

      this past May.

 

                It is a retrospective cohort study of paid

 

                                                                68

 

      medical claims data from the PharmMetrics

 

      Integrated Outcomes Database of 70 managed health

 

      plans for the period 1995 through March, 2003.

 

      Paid claims data include health care provided in

 

      which costs are incurred, such as for

 

      prescriptions, doctor visits, emergency room visits

 

      and hospitalizations.

 

                The investigators identified about 16,000

 

      adolescents aged 12-18 with the first major

 

      depressive disorder diagnosis.  They classified

 

      patients into cohorts by antidepressant

 

      prescription, those who received none over the

 

      entire follow-up period, which was the reference

 

      group, and those who received SSRIs, tricyclic

 

      antidepressants or other antidepressants within 30

 

      days of diagnosis.  They followed the cohorts for

 

      at least 6 months.

 

                The researchers used a Cox proportional

 

      hazards regression analysis to control for some 14

 

      covariates and to examine the multivariate

 

      relationship between antidepressant use and time to

 

      suicide attempt.  The majority of patients, 78

 

                                                                69

 

      percent, had no antidepressant filled in the 6

 

      months after diagnosis; 15 percent had SSRIs filled

 

      within 30 days of diagnosis.  And, 209, 1.3

 

      percent, of the 16,000 patients made at least one

 

      suicide attempt in the follow-up period.

 

                The investigators concluded that

 

      antidepressant treatment with any class of drugs

 

      did not increase the risk of suicide attempt.

 

      Antidepressant use for less than 6 months, compared

 

      with use for 6 months or more, was associated with

 

      a statistically significant 3-fold increased risk

 

      of suicide attempt.  Females, those who received

 

      psychotherapy within 90 days of major depressive

 

      diagnosis, patients with substance abuse,

 

      schizophrenia or another mental health disorder,

 

      patients with more chronic diseases and those in

 

      the Midwest and West were independently at greater

 

      risk of suicide attempt.

 

                Dr. Valuck and co-investigators recently

 

      expanded their study to include 24,000 eligible

 

      patients with a new diagnosed major depressive

 

      disorder.  This expanded study is currently being

 

                                                                70

 

      reviewed for publication.  The researchers added a

 

      propensity matching adjustment to control for

 

      predictors of treatment and to achieve greater

 

      balance among the antidepressant groups.  The

 

      proportion of suicide attempters, 1.4 percent, was

 

      about the same proportion as in their smaller

 

      study.

 

                In this expanded study the hazards ratio

 

      for SSRIs compared to no treatment was 1.58, not

 

      statistically significant.  The hazards ratio for

 

      tricyclics was not estimable due to small numbers

 

      and it was 1 for the other antidepressant category.

 

      The hazards ratio for multiple antidepressants was

 

      1.43, also not statistically significant.  The risk

 

      of suicide attempt declined with longer use of an

 

      antidepressant.  Compared with patients having less

 

      than 8 weeks of use, those with equal to or greater

 

      than 6 months of use had a statistically

 

      significant decline in the risk of suicide attempt.

 

                Concerning the limitations, the results

 

      are only as good as these paid claims data.  There

 

      are concerns about possible missing data, possible

 

                                                                71

 

      incomplete ascertainment and misclassification of

 

      patients, and possible uncontrolled selection

 

      biases by antidepressant group.  There were no

 

      interviews of patients so we don't have

 

      systematically collected information on medication

 

      compliance.  There are no data on the outcomes of

 

      the attempts.  We don't know how many of the

 

      attempters died, and there is no information on

 

      suicides.   Also, there is no information on

 

      individual antidepressants.  There were differences

 

      in study results between the poster and the

 

      expanded study, although this is probably due to

 

      the larger size of the expanded study.

 

                In conclusion, although most of the

 

      results for the individual antidepressants or

 

      classes of antidepressants were not statistically

 

      significantly associated with suicidal behavior, I

 

      do not believe that the Jick and Valuck studies

 

      completely rule out a possible increased risk of

 

      suicidal behavior with antidepressant use.  The

 

      studies reviewed were in agreement in showing that

 

      the risk of suicidal events occurred statistically

 

                                                                72

 

      significantly closer to diagnosis and onset of

 

      antidepressant treatment.  The studies did not

 

      provide data about the characteristics of patients

 

      who did not respond to antidepressants or whose

 

      illness worsened with them.  The studies had actual

 

      or potential methodological limitations.

 

                I conclude that more definitive studies,

 

      perhaps large randomized, controlled trials of

 

      sufficient length, are needed concerning the risk

 

      of suicidal behavior and suicide as related to

 

      antidepressant use in children and adolescents.

 

      With so much at stake, children and adolescents,

 

      their parents and physicians and society in general

 

      deserve to know which therapies and which

 

      individuals work best for treatment of depression.

 

      Thank you.

 

                DR. GOODMAN:  Thank you very much, Diane.

 

      My preference would be that you present those

 

      additional Jick data tomorrow.  I don't think we

 

      have time for it today.  Would that be an agreement

 

      by the committee as well?  So, I think we are in

 

      accord on that, if you could prepare to present

 

                                                                73

 

      that data tomorrow to us.  There is one question,

 

      yes, we will allow that.

 

                DR. PINE:  I am wondering if you could

 

      clarify in the Valuck 24,000 patient study, if you

 

      looked at the association in the less than 8-week

 

      treatment versus no treatment group.  Did that

 

      confidence interval exclude 1?  I mean, I saw that

 

      you gave less than 8 weeks versus prolonged

 

      treatment but I didn't see an odds ratio for less

 

      than 8 weeks versus no treatment.

 

                DR. WYSOWSKI:  I don't think that I have

 

      those data.  I don't think that they did that

 

      analysis.  Their results are still being considered

 

      for publication and we just got an abstract.  You

 

      saw the poster in your package.  Then, when they

 

      did the expanded analysis they only gave us an

 

      abstract of the results.  So, we don't have a lot

 

      of detail on the expanded study.

 

                DR. GOODMAN:  Dr. Fost, you had a question

 

      also?

 

                DR. FOST:  One of the theories of why

 

      suicide behavior might be increased shortly after

 

                                                                74

 

      prescribing is that the patients are at the worst

 

      then and that is why they are started on

 

      prescriptions.  If that were true, one might expect

 

      to see increased suicidal behavior in the week or

 

      two before prescribing.  Do either of these studies

 

      allow for that analysis?

 

                DR. WYSOWSKI:  I believe the Jick study

 

      did look at that.  Actually, no--no, I don't see

 

      any information on that; it is just after.

 

                DR. FOST:  And the data set doesn't allow

 

      itself for that reanalysis?

 

                DR. WYSOWSKI:  Well, it may.  I don't know

 

      whether either investigator has information on

 

      that.

 

                DR. GOODMAN:  I think that was an

 

      excellent question.  Now I would like to invite Dr.

 

      John March, from Duke University, to present

 

      results from the TADS trial.

 

                       Brief Report on TADS Trial

 

                DR. MARCH:  Thanks, it is a pleasure to be

 

      here and I would like to begin the presentation by

 

      thanking the committee for inviting the TADS team

 

                                                                75

 

      to present the TADS data, and also thank the FDA

 

      for the comprehensive and thoughtful way that it is

 

      approaching this question of enormous public health

 

      importance.

 

                The TADS trial, the Treatment for

 

      Adolescents with Depression Study, is an

 

      NIMH-funded comparative treatment trial, and I am

 

      going to present efficacy and safety data from the

 

      stage-1 outcomes that were published in The Journal

 

      of American Medical Association several weeks ago.

 

      We have a detailed safety paper in preparation.  We

 

      have a methods paper which has been published and a

 

      baseline paper which looks at the sample

 

      composition in press, and those of you who are

 

      interested in the TADS trial are referred to these

 

      papers for further information.

 

                As I mentioned, this is a study funded by

 

      the NIMH, coordinated by the Department of

 

      Psychiatry at Duke University and the Duke Clinical

 

      Research Institute, the DCRI.  It has had the

 

      benefit of oversight and consultation from numerous

 

      consultants, a scientific advisory board.  The

 

                                                                76

 

      NIMH, DSMB participants included 12 sites from

 

      around the United States.  Lilly provided

 

      fluoxetine under an independent educational grant

 

      to Duke University, had no input into the design of

 

      the study, the conduct of the study, the analysis

 

      of the data or the preparation of the manuscript.

 

      My sense is that the major credit for this work, as

 

      for all of the research on which we base

 

      evidence-based practice, goes to the children and

 

      families who are willing to participate in

 

      research.  Without their participation, we would

 

      have no evidence at all.

 

                The overall objective of the TADS trial

 

      was to examine the effectiveness of medication and

 

      cognitive behavioral psychotherapy alone and in

 

      combination for the acute and long-term treatment

 

      for adolescents with DSM-IV diagnosis of major

 

      depression.  The design of the trial was a

 

      balanced, randomized, controlled study that was

 

      masked by use of independent evaluators; four

 

      groups, placebo, cognitive behavioral

 

      psychotherapy, fluoxetine and their combination,

 

                                                                77

 

      and the study involved 36 weeks of treatment, of

 

      which I am going to present the stage-1 data for

 

      the first 12 weeks of treatment.  We also have a

 

      year of naturalistic follow-up and we have recently

 

      been funded to follow these youngsters out into

 

      young adulthood.  That data will not be discussed

 

      today.

 

                Now, it is important, in understanding the

 

      generalizability of the data, to know a little bit

 

      about the sample composition.  The inclusion

 

      criteria included outpatients, both boys and girls

 

      age 12-17, with a DSM-IV diagnosis of major

 

      depressive disorder and an IQ greater than or equal

 

      to 80.  Youngsters with severe conduct disorder or

 

      substance abuse, other than nicotine, pervasive

 

      developmental disorders, thought disorder, bipolar

 

      disorder, or history of suicidality or homicidality

 

      were excluded from the trial.

 

                Because suicidality is the question at

 

      issue today, I thought it important to say a little

 

      bit more about these exclusion criteria, kids were

 

      excluded if they had a hospitalization within the

 

                                                                78

 

      previous 3 months or if they were considered to be

 

      high risk, which meant a suicidal event of some

 

      sort within the past 6 months, the presence of

 

      active intent or plan, or if they had suicidal

 

      ideation in the context of a family which was so

 

      disorganized that we felt that even with the

 

      intensive monitoring in the TADS trial framework

 

      that it would not be reasonable to enter them into

 

      a randomized, controlled research study.

 

                This was a moderate to moderately severely

 

      ill population.  We had 439 kids, as you can see,

 

      randomized equally into the 4 groups.  The total

 

      sample on the Children's Depression Rating Scale

 

      had a CDRS scale score of 60.  That, again, is a

 

      mean score of moderate to moderately severely

 

      depressed, with a range from mild to severe

 

      depression.  The T score for that mean score is 75.

 

      That means that these kids were more than 2

 

      standard deviations out from normal with respect to

 

      severity of depression.  The sample was multiply

 

      comorbid, as is characteristic of patients in the

 

      clinical samples.  This was the first major

 

                                                                79

 

      depressive episode for about 90 percent of the

 

      sample.  Ten percent of the sample had had more

 

      than one depressive episode.  The mean duration of

 

      major depression was 42 weeks.  Again, over 50

 

      percent of the sample was comorbid for another

 

      mental disorder, both internalizing and

 

      externalizing disorders; 14 percent of the sample

 

      had ADHD and half of those kids were on concurrent

 

      psychostimulant treatment.

 

                So, unlike the industry-funded trials, the

 

      TADS sample is largely representative of patients

 

      who are treated in clinical practice with major

 

      depressive disorder.  As you might expect, given

 

      the severity of illness and the pattern of

 

      comorbidity, these youngsters suffered significant

 

      functional impairment.  This is the children's CGAS

 

      rating and you can see that the mean CGAS score was

 

      between 40-50, so significant functional impairment

 

      associated with mental illness in this patient

 

      population.

 

                What did we learn in the trial?  This is a

 

      take-home efficacy message.  Four groups, again,

 

                                                                80

 

      began at a CDRS raw score of 60.  These are random

 

      regression analyses looking at the adjusted or

 

      predicted means at baseline, week 6 and week 12 of

 

      treatment.  Actually, all 4 treatments showed

 

      significant improvement, a characteristic of major

 

      depression.  The placebo group and the cognitive

 

      behavioral psychotherapy group were superimposed,

 

      one on top of the other.  There is no additional

 

      benefit from receiving cognitive behavioral

 

      psychotherapy, either in the slope term or at the

 

      end point, over receiving placebo.  There was

 

      significant benefit from fluoxetine alone, and the

 

      largest effect was associated with the combination

 

      condition.  The combination condition beat the CBT

 

      condition and the placebo condition on all 4

 

      efficacy measures.  Fluoxetine beat these CBT on

 

      all 4 measures and placebo and placebo on 3 of the

 

      4 measures.

 

                If we look at the impact of treatment

 

      using effect size calculations, the mean of the

 

      control condition minus the mean of the placebo

 

      condition, divided by the pooled standard

 

                                                                81

 

      deviation, the effect size for the combination was

 

      close to 1.  This is a very large effect.  For

 

      fluoxetine it was around 0.6, a moderate to large

 

      effect.  For CBT there was no difference between

 

      CBT and placebo, effect size calculated relative to

 

      placebo.

 

                If we look at response rates, defined as a

 

      clinical global importance measure rated by the

 

      independent evaluator of much improved or very much

 

      improved, 71 percent of the combination kids

 

      improved; 61 percent of the fluoxetine-treated

 

      patients improved; 43 percent of the cognitive

 

      behavioral psychotherapy treated patients and 35

 

      percent of the placebo-treated patients improved.

 

      Combination statistically was no different than

 

      fluoxetine.  CBT was no different than placebo.

 

      The two drug-containing conditions were superior to

 

      the two non-drug-containing conditions on responder

 

      analysis.

 

                If we look at the effect size calculated

 

      as a derivative of the odds ratio of improvement

 

      for the active treatments relative to placebo for

 

                                                                82

 

      the responder analyses, the results parallel the

 

      analyses on a scale or outcome variable, the

 

      Children's Depression Rating Scale.  The effect

 

      size for combination was 0.8, almost 0.6 for

 

      fluoxetine, and about 0.2 for cognitive behavioral

 

      psychotherapy.  So, again, clear superiority for

 

      the drug-containing conditions, with the largest

 

      effect reserved for the combination of medication

 

      management and CBT.

 

                Now, of interest here are the safety

 

      outcomes from the TADS trial.  Although we spent an

 

      enormous amount of time and energy on measuring

 

      adverse events, particularly measuring the impact

 

      of the treatments on the potential for harm, I

 

      think it is important to point out that the study

 

      did not have safety as a primary outcome.  With 439

 

      subjects randomized to 4 conditions, it is easy to

 

      see that for these outcomes the study is clearly

 

      under-powered.

 

                It is I think important to separate

 

      ideation from behavior.  Despite the exclusion, we

 

      had significant suicidal ideation in the TADS

 

                                                                83

 

      sample.  This is looking at the Reynolds Adolescent

 

      Depression Scale, item 14, and 7.5 percent of the

 

      sample exhibited a score of 4 or above which is the

 

      threshold for clinical investigation on the RADS.

 

      CDRS item 6, serious suicidal ideation, the kind of

 

      suicidal ideation that leads you to consider

 

      hospitalization, 2 percent of the sample met CDRS

 

      item 13 criteria for severe suicidal ideation.  On

 

      the suicidal ideation questionnaire 2 measures, the

 

      SIQ flag which is to prompt clinical investigation,

 

      or elevated scores on any one of the items on the

 

      SIQ that would also prompt suicidal ideation, 29

 

      percent and 10 percent of the sample respectively

 

      on these measures exhibited clinically significant

 

      suicidality.  So, although suicidality was an

 

      exclusion criterion, there was plenty of suicidal

 

      ideation exhibited in the TADS sample at baseline.

 

                Now, as expected, suicidal ideation

 

      occurred across all 4 groups taken in the

 

      aggregate.  One sees here, looking at the CDRS item

 

      13 score, in this case greater than 1, or SIQ

 

      score, SIQ flag greater than 31, significant

 

                                                                84

 

      suicidal ideation at baseline.  It came down at

 

      week 6 and was significantly reduced overall at

 

      week 12.  This is the aggregated data across all 4

 

      treatment groups.

 

                Random regression analyses looking at

 

      between group differences on the SIQ, although it

 

      looks like these groups might be different at

 

      baseline, in fact there were no statistically

 

      significant differences on the SIQ in all 4

 

      treatments.

 

                One sees a different result than we found

 

      in the pattern for depression.  This is placebo;

 

      this is fluoxetine; this is CBT and this is

 

      combination.  The take-home messages here are

 

      three.  First, as we saw in the previous slide,

 

      suicidal ideation improves with treatment

 

      irrespective of which treatment one gets.  Second,

 

      fluoxetine and placebo are indistinguishable with

 

      respect to suicidal ideation, either with respect

 

      to the slope or at entry, indicating that

 

      fluoxetine, at least on average, is not provoking

 

      suicidal ideation.  Finally, the only treatment

 

                                                                85

 

      which separated from placebo with respect to

 

      reducing suicidal ideation was the combination of

 

      fluoxetine and CBT.  So, here the combination

 

      offers a significant advantage over medication

 

      monotherapy.

 

                Moving on to behavior, using a

 

      comprehensive adverse event monitoring procedure,

 

      we looked at the incidence of three kinds of

 

      harm-related adverse events.  Harm-related events,

 

      the broadest category, were defined as harm to

 

      self.  This could involve no suicidal ideation,

 

      ideation or attempt.  Or, harm to others which

 

      required aggressive ideation or actual behavior

 

      involving harming another person or physical

 

      property.  These events are subsumed one within the

 

      next.  So, a suicide-related event, which is a

 

      subset of harm-related events, involves harm to

 

      self, either ideation or attempt.  Then we had

 

      suicide attempts themselves.  What differentiates

 

      harm-related events from suicide-related events is

 

      primarily one subject with aggression and 7

 

      subjects, I believe, who exhibited self-injury

 

                                                                86

 

      without ideation, primarily cutting.

 

                These are the actual rates of harm-related

 

      and suicide-related events divided by treatment

 

      group.  One sees that there is a larger number of

 

      harm-related events and suicide-related events in

 

      fluoxetine-treated kids relative to placebo-treated

 

      kids.  The combination group is intermediate

 

      between harm-related and suicide-related events,

 

      intermediate between fluoxetine and placebo.  The

 

      cognitive behavioral psychotherapy group was

 

      roughly comparable to the placebo group.

 

                If you look at children who received drug,

 

      that is, combining the fluoxetine and the

 

      combination groups and comparing them to the

 

      placebo group, 10 percent, 22 of those

 

      fluoxetine-treated kids exhibited a harm-related

 

      event; 7 percent exhibited a suicide-related event;

 

      and the rates of these events overall were quite

 

      low, 7.5 percent of 439 kids, or 33 kids had a

 

      harm-related event, 24 of 439 kids, or 5.5 percent

 

      exhibited a suicide-related event.

 

                I think it is very important as we move

 

                                                                87

 

      through this discussion to understand that the base

 

      rates of these events are extremely low relative to

 

      the rates that we see for benefit.

 

                If we look at the odds ratios calculated

 

      from the actual rate data, the relative risk is 1.5

 

      for combination, 2 for fluoxetine, less than 1 for

 

      CBT.  Those is calculated relative to placebo.  For

 

      the collapsed category of fluoxetine and

 

      combination the relative risk is slightly greater

 

      than 2.  This is the only statistically significant

 

      relative risk in which the confidence interval

 

      crosses 1.  That is largely because these events

 

      are so rare so the power is quite low to identify

 

      these events.  In fact, the power for detecting a

 

      20 percent difference is about 10 percent.

 

                For suicide-related events there are no

 

      statistically significant differences although, as

 

      you can see from the graph, the odds ratios pretty

 

      closely track the odds ratios for harm-related

 

      events.

 

                The take-home message from this

 

      presentation actually is in this table--no, it is

 

                                                                88

 

      in the next table.  This is the table that looks at

 

      the suicide attempts in the trial.  We had 7 of

 

      them out of 439 kids, or slightly less than 2

 

      percent of the sample,  Two fluoxetine-treated

 

      kids, 4 combination-treated kids, 1 CBT and no

 

      placebo-treated patients made a suicide attempt.

 

      There probably is an imbalance in randomization

 

      which may in part be responsible for this.  There

 

      were more kids with an elevated SIQ flag randomized

 

      to the drug-containing than the non-drug-containing

 

      conditions so it is not clear what to make of this

 

      data.

 

                Here I think is where the take-home

 

      message lies relative to safety.  This is looking

 

      at the benefit/risk ratio using analyses for the

 

      number needed to treat and number needed to harm.

 

      What we see here is fluoxetine compared to placebo,

 

      in the first column; combination compared to

 

      placebo; and the collapsed category, SSRI versus no

 

      SSRI.  The absolute benefit increase is calculated

 

      as the control, the experimental event rate minus

 

      the control event rate.  So, it is the absolute

 

                                                                89

 

      benefit increase for receiving the treatment.  The

 

      absolute risk increase is calculated, again, as the

 

      experimental event rate minus the control event

 

      rate, that is, the risk increase attributable to

 

      the treatment over the placebo condition in

 

      absolute numbers.

 

                The NNT and the NNH are the reciprocal of

 

      the absolute benefit increase and the absolute risk

 

      increase respectively, 1 over the absolute benefit

 

      increase or 1 over the absolute risk increase.

 

      These are defined as the number of patients for

 

      benefit that would need to be treated to find one

 

      patient who had benefit over the benefit occurring

 

      from the control condition or, in the case of the

 

      NNH, the number of patients who would need to be

 

      treated to find one patient who would be harmed

 

      over treatment with the control condition.  So, it

 

      is a nice metric that combines both the absolute

 

      rate and the magnitude of the effect.

 

                One sees clearly here that 27 percent of

 

      patients benefit from treatment with fluoxetine,

 

      with an NNT of 4 which is a large effect; 27

 

                                                                90

 

      percent of patients benefit from treatment with

 

      combination, with an NNT of 3, again a very large

 

      effect; for SSRI versus no SSRI, combining the

 

      categories, 31 percent of patients benefit, again

 

      with an NNT of 3.

 

                The absolute risk increase for a suicidal

 

      event with respect to fluoxetine is 4.7 percent as

 

      compared to placebo; 2 percent for combination over

 

      placebo; and 3 percent for the combined category.

 

      NNH is number of patients that you would need to

 

      treat with the active treatment to find one patient

 

      who would be harmed over the control condition of

 

      21, 50 for the combination condition and 4 for the

 

      collapsed categories.

 

                From a clinical point of view, these

 

      patients would be easy to pick out in a crowd,

 

      easily identifiable who is getting better and who

 

      is not getting better, active treatment versus

 

      control.  These effects are so small and so

 

      uncommon that one could not possibly pick out

 

      patients who would be harmed by the medication

 

      versus patients who would commit these

 

                                                                91

 

      suicide-related event behaviors with placebo

 

      treatment.  If you calculate the NNT to NNH ratio

 

      looking at benefit to risk, one sees clearly here

 

      that the benefit tilts in favor of the treatment,

 

      and particularly the combination treatment.

 

                So, we conclude that the combination of

 

      fluoxetine and CBT is the most effective treatment

 

      for adolescents with major depression.  Fluoxetine

 

      alone is effective but not as effective as the

 

      combination of the two treatments.  CBT alone is

 

      less effective than fluoxetine and not

 

      significantly more effective than placebo.  We also

 

      conclude that placebo is acceptable in randomized,

 

      controlled trials of adolescent major depression

 

      and, in fact, is essential for looking at the

 

      adverse event outcomes, at least in this study.

 

      Suicidality decreases substantially with treatment.

 

      The improvement in suicidality is greatest with the

 

      combination and least for fluoxetine alone.

 

      Fluoxetine does not increase suicidal ideation.

 

      Suicide-related adverse events which are uncommon

 

      may occur more often in fluoxetine-treated

 

                                                                92

 

      patients.  CBT may protect against suicide-related

 

      adverse events in fluoxetine-treated patients.

 

      Taking both risk and benefit into account, the

 

      combination of fluoxetine and CBT appears superior

 

      as a short-term treatment for major depression in

 

      adolescents.

 

                Now, the most practical clinical trialist,

 

      the kind of trial models that are used in other

 

      areas of medicine--cardiology, oncology, infectious

 

      disease for example, would much prefer a large

 

      simple or practical clinical trial in 2000 subjects

 

      to a meta-analysis of 10 under-powered subject

 

      trials.  So, it is our sense from looking at the

 

      FDA data and also the TADS data that we have a

 

      significant signal for drug treatment relative to

 

      suicidality but the evidence is not conclusive.  In

 

      fact, a definitive study has not been done and we

 

      would, as a field and as consumers of this

 

      information, much benefit from a

 

      placebo-controlled, practical clinical trial

 

      comparing fluoxetine to another SSRI, perhaps

 

      sertraline or citalopram.  This trial could be run

 

                                                                93

 

      easily on the child and adolescent psychiatry

 

      trials network, which is a clinical trials network

 

      that we are now putting in place to run these kinds

 

      of trials in the pediatric population.  Thank you.

 

                DR. GOODMAN:  Thank you, John.  My first

 

      question is will you be here tomorrow?

 

                DR. MARCH:  No.

 

                DR. GOODMAN:  Oh, you won't?  That may

 

      affect my subsequent questions because I am sure,

 

      besides myself, there will be a number of questions

 

      for you.  I don't know if we have time to take them

 

      all right now.  I wonder if there is any other

 

      option, Anuja.  What time do you leave today, Dr.

 

      March?

 

                DR. MARCH:  Noon.

 

                   Committee Discussion on TADS Trial

 

                DR. GOODMAN:  I am going to ask a

 

      question.  My understanding in looking at your

 

      results is that on the categorical measures of

 

      response fluoxetine is superior to placebo.

 

      However, if you look at a comparison of the mean

 

      scores on the CDRS fluoxetine is not superior to

 

                                                                94

 

      placebo.  Is that correct?

 

                DR. MARCH:  The slope term on the random

 

      regression analysis for the CDRS, the p value was

 

      0.08 for fluoxetine versus placebo.  Fluoxetine was

 

      statistically significantly different than CBT but

 

      not placebo.

 

                DR. GOODMAN:  So, from a standpoint of

 

      FDA, if this trial had been submitted to the FDA

 

      and you didn't have the CBT group, it seems to me

 

      that it might be classified as a negative study.

 

                DR. MARCH:  It would have been classified

 

      as a negative study using the CDRS as the primary

 

      endpoint, the slope term.

 

                DR. GOODMAN:  Do you have any comments

 

      about the methodology or outcome measures we are

 

      using and whether we are using the most appropriate

 

      ones in your opinion?

 

                DR. MARCH:  Well, I think it is actually a

 

      very important question.  If you look at the

 

      fluoxetine outcomes on the predicted endpoint, the

 

      week 12 endpoint on the CDRS predicted by the CDRS

 

      slope, on the clinical global improvement measure

 

                                                                95

 

      dichotomized and on the Reynolds Adolescent

 

      Depression Scale fluoxetine was statistically

 

      better than placebo.  It was simply a near miss on

 

      the slope term, which probably relates to the way

 

      the random regression analyses handle standard

 

      errors.  So, my sense of the story that the data is

 

      telling us is that fluoxetine--and also if you look

 

      at the effect size calculations--the story the data

 

      is telling is that fluoxetine is an effective

 

      treatment and it would be a mistake to consider

 

      this a negative trial.  On the other hand, the

 

      technical definition used by the FDA would require

 

      that the study be considered negative.

 

                DR. GOODMAN:  Dr. Perrin?

 

                DR. PERRIN:  On the other side of the

 

      equation, you made comments that the sample was

 

      somewhat different from some of the trials that

 

      have been sponsored by industry.  Could you be a

 

      little bit more specific about what the differences

 

      are, and how they might have affected the results,

 

      and what are the implications for meta-analyses of

 

      these studies?

 

                                                                96

 

                DR. MARCH:  I think it is a very important

 

      question, and we have a paper that is in press in

 

      The Journal of Child Medicine Psychiatry, the

 

      "orange journal," which describes the TADS sample

 

      in some detail and compares the TADS sample to

 

      epidemiologic samples and to treatment samples,

 

      both on the pharmacotherapy side, primarily the

 

      industry data sets, and also the cognitive

 

      behavioral psychotherapy trials, of which there are

 

      13 published at this point.  In general, our sample

 

      is not substantially different from either the

 

      epidemiologic or the treatment seeking samples,

 

      with the caveat that we are slightly sicker and

 

      slightly more comorbid, particularly relative to

 

      the CBT samples.

 

                So, given that the range of depression

 

      goes from mild to severe and half the sample is

 

      comorbid, there are plenty of patients in the data

 

      set who resemble the mildly ill patient all the way

 

      up to the severely ill, multiply comorbid patients.

 

      So, I think the result of the TADS trial is

 

      generalizable to the total sample.

 

                                                                97

 

                With respect to meta-analyses, it would be

 

      better to have a very large sample including all

 

      these variations, but by combining the data sets

 

      one gets a better picture, I believe, of the total

 

      variation in the patient population than simply

 

      using the industry data sets which tend to exclude

 

      the more complicated and clinically ill patients.

 

                DR. GOODMAN:  Dr. Newman, did you have a

 

      question?

 

                DR. NEWMAN:  Normally when you use a

 

      continuous outcome you have greater power than when

 

      you dichotomize.  I assume that is why you

 

      specified that as the endpoint at the beginning of

 

      the trial.  A reason why you might not is that the

 

      medication helps maybe a majority but actually

 

      harms a minority and then you could actually see

 

      that if you dichotomize the percent to improve is

 

      statistically significantly greater in the

 

      fluoxetine group but the mean may not be

 

      significantly greater because there are some people

 

      who are harmed and they drag the mean down, whereas

 

      they have no effect on the dichotomized variable. 

 

                                                                98

 

      Did you look to see whether there was evidence that

 

      the variation in the standard deviation in the

 

      effect size differed between the two groups and

 

      might have been greater with fluoxetine?

 

                DR. MARCH:  That was actually the point

 

      that I made, that the standard errors are larger in

 

      the fluoxetine-treated group than in the combo or

 

      the placebo--

 

                DR. NEWMAN:  Actually, not just the

 

      standard errors but the standard deviation, meaning

 

      that, in fact, there are some people who are harmed

 

      by it and that diminishes the apparent benefit when

 

      you average.

 

                DR. MARCH:  It would be impossible to look

 

      at the standard errors or the standard deviation

 

      and make a judgment about harm because there is a

 

      fair amount of variability data point to data point

 

      which is intrinsic to the disorder.  Some patients

 

      get better; some patients deteriorate.

 

                We do have in the safety paper a whole set

 

      of analyses looking at shifts, which I am not

 

      confident enough in to have wanted to present

 

                                                                99

 

      today.  We will try to examine what percentage of

 

      patients are getting worse with respect to ideation

 

      and behavior, and how that relates to treatment

 

      classification and also how it relates to other

 

      adverse events like mania activation, anxiety

 

      disinhibition and so on.  I think the secondary

 

      paper is going to shed a fair bit of light on these

 

      questions.

 

                DR. GOODMAN:  Dr. Pfeffer?

 

                DR. PFEFFER:  I have two questions.  One

 

      is were there any differential dropout rates in the

 

      samples?  This also relates to the question of

 

      compliance in the different treatments.  My last

 

      question is these were intent-to-treat analyses?

 

                DR MARCH:  All very good questions.  The

 

      analyses are all intent-to-treat.  Although I

 

      didn't present the data, if we look at observed

 

      cases analyses, those who were still on their

 

      assigned arm at any given assessment point or

 

      completer analysis, the results are exactly the

 

      same.  About 10 percent of the kids in each

 

      treatment overall dropped before the week-12 data

 

                                                               100

 

      point.  Another 10 percent were what we call

 

      prematurely terminated.  That is, for ethical

 

      reasons.  They received an out of protocol

 

      treatment at some point during the first 12 weeks

 

      of the study.  There were no statistically

 

      significant differences across treatment groups in

 

      either the rate of dropping out or the rate of

 

      receiving an ancillary treatment, that is, a

 

      premature termination.

 

                You will see this afternoon when FDA

 

      presents its analysis of the TADS data that the

 

      odds ratios for being harmed by receiving

 

      fluoxetine are greater than we presented, or I

 

      presented this morning on behalf of the TADS team.

 

      That is because the FDA data set excluded those

 

      kids who were prematurely terminated and received

 

      another treatment.  That actually represented two

 

      kids in the placebo group and that, in turn,

 

      inflated the odds ratios in the FDA results versus

 

      the TADS results.  So, there is some method

 

      variance in there which accounts for the

 

      differences between the two findings.

 

                                                               101

 

                DR. GOODMAN:  Dr. Gorman?

 

                DR. GORMAN:  Does analysis of your data

 

      set allow interpretation of the time of onset of

 

      treatment to behaviors that we are studying today?

 

                DR. MARCH:  That is a very good question

 

      and, in fact, one that we will address in the

 

      secondary paper.  I can tell you that the majority

 

      of the events occurred within the first 6 weeks but

 

      not within the first 2 weeks.  But I don't have

 

      that data presented in slide form so I can show it

 

      to you, but it will be in the safety paper that we

 

      are currently preparing.

 

                DR. GOODMAN:  So you don't know what the

 

      differential rates are between the groups at this

 

      point, particularly between fluoxetine and placebo?

 

                DR. MARCH:  In terms of time?

 

                DR. GOODMAN:  Yes, in terms of the early

 

      events.

 

                DR. MARCH:  My general impression, looking

 

      at the data, is that the fluoxetine events occurred

 

      early and the placebo events occurred later, which

 

      is kind of what you would expect given what we know

 

                                                               102

 

      about the compounds.  But I wouldn't want to cite

 

      chapter and verse or have you base your decisions

 

      based on that because we haven't completed the

 

      final analyses of the data.

 

                DR. GOODMAN:  Dr. Rudorfer, you will be

 

      the penultimate questioner.

 

                DR. RUDORFER:  If I can go back to the

 

      characteristics of the patients for a moment, we

 

      will be looking at a number of studies that were

 

      submitted to the FDA by various sponsors, and it

 

      seems to me that the TADS inclusion criteria go

 

      beyond DSM-IV in terms of length of illness and

 

      degree of dysfunction in various spheres of life.

 

      Could you comment on that?

 

                DR. MARCH:  Sure, it is a very good

 

      question.  That is, the exclusion criteria included

 

      requirements that were designed to ensure a stable

 

      baseline.  So, we required at least six weeks of

 

      mood disorder symptoms that crossed two or three

 

      contexts--home, peers in school--which, of course,

 

      are not required in the DSM-IV criteria.  This was

 

      done in part to minimize the chance of a placebo

 

                                                               103

 

      response and to ensure that we had a sick patient

 

      population that would be both ethical to randomize

 

      and would offer some opportunity for the

 

      combination treatment to separate from the two

 

      monotherapy conditions.

 

                I think we actually designed a very good

 

      experiment, and looking at a 35 percent placebo

 

      response rate did exactly what we had intended to

 

      do.  But in that sense, this population is sicker

 

      perhaps than what is seen in the industry data sets

 

      and certainly sicker than what is seen in the CBT

 

      data sets on average.

 

                DR. GOODMAN:  I will permit two final

 

      questions and that is it, one from Dr. Pine and

 

      then Dr. Fant.

 

                DR. PINE:  I want to return to the first

 

      question from Dr. Goodman as far as how this study

 

      would be evaluated from an FDA perspective.  My

 

      sense from reading the paper is that there were two

 

      primary outcome measures and three analyses, and

 

      two of the three were positive so that the CGI

 

      analysis done categorically was positive, the CDRS

 

                                                               104

 

      analysis done categorically was positive, and it

 

      was only the third analysis, the CDRS continuous

 

      measure, that was not positive.

 

                So, my take on that from an FDA standpoint

 

      of, you know, do the primary outcome measures make

 

      it or not is that it would be closer to positive

 

      than negative.  Do I have that right?

 

                DR. MARCH:  You have it partially right.

 

      There is a CDRS slope analysis, and whether that is

 

      positive or negative depends on whether you treat

 

      the intercept term as random.  I mean, there is a

 

      fair amount of method variance on a subtle level

 

      that can tilt these things one way or the other

 

      when it is a near miss.  There is a CDRS endpoint

 

      analysis based on the predicted or marginal mean.

 

      There is a categorical analysis, logistic

 

      regression, and there is a self-report scale which

 

      was included, the Reynolds Adolescent Depression

 

      Scale, which was included because the CBT

 

      literature relies heavily on patient self-report.

 

      Three of those four measures, all but the CDRS

 

      slope analysis, were positive for the

 

                                                               105

 

      fluoxetine-placebo comparison.  The CDRS slope

 

      analysis, again, was a p value of 0.08, a near

 

      miss.

 

                So, I think the take-home message is

 

      actually in the effect sizes, not in whether you

 

      are looking at p values or not.  Clearly,

 

      combination and fluoxetine have larger effect

 

      sizes, meaningful effect sizes relative to placebo

 

      as contrasted to CBT.

 

                DR. FANT:  For the sake of the study I

 

      know it was necessary to exclude certain patients

 

      to optimize the conditions for the study, but I

 

      think in real-world practice a lot of the patients

 

      who were excluded would be patients who would be

 

      prescribed medication under various conditions.  Is

 

      there any reason, from your standpoint or

 

      perspective, to think that that population of

 

      patients may be at a different risk for fulfilling

 

      suicidal attempts, ideation, and carrying it

 

      through to the ultimate result, or may be affected

 

      differently by the medication than patients that

 

      are not excluded from the study?

 

                                                               106

 

                DR. MARCH:  That is a very good question.

 

      That is, is there some issue to believe that there

 

      would be a differential treatment response in

 

      patients who would be excluded, particularly

 

      excluded for harm to self or others, as compared to

 

      the TADS sample of patients?  I don't know of any a

 

      priori reason to believe that there would be a

 

      differential treatment effect relative to the TADS

 

      sample.  I do think it is quite clear from the

 

      treatment and epidemiologic literature that they

 

      would be at higher risk for adverse harm-related

 

      outcomes but whether they would be more at risk

 

      than, say, the TADS sample patient I don't think we

 

      know.  My guess as a clinician is probably not.

 

                We are actually doing an NIMH-funded study

 

      called the Treatment of Adolescent Suicide

 

      Attempter Study, in which we are comparing a

 

      medication algorithm to cognitive behavioral

 

      psychotherapy to the combination--no untreated

 

      control condition obviously in this sample--to try

 

      to understand something about treatment for this

 

      particular patient population precisely because

 

                                                               107

 

      they have been excluded from these other trials,

 

      and we need additional data on their care.  So,

 

      that trial is now under way and should be completed

 

      in the next couple of years.

 

                DR. GOODMAN:  Thank you very much, John.

 

      The final question that I will take the chairman's

 

      privilege to ask is do you have data that you

 

      haven't presented yet on long-term outcome based on

 

      this trial?

 

                DR. MARCH:  The final subjects in the

 

      trial are out in a naturalistic follow-up window so

 

      that 36-week data is in the can, but that data set

 

      has not been cleaned and locked yet.  We expect it

 

      will be cleaned and locked and ready for analysis

 

      in the spring, and we hope to have that data in

 

      press by this time next year.

 

                DR. GOODMAN:  Thank you very much, John.

 

                DR. MARCH:  Thank you.

 

                DR. GOODMAN:  I would like to ask our next

 

      speaker to come forward, Dr. Greg Dubitsky.

 

           Characteristics of Pediatric Antidepressant Trials

 

                DR. DUBITSKY:  Good morning.  You just

 

                                                               108

 

      heard about the TADS trial from Dr. March.  I would

 

      like to now go on and briefly summarize the other

 

      studies that were included in the FDA's primary

 

      analysis of suicidality.

 

                I do want to emphasize that my review and

 

      this presentation are really descriptive only.  I

 

      am not going to touch on efficacy outcomes or

 

      safety outcomes.  The discussion of the risk of

 

      suicidality in these trials with be presented later

 

      this morning by Dr. Hammad.

 

                The study pool, again excluding the TADS

 

      study, consisted of 23 placebo-controlled studies

 

      which were conducted between 1984 and 2001.  Each

 

      study was done with one of nine different

 

      antidepressant drugs and studied patients with one

 

      of five different diagnostic indications; major

 

      depression, obsessive compulsive disorder,

 

      generalized anxiety disorder, social anxiety

 

      disorder or attention deficit disorder.

 

                These studies all had some features in

 

      common.  They were all randomized, double-blind,

 

      placebo-controlled.  They utilized a parallel group

 

                                                               109

 

      design and a flexible dosing regimen.

 

                I did prepare a handout to go with this

 

      talk which should be in your packets, at least for

 

      the advisory committee members.  It consists of two

 

      tables which summarize some of the design

 

      characteristics of these 23 studies.  Table 1 has

 

      some basic study information, to include the

 

      diagnostic indication, the age range that was

 

      studied, number of patients by treatment group, the

 

      duration of double-blind treatment, and the dose

 

      range that was used in the particular study.

 

                I would like to point out though that I

 

      don't intend for everybody to read this and

 

      memorize it; this is really for reference for later

 

      this morning when you hear about the analysis of

 

      these trials.

 

                The second table in the handout includes

 

      some information on screening and exclusionary

 

      criteria.  Some of the studies used very extensive

 

      diagnostic screening.  I have indicated those in

 

      the table.  I have also indicated information on

 

      whether there was a placebo lead-in, and also

 

                                                               110

 

      whether certain exclusionary criteria were employed

 

      in the various studies to include whether people

 

      were excluded who had a history of treatment

 

      resistance, current suicide risk, history of a

 

      suicide attempt, bipolar disorder, or family

 

      history of bipolar disorder.

 

                My review of these studies did include a

 

      number of other variables.  I have listed in these

 

      two tables the most relevant ones but in my review

 

      that is on the Internet I do describe some other

 

      characteristics which you might be interested in,

 

      such as the location and number of sites, whether

 

      stratified randomization by age group was utilized

 

      and other exclusionary criteria such as homicidal

 

      risk or the presence of psychotic symptoms.

 

                There were a few notable differences

 

      between these studies that I would like to point

 

      out.  One study with Prozac, HCCJ, was a very small

 

      study.  It was the smallest of the 23 studies, with

 

      only about 40 patients and it was terminated early.

 

                Only one of the 23 studies included an

 

      active control arm.  That was study 329 with Paxil

 

                                                               111

 

      in major depressive disorder.  That included an

 

      imipramine control arm.  The others only had a

 

      placebo control.

 

                Two of the studies did include inpatients

 

      as well as outpatients, the Celexa study, 94404,

 

      and Wellbutrin, 75.

 

                Last, I did want to point out that three

 

      of the studies did use a rather extensive

 

      diagnostic screening of the patients, much more so

 

      than the other studies, Prozac studies X065 and

 

      HCJE, and Paxil study 329.  Those three studies

 

      were done in major depressive disorder.

 

                One other difference involves the

 

      treatment options after patients completed the

 

      acute phase of double-blind treatment.  This was

 

      quite variable across the trials.  In eight studies

 

      there was a taper of acute treatment before

 

      discontinuation.  Seven other trials just abruptly

 

      discontinued treatment, and there was no provision

 

      for continued treatment.  Five trials did allow for

 

      continuation of open-label treatment, and in three

 

      trials patients could continue double-blind

 

                                                               112

 

      treatment.

 

                However, this was also very variable

 

      within trials.  For instance, in Paxil 329

 

      responders could continue double-blind treatment

 

      but non-responders were tapered off treatment.

 

      This variability in the follow-up treatment

 

      following the acute phase made it very difficult to

 

      do any analysis of suicidality-related events post

 

      double-blind treatment.

 

                I would like to point out that none of

 

      these studies was specifically designed to assess

 

      suicidality.  Suicide attempts and ideation were

 

      detected only through routine safety monitoring,

 

      that is, through treatment emergent adverse events

 

      and through suicide-related items on various

 

      depression scales, such as the HAM-D and the CDRS.

 

      One problem with this is that often descriptions of

 

      possibly suicide-related events were rather vague

 

      or incomplete and often made it difficult to reach

 

      a classification.

 

                I have no specific conclusions since this

 

      is really a descriptive review and overview of the

 

                                                               113

 

      studies.  I think one of the important questions

 

      that arises from this information though is whether

 

      any of these differences in design characteristics

 

      could contribute to any observed differences in

 

      suicidality risk that we observed across these

 

      studies.  That is a question that will be addressed

 

      later this morning by my colleague, Dr. Hammad.

 

      So, that is all I have.

 

                DR. GOODMAN:  Thank you for being concise

 

      and providing us with an outstanding handout for

 

      our reference.  We have one question.  Dr.

 

      Rudorfer?

 

                DR. RUDORFER:  Thank you.  Could you

 

      clarify, of the 23 trials how many were submitted

 

      in response to the pediatric exclusivity rule?

 

                DR. DUBITSKY:  I don't have the exact

 

      number.  I believe most of them were but some of

 

      them were submitted well before pediatric

 

      exclusivity took place or came into effect.  I

 

      don't have the exact number off the top of my head.

 

                DR. GOODMAN:  Dr. O'Fallon?

 

                DR. O'FALLON:  Asking the question in a

 

                                                               114

 

      somewhat different way, the data that you have for

 

      this reanalysis, does any of that data come from

 

      outside, beyond the data that was submitted to the

 

      FDA?  That is, were you able to go in and obtain

 

      data from studies that were never submitted to the

 

      FDA for approval or whatever?

 

                DR. DUBITSKY:  Well, to my knowledge,

 

      there was one study that had not been submitted as

 

      part of an efficacy supplement or an approval

 

      package.  The other ones, I believe, were.  Dr.

 

      Hammad actually requested data sets for all these

 

      studies.  Correct me if I am wrong, but I believe

 

      we had relatively complete data sets to allow

 

      reasonable analysis for all these studies.

 

                DR. O'FALLON:  But I am asking whether

 

      there are, as some are claiming, studies that were

 

      done but were never submitted to the FDA.  Are

 

      there any of those data here, if they exist?

 

                DR. DUBITSKY:  There are some studies that

 

      are not included in this analysis, but those are

 

      mainly open-label continuation studies of the acute

 

      studies.  Also, there were a number of pediatric

 

                                                               115

 

      pharmacokinetic studies but I think for obvious

 

      reasons we didn't include those in the analysis.

 

      But, to my knowledge, I think we have everything.

 

                DR. GOODMAN:  Dr. Laughren, do you also

 

      want to respond to the question?

 

                DR. LAUGHREN:  I think I can respond to

 

      that.  The vast majority of these programs were

 

      submitted under pediatric exclusivity so the

 

      companies were required to submit every scrap of

 

      data they had as part of those supplements.  The

 

      only trial here that was not submitted as part of

 

      an application, in terms of a company trial, was

 

      the ADHD study for Wellbutrin.  The other study

 

      that we have included safety data for is the TADS

 

      trial and, of course, that was also independent.

 

      But those are the only two trials of the 24 that we

 

      looked at that were not submitted as part of an

 

      application.

 

                DR. GOODMAN:  Dr. Marangell?

 

                DR. MARANGELL:  How many of the studies

 

      excluded family history of bipolar disorder?

 

                DR. DUBITSKY:  Let's see, actually I think

 

                                                               116

 

      I have that in table 2.  I don't know the number

 

      off the top of my head.  It looks like about ten of

 

      the studies excluded a family history of bipolar

 

      disorder.

 

                DR. MARANGELL:  Thank you.

 

                DR. GOODMAN:  Dr. Perrin?

 

                DR. PERRIN:  You are saying basically that

 

      the extensive diagnostic screening occurred only in

 

      three studies, I believe.  Is that right?

 

                DR. DUBITSKY:  I am sorry?

 

                DR. PERRIN:  The extensive diagnostic

 

      screening occurred only I think in three

 

      studies--one of the points that you made.  Does

 

      that give us some information about the potential

 

      diagnostic heterogeneity and also raise questions

 

      about whether entrance into these studies of

 

      children, ages 7-17, might not have had MDD in the

 

      MDD studies?  My last related question is, since I

 

      am not a psychiatrist at all, what do we know about

 

      the ability to distinguish bipolar disorder from

 

      MDD in the 7-17 year-olds?

 

                DR. DUBITSKY:  Well, it is true that in

 

                                                               117

 

      looking across all 23 studies, those three studies

 

      did stand out as far as using more extensive

 

      diagnostic criteria.  I believe that it certainly

 

      is possible that we might have more confidence that

 

      those patients did actually have the diagnosis

 

      under consideration.  Whether that is actually true

 

      or not, I don't know and I don't know any good way

 

      of figuring that out.

 

                I am not a child psychiatrist so I can't

 

      answer your last question about the ability to

 

      diagnose.  I understand it is very tricky though.

 

                DR. GOODMAN:  Thank you again.  I would

 

      like to ask Dr. Kelly Posner to come up to the

 

      podium to present.  Dr. Posner is from Columbia

 

      University and she will be talking to us about the

 

      reclassification of the clinical trials data

 

      according to suicidality.

 

                  Classification of Suicidality Events

 

                DR. POSNER:  I would like to start by

 

      introducing my expert work group from Columbia that

 

      included myself, Dr. Maria Oquendo, Dr. Barbara

 

      Stanley and Dr. Madelyn Gould.  Dr. Stanley and Dr.

 

                                                               118

 

      Gould are here with me today.  Our statistical

 

      consultant was Mark Davies.

 

                Why was reclassification needed?  The

 

      problem is that the field is challenged by a lack

 

      of well-defined terminology and common language to

 

      refer to suicidal behavior, and this was reflected

 

      in the lack of standardized language used in the 25

 

      trials in question.  That is why there was

 

      difficulty in interpreting the meaning of all of

 

      these reported adverse events that occurred in

 

      these trials.  So, AEs that should have been called

 

      suicidal may have been missed and there may have

 

      been AEs that were inappropriately classified as

 

      suicidal.

 

                Here are some illustrative examples of the

 

      difficulties in adverse event labeling in the

 

      field.  I want to make sure to note that these

 

      labels have nothing to do with the labels the

 

      sponsor gave these events, but just original

 

      investigators at the site.  Again, they are extreme

 

      examples just to reflect the problem.

 

                You see the first one, it says patient

 

                                                               119

 

      attempted to hang himself with a rope after a

 

      dispute with his father.  Investigator did not

 

      consider this event to be a suicide attempt but

 

      called it a personality disorder in this 10

 

      year-old patient.

 

                The second one is one we have all heard a

 

      lot about.  The patient is reported to have engaged

 

      in an episode of auto-mutilation where she slapped

 

      herself in the face, called a suicide event.  Then,

 

      the patient took 11 tablets impulsively then went

 

      to school--called a medication error.

 

                So, how do we address this problem?  Well,

 

      a common set of guidelines needed to be applied and

 

      we needed to look at the data consistently across

 

      trials using research-supported definitions and

 

      concepts that had reliability and validity.  We

 

      also needed to broaden the range of adverse events

 

      that we were looking at.  This was for two reasons.

 

      The first one is to avoid bias in readings.  We

 

      wouldn't have wanted the expert raters only to have

 

      had what the sponsors had identified as possibly

 

      suicidal.  Also, to identify suicidal events that

 

                                                               120

 

      may have been missed.

 

                So, what was included in this broadened

 

      range of events?  Of course, the events originally

 

      identified by the sponsors as possibly suicide

 

      related, all accidental injuries which included

 

      accidental overdoses, and serious adverse events

 

      which includes life-threatening events and all

 

      hospitalizations.

 

                Why did we need experts in suicide?  Well,

 

      you all heard about the limited information

 

      provided in the narratives, particularly frequent

 

      lack of stated suicidal intent.  So, only experts

 

      in suicide would have allowed for inference based

 

      on details of behaviors and related clinical

 

      information.

 

                This is the list of our very distinguished

 

      international panel of experts.  I will just read

 

      their names very quickly, Drs. Bautrais, Brent,

 

      Brown, Van Herringen, King, Mazark, O'Carroll,

 

      Rudd, Spirido and Miller.

 

                So, what was the Columbia classification?

 

      I want to move to this slide because it goes

 

                                                               121

 

      through the definitions which I will just go

 

      through briefly.  Suicide attempt, of course, which

 

      is defined as a self-injurious behavior associated

 

      with some intent to die.  Intent can be stated or

 

      inferred by the rater.  It is important to know

 

      that no injury is needed.

 

                Then there was preparatory actions towards

 

      imminent suicidal behavior.  So, the person takes

 

      steps to injure himself but is stopped by self or

 

      other, anything beyond the threshold of a

 

      verbalization but not quite making it to a suicide

 

      attempt.

 

                Then we had self-injury behavior, intent

 

      unknown.  These are cases where we know there was

 

      some self-injury but we don't know what the intent

 

      was.  So, the associated intent to die is unclear

 

      and cannot be inferred.

 

                Self-injurious behavior with no suicidal

 

      intent is the next category.  That is where, again,

 

      we know there was deliberate self-harm but there

 

      was no intent to die so behavior is intended to

 

      affect other things.  This is what we think of

 

                                                               122

 

      self-mutilation typically.

 

                Suicidal ideation was the next relevant

 

      category, which can be passive or active thoughts,

 

      passive thoughts of wanting to be dead or active

 

      thoughts about killing oneself.

 

                Then we had all the other categories.

 

      That is essentially one rating, anything other than

 

      deliberate self-harm or something suicidal.  That

 

      could include accidents, psychiatric events or

 

      medical events.

 

                Finally, we had not enough information,

 

      which meant that there was insufficient information

 

      for a rater to be able to say whether or not there

 

      was some deliberate self-harm or something

 

      suicidal.

 

                The scheme is laid out conceptually here

 

      for you.  I think it helps make a little more sense

 

      of it.  The blue boxes refer to what you will hear

 

      later as the FDA's primary outcome.  These are

 

      ratings that are considered definitively suicidal,

 

      suicidal behavior and suicidal ideation.  You see

 

      codes 1, 2 and 6.  Suicide attempt, preparatory

 

                                                               123

 

      actions and ideation.  The next are non-suicidal

 

      events, all the other events and the self-injurious

 

      behavior without suicidal intent, and then

 

      indeterminants.  The green boxes are what will be

 

      referred to as the sensitivity analysis, things

 

      that could have been suicidal but there is no way

 

      to know.

 

                So, what was done?  The classification

 

      methodology involved, of course, choosing the

 

      expert panel who had expertise in adolescent

 

      suicide and suicide assessment, based on reputation

 

      and publications.  They had no involvement in

 

      industry youth depression trials in question, and

 

      no expert rater was employed by Columbia

 

      University.

 

                We had a training teleconference to review

 

      classification parameters, then training

 

      reliability exercise to ensure appropriate

 

      application of classifications.  All case

 

      narratives were blinded to any potentially biasing

 

      information, and I will review that in a minute.

 

      There was random distribution of 427 events to 9

 

                                                               124

 

      expert raters.  Each case was independently rated

 

      by 3 raters.  Each rater received approximately 125

 

      events to rate, and any group of 3 raters shared

 

      only 5 cases.  All ratings were reviewed for

 

      quality assurance and identification of

 

      non-agreement cases.  Consensus teleconferences

 

      were held for any disagreement cases, and there was

 

      double data entry for quality assurance.

 

                Now, what was the consensus process I

 

      referred to?  If ratings did not have unanimous

 

      agreement, then a consensus discussion was held.

 

      Each case was discussed by the three raters

 

      involved only.  Discussion of each case was led by

 

      an expert other than those originally assigned the

 

      case.  The goal of the discussion was to reach 100

 

      percent agreement.  If 100 percent agreement could

 

      not be reached, the case then became indeterminate.

 

      Sometimes the original majority opinion did not

 

      always end up as the final consensed

 

      classification.  In other words, if there was a

 

      minority rating, sometimes that ended up being the

 

      final outcome.

 

                                                               125

 

                Now, what was rated?  Blinding of event

 

      narratives to avoid bias included--we received the

 

      narratives from the FDA blind to all potential drug

 

      identifying information.  This included drug name,

 

      company sponsor name, patient identification

 

      numbers, whether they were on an active or placebo

 

      arm, and any and all medication names and types

 

      because there could be associated treatments that

 

      might bias somebody or tip them off as to what drug

 

      was being talked about and, of course, primary

 

      diagnosis.  We also did some additional blinding of

 

      potentially biasing information which included the

 

      original label of the event given by the

 

      investigator or sponsor and serious or non-serious

 

      labels.

 

                Rating guidelines--how was the

 

      classification scheme applied?  We wanted the

 

      experts to apply concepts using their clinical

 

      expertise and judgment; to use their experience to

 

      integrate clinical information and infer when

 

      appropriate.  We wanted them to have a reasonable

 

      certainty in order to commit to a rating, and

 

                                                               126

 

      rating was based on what was probable, not what was

 

      possible.

 

                The guidelines for intent inference

 

      involved inferring if something was clinically

 

      impressive, and I am going to give you an example

 

      of that in a moment, or using two smaller pieces of

 

      clinical information.  The clinical information

 

      that could inform inference of intent included

 

      clinical circumstances.  That could be method used,

 

      number of pills; past history of suicide attempt;

 

      past history of self-injurious behavior or

 

      self-mutilation; and family history of suicide or

 

      suicide attempts.

 

                Now, here is a case example of inferred

 

      intent, what we call clinically impressive

 

      circumstances.  This is the first time you are

 

      actually seeing one of these real narratives.  In

 

      this case clinical impressiveness actually

 

      overruled stated intent, so you see the subject

 

      attempted suicide by immolation.  Her siblings

 

      doused the flames immediately.  She was left with

 

      minor burns on her abdomen and on her left

 

                                                               127

 

      shoulder.  The subject admitted she was angry with

 

      her parents for going away and leaving her alone at

 

      home because she was fearful.  The subject admitted

 

      that she had acted impulsively and had not intended

 

      to kill herself.

 

                Here are more examples.  This is another

 

      example actually of clinically impressive

 

      circumstances which was ultimately called a suicide

 

      attempt.  It is also important to know that we had

 

      no idea what the sponsor ratings were but both

 

      these cases were consistent with what the sponsor

 

      had said as well.

 

                This case involved a 16 year-old who

 

      claimed to have ingested 100 tablets of study med

 

      after a fight with her mother.  The patient

 

      informed her mother.  The mother brought her to the

 

      ER.  The patient reported feeling shaky.  Emergency

 

      room physician said she was slightly tachycardic

 

      with a pulse of 100.  The tox. screen was negative

 

      but the patient did have some illness and she

 

      stayed in the ER until she was asymptomatic, and

 

      then was later admitted to the psych. unit.

 

                                                               128

 

                Another example of a suicide attempt, a

 

      patient age 17 took an overdose of 20 tablets.  In

 

      the father's opinion the overdose was 5 tablets.

 

      The patient didn't have any symptoms of an

 

      overdose, not even nausea, but it was classified as

 

      a suicide attempt, of course.

 

                More overdose examples.  You see in this

 

      first example there were 113 tablets and it

 

      exemplifies how medication types were blinded so

 

      you see all the different numbers there.  Then, the

 

      next one is patient aged 15, impulsively slit her

 

      wrist following an altercation with her mother.

 

      Finally, age 17, attempted suicide by taking 8

 

      tablets after a fight with her father, whom she

 

      considered harsh and rejecting.

 

                Now, these are examples of self-injurious

 

      behavior, intent unknown.  So, this is where we did

 

      some harm but we just don't know why.  A patient

 

      aged 10 had superficial scratches, left arm,

 

      scratched herself with scissors.  That was all the

 

      information that was there essentially.

 

                Patient, aged 14, ingested or simulated

 

                                                               129

 

      ingestion of 2-3 cigarettes.  The patient was

 

      reported as feeling tired and playing a theatrical

 

      role.  Subject, aged 9, reported he had ingested 4

 

      of his brother's tablets on a dare.  Finally,

 

      patient, aged 10, swallowed a small amount of

 

      after-shave lotion while angry.  It is hard to know

 

      what to make of those without information.

 

                Examples of preparatory actions, age 16,

 

      tried to hand herself and was prevented from doing

 

      so by her family.  The next case, age 18, a voice

 

      commanded him to jump from the roof.  Although he

 

      went up, he did not jump.  Next one, age 10, held a

 

      kitchen knife to her neck while alone but did not

 

      cut herself.  Event was not witnessed.  Finally, a

 

      patient, age 18, was noted to be hostile, hopeless

 

      and helpless and had written suicide notes.  As I

 

      said, anything beyond a verbalization was

 

      considered a preparatory action, including writing

 

      a suicide note.

 

                These are good examples of self-injurious

 

      behavior, no suicidal intent.  In the first case

 

      the patient stated there was increased family

 

                                                               130

 

      tension.  She made superficial cuts on her wrist

 

      with an Exacto knife.  The patient and mother

 

      reported the cuts weren't deep and they looked like

 

      cat scratches.  Patient adamantly denied any

 

      suicidal gestures or intent.  She stated she only

 

      wanted a release and that cutting and hitting her

 

      legs offers her a release.

 

                The second case, denied suicidal thoughts.

 

      The first time she cut herself was age 16.  She

 

      stated she did it for attention.  Today her cutting

 

      was more spontaneous. She reported that cutting

 

      gives her a good weird feeling.

 

                So, what were the results?  This slide

 

      just refers to what we are talking to in our

 

      results, or referring to, and there were 427 events

 

      but some patients had more than one event so we

 

      ultimately, in the reliability data, used 378

 

      cases.  We employed the same severity hierarchy

 

      that the FDA used.  So, we just took the most

 

      severe event for cases that had multiple events.

 

                Expert rater consensus--only two of 427

 

      cases had no agreement among the three raters.  So,

 

                                                               131

 

      each rater had a different rating in only two

 

      cases.  Fifty-nine cases had agreement among two of

 

      three raters, and those had to go to

 

      teleconference.  There were no cases in which

 

      consensus was not able to be reached during the

 

      teleconference and they, of course, had that

 

      option.

 

                Now, discordant cases between the sponsor

 

      and Columbia classifications, there were 40 out of

 

      the 427 cases in which the sponsor and the Columbia

 

      classification differed.  Twenty-six new cases were

 

      identified that had not been identified by the

 

      sponsor as possibly suicide-related.  There were

 

      two new cases of self-injurious behavior without

 

      suicidal intent that had been labeled something

 

      other than deliberate self-harm, and 12 cases were

 

      originally called possibly suicidal and were

 

      changed to something other than possibly suicidal.

 

                Here it breaks it down for you further.

 

      Of the 26 new possibly suicide-related events, one

 

      was a suicide attempt; one was a preparatory act;

 

      13 were ideation events; four were intent unknown

 

                                                               132

 

      acts; and seven were not enough information to say

 

      whether there was deliberate self-harm.

 

                Here is an example of one of the newly

 

      identified suicidal events.  This is a preparatory

 

      act.  The patient, age 11, held a knife to his

 

      wrist and threatened to harm himself.  The patient

 

      was hospitalized with an acute exacerbation of

 

      major depressive disorder.  The reason we have this

 

      is because, as I mentioned before, every

 

      hospitalization is a serious adverse event so that

 

      is why this preparatory act was caught.

 

                The events that were changed from suicidal

 

      to something other included two changed to

 

      psychiatric; one changed to an accident; and nine

 

      changed to self-injurious behavior with no suicidal

 

      intent.  Again, our famous example, a patient

 

      reported to have engaged in an episode of

 

      auto-mutilation where she slapped herself in the

 

      face.  The event resolved the same day without any

 

      intervention.

 

                These are actually the kappa, the

 

      agreement between the sponsor and Columbia.  So,

 

                                                               133

 

      Columbia's classification of possibly

 

      suicide-related and the sponsor's classification of

 

      possibly suicide-related, the kappa was 0.77.  You

 

      see in the 2 X 2 table that the numbers correspond

 

      to the numbers that I just went through with you.

 

                Now, if you want to look somewhat more

 

      specifically or at least what we think is more

 

      specific, we did a comparison of what Columbia said

 

      was definitively suicidal and what the sponsor said

 

      was possibly suicidal, and the kappa was 0.69.

 

                Here are the reliability results of the

 

      ratings with the nine expert raters.  The median

 

      ICC was 0.86 and what the FDA will refer to as the

 

      primary outcome variables, you can see the numbers

 

      here, suicide attempts is 0.81; preparatory

 

      actions, 0.89; suicidal ideation, 0.97.

 

                Where do we go from here?  We need to

 

      improve our adverse event reporting for

 

      suicide-related events by developing consistent

 

      terminology; developing guidelines for

 

      classification of suicidality so that adequate

 

      information is provided by the clinician;

 

                                                               134

 

      utilization of research assessment tools, what

 

      questions to ask, how to ask, and what measures aid

 

      this; finally, hopefully, that will lead to

 

      improved, more valid identification and

 

      documentation of suicidality.

 

                DR. GOODMAN:  Thank you very much.  Dr.

 

      Chesney?

 

                DR. CHESNEY:  Thank you.  This is a little

 

      bit of thinking outside the box, but we heard at

 

      the February meeting a number of examples of

 

      homicidal behavior.  I wonder if, in your

 

      speciality, homicidal behavior is ever identified

 

      as being self-injurious primarily to affect

 

      circumstance or to affect an internal state.

 

                DR. POSNER:  No, that did not represent

 

      any of those self-injurious, no suicidal intent

 

      ratings.  So, the classifications that you are

 

      referring to, internal state and circumstance, are

 

      not synonymous at all with the cases that had

 

      homicidal ideation or any kind of aggressive

 

      behavior.  It doesn't mean that it couldn't be

 

      looked at in another analysis but it wasn't

 

                                                               135

 

      represented in these cases.

 

                DR. CHESNEY:  I guess my more general

 

      question, I just wonder in the bigger question,

 

      homicidal behavior outcome is bound to be bad and

 

      self-injurious, and if it is just another factor

 

      that we should consider in this whole picture.

 

      Thank you.

 

                DR. GOODMAN:  Dr. Robinson?

 

                DR. ROBINSON:  Do you know how many of the

 

      events led to hospitalization and how it breaks

 

      down in terms of your classification?

 

                DR. POSNER:  Dr. Laughren, do you know?

 

                DR. LAUGHREN:  I don't have that figure

 

      off the top of my head.  There were a substantial

 

      number of events leading to hospitalization, I

 

      believe somewhere in the ballpark of maybe 40

 

      percent.  I don't have the exact number.  It was a

 

      common outcome.

 

                DR. POSNER:  It is important to know that

 

      we were very narrowly just looking at obtaining the

 

      most appropriate label for the particular event in

 

      question, and we didn't have any of the surrounding

 

                                                               136

 

      information or follow-up information in this

 

      particular piece of the project.

 

                DR. GOODMAN:  Dr. Wang?

 

                DR. WANG:  I have a question.  Do you know

 

      how many of the sponsors originally submitted

 

      reports that were categorized as serious?  The

 

      reason I am asking is to get a sense of how many

 

      cases may not have been originally reported.  I

 

      know you had these serious cases sent to you for

 

      adjudication, just to check in case there were

 

      cases that were being missed in what the sponsors

 

      were reporting, but did you look as to how many

 

      cases were not considered serious by the sponsor,

 

      jut to give us a sense of how many may be sort of

 

      out there in the non-serious pool?

 

                DR. POSNER:  Again, we were blinded to

 

      sponsors' classifications throughout the entire

 

      process.  I don't know if somebody from the FDA can

 

      answer that question for you.

 

                DR. LAUGHREN:  Again just a ballpark

 

      figure, I think it is probably somewhere in the

 

      vicinity of maybe 65-70 percent.  But you have to

 

                                                               137

 

      understand that a designation of serious is a

 

      judgment that is made by the sponsor fairly

 

      subjectively.  I mean, there are criteria for

 

      regulatory serious.  It is fatal, life-threatening,

 

      seriously disabling, leading to inpatient

 

      hospitalization.  But even though, you know, that

 

      on face appears to be fairly definitive, sponsors

 

      in many cases, in my view, made the judgment that

 

      if it was considered to be suicide-related it was,

 

      by definition, serious.

 

                So, if you look at many of the narratives

 

      that were classified as serious, I think no

 

      reasonable person looking at those would consider

 

      that, in a common sense notion, as a serious event.

 

      But the point is that that designation--how that

 

      judgment was made varied from sponsor to sponsor.

 

      So, you know, some of them classified many more of

 

      the events as serious than other sponsors.  But the

 

      answer to the question is that overall roughly

 

      two-thirds of these events that were included in

 

      the analysis were designated as regulatory serious.

 

                DR. GOODMAN:  Ms. Griffith?

 

                                                               138

 

                MS. GRIFFITH:  I have a question about

 

      cutting specifically.  It seems to me that most of

 

      the examples of cutting fall into self-injurious

 

      behavior, intent unknown or self-injurious behavior

 

      with no intent.  I am just curious as to are you

 

      confident that the reporting that you received and

 

      reviewed actually got to whether or not there was

 

      intent or no intent, and how subjective is the

 

      reporting likely to be?

 

                DR. POSNER:  Again, as you can see,

 

      cutting is a method that is used both in suicidal

 

      behavior and self-injurious behavior without

 

      suicidal intent.  If you remember the conceptual

 

      scheme, there was the category self-injurious

 

      behavior, intent unknown, because cutting can be

 

      used both ways.  I forget the exact number but

 

      there were 20-something cases in which they cut but

 

      we don't know if it was suicidal or not.  That is

 

      why we had to come up with a category just to

 

      categorize and deal with that issue.  The FDA will

 

      point out that the included that in the sensitivity

 

      analysis so just in case all of those were

 

                                                               139

 

      suicide-related events, they have those numbers.

 

                DR. LAUGHREN:  If the question you are

 

      asking were the narratives lacking in detail, they

 

      absolutely were.  These were not by any sense

 

      complete descriptions.  Ideally, many more

 

      questions would have been asked when these events

 

      occurred to help flesh them out.  That is why it

 

      was necessary to use inference as one approach to

 

      try and get at intent because intent was not

 

      included for the vast majority of these.

 

                DR. POSNER:  Which is why only experts in

 

      the field could have been able to infer from the

 

      surrounding information.  The narratives were

 

      limited with respect to suicidal intent often but