FOOD AND DRUG ADMINISTRATION










                         PULMONARY-ALLERGY DRUGS


                            ADVISORY COMMITTEE


















                         Wednesday, July 13, 2005


                                8:00 a.m.







                           Gaithersburg Hilton

                              The Ballrooms

                            620 Perry Parkway

                          Gaithersburg Maryland





          Erik R. Swenson, MD, Chairman

          Teresa Watkins, R.Ph., Executive Secretary




          Mark L. Brantly, M.D.

          Steven E. Gay, M.D., M.S.

          Carolyn M. Kercsmar, M.D.

          Fernando D. Martinez, M.D.

          I. Marc Moss, M.D.

          Lee S. Newman, M.D.

          Calman P. Prussin, M.D.

          Michael Schatz, M.D.

          David A. Schoenfeld, Ph.D.




          Karen Schell, RRT, Consumer Representative

          Jacqueline S. Gardner

          Nancy J. Sander, Patient Representative




          Christine Sorkness, Pharm.D.


          FDA STAFF:


          Robert Meyer, M.D.

          Badrul Chowdhury, M.D.

          Ann Trontell, M.D., M.P.H.

          Sally Seymour, M.D.

          J. Harry Gunkel, M.D.

          Eugene J. Sullivan, M.D., FCCP



                             C O N T E N T S


       Call to Order

                  Erik R. Swenson, M.D., Chairman                 5


       Introductions                                              6


       Conflict of Interest Statement

                 Maryanne Killian                                 8


       FDA Introductory Remarks

                 Badrul Chowdhury, M.D., Division of

                   Pulmonary-Allergy Drug Products               14


       Guest Speaker Presentation:


          An Overview of Long-Acting Beta Agonists

                 Christine Sorkness, Pharm.D.,

                   University of Wisconsin                       21


       Questions by the Speaker                                  68


       GlaxoSmithKline Presentation:


          Opening Remarks

                 C. Elaine Jones, Ph.D.                          82


          Salmeterol Review

                 Katharine Knobil, M.D.                          87


          Closing Remarks

                 C. Elaine Jones, Ph.D.                         115


       Questions by the Committee                               116


       Novartis Presentation:



                 Eric A. Floyd, M.S., M.B.A.                    136


          Efficacy and Safety of Foradil

                 Gregory P. Geba, M.D.                          139


          Clinical Implications

                 James F. Donohue, M.D., Chief, Pulmonary

                   Division, University of North Carolina       155



                       C O N T E N T S (Continued)




       Questions by the Committee                               168


       FDA Presentation:



                 Sally Seymour, M.D., Division of

                   Pulmonary-Allergy Drugs                      183



                 J. Harry Gunkel, M.D., Division of

                   Pulmonary-Allergy Drugs                      207


       Questions by the Speakers                                224


       Opening Public Hearing:


                 Chris Ward, Asthma and Allergy

                   Foundation of America                        233


       Committee Discussion                                     238



                          P R O C E E D I N G S


                 DR. SWENSON:  Good morning, everyone.  I


       am Dr. Erik Swenson.  I am the Chairman of this


       Pulmonary-Allergy Drug Advisory Committee meeting,


       meeting today here to discuss the implications of


       recently available information and data related to


       the safety of long-acting beta agonist




                 Before we go around and introduce the


       members of the panel, I would like to ask them to


       remember that we have microphones here that have


       dual functions.  One is to show that you wish to


       raise a question.  That is the "request" option


       there; then to speak is on the right-hand side.


       So, in raising questions, would you please first


       hit the "request" button.  We will be monitoring


       and call you in turn.  Please do remember to use


       the "speak" button when you do speak since


       transcribers will need to hear you on the tapes.


                 With that having been said, I would like


       to have members of the panel here go around and


       introduce themslves.  We will start with Bob Meyer



       and have you introduce yourself in turn.




                 DR. MEYER:  I am Bob Meyer.  I am the


       Director of the Office of Drug Evaluation II in the


       Center for Drugs.


                 DR. CHOWDHURY:  I am Badrul Chowdhury, the


       Division Director, Division of Pulmonary and


       Allergy Drug Products.


                 DR. TRONTELL:  Ann Trontell, the Deputy


       Director of the Office of Drug Safety.


                 DR. SULLIVAN:  My name is Gene Sullivan.


       I am the Deputy Director of the Division of


       Pulmonary and Allergy Drug Products.


                 DR. SEYMOUR:  I am Sally Seymour, medical


       officer in the Division of Pulmonary and Allergy


       Drug Products.


                 DR. GUNKEL:  Harry Gunkel, medical officer


       in the Division of Pulmonary and Allergy Drug




                 MS. SANDER:  Nancy Sander, President,


       Allergy and Asthma Network, Mothers of Asthmatics.


                 DR. GARDNER:  Jacqueline Gardner,



       Professor of Pharmacy at the University of


       Washington, and a member of the Drug Safety and


       Risk Management Advisory Committee to FDA.


                 DR. SCHATZ:  Michael Schatz.  I am an


       allergist/immunologist from Kaiser Permanente San




                 MS. WATKINS:  I am Teresa Watkins,


       executive secretary for this committee.


                 DR. GAY:  I am Steven Gay.  I am medical


       director of critical care support services,


       assistant professor at the University of Michigan.


                 DR. MOSS:  Marc Moss, associate professor


       of medicine, Emory University in Atlanta.


                 DR. NEWMAN:  Lee Newman, professor of


       medicine, National Jewish Medical and Research


       Center and University of Colorado Denver.


                 DR. BRANTLY:  Mark Brantly, professor of


       medicine, University of Florida.


                 DR. MARTINEZ:  I am Fernando Martinez,


       professor of pediatrics at the University of




                 DR. KERCSMAR:  Carolyn Kercsmar, professor



       of pediatrics, Rainbow Babies and Children's


       Hospital, Cleveland, Ohio.


                 MS. SCHELL:  I am Karen Schell.  I am the


       consumer representative.  I am a respiratory


       therapist from Emporia, Kansas.


                 DR. PRUSSIN:  Calman Prussin.  I am senior


       clinical investigator in the Laboratory of Allergic


       Diseases, NIAID, NIH.


                 DR. SCHOENFELD:  David Schoenfeld,


       professor of medicine at the Harvard Medical School


       and professor of statistics at the Harvard School


       of Public Health.


                 DR. SWENSON:  Thank you.  I would like now


       to call Maryanne Killian, of the FDA.  She has a


       statement on conflict of interest to read.


                      Conflict of Interest Statement


                 MS. KILLIAN:  Good morning, everybody.


       The Food and Drug Administration is convening


       today's meeting of the Pulmonary-Allergy Drugs


       Advisory Committee under the authority of the


       Federal Advisory Committee Act.  With the exception


       of the industry representative, all members of this



       committee are special government employees or


       regular federal employees from either agencies,


       subject to the conflict of interest laws and




                 FDA has determined that the members of


       this advisory committee are in compliance with


       federal ethics and conflict of interest laws,


       including but not limited to 18 USC Section 208 and


       21 USC Section 355(n)(4) which applies to FDA


       people.  Congress has authorized FDA to grant


       waivers to special government employees who have


       financial conflicts when it is determined that the


       agency's need for a particular individual's


       services outweighs his or her potential financial


       conflict of interest.


                 Members who are special government


       employees at today's meeting, including special


       government employees appointed as temporary voting


       members, have been screened for potential financial


       conflicts of interest of their own, as well as


       those imputed to them including those of their


       employers, spouse or minor child related to the



       discussions on July 13, 2005 regarding implications


       of recently available data related to the safety of


       long-acting beta agonist bronchodilators, and on


       July 14, 2005 regarding the continued need for the


       essential use designation of prescription drugs for


       the treatment of asthma and chronic obstructive


       pulmonary disease under 21 CFR 2.125.  These


       interests may include investments, consulting,


       expert witness testimony, contracts, grants,


       CREDAs, teaching, speaking, writing, patents and


       royalties and primary employment.


                 In accordance with 18 USC Section


       208(b)(3), four waivers have been granted to the


       following participants.  Please note that all


       interests are in firms that could be potentially


       affected by the committee's deliberations.  With


       regard to the July 13th meeting, Dr. Carolyn


       Kercsmar for activities on a speaker's bureau.  She


       receives less than $10,001 per year for a grant


       which is valued at less than $100,000 per year, and


       for a grant for which the firm supplies products


       worth approximately less than $100,000 per year;



       Ms. Nancy Sander for ownership of stock currently


       valued at between $25,001 and $50,000, and for


       unrelated advisory board activities for which she


       receives less than $10,001 per year; Dr. Steven Gay


       for speaker bureau activities with four firms, from


       three of which he receives less than $10,001 per


       firm per year, and one for which he receives from


       between $10,001 to $50,000 per firm per year.  We


       would also like to disclose that Dr. Erik Swenson


       owns stock worth less than $5,001.  A waiver under


       USC 208(b)(3) is not required because the de


       minimis exemption under 5 CFR 2640.202 applies.


                 With regard to the July 14th discussions,


       Dr. Carolyn Kercsmar for activities on a speakers


       bureau.  She receives less than $10,001 per year


       for two grants which are valued at less than


       $100,000 per year, and for a grant for which the


       firm supplies products worth approximately less


       than $100,000 per year.  She also owns stock less


       than $5,001.  A waiver under the USC 208(b)(3) is


       not required because the de minimis exemption under


       5 CFR 2640.202 applies.  Dr. Fernando Martinez for



       his membership on a speakers bureau.  He has not


       lectured or received remuneration in the past 12


       months, and for membership on a related advisory


       board.  He has not participated or received any


       remuneration to date.  Dr. Michael Schatz for his


       activities on a speakers bureau.  He receives less


       than $10,001 per year, and for a grant for which


       the firm supplies product worth approximately less


       than $100,000 per year.  Miss Nancy Sander for


       ownership of stock currently valued between $25,001


       and $50,000, and for unrelated advisory board


       activities for which she receives less than $10,001


       per year.  Miss Sander also owns stock worth less


       than $5,001, again a de minimis waiver is not


       required because 5 CFR 2640.202 applies.  Dr.


       Steven Gay for speakers bureau activities with five


       firms, from three of which he receives less than


       $10,001 per year, and two of which he receives from


       $10,001 to $50,000 per firm per year.


                 We would also like to disclose that Dr.


       Marc Moss' spouse owns stock less than $5,001.  A


       waiver under 18 USC 208(b)(3) is not required



       because the de minimis exemption under 5 CFR


       2640.202 applies.


                 A copy of the written waiver statements


       may be obtained by submitting a written request to


       the agency's Freedom of Information Office, Room


       12A-30 of the Parklawn Building, 5600 Fishers Lane,


       Rockville, Maryland.


                 In addition, Dr. Christine Sorkness is


       participating as FDA's invited guest speaker on


       July 13th.  She would like to disclose that she is


       a researcher with regards to GlaxoSmithKline's


       Advair and Novartis' formoterol.  She also lectures


       for GlaxoSmithKline concerning Advair and receives


       less than $10,000 per year.


                  Lastly, Dr. Theodore Reiss is the


       industry representative on the committee at the


       meeting.  He is acting on behalf of all related


       industry.  He is employed by Merck.  Thank you.  I


       am done.


                 DR. SWENSON:  Thank you, Miss Killian.  I


       would like now to turn the microphone over to Dr.


       Robert Meyer of the FDA.



                 DR. MEYER:  Thank you.  Prior to more


       formal introduction by Dr. Chowdhury, I wanted to,


       first off, thank the advisory committee in advance


       for your attendance today and for what I am sure


       will be a very careful deliberation.


                 One of the things I wanted to mention was


       that there was some speculation in the trade press


       yesterday that there was a very specific purpose


       and outcome hoped for by the agency in holding this


       meeting today.  I just wanted to be clear that the


       FDA looks forward to a very open discussion of the


       data available on the safety experience with the


       long-acting beta agonists and any potential future


       regulatory actions that might be recommended coming


       out of this committee.  So, thank you very much for


       your attendance today.


                 DR. SWENSON:  Thank you, Dr. Meyer.  Now


       Dr. Chowdhury, from the FDA, is going to give us


       some introductory remarks pertinent to our


       discussion today.


                         FDA Introductory Remarks


                 DR. CHOWDHURY:  Good morning.  Honorable



       Chairperson, members of the Pulmonary-Allergy Drugs


       Advisory Committee, representatives from GSK and


       Novartis and others in the audience, I welcome you


       to this meeting.


                 In this brief presentation I will


       introduce you to the subject matter of this


       advisory committee meeting.  Members of the


       committee, the objective of this meeting is to


       discuss the implications of the available data


       related to the safety of long-acting beta agonist


       bronchodilators.  There are two long-acting


       bronchodilators marketed in the United States that


       will be discussed in this meeting.  These are


       salmeterol from GSK and formoterol from Novartis.


       Products containing salmeterol and formoterol are


       indicated for use as bronchodilators in patients


       with asthma and COPD as maintenance treatments.


                 These are effective drugs and form


       important components of the treatment options


       available for patients with asthma and COPD.  But


       an important array of adverse effects that has been


       observed with these drugs is the occurrence of



       severe asthma exacerbation.  The intent of this


       advisory committee meeting is to discuss this


       specific finding of severe asthma exacerbation


       related to these two drugs.  Since the available


       data pertain to asthma, the focus of this meeting


       is on asthma and not COPD.


                 Surrogates of short-acting beta agonist


       bronchodilators, such as albuterol, is not a


       subject of this meeting.  As you discuss and


       deliberate on the safety of thee two drugs, keep in


       mind the well-established efficacy of these drugs


       because the use of these drugs, like any other


       drug, is dependent on the risk/benefit ratio.


                 As you can see in the agenda, the first


       presentation will be by Dr. Christine Sorkness.


       Dr. Sorkness is a professor of pharmacy and


       medicine in the University of Wisconsin.  She will


       give an overview of long-acting beta agonist


       bronchodilators.  We are very fortunate that Dr.


       Sorkness, and expert in pharmacological drugs used


       in the treatment of asthma, has agreed to speak at


       this meeting.  I thank her on behalf of the agency.



                 Following Dr. Sorkness, GSK and Novartis


       will make presentations on salmeterol and


       formoterol respectively, followed by FDA


       presentations on these two drugs.  This will be


       followed by an open public hearing and committee




                 As you hear these presentations you will


       note that the safety signal of severe asthma


       exacerbation with salmeterol was seen in


       postmarketing studies, specifically the recently


       halted large controlled study called the salmeterol


       multicenter asthma research trial, acronym SMART,


       conducted by GSK.  In contrast, the safety signal


       of severe asthma exacerbation with formoterol was


       seen in the studies conducted by Novartis to


       support registration of formoterol in the United


       States.  Novartis also conducted a Phase 4 study


       with formoterol that did not show a clear signal of


       severe asthma exacerbation, but the formoterol


       Phase 4 study was much smaller compared to the


       SMART study.


                 We are choosing to have this meeting now



       because all pertinent data on salmeterol and


       formoterol have only become recently available.  We


       also decided that it would be fruitful to discuss


       these two related drugs together in one meeting.


       Although salmeterol has been approved for marketing


       in the United States since 1994, the study relevant


       to this meeting, the SMART study, was halted by GSK


       in January of 2003 and the data has been recently


       fully analyzed.


                 Formoterol was approved for marketing in


       the United States in 2001.  The Phase 4 study for


       formoterol was completed in March, 2004 and the


       data from the study also has been recently




                 The significant regulatory actions that


       the FDA has taken so far pertaining to these two


       drugs, based on the available data, are in


       cooperation of the results of the SMART study in


       all salmeterol-containing product labels, including


       the addition of a boxed warning, and not approving


       formoterol 25 mcg twice daily dose for marketing in


       the United States.  Formoterol is currently



       approved at a dose of 12 mcg twice daily.  Please


       note that the formoterol drug label does not


       currently have warnings similar to salmeterol


       because of lack of specific data related to the


       marketed formoterol 12 mcg twice daily dose.


                 In the presentations from the industry and


       the FDA you will see the data that led to the


       agency regulatory actions.  As you hear the


       presentations, I request that you keep in mind the


       questions that are in the FDA briefing book and


       also attached to the agenda since you will discuss


       and deliberate on these questions later in the day.


                 Here are the four questions that you will


       be asked to discuss and deliberate later in the day


       today.  Question one, the product labels of


       salmeterol-containing products have been modified


       to include warnings related outcome the SMART


       study.  Based on currently available information,


       what further actions, if any, do you recommend that


       the agency take to communicate or otherwise manage


       the risks of severe asthma exacerbations seen in


       the SMART study?



                 Based on the currently available


       information, do you agree that salmeterol should


       continue to be marketed in the United States?


                 Question two, the label of the


       formoterol-containing product does not include


       warnings comparable to the warnings that are


       present in the salmeterol-containing products.


       Based on the currently available information,


       should the label of formoterol-containing products


       include warnings similar to those in the salmeterol




                 Based on the currently available


       information, do you agree that formoterol should


       continue to be marketed in the United States?


                 Question three, what further


       investigation, if any, do you recommend to be


       performed by GSK that can improve the understanding


       of the nature and magnitude of the risk of




                 Question four, what further investigation,


       if any, do you recommend to be performed by


       Novartis that can improve the understanding of the



       nature and magnitude of the risk of formoterol?


                 These are the four questions.  We look


       forward to an interesting meeting and, again, I


       thank you for your time, effort and commitment to


       this important public health service.  Thank you.


                 DR. SWENSON:  Thank you, Dr. Chowdhury.


       At this point we would like to invite Dr. Christine


       Sorkness who was just introduced.  She has been


       kind enough to give us a broad overview of these


       drugs and I would like to turn the podium over to




                 An Overview of Long-Acting Beta Agonists


                 DR. SORKNESS:  Good morning.  I would


       first like to thank Dr. Chowdhury and Dr. Sullivan


       for inviting me to speak this morning, and most of


       all, for gathering this group of both clinicians,


       researchers, industry colleagues and the committee


       to review what I believe to be an incredibly


       important topic.  The risk versus benefit


       considerations for the long-acting beta agonists


       are the topic at hand and the committee has been


       asked to discuss the implications of the available



       data related to the safety of long-acting beta


       agonists, as Dr. Chowdhury articulated.


                 It is a little bit awesome to review this


       topic because of its breadth and depth, and also


       because I know many of the committee members and


       would acknowledge that they probably know more than


       I do about this particular topic.  So with that


       caveat, I am going to indicate that I am just going


       to review and try to set a tone for the discussions


       and in particular anchor some of the available


       data, at least as I see it as a researcher and a


       clinician, that you might use to answer the


       questions that you have been charged with.


                 The specific objectives that I have been


       asked to address are to provide an overview of the


       clinical pharmacology of the long-acting beta


       agonists; to discuss the selection of therapeutic


       outcomes which I believe are relevant for the


       assessment of risks versus benefits of the


       long-acting beta agonists; to review selected


       clinical trials, selected because there are so many


       which provide insight into the risk/benefit of the 



       long-acting beta agonists; and to outline the


       controversies and the remaining questions which I


       believe are related to the role.


                 First an overview of the clinical


       pharmacology of albuterol, salmeterol and


       formoterol.  We have come a long way from ephedra


       from China and its pharmacologic properties many,


       many years ago to, certainly ephedrine and


       epinephrine and isoproterenol.  The three major


       drugs that we use in our therapeutic armamentarium


       for asthma right now are albuterol, salmeterol and


       formoterol.  You can see in common that they all


       have a simple catecholamine ring, and there has


       been great novelty from the industry of adding a


       variety of different side chains to these products


       to affect their oral versus inhalation efficacy


       and, in particular, if you look at salmeterol and


       formoterol you see that there are very large side


       chains that have been attached to the basic


       molecule of albuterol.  This has allowed these two


       products to have an extended duration of action.


                 Both salmeterol and formoterol are highly



       lipophilic products, which may explain some of


       their long duration of action, salmeterol more than


       formoterol.  We know that salmeterol binds within


       the ligand binding cleft of the receptor which


       probably allows sensitivity stimulation of the


       receptor and its long duration, and there are other


       speculated mechanisms of action for the long


       duration of formoterol.  Formoterol is a raceme and


       only the RR and N tumor is active.


                 If you were to compare very globally the


       beta adrenergic agents, this table is probably


       relevant.  Most of the pharmacologic studies relate


       molar potency of these products to isoproterenol,


       which is designated as a potency of 1.  You can see


       that both formoterol and salmeterol are more potent


       products than isoproterenol.  The pharmacologic


       profile of the drugs is illustrated, with


       isoproterenol an formoterol classified as full


       agonists and albuterol and salmeterol as partial




                 You can see that in comparison to


       isoproterenol as its comparator, albuterol,



       formoterol and salmeterol all have the luxury or


       beta2 selectivity which is acknowledged to allow


       these drugs to have primarily effects on the lung


       versus the cardioselective effects that we see


       primarily with activation of the beta                                    

                                                         1 receptors.


       The duration of action clearly is different in


       these agents and, because of the long side chains


       and mechanisms of action of formoterol and


       salmeterol, we have known that these are the


       longest acting inhaled bronchodilators on the


       market today, with durations of action of at least


       12 hours after a dose, and the bronchoprotective


       effects, which specifically in this slide refer to


       the prevention of bronchoconstriction induced by


       exercise or non-specific bronchial challenges such


       as methacholine, have, indeed, a long


       bronchoprotective effect.


                 If you were to look at a more direct


       clinical comparison of formoterol and salmeterol


       based specifically on information in the package


       inserts, it is believed that equipotent


       bronchodilating doses of formoterol and salmeterol



       are listed as above, based specifically on the


       dosage form by which they are delivered.  So we


       believe, at least in clinical practice, that 12 mcg


       of Foradil aerolizer is clinically bronchodilating


       equipotent to 50 mcg delivered by Serevent Diskus.


       In order to deliver these equipotent doses, the


       recommended inspiratory flow rate is acknowledged


       to be about 60 L/min for both products over a time


       course of 2-3 seconds.  As you might expect,


       particularly because these drugs have been FDA


       approved for individuals with much more severe


       broncho-obstruction such as in COPD, probably an


       inspiratory flow rate much below that can get


       adequate delivery of drugs.


                 Both of these drugs are classified as


       pregnancy category C and, indeed, enjoy the same


       FDA approved indications based on the package of


       information submitted to the FDA, the only


       distinction being that salmeterol is approved for


       the treatment of asthma and prevention of


       bronchospasm for children over 4 years of age and 5


       years on formoterol.  Both of these agents have



       been approved for EIB prevention and for


       maintenance treatment of COPD, which is not part of


       the agenda today.


                 Now, the differentiation of formoterol and


       salmeterol, by and large, comes down to its


       acknowledged difference in onset of action.  You


       can find many, many studies that would classify


       different pharmacologic profiles.  In summary,


       formoterol probably achieves 80 percent of the


       maximum bronchodilation within 5-10 minutes.  It is


       thought to have an onset of action quite comparable


       to albuterol and acts within 3 minutes.  For


       salmeterol most of the data suggests that 90


       percent maximum bronchodilation occurs after one


       hour, with a median time to significant


       bronchodilation of 30-40 minutes, and an onset


       certainly at a time point of about 10 minutes.


                 This is a simple cartoon that segues to


       the issue of the long-acting beta agonists


       themselves in combination with clucocorticoids.


       This is a cartoon that suggests the proposed


       molecular interaction between the long-acting beta



       agonists and the inhaled corticosteroids.  The


       long-acting beta agonist, through their activation


       of the beta adrenergic receptor with adenylyl


       cyclase, cyclic AMP, protein kinase A and mitogen


       activated protein kinase may actually prime the


       glucocorticoid receptor for greater nuclear


       translocation and affinity for the binding to the


       glucocorticoid regulatory element, which is


       designated in this slide as GRE.  Therefore, it has


       been speculated by a variety of pharmacologic


       models--Ikleburg[?] and others who have done very


       elegant work--that actually the anti-inflammatory


       effect of glucocorticoids can be enhanced with the


       combination of long-acting beta agonist and,


       clearly, that is certainly the rationale that


       brought the combination products to the




                 Now, when we talk about risks versus


       benefits of any agent, it is best to talk about the


       outcomes of interest.  I am going to preface my


       remarks by the fact that I think the medical


       community and patients have all been led to hope



       for a 100 percent active and effective drug with


       absolutely no side effects.  I quite honestly


       believe that to not be realistic.  Therefore, when


       we talk about risk/benefits we need to put in


       perspective and weigh those issues, and I think it


       is important to recognize that we may have very


       safe medications that really have very poor


       clinical efficacy, and I would suggest that they


       have a distinct risk in their own right by their


       inability to treat the disease at hand.


                 So, I am going to try to illustrate some


       issues about what I believe to be important


       outcomes and talk about some of the clinical trials


       to date that teach us lessons about this as


       applying this drug class to asthma.


                 We traditionally have used lung function


       measures for management of asthma, both from the


       perspective of clinical decision-making and


       clinical research.  There are many longitudinal


       studies of lung health that have been enhanced by


       measurements of lung function, particularly FEV                         




       and FEV                                           1/FEC ratio.  Clearly,

it has been



       acknowledged that the gold standard for trial entry


       for the pivotal trials reviewed by the FDA have


       been traditional FEV                                                    

           1's of 60-80 percent predicted


       with 15 percent reversibility.  Therefore, there


       have been very uniform population groups that have


       been studied in our clinical trials.  I would


       actually conjecture now, and will come back to it,


       that we may need to broaden that a bit to capture a


       more generalizable population.


                 Clearly, lung function measures have been


       primary outcomes to measure efficacy because we can


       standardize those procedures both on site and with


       home measurements, and we have grown to believe


       that we can minimize variability around the


       measurements and can really get a handle and our


       arms around what outcomes are important.  Please


       recognize as I talk about different outcomes in


       asthma, I am not dispelling at all the value of


       lung function measurements.  I think they are still


       critical but I don't believe that they are enough.


                 Let's start talking about what I believe


       to be illustrative studies.  This is a study



       published by the Asthma Clinical Research Network


       in which I am one of the investigators, and it was


       affectionately called the SOCS trial.  This is a


       study that was intended to ask the question that in


       a patient who was well stabilized on an inhaled


       steroid and representative triamcinolone, and that


       had pretty stable FEV                                                   

             1's and peak flow variability,


       could this patient basically be transferred to


       placebo and do equally well; be converted to a


       salmeterol product and do equally well; or did they


       need to maintain continuance on an inhaled steroid


       as represented by triamcinolone?


                 This is a study that enrolled individuals


       whose mean FEV1 was 93 percent predicted, had very


       low peak variability of about 10 percent, and


       during the run-in period showed very good asthma


       stability.  The primary outcome of this study was


       morning peak flow.  That was selected because of


       experience that the Asthma Clinical Research


       Network had with what we believe to be an effect


       size that we could power our study of a difference


       of about 25 L/min, and because that effect size



       correlated with other more clinically robust


       endpoints in a variety of trials.


                 I think you can see that if you look at


       the primary outcome of this trial of AM peak flow


       you wee in the run-in period that all of the


       patients in ultimately the three arms improved


       during the run-in with triamcinolone, as you would


       expect.  You see the placebo group, once it was


       randomized at six weeks, had deterioration in that


       outcome; whereas, the triamcinolone and salmeterol


       groups both had maintenance and actually


       improvement in the primary outcome of peak flow,


       and there was not statistically significant


       difference between those two arms in this


       particular outcome.


                 Now, there was obviously a variety of


       secondary outcomes in this trial.  You can see that


       on the basis, in particular, of some markers of


       inflammation that there was both a clinically and


       statistically significant difference in favor of


       the inhaled corticosteroids.  Because of the


       multiple comparisons used by the statistician, a p



       value of 0.016 was that which was deemed to be of


       statistical significance.


                 This is important in that it translates to


       another very important secondary outcome of this


       trial, that being defined as treatment failure


       rates, on the left, and asthma exacerbation rates,


       on the right.  First, asthma exacerbation rates


       were defined as increases in albuterol use,


       decrease in peak flow, and the need for oral


       corticosteroids.  You can see with this particular


       outcome that triamcinolone is the only one by the


       Kaplan-Meier survival curve that, in essence, did


       not have significant asthma exacerbations.  Very


       similarly, if you looked at treatment failure


       rates, which was defined as an FEV1 less than 50


       percent predicted, at least one course of


       prednisone, the occurrence of emergency room or


       urgent care visits or hospitalization, the same


       trend could be seen.  The triamcinolone was very


       effective in preventing treatment failure rates but


       salmeterol was quite comparable to placebo.


                 The summary for the ACRN investigators was



       that patients with persistent asthma, well


       controlled by low doses of an inhaled steroid


       cannot be switched to salmeterol monotherapy


       without risk of clinically significant loss of


       asthma control.  I think this is one of the studies


       that clearly the asthma community has endorsed to


       support the fact that long-acting beta agonists in


       asthma should not be used as monotherapy, and I


       don't believe that there is particular debate on


       this issue and I think there are many studies that


       illustrate similar outcomes.


                 This study is also important in that it


       shows clear disparity between lung function


       measures and other outcome measures, and leads us


       to the conclusion from this study that multiple


       measurements and dimensions of control are needed


       to adequately assess therapies.


                 Therefore, I think, whether we broach


       studies that are industry sponsored or NIH


       sponsored, we are beginning to endorse more


       composite measures of asthma control.  This would


       include days of asthma control; treatment failure



       and asthma exacerbation criteria, as I have shown


       in this study and many others.  I would make as a


       caveat that it becomes oftentimes very difficult to


       compare trials because the specific definitions for


       treatment failure versus asthma exacerbations and


       mild, moderate and severe exacerbations may be a


       little bit different.  So, it is important for us


       to anchor the definitions when we evaluate.


                 Other composite measurements have actually


       been improvements or shifts in NAEEP defined NAEEP


       defined asthma severity classification; the


       achievement of total control or well controlled


       status, as defined by GINA and applied to the GOAL


       study; and certainly a variety of more patient


       specific surveys of asthma control and quality of


       life that have become important secondary outcomes


       in our clinical trials.


                 Now, in reflecting upon the issue of more


       composite clinical outcomes, the question needs to


       be raised in applying an appropriate risk/benefit


       relationship and assessment of how much benefit can


       actually be achieved by the combination of inhaled



       steroids and long-acting beta agonists.  I am going


       to focus my remarks on the combination because I


       have told you that at least my belief is that


       asthma is best treated by combination and,


       therefore, the relevant studies are those that use




                 A fairly early study that began to address


       the role of inhaled steroids and long-acting beta


       agonists in combination is the OPTIMA trial,


       entitled, low dose inhaled budesonide as a


       representative inhaled steroid an formoterol as a


       representative long-acting beta agonist.


                 This study had both a group A and a group


       B.  I am going to focus on group A as a


       representative trial of taking patients naive to


       being on inhaled steroids and ultimately, after a


       one-month run-in in which they were qualified to be


       in this trial, were then continued on placebo,


       Pulmicort or Oxis, as formoterol is called.


       Therefore, they continued on beta agonists alone


       versus being randomized to Pulmicort 100 mcg BID


       and Oxis placebo or Pulmicort 100 mcg BID and



       active Oxis 4.5 mcg BID.


                 The primary outcome of this trial was


       severe exacerbation, designated by the arrow.  This


       was defined as the need for oral corticosteroids or


       admission to a hospital or an emergency room visit


       or substantial decrease in peak flow.  This study


       group enrolled patients who were 12 years of age


       and older, not on inhaled steroids, who had to have


       an FEV                                         1 of at least 80 percent

predicted post


       bronchodilator, and actually enrolled a pre


       bronchodilator mean FEV                                                  

                   1 group of about 90 percent.


                 These are the two primary outcomes of this


       particular study.  If you look on the left-hand


       side in panel A, this is the Kaplan-Meier survival


       curve and you can see that both the budesonide


       alone versus the budesonide in combination win


       formoterol did much better in preventing the time


       to the first severe asthma exacerbation as compared


       to the placebo group, which is the last curve that


       you see on the slide.  When you plot this, on the


       right-hand side of the slide you see that actually


       the two active treatments were, indeed, better than



       placebo but were quite comparable to each other.


       However, if you look at the other important outcome


       of pulmonary function test, the morning peak


       expiratory flow, you see in the top curve that the


       combination product is superior to both budesonide


       and placebo.  So, whereas by one outcome the


       exacerbation rates of the two active products were


       not statistically significant, when you add in


       another important secondary outcome the


       combination, indeed, showed better outcomes.


                 Now, this same issue of looking at


       prevention of asthma exacerbations has been


       published by many, many authors.  This is just a


       representative study which looked at an analysis of


       asthma exacerbations, looking at available studies


       of higher dose fluticasone versus the addition of


       salmeterol to low dow fluticasone.


                 If you look at this particular slide,


       which is the probability of the time to the first


       exacerbation, you see that the top curve, in green,


       is salmeterol and, in red, the combination, and the


       combination was clearly superior in the outcome of



       time to first asthma exacerbation compared to the


       long-acting beta agonist alone.


                 The analysis in this study group culled


       out the different Ns of the spectrum of pulmonary


       function impairment at baseline.  As I mentioned,


       typically the pivotal trials enroll patients that


       have baseline FEV                                                       

   1's pre bronchodilator between


       40-85 percent predicted.  That is what you see on


       the left-hand side of all-comers that enrolled in


       those pivotal trials.  If you, instead, break down


       patients who present with less bronchoconstriction


       at baseline, for example, 60-85 percent predicted


       versus 40-60 percent, you see that the trends are


       not different and that either way, depending upon


       the severity of obstruction in these patients, the


       trend of the benefit of combination certainly could


       be seen.


                 Now, the FACET trial also showed I think a


       very important lesson about looking at the outcome


       of severe exacerbations and relationships of


       dose-response curves with inhaled steroids, as well


       as the benefit of long-acting beta agonists.  This



       goes back to budesonide and formoterol as the


       representative drugs in this study, and in this


       study severe exacerbations were defined as a need


       for oral steroids or a decrease in peak flow to


       more than 30 percent baseline.  So, severe


       exacerbations here are predominantly due to the


       need for oral beta agonists.


                 This is a large trial that randomized


       individuals to one of four arms.  If you look at


       the far left, in green is the budesonide 200 mcg or


       low dose inhaled steroid group; the purple bar is


       budesonide 200 mcg a day plus formoterol; in


       yellow, a higher dose of budesonide alone versus,


       in the orange bar, the addition to formoterol.  I


       think what you can see is the very logical


       dose-response curve that 800 mcg of budesonide


       fared better than 200 mcg of budesonide but, very


       importantly, you can see that the prevention of


       severe exacerbations in both groups could be


       enhanced by the addition of formoterol.  So, again,


       another study that suggests to us that combination


       therapy can achieve the prevention of asthma





                 Now, in brevity, rather than showing you


       the individual studies of exacerbations to date


       published, I am going to take advantage of a


       meta-analysis, published by Sinn and others in


       JAMA, in 2004 that looked at a systematic review


       and meta-analysis of a variety of pharmacologic


       therapies to reduce exacerbations.


                 This study clearly reviewed all of the


       drugs that we know that are on the marketplace but


       I am specifically going to look at two of the


       analyses.  This is the effect of long-acting beta


       agonists alone on exacerbations and the distinct


       trials that the meta-analysis chose.  You can see


       that the majority of these studies favored a


       long-acting beta agonist over placebo, and a pooled


       analysis showing a relative risk and confidence


       interval that favors the long-acting beta agonists.


                 This is the analysis that looks at many of


       what I believe to be the paradigm shifting trials


       that showed the addition of long-acting beta


       agonists to be better than either doubling or more



       than doubling the inhaled steroids, and includes


       the Matz and O'Byrne studies and Pauwels studies


       that I shared with you earlier.  I think you can


       see that we have at least somewhat mixed results


       here.  Certainly the majority of trials favor the


       combination of inhaled steroids and long-acting


       beta agonists together versus favoring the high


       doses of steroids.  Some of them are right on the


       line.  The pooled summary obviously, here by this


       graph, favors the steroids and the long-acting beta




                 I would suggest that certainly some of the


       differences are certainly on the basis of study


       design, size of study, construct, and so forth but,


       again, I think the meta-analysis supports the


       individual trials as far as evidence that suggests


       benefit of the combination.


                 Now, in switching gears, besides asthma


       exacerbations, I think that the issue of the


       capture of asthma control, as has been defined by


       GINA and the NAEEP, is a very important outcome


       that we have begun to carefully think about and to



       posture in our individual trials.  The GOAL trial


       asked a very simple but important question, is GINA


       NIH guideline based control achievable, and in what


       proportion of patients with a


       salmeterol-fluticasone combination compared with


       fluticasone alone?


                 So, this is going beyond the issue of just


       looking at exacerbations but overall asthma control


       as defined by the guidelines.  You can read this.


       There is both total control and well controlled,


       and it basically reflects what we, as clinicians,


       hope to achieve for our asthma patients.  And, the


       question is can this be achieved by the therapies


       that we have at hand?


                 The GOAL study design was very complex.


       It was a year study of three strata of patients


       based on whether they were either corticosteroid


       naive or free for six months; whether they were on


       a modest dose of a baclomethasone equivalent or


       higher dose of a beclomethasone equivalent.  These


       were individuals that had to be at least 12, not


       well controlled in the run-in period, and showed



       reversibility of 15 percent.  They were randomized


       to either the salmeterol-fluticasone combination or


       fluticasone alone via diskus, with a dose based on


       the stratum.


                 During this complex design in phase 1, the


       doses were either stepped up every 12 weeks until


       total control was achieved or a maximum dose was


       reached.  In study phase 2 a dose of total control


       or a maximum study dose was continued for 52 weeks.


                 It is important to recognize that all the


       patients in this trial deserved to be on control


       therapy.  Their FEV                                                     

        1's were about 75-80 percent


       predicted.  They had very, very obvious


       bronchodilator reversibility, averaging about 20


       percent, and what I would call were young adults.


       So, whatever the stratum, these individuals


       deserved to be stepped up with the therapies that


       were used.


                 These are the patients who achieved well


       controlled status.  The triangles in dark are the


       combination; the open circles are fluticasone


       alone.  You can see the run-in phase versus phase 1



       versus phase 2 on this graph.  You can see that


       both study groups had a fairly brisk improvement in


       achievement of well-controlled status.  This


       continued through the 52 weeks of the trial and was


       achieved by both study arms, but was achieved to a


       statistically significant greater extent with the


       combination therapy.


                 Also importantly is exacerbation rates as


       were studied in this trial as a secondary outcome.


       This exacerbation was defined in this study as


       either a burst of steroids or an ER or


       hospitalization.  You can see whether it was


       steroid naive, the low dose inhaled steroid or the


       moderate dose inhaled steroid stratum.  Clearly,


       all groups showed the trend that the combination


       therapy was better at achieving prevention of


       exacerbation rates as defined by the GOAL




                 The results of GOAL are very important in


       that significantly more patients achieved control


       with combination versus fluticasone in each stratum


       and in each stratum the time to achieve the first



       individual week of well-controlled asthma was


       significantly lower with combination than


       fluticasone alone.  More patients achieved control


       at the same or lower dose of inhaled steroid in


       each stratum for combination again verifying what


       had been previously published on the inhaled


       steroid-sparing effect.


                 I think very importantly in looking at


       outcomes, we know that the majority of patients who


       achieved well-controlled asthma in phase 1


       maintained the status when assessed in the last 8


       weeks of the study.  But, also, there were some


       patients that, additionally, were able to gain


       control with sustained therapy.  So, there may be,


       very importantly, subjects who initially are able


       to gain control but others that require longer


       exposure to achieve this particular outcome.


                 Now I am going to switch gears a little


       bit and talk briefly about a pediatric trial.  One


       of the things, at least in my mind, is that most of


       the data that we have in looking at inhaled


       steroids and long-acting beta agonists, whether



       they be as entry therapy or as add-on therapy in


       preventing the addition of inhaled steroids, has


       predominantly been done in adults.  Even those


       studies which have enrolled individuals greater


       than 12 years in age and up in general have not had


       a sizeable enough cohort of the 12-18 population


       that really have led to what I believe is a


       substantive subanalysis.  So, most of what we have


       I believe is in adult studies, and I think we will


       see more pediatric studies in the future.


                 This is a study that was recently


       presented at the American Thoracic Society meeting


       this summer, and was conducted by the CARE network


       of the NHLBI-sponsored network.  It is a one-year


       prospective comparison of three control or


       medications for the treatment of mild or moderate


       persistent asthma in children.


                 In brief, the study schematic is a proof


       of study concept.  All children were in a one- or


       two-week run-in period and then were either


       randomized to an inhaled steroid alone, an inhaled


       steroid at half the dose in combination with a



       long-acting beta agonist in comparison to a


       leukotriene receptor antagonist.  In order to


       achieve this particular proof of concept, the ICS


       group received fluticasone by morning and evening


       diskus and an evening capsule placebo.  The middles


       group of combination, and what I am going to call


       combination in the future, received an Advair


       diskus in the morning, a salmeterol diskus in the


       evening and a placebo capsule, and the leukotriene


       regimen active arm received montelukast at night


       and two placebos.


                 Because this study has not been published


       and there are responsibilities to editors, I am not


       going to be able to share with you in slide form


       all of the data, but I would like to summarize it


       for you as I did at the ATS.


                 Inclusion criteria for this study were


       children 6-14 years of age who had acknowledged


       mild to moderate persistent asthma, as defined by


       symptoms or beta agonist rescue use of peak flows


       in the yellow zone.  They needed to demonstrate


       asthma by a PC20 methacholine less than 12.5 mg/ml. 



       Bronchodilator reversibility was collected but it


       was not an entry criterion because we believed it


       would bias the outcomes because one of the study


       arms contained a long-acting beta agonist.  These


       were individuals who were naive to controller


       medications.  The issue was to look at whether


       these three arms and how asthma control was


       achieved in individuals with mild or moderate




                 The percent of asthma control days during


       the study period of 12 months was asthma control


       days defined as a day without albuterol rescue,


       without the use of non-study asthma medications, no


       daytime or evening asthma symptoms, unscheduled


       provider visits of school absenteeism, so a day in


       which a parent and a physician both would be happy


       that the asthma was well controlled and that was


       the defining outcome for this trial.


                 In summary, I am going to focus


       predominantly on the two outcomes related to the


       full dose inhaled steroid arm and the combination


       arm of the half dose fluticasone in combination



       with salmeterol.  Both of those study arms achieved


       improvement in the percent of asthma control days.


       At baseline this group of children had about 27


       percent of the days that were asthma


       controlled--so, very, very few.  This actually


       almost doubled or tripled during the active they


       and the fluticasone group gained asthma control


       days of 64 percent versus the combination of 60


       percent.  So, both groups adequately achieved


       asthma control and these were not statistically




                 Treatment failure was also a secondary


       outcome in this trial, defined by either the third


       burst of prednisone or a hospitalization or ER


       visit due to asthma.  There were only five


       treatment failures in the fluticasone arm and eight


       treatment failures in the combination arm.  That


       was not statistically significant.  Of that, there


       were no hospitalizations due to asthma in the


       fluticasone group and two hospitalizations with the


       combination group.


                 Overall, the comparison of the two groups



       showed in many outcomes that the inhaled steroid


       alone versus the inhaled steroid at half dose in


       combination with salmeterol were comparable, as I


       mentioned, in asthma control days; the time to


       prednisone bursts and treatment failure status.


                 There were some important differences in


       that if you looked at secondary outcomes such as


       change in PC                                                   20, the

improvement and ENO as a marker


       of inflammation, and actually changes in maximum


       bronchodilator response, the full dose of inhaled


       steroid was actually statistically better.


                 I mention this study from the point of


       view of one study looking at children that will,


       hopefully, soon be published and gives us some


       experience, I believe, with at least efficacy and


       safety in a pediatric population.


                 Now, let's switch gears to potential


       safety concerns that have been raised by the use of


       beta agonists.  That is what the committee has been


       asked to really put in perspective today.  It has


       not been just in the last few years that safety


       concerns with beta agonists have been raised. 



       Studies in the early '90s suggested that the


       regular use of a particular beta agonist,


       fenoterol, might produce adverse effects.  This is


       the number of subjects without exacerbation as a


       Kaplan-Meier curve and you can see those


       individuals treated with a regular dose of


       fenoterol had more asthma exacerbations than as


       needed.  This study, by Taylor and others, raised


       the specter of regular use of short to intermediate


       beta agonists producing adverse effects.


                 As you well know, fenoterol never made it


       to the U.S. market and albuterol has become clearly


       the drug of choice as the intermediate rescue beta


       agonist.  Therefore, Jeff Drazen and the Asthma


       Clinical Research Network felt it important as one


       of its missions to try to answer the question of,


       given that albuterol was the primary beta agonist


       used in the marketplace, did it matter whether


       patients were treated with regular beta agonists


       versus as needed beta agonists.  To achieve this


       trial, patients either received two puffs of


       albuterol four times a day plus extra as needed, or



       placebo inhaler two puffs four times a day and as


       needed, thus, sufficing the regularly scheduled


       versus as needed paradigm.  The study had a run-in,


       a 16-week treatment trial and then a run-out of 4




                 Now, whereas this group today is not here


       to debate the issues of safety of short and


       intermediate beta agonists, this trial basically


       has led to many of the questions that we have asked


       about long-acting beta agonists, and has led to


       what I believe is a series of trials that are in


       construct and will build on.


                 The summary from this particular study,


       using again peak flow as the primary outcome and


       power to find a difference of 25 L/min in the two


       study arms, suggested that whether you are on as


       needed albuterol or regular albuterol it really


       didn't make a difference in this outcome and,


       therefore, there was nothing evil about the use of


       regular beta agonists.  But the authors


       acknowledged that clearly based on the way the


       asthma community was moving, PRN beta agonists was



       the more rational approach.


                 Whereas this was a prospective trial, at


       the same time that this study was in the midst of


       being carried out, Steve Liggett's group at


       Cincinnati and others were working on cloning the


       beta receptor.  This is the beta receptor as a


       G-coupled protein.  As you well know there has been


       a lot of interest in single nucleotide


       polymorphisms at both the 27 position and the 16


       position in a variety of both in vitro and in vivo


       studies, looking at acute bronchodilator responses


       as well as a variety of other asthma outcomes.


                 So, when this was cloned, the Asthma


       Clinical Research investigators went back to the


       BAGS trial that was still ongoing and were able to


       get most of the participants to come back and be


       genotyped.  In that regard, the analysis showed


       that there was no effect in this primary outcome at


       the B27 locus.  There was no effect in the B16


       heterozygotes.  However, there was a signal.  When


       the B16 Arg/Arg patients were compared to the B16


       Gly/Gly patients, with a difference found in the



       primary outcome variable.


                 So, this is a retrospective look at the


       BAGS data that shows that if you were a group of


       patients who received regular albuterol and you


       were Arg/Arg, in yellow, your AM peak flow


       deteriorated during the course of the trial, in


       contrast to whether you received as needed beta


       agonists and were Arg/Arg, in red, or whether you


       received regular albuterol and were Gly/Gly.  This


       retrospective analysis was believed by the ACRN to


       be hypothesis generating, not definitive and,


       therefore, led to another study which I will share


       with you.


                 At the same time, Robin Taylor reported on


       the influence of beta adrenergic receptor


       polymorphisms in some studies he had done looking


       at, again, asthma exacerbations in this context.


       If you look at the far right of all-comers in this


       trial, you see that albuterol and salmeterol are


       comparable and superior to placebo in preventing


       exacerbations.  If you look at the Gly/Gly and the


       Gly/Arg groups, there were really no significant



       differences.  However, in those individuals that


       were Arg/Arg at the B16 locus, you can see that


       there were more exacerbations with those treated


       with albuterol but this was not seen with the


       salmeterol therapy.


                 So, we began to see in the asthma


       community some signals, some subtle signals in


       retrospective data about the issue of the potential


       relevance of polymorphisms at the beta receptor.


       Therefore, I told you that the Drazen trial,


       retrospective, was hypothesis generating to allow


       us to go forward to actually create a prospective,


       randomized, placebo-controlled, double-blind trial


       of regular versus minimal albuterol in each


       genotype.  This has affectionately been called the


       BARGE trial.


                 In this trial, in order to minimize beta


       agonist use, patients were provided with


       ipratropium for rescue as a primary inhaler and


       then had a backup to use albuterol of symptoms were


       not relieved by ipratropium.


                 This is a fairly complex study design but



       which we believed was important to answer the


       question.  First, individuals between the ages of


       18 and 55 years of age who had an FEV                                    

                                                         1 of at least


       70 percent predicted, and naive to inhaled


       steroids, were screened and genotyped.  If they


       were either found to be Arg/Arg or Gly/Gly at the


       B16 they were matched on the basis of FEV                               



       enrolled in the trial, went in a 6-week run-in


       period in which individuals were all on placebo


       with just rescue therapy.  They were then


       randomized to receive 16 weeks of active treatment


       or placebo; then had an 8-week run-out; were


       crossed over to the opposite trial; and then a


       following run-out arm.


                 So, a complex study design that allowed


       each patient to serve as their own control of being


       on scheduled albuterol versus placebo and using the


       backup rescue.  These are individuals who were


       about 31 years of age, had fairly normal FEV1's of


       about 90 percent predicted and were matched in


       pairs on the basis of the genotype of interest.


                 This is the data as published in Lancet. 



       This shows the curves of either the albuterol


       modeled or raw means data versus the placebo


       modeled and raw means data.  In particular, if you


       can look at the left-hand side of the slide, this


       is the Arg/Arg group.  The right-hand side is the


       Gly/Gly group.


                 Let's look at the Gly/Gly group first.  If


       you look at the Gly/Gly patients in the orange line


       on the top, you can see that, as you would expect,


       those patients on albuterol scheduled therapy


       improved by their morning peak flow during the


       course of the study.  In contrast, during the time


       they received placebo, in green, they really showed


       no improvement in their peak expiratory flow.  In


       contrast, the Arg/Arg patients behaved differently.


       In green is the placebo and you can see the Arg/Arg


       patients on placebo actually improved and those


       Arg/Arg patients on albuterol, in orange, failed to


       improve their peak flow during the course of the




                 The primary analysis with this study was


       to look at the treatment differences and the mean



       change in AM peak flow by genotype at week 16.  You


       can see that the albuterol versus placebo Arg/Arg


       patients had a difference in their mean peak flow


       of 10 L/min; the albuterol versus placebo Gly/Gly


       comparison, a difference of about 14.  Therefore,


       the treatment difference of the mean Arg/Arg minus


       the Gly/Gly was a difference of about 25 L/min,


       which is what this study was powered to find and


       what we had used in other studies to power it.  So,


       this was determined to be statistically




                 There were other outcomes that paralleled


       the change in peak flow.  This is looking at the


       difference between regular versus placebo changes


       in FEV                                         1 over the 16 weeks.  You

can see that the


       Gly/Gly subjects had an improvement in their FEV                        




       whereas the Arg/Arg patients had a deterioration in


       FEV                                    1.  The same thing could be seen with



       symptoms of an increase in the Arg/Arg patients


       versus a decrease in the Gly/Gly patients, and a


       complementary pattern of seeing a difference in


       inhaler use in the different groups, whether it be



       ipratropium as first-line rescue versus albuterol.


                 In summary, the BARGE data concluded that


       morning and evening peak flow, FEV1's, symptoms and


       rescue inhaler use improved significantly in


       Arg/Arg patients with asthma when beta agonists


       were withdrawn, and when ipratropium was


       substituted, as compared with regular albuterol


       used.  The pattern was reversed in the Gly/Gly


       patients who actually improved with regular beta


       agonist use.  The authors suggested that Arg/Arg


       patients, who are known to be one-sixth of


       asthmatics, may actually benefit from minimizing


       short-acting beta agonist use.


                 I included this study also because of the


       important caveats from the investigators and their


       conclusions.  They emphasized that this study was


       conducted in only individuals with mild disease,


       not patients with concomitant inhaled steroid doses


       and, therefore, whether this data can be


       extrapolated to more severe disease or to those


       patients who are on concomitant inhaled steroid


       doses just could not be answered by this particular



       trial, suggesting that both issues need to be


       studied more in the asthma community.


                 Obviously, the million dollar question is,


       indeed, do similar effects occur with long-acting


       beta2 agonists, and what is the impact of


       concurrent use of inhaled steroids?  Obviously, Dr.


       Chowdhury addressed the committee to really


       deliberate today to answer those questions.  I


       don't have the answers for you and, fortunately, I


       am not charged to do that.  That is your tough job




                 I would have some comments on what I


       believe to be future studies that may help you to


       answer those questions.  Much as the BAGS trial was


       hypothesis generating for BARGE, the SOCS and SLIC


       trial from the ACRN did retrospectively look at


       their two studies of long-acting beta agonists


       alone.  That was the SOCS trial that I shared with


       you, and the SLIC trial which looked at combination


       of inhaled steroid and long-acting beta agonists


       and the tapering of such.


                 The data from these two retrospective



       studies has been presented at meetings, suggesting


       that there was a signal of a same pattern of a


       difference in morning peak flow based on whether


       you were Arg/Arg or Gly/Gly at the 16 locus, and


       that the pattern with salmeterol, with or without


       the inhaled steroid, seems to be the same.


                 I carefully indicated that, indeed, these


       are retrospective studies, very small in design


       and, clearly, will be hypothesis generating for


       more robust, longer-term studies that the ACRN, and


       I believe the industry, will conduct.  Therefore,


       the ACRN now has a study called LARGE that is in


       the middle of operation that is very similar to the


       BARGE study but will look at an inhaled steroid,


       with or without the addition of a long-acting beta


       agonist, to answer the question of whether the same


       patterns in a prospective, carefully designed study


       can be extrapolated.


                 Now, we do have some data to answer the


       question on a safety issue about does regular use


       of long-acting beta agonists delay awareness of


       asthma progression or effect from recovery?  We



       have been concerned that if patients are so well


       controlled with symptoms with their long-acting


       beta agonists will they be aware that they are


       having an asthma exacerbation, or will they fail to


       recover from an exacerbation in the way that they


       expected to?


                 This is one representative study that I


       think illustrates the point.  This is the Matz


       article I showed you earlier of an accumulation of


       data from earlier published studies.  At the arrow,


       the day of diagnosis is the point in time at which


       the patient had an asthma exacerbation as defined


       by these authors.  You can see that if you look at


       the change in asthma symptom score about four days


       or so before the actual diagnosis of an


       exacerbation these individuals began to have an


       increase in symptoms, were treated in completion of


       an exacerbation satisfactorily, and you see that


       their symptoms decreased after the exacerbation.


       In this particular trial you actually see that


       there is a change in the asthma symptom score that


       was different in the two different study groups.



                 Now, one of the things that this provides


       I think is some reassuring issues that on the basis


       of symptoms patients are well able to detect a


       difference in their symptoms, and to know whether


       they are having an asthma exacerbation, and they


       recover as we expect.  There seems to be no adverse


       effect of the addition of salmeterol.  In fact,


       these patients seem, by symptoms, to recover even




                 We did the same analysis with the PACT


       pediatric trial that I shared with you just for


       interest, to do the same pattern looking at


       symptoms, the issue of albuterol use and the issue


       of peak flow.  We plotted the three arms of the


       study to look at whether the patterns were any


       different.  In relevance to you today, the patients


       who were on combination therapy as compared to


       inhaled steroid alone had no difference in their


       pattern.  So, all three groups were equally able to


       perceive symptoms of an exacerbation and to


       adequately recover in the same kind of a pattern.


       So, we are beginning to, I think, have more data



       that resolves this concern that has been raised.


                 Now, why we are here today in particular


       is to discuss the evidence for increased severe


       asthma exacerbations with long-acting beta


       agonists.  Indeed, for these studies, as Dr.


       Chowdhury outlined for you, the major issue at hand


       is, indeed, severe asthma exacerbations as has been


       defined by these trials.  I am not going to review


       them for you as you clearly have received


       preliminary information and I suspect you will have


       other members of the audience that will provide far


       better detail of these than I can do.  Suffice it


       to say that these are studies that have raised


       questions in the asthma community about the role of


       long-acting beta agonists, and my own particular


       comment on these is the fact that, whereas they are


       compelling for a signal and certainly warrant a


       very careful review of the trials of what they can


       tell us and what they cannot tell us, it is very


       difficult from these trials to discern whether


       these individuals were, indeed, using concurrent


       inhaled steroids during the course of the trial. 



       Therefore, it makes it certainly somewhat difficult


       to do a full analysis and, therefore, no questions


       are easily answered.


                 In summary, I think the committee today


       has a very important job of reconciling what I


       believe to be a very crucial question.  How do we


       all reconcile the finding of these very rare


       severe, life-threatening episodes that are reported


       in the SMART an formoterol trials with what I hope


       to have shown you is obviously the far more global


       evidence that the use of long-acting beta agonists,


       particularly in combination with inhaled steroids,


       results in a decrease of overall asthma


       exacerbations?  You all are faced with the data


       that I believe show that there is very strong


       evidence of the ability of inhaled steroids and


       long-acting beta agonists to both achieve asthma


       control and to reduce overall asthma exacerbations,


       as defined by the trials that I have shown you and


       others.  So, that piece of data needs to be kept in




                 I would comment that there clearly is more



       evidence in adults than children so most of the


       decisions made are based on adult data.  I believe


       that the remaining concerns about safety have to


       ask the question about whether, indeed, there is an


       influence of genotypic predictors, as has been


       picked up as the signal with the intermediate beta


       agonists.  I believe that we have to look at


       phenotypic predictors.


                 I think the era of treating all patients


       equally for asthma is gone and we need to gain


       insight about phenotypic predictors of responses to


       all our therapy.  I think this needs to include


       age, severity of disease, bronchodilator


       reversibility status, ethnicity and a variety of


       others.  Clearly, we have had some signals that


       there may be ethnic differences in responses to


       albuterol based on whether you happen to be Puerto


       Rican or Mexican-American.  So, we need to get more




                 We need to have larger and longer trials


       which incorporate multiple outcomes, including the


       concurrent use of inhaled steroids, and we need to



       be able to ultimately answer questions of whether


       this is a class effect of a dose effect.


                 I don't envy the committee.  I know that


       you will deliberate carefully.  And, I appreciate


       you allowing me to provide you an overview in


       anchoring your thoughts for your deliberation.


       Thank you very much.


                 DR. SWENSON:  Dr. Sorkness, I want to


       thank you for a very fine talk.  Since you are


       going to be leaving before the day is out, I wanted


       to particularly leave some time for members of the


       panel to ask you questions at this moment.  So, we


       will take questions from the panel on the talk or


       issues around it.


                        Questions for the Speaker


                 DR. MARTINEZ:  Thank you so much for that


       very, very nice presentation.  During your


       presentation you said that in the PACT trial you


       were the principal investigator within the CARE


       network.  The decision was made, you said, to use


       methacholine responsiveness as a criterion for


       inclusion into the trial and not reversibility.  I



       am trying to quote you as best as I can, because


       this could have introduced bias into the results,




                 DR. SORKNESS:  Yes.


                 DR. MARTINEZ:  Are you suggesting that


       some of the results of the studies that you have


       shown to us, including the GOAL study in which


       exacerbation was shown to be less in combination


       than in use of inhaled corticosteroids alone, may


       be explained by bias introduced by the fact that,


       for example, in the GOAL study 15 percent


       reversibility was a criterion for inclusion?


                 Or, a second question, has anybody tried


       to separate the studies in this meta-analysis that


       you presented to us between those that demanded 15


       percent reversibility and those which did not?


                 DR. SORKNESS:  It is a great question,


       Fernando, and I think it allows me to clarify my


       intent of saying that.  Clearly, because of a


       variety of reasons, whether it be historical of our


       belief that bronchodilator reversibility convinces


       us that this is, indeed, reversibly asthma and,



       therefore, the documentation of such allows


       enrollment into a clinical trial, or it convinces


       us that reversibility allows other drugs to show


       comparability.  The majority of trials, whether


       they be industry sponsored or not, clearly have


       used bronchodilator reversibility as entry


       criteria, and clearly most of that which I shared


       with you is that.  That has historically been the




                 My point in this the fact that I believe


       that there are a much broader group of asthmatics


       in the world today that don't have that much


       bronchodilator reversibility or may have very


       little and truly have asthma.  So, our assumptions


       of our outcomes in the therapies are predicated on


       the fact that we tend to enroll a fairly defined




                 I think, second to that, there is


       certainly some data from ACRN and other groups that


       bronchodilator reversibility as a phenotype clearly


       may be more predictive of response to long-acting


       beta agonists or for inhaled steroids, for that



       matter.  So, we have isolated a particular


       phenotype and enrolled them in our trials.


                 The ACRN, because of that and I think


       because of our mission of trying to more globally


       answer questions in a broader asthma population, in


       general have suggested that people can be in these


       studies whether you have a bronchodilator response


       or PC                                       20 as evidence of having asthma.

Both issues


       are collected by entry is not predicated on having


       simply a bronchodilator effect.


                 In the PACT trial I wanted to emphasize


       that I think, because of at least some concerns


       about the generalizability of the PACT results, we


       felt that PC                                                   20

predominantly was the right entry


       criteria.  Bronchodilator reversibility was


       collected.  And, clearly, the PACT data will have


       the capability of looking at both genotypic and


       phenotypic predictors of responses.  I can say that


       about the PACT data.  I haven't, Fernando, really


       been privy to know whether many of these other


       studies teased out bronchodilator responsiveness.


       So, that is my answer to the question.



                 DR. SWENSON:  Just for the record, that


       was Dr. Martinez that posed that question.  Dr.


       Sorkness, to what extent are the exacerbations, as


       they are detected in these multiple studies, based


       on the criteria of increased use of a short-acting


       beta agonist or the rescue use?  Because that seems


       pertinent to the question of whether long-acting


       beta agonists simply just, for a while, reduce the


       need for short-acting and so allow whatever


       underlying process toward exacerbation to go


       further without recognition.


                 DR. SORKNESS:  As a very general comment


       to this, it is a difficult question to answer


       simply because whether it be asthma exacerbations


       or treatment failure there is clearly, in my mind,


       not a uniform definition of either in the trials


       that have been described.  I think, in fairness,


       the vast majority of at least the mild and leading


       on to the use of prednisone exacerbations in


       general have been anchored by asthma action plans


       that have been a combination of symptoms, albuterol


       use and, on some occasions, peak flows below some



       safety criteria.  So, many of these studies have at


       least incorporated an asthma action plan of


       albuterol symptoms and peak flow leading to the use


       of prednisone.  So, I think it becomes kind of a


       composite decision that the patient makes in


       concert with the physician for those studies.


                 Having said that, the vast majority of the


       studies, at least in my mind, that have used the


       term asthma exacerbation in general have been


       defined by the need for prednisone, with or without


       in some cases either an ER visit or a


       hospitalization, but certainly the asthma


       exacerbation in many of the studies could have been


       achieved simply by the issue of prednisone by that


       action plan.  But the definitions are very variable


       and I think that does make it harder to bring all


       of these together to get the best insight.


                 DR. SWENSON:  Miss Sander?


                 MS. SANDER:  Thank you.  I need a little


       bit more information on what you just said.


       Whenever there is the term "rescue" medications


       used, that is any and all reasons not just rescue





                 DR. SORKNESS:  I am not sure I understand,


       Miss Sander.


                 MS. SANDER:  So, rescue would imply that


       they had an emergency need for that medication.


       Would it include all uses such as early


       intervention, prevention of exercise?


                 DR. SORKNESS:  In my mind, most of the


       studies have in their action plans specified that


       the use of albuterol to relieve symptoms and/or to


       treat a peak flow at a certain safety level were


       used in the definition of an action plan of going


       on to treat the exacerbation.  Most of the action


       plans in these trials, or at least the ones


       certainly from the ACRN and CARE, did not


       incorporate pre-exercise intended scheduled


       albuterol use in that paradigm.  It was strictly


       albuterol use for relief of those symptoms or


       relief of a drop in a peak flow to make it return


       to some baseline safety level.


                 MS. SANDER:  Thank you.  Also one other


       question, were there any expectant mothers in any



       of these?


                 DR. SORKNESS:  I can't say this with


       absolute confidence but I would be highly suspect


       that any of the trials were conducted that did not


       have a safety pregnancy test at entry and did not


       have some appropriate monitoring of pregnancy


       status during the trial.  The vast majority of


       studies that have been privy to even mandate that


       if a methacholine challenge procedure is being done


       at a study visit a pregnancy test be done.  There


       is a series of questions that coordinators and


       investigators ask about the chance of a pregnancy


       to make decisions as far as people continuing in


       trials.  So, I would be very surprised if


       individuals were enrolled being pregnant.


       Unfortunately, life is not perfect and I think that


       there are certainly trials where a woman became


       pregnant during the trial.  Most of the studies I


       know of, that actually required a mandated


       withdrawal because of the potential influence of


       pregnancy on stability of asthma.  So, I don't


       think there is much we can gain in insight, quite



       honestly, if that is some of what you are driving


       at.  I just don't think it exists in these trials.


                 DR. SWENSON:  Dr. Newman?


                 DR. NEWMAN:  Yes, thank you for what was a


       very clear presentation.  I wonder if you might


       comment about, from the benefit side, any


       differences in these trials based on race.


                 DR. SORKNESS:  That is a tremendous


       question.  I think the fairness of answering the


       question is that most of the trials that I am aware


       of--and I say this carefully because I don't know


       the literature in its extreme--probably did not


       have the ability to have a satisfactory subset of a


       particular racial or ethnic group to be able to


       cull out to do a reasonable racial analysis.  In


       the beta agonist trial by Drazen, et al., I know


       for a fact that because of NIH NHLBI guidelines of


       enrollment of at least a third of minority


       participants, that we did do a statistical analysis


       in that trial and it showed that the minority


       ethnic group did not do differently on any of the


       outcomes versus Caucasians, negative or benefit. 



       They had equal responses, as did actually a gender




                 I really do not know of any other trial


       that could answer your question explicitly but I


       think it is very important, especially given some


       of what we are learning about the potential role of


       ethnicity, and that mandates that we all make a far


       more serious effort for doing trials big enough


       with groups to answer the question.


                 DR. SWENSON:  Dr. Brantly?


                 DR. BRANTLY:  Dr. Sorkness, as I recall


       there were a number of bronchial biopsy studies


       using ICSs.  I don't recall any regarding using


       either long-acting beta agonists or short-acting


       beta agonists.  Do they exist?


                 DR. SORKNESS:  I am not sure I can answer


       that with comfort.  I actually do believe that


       there are bronchial biopsy studies in individuals


       on long-acting beta agonists alone and certainly on


       combination.  That is not clearly my area of


       expertise and I really think I would be remiss in


       trying to answer the question of what I know about



       those studies.  I am a clinical researcher.


       Certainly, some of my partners do those kinds of


       studies but that is clearly not an expertise that I


       would feel comfortable answering.  And, there may


       be somebody else on the committee that clearly


       knows that data far more than I.


                 DR. SWENSON:  Dr. Prussin?


                 DR. PRUSSIN:  Chris, on your last slide


       you have a note that says, "need for larger and


       longer trials which incorporate multiple outcomes."


       My question is, you know, clearly long-acting beta


       agonists decrease exacerbations and, yet, we have


       very good data that severe pulmonary events and


       death are increased.  So, you can't use a trial


       that is looking at exacerbations to answer the


       outcome that we are interested in here.  Since I


       work more in a smaller frame in terms of allergic


       disease, not large clinical trials, can you give me


       more of an idea of what you think a large clinical


       trial and multiple outcomes that we should be


       looking at for these endpoints of death,


       intubation, severe pulmonary outcomes?



                 DR. SORKNESS:  Cal, I think it is


       difficult and I will try to answer as best I can.


       I do believe we are in an era where the most


       important studies are not monotherapy in asthma


       with long-acting beta agonists but with


       combination.  So, that is the first issue.


                 I believe that whereas the SMART trial and


       some of the formoterol pivotal trials and others


       that have raised the signal of concern are helpful


       and we need to take that under consideration.  I


       find that the way that those studies were


       constructed leave me wanting more.  The methods by


       which patients were accrued; the issue of whether


       you really knew whether people were on inhaled


       steroids concurrently and were adherent with such;


       that you took into account and balanced severity of


       disease at the beginning; that you truly looked at


       what we believe clinically as the best that we can


       ask of this array of overall asthma exacerbations


       and control of disease; a year long study to deal


       with seasonality, especially in kids; looking at


       some markers of inflammation.



                 I think that we are at a stage that we


       would feel better and have more confidence in the


       risk/benefit relationship if we had those kinds of


       trials done both in adults and children, and


       particularly were able to answer in our own minds


       whether the combination together--adherence, people


       taking them, being on them, controlling for the


       issues--that we really knew what we were doing with


       those particular trials.  And, I think that is the


       best that we can do.


                 DR. PRUSSIN:  Let me just follow that up.


       The SMART trial was stopped because of difficulties


       with accrual and slow accrual.  Again, we are


       talking about a huge clinical trial.  In your


       estimation, since this is what you do, is it


       possible to do that large a trial and get the


       information in a much more rigorous way, as you are


       proposing?  I mean in terms of accrual.  Is this a


       feasible endeavor to go into?  Because we have been


       told that SMART simply became impossible to carry




                 DR. SORKNESS:  Yes, I think the reality is



       such that it is I believe, and I am investigator so


       I am asked to do these things--I think it is


       impossible in this day and age to recruit large


       enough subjects even in a multicenter study that


       are naive to either inhaled steroids or long-acting


       beta agonists at entry so that you are bringing in


       this naive population to answer the question.  I


       don't think those patients are out there anymore


       because we have done such a good job with


       guidelines and because all these trials showing


       that when you give people good medicines, by golly,


       they get better.


                 So, I think that if you, indeed, enroll a


       far broader population of phenotypes, of patients


       that have certain entry criteria, and then you


       randomized them to an inhaled steroid with and


       without a long-acting beta agonist, and followed


       them for long enough, I think those studies can be


       constructed.  And, I think that is one of the


       challenges to do and I suspect that they will be




                 DR. SWENSON:  Well, thank you, Dr.



       Sorkness.  We appreciate very much that fine talk


       and discussion.  At this point we will turn the


       program now over to GlaxoSmithKline and, to do so


       and to introduce her colleagues, Elaine Jones will


       take over.


                       GlaxoSmithKline Presentation


                             Opening Remarks


                 DR. JONES:  Good morning.  My name is


       Elaine Jones and I am Vice President of Regulatory


       Affairs at GlaxoSmithKline.  On behalf of


       GlaxoSmithKline, I would like to thank the agency


       and the advisory committee for this opportunity to


       review data pertinent to the discussion of the


       safety of long-acting beta                                              

                           2 agonists in the


       treatment of asthma.


                 Our presentation today will focus on our


       data with the inhaled long-acting beta                                  

                                                           2 agonist


       salmeterol.  As we begin the presentation today, I


       would like to set the stage by speaking first about


       the burden of asthma.  As the committee members are


       well aware, asthma is a chronic disease associated


       with significant morbidity and mortality.  In the



       United States alone asthma affects approximately 20


       million patients.  Asthma exerts a tremendous


       societal burden as evidenced by the half million


       hospitalizations and over 4,000 deaths in the U.S.


       in 2002.


                 There are many risk factors that have been


       identified that put patients at risk for an


       asthma-related death.  Some of these include


       excessive reliance on rescue medications and use of


       inhaled corticosteroids, disease severity and a


       delay in seeking care.  Ethnic origin is also an


       important risk factor, demonstrated by the fact


       that the rate of asthma deaths in African Americans


       is approximately 2.5-fold higher than that of




                 The tremendous burden of asthma has fueled


       a continual development of new medications to treat


       this disease and GSK has a long history in the


       development of respiratory medicines.  Salmeterol


       was the first inhaled long-acting bronchodilator,


       and its approval in the United Kingdom over a


       decade and a half ago represented an important



       advance in the management of asthma.


                 To date, regulatory authorities have


       granted approval to market salmeterol in over 100


       countries.  In the United States there are three


       salmeterol-containing products that have been


       approved for marketing.  These are Serevent


       inhalation aerosol, Serevent diskus and Advair


       diskus which contain salmeterol as one of its


       components.  Each one of these products has been


       approved for use in patients with asthma or COPD,


       and each of these approvals required a full


       clinical development program.


                 It should be noted that the inhalation


       aerosol formulation, which contained


       chlorofluorocarbons, has been discontinued by GSK


       as part of the phase-out of CFC-containing products


       consistent with the Montreal protocol.


                 Worldwide approvals by regulatory


       authorities have led to a great deal of clinical


       experience with salmeterol.  Over the last 15 years


       the exposure to salmeterol is the result of the use


       of salmeterol formulated as a single agent and the



       use of salmeterol formulated with fluticasone


       propionate in a single device.  In total the


       worldwide exposure is now estimated at 45.2 million




                 Based on extensive clinical experience and


       a systematic review of numerous clinical trials,


       evidence-based guidelines from the National Heart,


       Lung and Blood Institute's expert panel report


       recognize the pivotal role of long-acting beta


       agonists in the treatment of asthma.  While the


       safety of long-acting beta                                              

                           2 agonists is the topic


       of today's meeting, it is important to consider the


       safety of these medications in the context of their


       overall benefit/risk profile.  Part of the context


       is provided by current asthma treatment guidelines


       which position the use of inhaled long-acting beta


       agonists with inhaled corticosteroids as the


       preferred treatment option for patients with


       moderate to severe persistent asthma.


                 Asthma is a serious disease with


       significant morbidity and mortality and salmeterol


       has become a well-established pharmacological



       therapy in the management of this disease.  As you


       know, no medication is without risk and today's


       meeting provides an important opportunity to review


       safety data for inhaled long-acting beta agonists.


       We look forward to discussing the safety of


       salmeterol with the committee.


                 Salmeterol has been shown to be highly


       effective in the treatment of asthma and, since its


       approval 15 years ago, clinicians have accrued


       considerable experience with its use.  Based on the


       extensive body of evidence in patients with asthma,


       including 64 studies in approximately 45,000


       patients in the U.S. alone, GSK believes that


       salmeterol continues to exhibit a favorable benefit


       to risk profile.


                 Dr. Kate Knobil will now provide a brief


       overview of the efficacy of salmeterol, followed by


       a discussion of safety data.  Following Dr.


       Knobil's presentation, I will return to the podium


       to introduce the experts here with us today and


       then we will take questions from the committee.


                            Salmeterol Review



                 DR. KNOBIL:  Good morning, everyone.  For


       my presentation today I will first present a brief


       overview of the benefits of salmeterol for the


       treatment of asthma, followed by a review of the


       salmeterol safety data.  My review of the safety


       data will focus on the postmarketing safety


       surveillance studies, SNS and SMART, and the


       results from epidemiology studies of salmeterol.


       In addition, I will describe the ongoing studies


       currently being conducted by GSK to further


       evaluate the efficacy and safety of salmeterol.


       Finally, I will close with an overall assessment of


       salmeterol for the treatment of patients with


       asthma.  Given time limitations, I will not be able


       to cover all of the information that is in your


       briefing document.  However, any questions you may


       have may be addressed during the Q&A.


                 For several decades beta agonists have


       been widely used to treat bronchoconstriction.


       This slide shows the structures of albuterol and


       salmeterol, and you have seen these already today,


       and highlights the long lipophilic tail that helps



       anchor salmeterol in the beta adrenergic receptor.


       Albuterol is highly selective for beta                                  

                                                           2 receptor,


       thus having fewer cardiovascular effects than


       earlier less selective beta agonists.  Short-acting


       beta agonists are very effective but are limited by


       their relatively short duration of action of 4-6




                 This limitation was largely overcome by


       the development of selective long-acting beta                           



       agonists, such as salmeterol, which are effective


       for at least 12 hours.  In addition to having a


       longer duration of action, in vitro studies have


       shown salmeterol to be at least 50 times more


       selective for the airway beta                                           

                                   2 receptor than




                 The benefit of the longer duration of


       action of salmeterol can be seen in the data pooled


       from the two registration studies for salmeterol


       metered dose inhaler.  At the time that these


       studies were conducted regular albuterol use was a


       common treatment for asthma and so was included as


       an active comparator.  For salmeterol, shown in



       green, a single dose results in a clinically


       significant improvement in FEV                                          

                                      1 within 30 minutes,


       with maintenance of effect for at least 12 hours.


       This is in contrast to albuterol, shown in grey,


       which has a more rapid onset of effect but the


       bronchodilator effect lasts only 4-6 hours.


       Additionally, as shown on the right, the


       bronchodilator effect of salmeterol was maintained


       after 12 weeks of treatment with no diminution of


       FEV                                    1 response over time.


                 Studies of up to one year in duration have


       confirmed that the bronchodilator properties of


       salmeterol are maintained with long-term use.  In


       this study, 12-hour FEV                                                 

                   1 area under the curve, or


       AUC, was obtained after the first dose and


       following 8, 20 and 48 weeks of treatment.  For


       salmeterol the mean FEV                                                 

                   1 AUC was similar at all


       time points, and in all cases was significantly


       greater than placebo, demonstrating maintenance of


       bronchodilator effect.   In addition to important


       bronchodilator effects, salmeterol is very


       effectiveness at reducing the symptoms of asthma. 



       The data shown here are from the same two studies


       for salmeterol MDI that I showed previously.  Over


       12 weeks treatment with salmeterol resulted in a


       significant reduction in asthma symptoms scores for


       chest tightness, shortness of breath, wheezing and


       cough compared with placebo and albuterol given 4


       times daily.  Although not shown here, in these and


       other studies salmeterol also reduced nocturnal


       symptoms associated with asthma.


                 Salmeterol has also been shown to be an


       important treatment option for patients with asthma


       who are not adequately controlled on inhaled


       corticosteroids.  This landmark study by Greening


       and colleagues examined the effect of adding


       salmeterol to inhaled corticosteroid therapy, in


       this case beclomethasone, as compared to increasing


       the dose of inhaled steroids.  The addition of


       salmeterol to a low dose of inhaled corticosteroid


       was shown to result in a greater improvement in


       lung function, as shown by peak expiratory flow,


       then when compared to the higher dose of inhaled


       steroids.  In addition to the improvements in lung



       function, the use of salmeterol resulted in greater


       improvements in symptoms and rescue albuterol use.


                 The addition of salmeterol to a low to


       medium dose of inhaled steroid has also been shown


       to reduce the recurrence of asthma exacerbations.


       Shown here again is the study by Matz and


       colleagues, and was of similar design to the study


       that I showed previously.  When compared to


       increasing the dose of fluticasone propionate, or


       FP, the addition of salmeterol to the low dose of


       FP significantly increased the time to the first


       asthma exacerbation requiring oral corticosteroids.


       Further, significantly fewer salmeterol-treated


       patients experienced one or more exacerbations, 8.8


       percent compared to the increased dose of FP at


       13.8 percent of patients.


                 Another means of evaluating the patient


       benefit of a medication is to assess the impact on


       quality of life.  In this 12-week study that was


       designed to assess asthma-specific quality of life,


       patients with asthma were randomized to either


       salmeterol or placebo, with all patients receiving



       albuterol as needed to use for symptoms.


       Salmeterol MDI was shown to significantly improve


       quality of life compared with placebo, and the


       minimally clinically important difference of 0.5


       was achieved for each domain as well as the global




                 To summarize, the benefits of salmeterol


       have been well established and salmeterol has been


       accepted as having an integral role in the


       treatment of asthma.


                 I will now move on to the safety portion


       of the presentation, beginning first with the


       postmarketing surveillance studies for salmeterol.


       These studies are of interest because at the time


       of launch of salmeterol in both the U.K. and in the


       U.S. there was concern that the regular use of beta


       agonists may lead to deterioration of asthma


       control.  This was based primarily on studies of


       short-acting beta agonists, particularly fenoterol,


       that suggested worsening of asthma with scheduled


       use.  These studies could not determine a cause and


       effect relationship, however, they did bring



       significant attention to the appropriate use of


       this class of medications.


                 The first study that I will discuss is the


       Serevent Nationwide Surveillance Study, or SNS,


       which was performed in the U.K. between 1990 and


       1992.  This 16-week randomized, double-blind study


       evaluated over 25,000 patients with moderate to


       severe asthma.  The study compared salmeterol MDI


       to albuterol given 4 times daily in patients 12


       years of age and older.  Both treatments were added


       to the patient's current asthma therapy.


                 At visit 1 patients were randomized in a


       2:1 fashion to either salmeterol or albuterol.  The


       primary endpoint for SNS was combined serious


       adverse events and all medical and non-medical


       withdrawals.  This very broad endpoint was not


       restricted to respiratory events.  For this


       endpoint the percentage of events was similar for


       the salmeterol and albuterol groups.  Additional


       endpoints of interest included asthma-related


       deaths, hospitalizations and withdrawals.  Based on


       national health statistics in the U.K. and on the



       2:1 randomization, 10 and 5 asthma-related deaths


       were predicted in the salmeterol group and


       albuterol group respectively.


                 In this study, 14 asthma-related deaths


       occurred, with 12 in the salmeterol group and 2 in


       the albuterol group, resulting in a relative risk


       of 3.  This difference was not statistically


       significant but did raise concern.  The results for


       asthma-related deaths were not consistent with the


       data for asthma-related hospitalizations.  As you


       can see here, the data for this endpoint did not


       indicate an increase in risk with salmeterol.  The


       only statistically significant difference between


       the groups was seen for the percentage of


       withdrawals due to worsening asthma, with a lower


       percent observed in the salmeterol group compared


       with albuterol.


                 In light of the results of SNS, including


       the asthma-related deaths and spontaneous reports,


       GSK, in conjunction with the FDA, designed the


       Salmeterol Multicenter Asthma Research Trial, or


       SMART.  The study was initiated in 1996.  SMART was



       a randomized, double-blind surveillance study of 28


       weeks duration that was conducted at over 6,000


       sites in the United States.  Patients with asthma


       who were 12 years of age or older, with no previous


       use of inhaled long-acting beta agonists, were


       included.  All other asthma medications were


       allowed during the study.


                 SMART consisted of a single clinic visit


       at which patients were assessed for eligibility and


       then randomized to receive either salmeterol or


       placebo which was added to their usual asthma care.


       Subjects were given a 28-week supply of study


       medication and were not required to return for


       clinic visits.  Instead, patients were contacted


       every 4 weeks by phone primarily to collect


       information on serious adverse events, including


       respiratory-related events.


                 The combined endpoint of


       respiratory-related deaths or life-threatening


       experiences was chosen as the primary endpoint.


       Asthma-related death was also of interest but


       because this is a rare event the sample size



       required for this to be the primary endpoint was


       too large to be feasible.  Even with the broader


       combined endpoint, it was determined that a sample


       size of 30,000 patients would be required.


       However, after 15,000 patients were enrolled in the


       study the actual rate of primary events was found


       to be approximately half of what was expected and


       the target sample size was increased to 60,000




                 Key secondary endpoints were


       respiratory-related deaths, combined asthma-related


       deaths or life-threatening experiences, and


       asthma-related deaths, all of which were subsets of


       the primary endpoint.


                 Two independent review committees were


       involved with SMART.  They were the mortality and


       morbidity review committee, or MMRC, and the data


       safety monitoring board, or DSMB.  Each serious


       adverse event was adjudicated in a blinded fashion


       by the MMRC to determine if it was respiratory


       related and, if so, whether it was asthma related.


       The categories for this adjudication were



       unrelated, unlikely related, possibly related or


       almost certainly related.  Only respiratory- and


       asthma-related events considered possibly related


       or almost certainly related comprised the primary


       and secondary endpoints.  The DSMB met regularly to


       evaluate blinded aggregate data which included the


       cases adjudicated by the MMRC.


                 An interim analysis was planned when


       approximately one-half of the patients had been


       enrolled.  At the interim analysis the study did


       not reach predetermined stopping criteria, however,


       there was a suggestion of worse outcomes in


       salmeterol-treated patients, especially African


       Americans.  For this reason, the DSMB recommended


       that ideally the study should be completed within 2


       years or, if that was not possible, the study


       should be terminated and the results disseminated.


       Following discussions with the DSMB, GSK made the


       decision to stop the study due to difficulties in


       enrollment and the findings in African Americans.


                 I will now move on to the results of


       SMART.  Overall, the baseline characteristics of



       age, sex, ethnic origin and baseline peak


       expiratory flow were well matched between the


       treatment groups.  Approximately 70 percent of the


       population was Caucasian and 18 percent was African


       American.  For reference, approximately 15 percent


       of the patients with asthma in the United States


       are African American.


                 Asthma medications were reported at


       baseline and were similar between the treatment


       groups.  The most commonly reported asthma


       medications were inhaled or oral beta agonists


       which were reported in over 90 percent of patients.


       Forty-seven of the patients reported use of inhaled


       corticosteroids at baseline.


                 While baseline characteristics were


       similar between the treatments for the total


       population, this was not the case when comparing


       baseline characteristics between Caucasians and


       African Americans.  The baseline characteristics


       indicate that African Americans had more severe


       asthma as measured by peak expiratory flow,


       nocturnal symptoms, and history of hospitalizations



       and intubations.  For example, the proportion of


       African Americans reporting a hospitalization for


       asthma during the previous 12 months was more than


       twice the percentage reported for intubation for


       asthma in their lifetime.  In addition to these


       markers of increased severity in African Americans,


       the reported use of an ICS at baseline was lower


       than that in Caucasians.


                 The results for the primary and key


       secondary endpoints will be shown on this slide.


       Due to the amount of information, I will take a few


       moments to summarize the data.  These figures are


       also available in your briefing document for




                 First let me orient you to the slide.  The


       relative risk point estimate and corresponding 95


       percent confidence intervals for the primary and


       secondary endpoints will be displayed graphically.


       The values that correspond with these data will be


       shown on the right side of the slide.  The total


       population will be represented in yellow, the


       Caucasian subgroup in green, and the African



       American subgroup in orange.


                 I will start by showing the results for


       the total population as this was the primary


       analysis.  Then I will show the results for


       Caucasians and African Americans as this post hoc


       analysis was requested by the DSMB at the time of


       the interim analysis.


                 The number of primary events, combined


       respiratory-related death or life-threatening


       experiences, was approximately two-thirds of what


       was expected.  The primary endpoint for the total


       population, as shown here on the slide, was not


       statistically significantly different between


       treatment groups as the confidence interval


       includes 1.  As I review the key secondary


       endpoints for the total population, which are


       respiratory-related deaths, combined asthma-related


       deaths or life-threatening experiences and


       asthma-related deaths, it is important to remember


       that each is a subset of the primary endpoint.


                 For the secondary endpoints, statistically


       significant differences were observed between



       treatment groups for the total population,


       including asthma-related death which I will discuss


       in more detail in a moment.  The numbers of primary


       events in the Caucasian subgroup, shown in green,


       were similar between the treatment groups.


       However, in the African American population, shown


       here in orange, a significantly greater number of


       primary events occurred in the salmeterol treatment




                 For the key secondary endpoints the


       relative risk of events was higher in African


       Americans compared with Caucasians.  In particular,


       a significantly greater number of combined


       asthma-related deaths or life-threatening


       experiences occurred in the salmeterol group in the


       African American population, while there was no


       difference between treatment groups in the


       Caucasian population.


                 The number of asthma deaths in SMART was


       approximately half of what was expected.  There was


       a significantly higher number of asthma-related


       deaths seen in the overall population for patients



       receiving salmeterol compared with placebo and the


       same pattern was seen in the ethnic subgroups.


       While the relative risk of asthma deaths appears


       similar between ethnic groups, note that there were


       approximately 4 times as many Caucasians in this


       study than African Americans.  Therefore, the rates


       for all asthma-related endpoints were much higher


       in the African American population.


                 The effect of inhaled corticosteroids was


       also of particular interest to the DSMB at the time


       of the interim analysis.  A post hoc analysis was


       conducted to explore the association of baseline


       use of ICS with the primary and key secondary


       outcomes.  As I mentioned previously, 47 percent of


       the patients reported using inhaled steroids at


       baseline.  The results for the total population are


       shown here, again in yellow, for reference.


       Results for subjects reporting inhaled


       corticosteroid use at baseline will be shown in


       blue, and those not reporting ICS use at baseline


       will be shown in white.


                 For subjects reporting ICSs at baseline



       there were not statistically significant


       differences between the treatment groups for the


       primary and secondary outcomes.  Patients receiving


       salmeterol who did not report the use of inhaled


       corticosteroids at baseline, here in white,


       experienced significantly more combined


       asthma-related events than those receiving placebo.


       The number of deaths in those patients not


       reporting inhaled corticosteroid use at baseline


       was 9 in the salmeterol group versus zero in the


       placebo group so direct calculation of relative


       risk cannot be performed.  In the patients


       reporting corticosteroid use at baseline the


       numbers were 4 and 3 respectively.


                 Although SMART was not designed to assess


       the effects on inhaled corticosteroid use, these


       data suggests that ICS may have had a beneficial


       effect on asthma outcomes in SMART.


                 Finally, the data were analyzed by both


       ethnicity and inhaled corticosteroid use reported


       at baseline.  Caucasians are shown, again, in green


       and African Americans in orange.  Patients



       receiving inhaled corticosteroids are represented


       by solid lines while dotted lines represent


       patients not reporting inhaled corticosteroid use


       at baseline.  The relative risk for the primary


       endpoint was higher in African Americans than


       Caucasians, and those not reporting inhaled


       corticosteroids at baseline had higher relative


       risks within those populations.


                 Similar to the primary endpoint, the


       relative risk for combined asthma-related events


       was higher in the groups that did not report


       inhaled corticosteroids at baseline independent of




                 If we focus specifically on the number of


       asthma-related deaths, shown here at the bottom of


       the slide, it is evident that these events were


       rare.  There were more asthma-related deaths in


       patients receiving salmeterol who did not report


       ICS use at baseline in both Caucasians and African


       Americans.  Again, direct calculation of relative


       risk cannot be performed for asthma-related deaths


       for patients not reporting inhaled corticosteroids



       at baseline since there were no deaths in the


       placebo group for this endpoint.


                 SMART was not designed to determine the


       effect of inhaled corticosteroids and ethnicity on


       these endpoints, and the number of events in each


       subgroup is quite small.  Therefore, these data


       should be interpreted carefully.


                 In summary, there were more events,


       including asthma-related deaths, reported in the


       patients receiving salmeterol.  There was also a


       suggestion that both African Americans and patients


       who did not report using inhaled corticosteroids at


       baseline had a higher risk of asthma-related


       events.  However, the number of events in SMART was


       lower than expected, preventing definitive


       conclusions from the data.


                 A careful review of the data did not


       reveal any clear explanation of the results.


       Possible explanations include a direct


       pharmacologic effect of salmeterol; the presence of


       polymorphisms in the beta receptor gene; or


       patient-related factors, including a delay in



       seeking medical care.  It is well accepted that the


       prevalence of patient-related risk factors is not


       equally distributed across ethnic groups so the


       differences in outcomes seen between the ethnic


       groups in SMART may be associated with disparities


       in access to medical care and asthma management and


       may not reflect biological differences between the


       groups.  Unfortunately, none of these hypotheses


       can be confirmed or refuted by the data from SMART.


       While there are no clear explanations for the data,


       the findings were communicated to physicians to


       allow for informed treatment decisions.


                 In collaboration with the FDA, a number of


       activities were undertaken to communicate the


       results in order to inform physicians about SMART.


       On the day the study was stopped a "dear healthcare


       professional letter" was delivered by overnight


       mail to the 229,000 healthcare professionals in the


       United States who had prescribed salmeterol or


       salmeterol-containing products within the previous


       year.  Simultaneously, notices on both the FDA and


       GSK web sites were posted.



                 A second letter was sent out to health


       professionals when the prescribing information for


       Serevent and Advair was changed to include the


       preliminary results of the interim analysis of


       SMART.  The information was elevated to the highest


       level of prominence in the form of a boxed warning.


       When the final results were obtained the labeling


       for both products was updated.


                 This is the boxed warning that was added


       to the prescribing information for Serevent and


       Advair.  It describes the final results of the


       interim analysis of SMART.  For your reference, the


       full label, including the boxed warning, is


       available in your briefing package.


                 The results of the interim analysis were


       presented at the American College of Chest


       Physicians meeting as a late-breaking abstract.


       This was the first available national meeting with


       high attendance of respiratory physicians.  The


       manuscript is now in press at Chest, the journal of


       the American College of Chest Physicians.


                 Epidemiology studies offer an additional



       method to investigate associations between drug


       exposure and serious outcomes.  The major advantage


       of these studies is the utilization of


       comprehensive medical and pharmacy databases.


       These databases allow identification of a greater


       number of events than can be achieved in


       traditional randomized clinical trials.  The


       primary limitation of observational studies is the


       fact that assignment of treatment is not random and


       treatment effects may be confounded by differences


       in baseline characteristics, including co-morbid


       disease, differences in asthma severity and


       selective prescribing.  Since many more events can


       be evaluated in an observational design, this may


       be a more informative way to assess treatment


       effects on the rare endpoint of asthma-related




                 This figure displays the relative risks


       from all large published cohort and case-control


       studies that evaluated whether salmeterol use was


       associated with the occurrence of severe


       respiratory and asthma-related outcomes.  The



       dotted line represents a relative risk of 1.


                 The first study determine the relative


       risk of respiratory-related death among patients


       with asthma receiving salmeterol compared with


       those receiving theophylline on the left side of


       the highlighted area, or those receiving


       ipratropium, which is on the right side of the


       highlighted area.


                 The second study, which was conducted in


       the United States, evaluated three endpoints,


       asthma-related emergency room visits,


       hospitalizations and ICU admissions, comparing


       salmeterol with theophylline recipients.


                 The last two were separate case-control


       studies that evaluated the relative risk of


       asthma-related ICU admission or asthma-related


       death associated with salmeterol use relative to no




                 This last and most recent study, shown


       here on the far right, included 532 pairs of asthma


       deaths and matched controls and is the largest


       case-controlled study evaluating asthma-related



       death ever conducted.  Notably, none of these


       studies showed a significant increase in the


       relative risk of these serious outcomes for




                 GSK is committed to a comprehensive


       research plan to further evaluate the safety and


       efficacy of salmeterol.  We believe that these


       currently ongoing studies will provide valuable


       information regarding the safety and efficacy of


       salmeterol in patients with either asthma or COPD.


       In order to address some of the issues raised by


       SMART, two studies are under way.


                 The first is a year long clinical study


       evaluating the incidence of asthma exacerbations in


       460 African American subjects.  Results are


       expected in 2007.  The second is an epidemiology


       study utilizing data from 7 Medicaid plans to


       examine racial variation and association of


       asthma-related prescription medication use with


       asthma morbidity and mortality.  The results are


       expected in 2006.


                 Studies have suggested that response to



       short-acting beta agonists may be affected by


       genetic polymorphisms in the beta                                       

                                              2 receptor.


       However, there is one study that has suggested


       there is no similar association with salmeterol and


       clinical outcomes including exacerbations.  This


       was the Taylor study that Dr. Sorkness showed


       earlier.  Since there were no genetic samples


       collected in SMART we are conducting two studies to


       address whether clinical outcomes in patients


       receiving salmeterol are affected by genotype.


                 The first study is a 38-week clinical


       trial evaluating response by beta                                       

                                              2 receptor


       genotype in 540 subjects with asthma.  The results


       are expected in 2007.  The second study will


       evaluate polymorphisms in beta                                          

                                      2 adrenergic


       glucocorticoid pathways with respect to clinical


       response in approximately 1,000 subjects from


       completed GSK clinical trials.


                 Finally, while asthma has been the focus


       of today's discussion, there are ongoing studies in


       patients with COPD that will help address whether


       the results of SMART are relevant for patients with



       COPD.  The first is a 3-year study of all-cause


       mortality in approximately 6,200 subjects with


       COPD.  Results from this study are expected in


       2006.  In addition, we are conducting 2 year-long


       replicate studies examining the rate of moderate to


       severe COPD exacerbations, with results expected in




                 Asthma is a serious chronic disease with


       significant morbidity and mortality.  Salmeterol is


       one of the most thoroughly studied medications for


       asthma and has been shown to provide substantial


       therapeutic benefit, including improvements in lung


       function, and asthma-related quality of life, and


       reduction in symptoms, rescue medication and asthma




                 The extensive clinical trials have led


       evidence-based asthma treatment guidelines to


       recommend long-acting beta agonists with ICS as the


       preferred option for patients with moderate,


       persistent asthma.  There are conflicting data for


       salmeterol.  SMART and SNS suggest that salmeterol


       may be associated with an increased risk of rare



       serious asthma-related events including


       asthma-related death.  But when large cohorts of


       patients are evaluated in epidemiology studies this


       association is not observed.  The low number of


       serious asthma events in SNS and SMART does not


       allow for definitive conclusions, and the fact that


       the events of concern are also those that are


       experienced by patients with asthma, regardless of


       treatment, makes assessment of cause and effect


       relationships difficult.


                 Ultimately, what are the implications of


       the data for patients with asthma?  Well, specific


       treatment decisions for an individual patient can


       only be made by their physician.  It is our


       responsibility to provide the complete information


       so that the physician can make well-informed


       treatment decisions.  We have done this in the


       prescribing information for products containing




                 From the time that salmeterol was


       introduced in the United States in 1994 the


       prescribing information has provided specific and



       appropriate guidance on its use.  This includes


       that salmeterol should not be used to treat acute


       symptoms.  It is not a substitute for inhaled or


       oral corticosteroids, and consideration should be


       given to adding anti-inflammatory agents, for


       example corticosteroids.


                 In 1995 further information was added,


       including a warning that salmeterol should not be


       initiated in patients with significantly worsening


       or acutely deteriorating asthma.  As I have already


       mentioned, detailed information on the rare


       asthma-related events seen in SNS and SMART had


       been incorporated in the prescribing information so


       that informed treatment decisions can be made.


       This includes the boxed warning, as well as other


       language to address the inconclusive nature of the


       results and the potential for a class effect of


       inhaled long-acting beta agonists.


                 In conclusion and based on the weight of


       evidence, we firmly believe that salmeterol remains


       a valuable medication that has improved the level


       of care for patients with asthma and COPD. 



       Additionally, GSK is committed to further research


       that will not only help better characterize the


       efficacy and safety of salmeterol but will also


       help better understand asthma in general.  There is


       a large volume of data from clinical trials of


       salmeterol, as well as extensive clinical


       experience with this medication.  Taken together,


       these continue to support a favorable benefit to


       risk profile and, therefore, salmeterol should


       remain available to physicians and patients.


                 I thank you for your time and I will now


       turn the podium back over to Dr. Jones.


                             Closing Remarks


                 DR. JONES:  I would like to introduce


       three additional experts here with us today.  Prof.


       Richard Beasley is the director of the Medical


       Research Institute of New Zealand.  He is also an


       international expert on beta agonist safety and


       asthma epidemiology.  Dr. Eugene Bleecker is


       professor and section head, Pulmonary, Critical


       Care, Allergy and Immunologic Diseases at Wake


       forest University Health Sciences.  Dr. Bleecker is



       also the co-director of the Center for Human


       Genomics, and was a member of the MMRC for SMART.


       Finally, Dr. George O'Connor is professor of


       medicine in the Division of Pulmonary and Critical


       Care Medicine at Boston University School of


       Medicine, and is director of the Adult Asthma


       program at Boston Medical Center.  In addition, he


       was the chairman of the DSMB for SMART.


                 We would be happy to address any points of


       clarification and questions.  Thank you.


                        Questions by the Committee


                 DR. SWENSON:  Dr. Schatz?


                 DR. SCHATZ:  I think one possible


       hypothesis based on the SNS study, where there was


       increased withdrawal due to asthma in the placebo


       patients, is the possibility that there ended up


       being an imbalance in severity by the end of the


       study due to disproportionate withdrawal of more


       severe patients from the placebo group.  That is


       obviously not so easy to tease out but I would


       question whether, in fact, one looked at baseline


       severity in those who withdrew versus those who



       didn't in both groups but particularly in the


       placebo group.


                 DR. KNOBIL:  For SNS we don't have that


       cut of the data to provide for you, but it would


       probably make a lot of sense that the patients who


       withdrew were more severe.


                 DR. SCHATZ:  But for SMART data--


                 DR. KNOBIL:  Oh, for SMART we saw the same


       thing in that there was greater withdrawal in the


       placebo group than in the salmeterol group but,


       again, we don't have that cut of the data for you




                 DR. SWENSON:  Dr. Gardner?


                 DR. GARDNER:  I have two questions, one


       related to measures of adherence within or between


       the groups and whether anything has been done with


       that.  The second is would you give us the status


       of the ongoing studies at this time?  You have


       listed four with expected results.  Can you tell us


       the enrollment at this time; how far along you are?


       I understand the word "commitment" but can we see


       something about progress at this time and status?



                 DR. KNOBIL:  All of the studies that I


       mentioned to you are right now currently enrolling.


       As you might imagine, enrolling limited populations


       of only African Americans takes a little bit longer


       than general populations.  As well, in the genetic


       studies we are making sure that we have balanced


       groups with the different genotypes.  So, that does


       take some time.  So, what I can tell you is that


       they are enrolling but I can't tell you when they


       will be done enrolling.


                 DR. GARDNER:  And adherence?


                 DR. KNOBIL:  Oh, I am sorry.  During the


       monthly telephone contacts we did ask the


       question--if you could show the slide, please?  On


       a scale of 0-10, with zero meaning you missed all


       of your doses and 10 means you took all of your


       doses, what number represents how well you followed


       the study physician's instructions?  In the study


       the mean response, patient reported response was 8


       for each group.  Again, this is patient reported


       and patients did not return to the clinic so we


       cannot verify this.  We did not ask them to bring



       their cans in.  We did not weigh canisters and we


       didn't have a chronolog.  But I think from studies


       of chronolog, we know that patients sometimes


       report taking more medication than they actually


       do.  So, I can't really verify the actual


       compliance of the patients.


                 DR. GARDNER:  As far as you can tell, was


       there similar adherence between the African


       American subjects and the Caucasian subjects?


                 DR. KNOBIL:  Yes, as far as can tell there


       was no difference between any specific group.


                 DR. GARDNER:  Finally, on the Medicaid


       study, that doesn't require enrollment.  Can you


       tell me where that is being done and how far along


       that one is?


                 DR. KNOBIL:  Yes, that one is being done


       in seven states.  What we are doing right now is


       the first phase of the study, which is to see if we


       can identify enough patients with high enough


       proportion of African Americans to make an analysis


       feasible.  So, that is where that study is right





                 DR. SWENSON:  Dr. Moss?


                 DR. MOSS:  I have two separate questions.


       The first one has to do with the dissemination of


       the information in your letters and the labeling


       revisions.  I think the goal of this meeting is to


       disseminate information properly to the physicians


       and the community.  The question I have is when you


       sent out those letters and put on the box labels,


       do you know if that changed the prescription


       practices for the physicians that received those


       letters or for the general community in terms of


       the prescription of your medication?


                 DR. KNOBIL:  I don't know if that


       information changed the way physicians prescribe


       the medications.  We don't have any specific data


       to that effect.


                 DR. MOSS:  Did overall use of your


       compound decline after those letters were sent out?


                 DR. KNOBIL:  The rate of use did not


       change after those letters were sent out.  But, you


       know, we can't guess what would have happened; all


       we saw was no change.



                 DR. MOSS:  You saw no change.  I think


       that raises a concern about, you know, we are


       trying to disseminate information and are we doing


       that in the proper way?


                 The next question I have may be answered


       by you but might also be answered by either Dr.


       Bleecker or O'Connor.  I think it is really hard to


       figure out why someone died.  Taking care of people


       I see this.  It is really hard to define


       specifically what a cause of death is.  I was just


       wondering if you can give us some insight into how


       you define respiratory versus asthma deaths in the


       SMART study.


                 DR. KNOBIL:  Dr. Bleecker, would you like


       to address that?


                 DR. WEAN:  While Dr. Bleecker is coming


       up, I want to get back to the earlier question you


       raised.  I am David WEAN, Senior Vice President of


       Regulatory Affairs at GlaxoSmithKline.  I don't


       think it appropriate to say that because we can't


       posit a change in prescribing habits because of the


       letters and the label that they were not effective.



       The important thing is that the information was


       disseminated in a very large way to prescribers


       and, thereby, to patients.  To expect that you


       would see a drop-off in use of these drugs that


       have therapeutic benefit because of that


       communication I think would be an inappropriate


       assessment about the effectiveness of that




                 DR. MOSS:  But I think one thing we have


       learned is that publishing articles and papers does


       not get information out to the people that need the


       information to be received.  In the same way, I am


       not sure that sending letters out to physicians who


       get a lot of mail is an effective way of


       communicating information.


                 DR. SWENSON:  Dr. Bleecker?


                 DR. BLEECKER:  I was asked to talk a


       little bit, and George O'Connor could complement on


       how the mortality review committee in SMART


       adjudicated cases.  I agree, it is often very


       difficult, and as we did this we probably all


       learned.  I joined the mortality review committee



       after about a third of the cases had been done.


       The members of the committee before that had been


       Roland Ingrham who was professor of medicine at


       Emory.  He had retired.  The chairman of the


       committee was Hal Nelson who is a professor of


       medicine at Colorado, and the third member was


       Scott Weiss who is professor at Brigham and


       Williams and also head of their pulmonary and


       respiratory epidemiology program.


                 We had data on most patients available


       both from death certificates and from the medical


       monitors who worked with Covance.  We used that to


       answer questions which were related to the cause of


       death; was this respiratory related; and you heard


       before the likelihood or unlikelihood of that


       during Kate Knobil's presentation; and then was it


       asthma related.  All three of us adjudicated this


       independently.  If we agreed on all of these


       characteristics the case was not discussed further.


       All of the cases in which there was any


       disagreement, ranging even between "unrelated" and


       "unlikely" were discussed in detail in timely



       conference calls.  I think at times we did the best


       we could on relating to that.


                 The least amount of information was


       available on deaths toward the end of the study


       that were picked up from the national death


       registry.  On those deaths we had to rely on death


       certificates or more limited information.


                 DR. MOSS:  Can I just follow-up on that?


       Can you explain a little bit how you differentiate


       asthma from respiratory deaths?  A respiratory


       death that is not asthma, is that pneumonia?  What


       were criteria to differentiate those specific




                 DR. BLEECKER:  Well, often asthma entered


       into those deaths.  Let me give you an example of


       respiratory death.  Because someone during an auto


       accident had trauma to the chest--and this did


       occur in some of the younger individuals--and died,


       and that clearly was related to an automobile


       accident and because of the nature of the event and


       other things was not related to asthma.  It was


       more difficult if someone entered the hospital with



       pneumonia, was intubated and in an intensive care


       unit.  I think under those circumstances, some of


       those deaths were adjudicated depending on the


       course on the ventilator or those serious events as


       possibly related to asthma.  So, at times it is


       very hard to distinguish that from the records.


       Again, I think the fact that they were performed


       independently and you needed to look for


       concurrence and discussion, I think a reasonable


       approach was performed.


                 DR. SWENSON:  Dr. Gay?


                 DR. GAY:  Based on the appropriate


       emphasis that you have begun to make on genetic


       testing, as we have seen based on the information


       that Dr. Sorkness elegantly presented before, do


       you have any preliminary estimates of what you feel


       would be the prevalence of Arg/Arg gene


       presentation in patients with greater severity of


       asthma or based on ethnic differences?


                 DR. KNOBIL:  Yes, I would not be able to


       predict the different prevalences of those genetic


       subtypes and I would ask Dr. Bleecker to comment on



       that.  The one thing I would add is that in the one


       genetic study we are making sure that there are


       equal numbers of each genetic subtype so that we


       don't have a predominance of one and very few of


       another.  Dr. Bleecker?


                 DR. BLEECKER:  I would like to add a few


       comments to that because I think there are some


       important aspects of this.  First of all, we have


       centered on basically one variation within the


       beta2  adrenergic gene.  Looking at that gene more


       carefully--and there have been published studies on


       this, especially from the Liggett group as well as


       some work from our laboratory which has been


       presented at last year's ATS and Academy of Allergy


       meetings--there is a good deal more variation in


       that gene, and there are relationships between the


       arginine genotype and some of that other variation.


       Some of that may be very important in trying to


       sort out the hypothesis of whether variation in


       this gene affects therapeutic response and


       potentially affects outcome.


                 The second important issue is there is



       more variation, and African Americans have a higher


       prevalence of the Arg/Arg genotype, about 22


       percent, and that is what was seen during the


       screening for the ACRN BARGE trial versus about


       12-14 percent in Caucasians.  The implications of


       that on outcome are difficult, and I think it is


       very important that the studies that were outlined


       by Dr. Knobil on studying specifically an African


       American cohort in which they are going to look at


       outcome such as exacerbations and genotype are


       critical because those kinds of studies are not


       being done because of the limitations in


       recruitment by the NIH NHLBI asthma clinical




                 DR. SWENSON:  Dr Prussin?


                 DR. PRUSSIN:  I have a similar question


       that I raised before with Dr. Sorkness.  You have


       some very nice studies that are undergoing, but do


       you think they are going to address the question at


       hand in terms of asthma deaths or severe pulmonary


       endpoints?  They are fairly small studies relative


       to the SMART study and, when all is said and done



       in three or four years, it is not clear to me at


       least that you are going to have any kind of handle


       on asthma death.  Could you comment on that?


                 DR. KNOBIL:  Yes, as I mentioned, it is


       very difficult to prospectively study


       asthma-related death because the rate is relatively


       low and you need a very large N to get conclusive


       results.  The one study that will help us get there


       though, I believe, is the Medicaid study in that we


       can get information from the 7 Medicaid plans and


       look at the different racial subgroups, look and


       see what medications they were on and see what


       contributed to those patient's deaths, whether it


       is a long-acting bronchodilator, short-acting


       bronchodilator or some other related medication.


       So, I do believe that in that observational design


       we will be able to get some more information about


       asthma-related death.  The other studies are mainly


       going to address asthma exacerbations and other


       responses such as lung function, rescue albuterol


       use, and we will be able to look at those in


       relationship to genetic makeup as well.



                 DR. PRUSSIN:  The difficulty I have with


       that though is that there is a lot of data showing


       that salmeterol improved asthma exacerbations.  You


       have shown that.  Clearly, the more severe


       endpoints of death are tracking differently.  So,


       you are thinking of it as the tip of the iceberg,


       that however exacerbations are going to track


       asthma deaths are going to track and clearly they


       are behaving differently.  So, just because you


       have certain data on asthma exacerbations in


       certain subgroups, that may not be proportional or


       relate to asthma.  We don't have any prospective


       studies ongoing or planned to really address the


       endpoint that I think this committee has been asked


       to address, which is asthma death.  So, it could


       very well be a different phenomenon than what is


       causing asthma exacerbations.


                 DR. KNOBIL:  That is true, and there are


       other factors beyond asthma treatment that may be


       contributing to asthma-related death as well,


       including access to care or how a patient's asthma


       is managed.  We don't know the answers to that



       either.  So, that is why in order to actually even


       look at asthma-related death an observational


       design is probably going to provide more


       information more quickly.


                 I take your point that it seems that a


       prospective study would be the gold standard.  The


       issue there is that if the rate of asthma-related


       death was similar to what we saw in SMART, in order


       to see an effect you might have to have a study as


       large as 800,000 patients.  As you can see, that


       would be very difficult to do.  So, yes, it would


       be nice to be able to do it prospectively but it is


       going to be more difficult than doing it in an


       observational design.


                 DR. JONES:  Actually, I think Dr. Beasley


       wanted to make a comment.


                 DR. BEASLEY:  Yes, in response to that I


       would like to caution in terms of considering a


       prospective clinical trial as being the gold


       standard when you are looking at a rare outcome


       such as mortality.  I think we saw that in the


       SMART study, where it was required to study



       approaching 30,000 subjects to obtain around 35


       respiratory-related deaths, and there was a real


       compromise in terms of design of the study to


       actually achieve that and individuals were given 7


       canisters at the beginning of the study without any


       formal clinical follow-up, which is something which


       would not be done on label in terms of salmeterol


       therapy. In terms of the other issues of the label,


       many of the subjects had asthma severity where


       salmeterol would be inappropriate as sole therapy.


                 So, I think that when you try a


       prospective controlled trial to look at a rare


       outcome such as death you often have to incorporate


       a methodology that clearly is outside the spectrum


       of what is recommended clinical practice.  That


       clearly happened with the SMART study so that we


       have to consider the SMART study results not


       applicable to what would be considered good


       clinical practice, and the alternative is actually


       to increase the number of deaths through


       case-control methodology and that was the method we


       used in New Zealand to identify the risks



       associated with fenoterol.


                 In that regard, I think that the Anderson


       study in the BMJ is considerably more powerful in


       terms of looking at the role of bronchodilator


       therapy and the rare outcome of asthma mortality.


       They were able to look at over 500 deaths, match


       them with subjects who came from the same


       proportion of the population of patients in terms


       of severity, who were subjects with a previous


       hospital admission, and then they were able to


       stratify their analysis by looking at an even more


       severe subgroup.  When they did that there was


       actually no increased risk associated with the use


       of salmeterol therapy.


                 So, I think that in terms of the


       epidemiological approach to a rare outcome of


       asthma mortality looking at the role of medication


       use, the epidemiological view is very clearly that


       the case-control methodology is more powerful and


       more accurate than a prospective clinical trial.  I


       think in some respects almost the learning point


       from the experience with the SMART study was the



       compromise that had to be obtained in terms of


       clinical practice to achieve a study which, even


       with 30,000 people, did not have sufficient power.


                 DR. SWENSON:  Dr. Schoenfeld?


                 DR. SCHOENFELD:  Do you have a tabulation


       for the whole group and for each of these subgroups


       and for each of the endpoints  of the absolute risk


       or the attributable risk?  Because from a


       risk/benefit point of view the relative risk isn't


       very informative because, of course, you can have a


       3-fold relative risk of a very, very rare event and


       that may be important if you are judging something


       like an environmental pollutant that you can remove


       but is not really important when you are judging a


       very effective drug which improves people's quality


       of life.  So, I wonder if you have that tabulated.


       I have done some back-of-the-envelope calculations


       but it would be helpful to have the actual numbers


       where we looked at the percent of deaths per


       patient-year or number of deaths per 1,000


       patient-years, or something of that sort that would


       be attributable, assuming that there is some



       attributable risk to the use of the drug, or even


       just the total risk among asthma patients per 1,000


       patient-years or 10,000 patient-years, or whatever.


                 DR. KNOBIL:  Unfortunately, we don't have


       those data.  We have not done those calculations.


                 DR. SWENSON:  Dr. Martinez?


                 DR. MARTINEZ:  Would you please clarify


       what the entry criteria were for both studies, the


       British one and the one done in the United States?


       How was asthma defined?  Were any of the usual


       parameters--response to beta                                            

                                2 agonists or


       methacholine used to define asthma in these two




                 DR. KNOBIL:  These studies were a little


       different than normal clinical trials, say, to


       register a medication for asthma.  It was the


       opinion of the investigator if the patient have


       asthma.  They did not have to show reversibility.


       There was no smoking criterion.  In this case they


       had to be 12 years of age or older.  In SNS they


       had to have moderate to severe asthma and in SMART


       they were not allowed to have concomitant



       administration of beta blockers.  That is about




                 DR. MARTINEZ:  Was response to


       bronchodilators measured?


                 DR. KNOBIL:  It was not measured.


                 DR. SWENSON:  Miss Sander?


                 MS. SANDER:  Yes, when we are looking at


       possible causes of death, are you able to know if


       patients who died were using salmeterol as if it


       was an acute bronchodilator?


                 DR. KNOBIL:  In SMART the only question we


       asked was if the patient was using the medication


       as the physician told them to.  That is the data


       that I showed you earlier.  We do not have any data


       on whether they were using it as a rescue




                 DR. SWENSON:  I realize that there are


       more questions to be posed to GSK but we need to


       take a break at this point.  We have a general


       question session in the afternoon and the rest of


       these questions should be addressed at that time.


       We will be back again in 15 minutes.



                 [Brief recess]


                 DR. SWENSON:  We will resume the meeting


       with Dr. Eric Floyd, from Novartis, to discuss


       formoterol and its place in this controversy.


                          Novartis Presentation




                 DR. FLOYD:  Good morning.  Dr. Swenson,


       members of the FDA advisory committee, Dr.


       Chowdhury, members of the Food and Drug


       Administration and guests, my name is Eric Floyd.


       I am Vice President and Global Head of Drug


       Regulatory Affairs for the therapeutic areas of


       dermatology, respiratory and infectious diseases.


       On behalf of Novartis Pharmaceuticals Corporation,


       I thank you for the opportunity to review the


       current safety experience today for a long-acting


       beta agonist, Foradil.


                 There have been numerous safety concerns


       expressed publicly regarding use of long-acting


       beta agonists.  The purpose of our presentation


       today is to discuss implications of recently


       available data related to the safety profile of



       long-acting beta agonists.  This morning Novartis


       would like to present to you the results and


       conclusions of our review of available safety data


       for Foradil.  Based upon our review of clinical


       trial data and postmarketing experience, we would


       like to demonstrate that formoterol exhibits a


       favorable risk/benefit profile.


                 To provide a brief regulatory overview,


       Foradil was first approved for marketing in France


       in 1990, and subsequently approved in other


       European countries.  Foradil received FDA approval


       to market in the United States in 2001.  Foradil is


       currently approved in over 80 countries worldwide


       and to date, we currently have over 13 million


       person-years of exposure.


                 Foradil Aerolizer was marketed as a dry


       powder capsule for oral inhalation and was approved


       in 2001 for a dosage regime of 12 mcg twice daily


       for maintenance therapy for individuals with asthma


       5 years of age and above for the following


       indications, acute prevention of exercise-induced


       bronchospasm in individuals 5 years of age and



       older when administered on an occasional as needed


       basis, and for chronic obstructive pulmonary


       disease, including chronic bronchitis and




                 Outside of the United States, Foradil is


       approved at 12-24 mcg twice daily and is a highly


       prescribed bronchodilator, and is also indicated


       for the maintenance therapy of asthma specifically


       in adults and children 5 years of age and older;


       for the prophylaxis and treatment of


       bronchoconstriction in patients with asthma;


       prophylaxis of bronchospasm induced by inhaled


       allergens, cold air or exercise; prophylaxis and


       treatment of bronchoconstriction in patients with


       reversible or irreversible COPD, including chronic


       bronchitis and emphysema.  In some countries it is


       also approved as metered dose inhaler and


       multi-dose dry powder inhaler, 10 mcg, Certihaler.


                 In order to provide a more detailed review


       of our current data to date, I would like to


       introduce our speakers today.  Dr. Gregory Geba,


       who is the Vice President and U.S. Head,



       Respiratory, Dermatology and Infectious Diseases,


       Clinical Development and Medical Affairs for


       Novartis Pharmaceuticals, will present the safety


       profile of Foradil.


                 He will be followed by Dr. James Donohue,


       who is Professor of Medicine, Chief of Pulmonary


       Division, University of North Carolina, Chapel


       Hill, who will present the clinical implications.


                 In addition to the key speakers, we also


       have additional advisors available to address any


       specific questions you may have.  Specifically from


       a statistical perspective we have Dr. Gary Koch,


       Professor of Biostatistics, School of Public


       Health, University of North Carolina, Chapel Hill.


       I would now like to turn the podium over to Dr.




                      Efficacy and Safety of Foradil


                 DR. GEBA:  Thank you, Dr. Floyd.  Dr.


       Swenson, committee members, members of the FDA and


       interested attendees from the community, it is my


       role to review with you all of the clinical data


       and postmarketing data we have available for



       Foradil, the other long-acting beta agonist being


       discussed today.  We firmly believe that Foradil is


       a drug that provides clinical benefit with a


       favorable benefit/risk profile.


                 As indicated in the previous presentation


       by Dr. Floyd, and later on in this presentation by


       Dr. Donohue, the clinical features of Foradil have


       led to its inclusion in both U.S. and international


       guidelines in the treatment of moderate to severe


       persistent asthma.


                 These are the key points of the


       presentation.  We will describe pharmacologic


       differences that exist between formoterol and


       salmeterol, which may or may not have clinical


       impact; a Phase 4 trial, 2307, which examined


       asthma-related serious adverse events in


       adolescents and adults; our integrated Foradil


       clinical database; and postmarketing adverse event


       data.  The totality of the evidence does not


       elevate concern for a safety signal for Foradil


       which continues to support its favorable


       benefit/risk profile.



                 These are the chemical structures of


       formoterol on top and salmeterol on the bottom.


       Please note that at the end of the molecule that


       interacts with the beta receptor, the catechol end


       of the molecule, the molecules are actually similar


       in having a hydroxyl group at position 5.  However,


       they are different at positions 6 where you see


       different side chains.  Most importantly, at


       position formoterol has a much longer allophanic


       chain, as previously shown.  Thus, although both


       molecules bind to the beta                                              

                           2  receptor, differences


       in structure allow salmeterol to bind to an


       additional site within the beta                                          

                                        2  receptor termed


       an exosite.  Prolonged salmeterol activity depends


       on binding with the exosite when formoterol's


       activity is independent of an exosite.  In


       addition, mutation of the exosite region could


       affect the duration of action of salmeterol.  One


       of the consequences of structural differences is


       that formoterol is a full agonist at the beta                           



       adrenergic receptor, whereas salmeterol is a


       partial agonist.



                 Typical experiments illustrating this


       point are performed with human bronchial explants


       which show that the bronchodilatory effects of


       isoprenaline are decreased by prior incubation of


       tissues with salmeterol but not formoterol.  The


       potential clinical implication of this difference


       is that in the setting of beta                                          

                                      2  receptor


       down-regulation the effect of rescue


       bronchodilators may be greater for full agonists


       than for partial agonists.


                 I would like to now move on to a


       discussion of the Phase 4 trial 2307.  This trial


       was recently completed and is detailed in your


       briefing book.  Why was this study done?  Protocols


       040, 041 and 049 were 3 pivotal studies conducted


       to support registration of Foradil in the United


       Sates and 040 and 041 were 12 weeks in duration and


       were conducted in adolescents and adults; 049 was


       conducted for one year in children ages 5-12.


                 Trial participants were randomized to


       receive Foradil 12 mcg BID, 24 mcg BID or placebo.


       In 040 and 041 there was also a comparison to



       regular doses of albuterol 180 mcg QID.  Please


       note that all groups were allowed to take


       additional doses of albuterol as needed for


       residual symptoms and all groups were allowed


       anti-inflammatory agents.


                 Shown here are the proportion of patients


       with asthma-related serious adverse events.  These


       studies showed more serious asthma-related serious


       adverse events in the higher don formoterol arms


       compared to the lower doses or the approved


       formoterol arm as well as the placebo arm.


                 In light of these findings, the agency did


       not approve the higher dose of Foradil.  After


       discussing this observation with the agency and


       with their guideline, we pursued a safety study


       whose primary endpoint was asthma-related SAEs.


       The inclusion and exclusion criteria for protocol


       2307 were identical to protocols 040 and 041 which


       were our pivotal trials.  Indeed, the resulting


       population studied in protocol 2307 was similar to


       that of the pivotal trials in adolescents and


       adults, which was the population studied and



       requested, as shown in this slide.


                 Apart from the trial duration which is


       different, age differences were pretty similar.


       FEV                                    1 at baseline was fairly similar.

Please note


       that the proportion of black patients in this trial


       mimicked the proportion of patients in the U.S.


       population.  There are some subtle differences in


       terms of ICS use and reversibility.  Actual


       reversibility for 2307 was slightly less than 040


       and 041 but otherwise the study populations were


       very, very similar.


                 The design of this safety study is shown


       here.  Using identical entry criteria--again, these


       are identical entry criteria to those employed in


       our pivotal studies--after a 2-week run period,


       shown here, patients were randomized to receive, in


       a double-blind fashion, one of the following


       treatments, either formoterol 12 mcg BID formoterol


       24 mcg BID--again, 12 mcg BID was the approved


       dose; 24 mcg was the higher dose.  They received


       either of those two doses or placebo in another


       group or, in an open-label group, received



       formoterol 12 mcg BID plus an additional up to 2


       rescue doses of formoterol 12 mcg BID, which


       constituted the intermediate dose arm.


                 Please note that to increase the rigor of


       this study, after 16 weeks of treatment we planned


       to contact all patients, including those who


       discontinued, to record all adverse events.  This


       assured that all patients would be evaluated for


       AEs irrespective of treatment efficacy and trial




                 Results of the study are shown here.


       Please note that there was a correction made to the


       briefing book provided by the agency.  The lower


       dose arm had a somewhat higher number of patients


       who reported serious asthma-related AEs.  However,


       after review of the specifics of these cases, the


       agency excluded 2 of these events in the Foradil 12


       mcg arm, reducing that number from 5 to 3.  Thus,


       the final event rates were 0.2 percent in the


       placebo group; 0.6 percent in the low dose group;


       0.2 percent in the intermediate dose group; and 0.4


       percent in the high dose group.  Overall, there



       were far fewer events than expected based on the


       pivotal trials.


                 Displayed on this slide are the point


       estimates and 95 percent confidence


       intervals--which I hope you can see as red on a


       blue background--for the proportion of patients


       experiencing asthma-related SAEs in each treatment


       group.  As previously mentioned, the rates are low


       for all treatment groups.  The 95 percent


       confidence interval for all Foradil doses combined


       overlaps the 95 percent confidence interval for


       placebo and excludes 1 percent.  These data reflect


       the revised rates after an FDA adjudication.


                 In conclusion, the observed rate of


       adverse events was far lower than we expected from


       our pivotal trials despite demographics that were


       similar to protocols 040 and 041.  Absolute


       differences between groups were very small.  Higher


       SAE rates in the higher dose of Foradil arm,


       previously observed in adolescents and adults in


       protocols 040 and 041, were not observed in this


       larger, specifically designed safety study.



                 Now I would like to review with you a


       comprehensive safety analysis of our clinical trial


       database.  We performed an extensive review of


       safety based on our clinical trial database which


       focused on deaths and asthma-related adverse


       events.  In terms of deaths, we examined all


       controlled and uncontrolled trials in the Aerolizer


       and Certihaler databases.  All studies irrespective


       of trial duration were examined to assure that the


       very rare case of sudden paradoxical asthma,


       culminating in the demise of a patient, was




                 For the analysis of asthma-related adverse


       events we focused on controlled trials of greater


       than or equal to 4 weeks duration so as not to


       dilute the denominator for adverse events in trials


       of longer duration with very short trials,


       sometimes as short as 24 hours, which were


       performed to assess short-term changes in lung




                 For controlled trials the database


       included nearly 6,000 patients on Foradil, shown



       here; for uncontrolled trials over 2,700.  For


       trials of 4 weeks or greater in duration the number


       was over 5,000 on Foradil, whereas the


       placebo-controlled trial database was comprised of


       over 3,700 patients who had been randomized to




                 Looking at our controlled trials, there


       were 3 deaths overall, one death in the Foradil


       group, representing over 1,600 patient-years of


       experience; one death in the albuterol group,


       representing 241 patient-years of experience; and


       one death in the placebo group, representing nearly


       600 patient-years of experience.  The rates of


       death, therefore, was 0.41, 0.17 and 0.06 in the


       albuterol, placebo and Foradil arms respectively.


       The Foradil death was asthma related, shown here,


       representing a rate of 0.06 asthma deaths per 100


       patient-years of exposure.  So, here we are


       expressing it as a rate per years of exposure to


       the drug, which represents less than one asthma


       death per 1,000 years of treatment.


                 In reviewing the uncontrolled clinical



       database, it is important to note that these


       studies were those that did not incorporate


       comparator arms.  They were all open-label and


       included trials conducted as part of compassionate


       use programs.  In addition, patients tended to be


       older, with a high proportion of elderly subjects;


       had more severe asthma; used more beta agonists at


       baseline; and exhibited noon-asthma-related


       mortality at a higher rate, indicating a higher


       degree of general medical morbidity compared to the


       control database.  There were 5 deaths overall, 3


       of which came from one study in France which


       allowed entry of severely ill patients.


                 Now I would like to move on to address


       significant asthma exacerbations.  This term


       includes asthma-related adverse events which were


       meaningful enough to prompt patient discontinuation


       whether severe or not and, to get to Dr. Schatz'


       point, included asthma-related adverse events


       reported as serious whether or not they caused a


       discontinuation.  So, we are looking at patients


       that dropped out of the trial due to an



       asthma-related event that was meaningful enough to


       stop therapy.


                 Displayed are the discontinuation rates


       due to an asthma-related adverse event in multiple


       dose, placebo-controlled trials of greater than or


       equal to 4 weeks in discontinuation.  These are the


       discontinuation rates.  Please note that there were


       fewer asthma-related discontinuations in the


       Foradil arms compared to the placebo arm or to


       albuterol.  So, the rate overall for an formoterol


       doses was 7.1; for placebo it was 10.7; for


       albuterol 8.1.  Please note that this was


       especially notable for the approved dose of


       Foradil, 5.6 versus 10.7.


                 A reverse pattern was observed for


       asthma-related serious adverse events.  Foradil


       patients experienced more events than placebo.


       Here the imbalance was greater for the higher


       Foradil dose.  The numbers are shown here, 3.5 for


       the approved dose versus 3.1 for albuterol, 0.9 for


       placebo and a higher rate for the albuterol 48 mcg


       dose.  Again, this dose is the approved dose in the





                 When both types of events are taken into


       consideration, the rate of significant asthma


       exacerbations, that is, asthma-related AEs


       meaningful enough to cause the patient to


       discontinue and asthma-related events that were


       labeled as serious whether or not they caused


       discontinuation, was actually lower for Foradil at


       its approved dose than the rate for placebo or for


       albuterol.  Note that at the highest dose of


       Foradil that rate was similar to the placebo rate.


       But for Foradil at its approved dose the rate was


       7.1 versus a placebo rate of 10.9.


                 In summary, based on this analysis of our


       clinical trial database for asthma-related adverse


       events, we observed a rate of significant asthma


       exacerbations for the approved Foradil dose that


       was lower than placebo rates.


                 I would like to now move on to a review of


       postmarketing data.  In order to explore the


       adverse event profile of Foradil on the market and


       to provide some estimates as to how it might



       compare to other drugs in its class, we performed


       an analysis of postmarketing data based on FDA AERS


       database, that is the FDA's adverse events


       reporting system.


                 We must recognize that this type of


       postmarketing analysis has its limitations.


       Although spontaneous reporting of ADRs remains the


       most common method used for monitoring the safety


       of marketed drugs and is useful for detecting


       safety signals, it is limited by the fact that a


       substantial percentage of ADRs are not reported.


       The reporting rate also tends to be lower the


       longer a drug is on the market.  This is a


       well-known phenomenon and is known as the Weber


       effect.  In addition, targeting drugs to lower or


       higher risk patients may alter apparent ADR


       occurrence, and notoriety associated with a drug or


       class may alter reporting rates.  A final concern


       is that the ADR that is being reported in this


       instance is the disease itself,  It is a


       manifestation of the disease itself.


                 We examined FDA adverse reports for death



       or outcome of death and found that rates were


       highest in the first years after launch and


       declined each year thereafter.  This analysis is


       shown in your briefing book.


                 We will now review the reporting rates for


       these and other events of interest.  Please note


       that reporting rates are reports with case


       definition on the drug of interest divided by the


       exposure worldwide since the drug was marketed in


       the U.S. per 100,000 patient-years.


                 Relative rates of reporting can also be


       assessed by simply calculating the percentage of


       reports at case definition by the total number of


       adverse events reported.  This does not take into


       consideration the exposure to the drug of interest


       and may inflate the Weber effect if there is a


       difference of time on the market.  Shown here are


       the reporting proportions, on the left, and the


       reporting rates per 100,000 patient-years, on the


       right, for Foradil and salmeterol.


                 As you can see, the reporting proportion


       for formoterol, a drug that was established on the



       U.S. market in 2001, compared to salmeterol, which


       was on the market since 1994, is somewhat higher.


       However, one must adjust for the exposure to the


       drug which was far greater for salmeterol.  When


       adjusting for this higher number of exposures for


       salmeterol the result ratio flips.  That is, when


       we adjust for actual exposure to the drug we note


       that the rate for Foradil was somewhat lower than


       that of salmeterol.


                 In conclusion, well-described


       pharmacologic differences exist between formoterol


       and salmeterol although the clinical relevance is


       not known.  In pivotal trials conducted for U.S.


       registration, a potential safety signal emerged in


       the Foradil high dose group, leading only to the


       approval of the 12 mcg dose, that is 12 mcg BID,


       and the request for postmarketing asthma safety


       study 2307.  Study 2307 examined asthma-related


       serious adverse events in adolescents and adults


       and did not provide evidence of a safety signal for


       Foradil at any dose.


                 An analysis of the pooled Foradil clinical



       trial database and a review of postmarketing


       adverse event data, with its limitations, do not


       provide evidence of a safety signal for Foradil.


       The totality of the evidence, therefore, does not


       elevate concern for a safety signal and continues


       to support the favorable benefit/risk profile of


       Foradil in the treatment of asthma.


                 Thank you for your attention.  Dr. James


       Donohue will now present the clinical implications.


                          Clinical Implications


                 DR. DONOHUE:  Thank you, Dr. Geba.  Dr.


       Swenson, members of the advisory committee, Dr.


       Chowdhury and Dr. Meyer and members of the FDA,


       ladies and gentlemen, I am here today as a


       clinician investigator to talk about the clinical


       implications of the long-acting beta agonist class.


       As an older physician, I can talk a little bit


       about life before the introduction of the


       long-acting beta agonist class into our clinical


       practices.  While the alternatives were


       short-acting beta agonists, theophylline, various


       epinephrine agents, oral beta agonists, each of



       these treatments had different benefit/risk ratios


       or profiles from the long-acting beta agonist


       class.  I would like to discuss a little bit the


       implication of the roller-coaster effect on our


       patients with asthma's lives, the lack of nocturnal


       coverage with most of these shorter-acting agents


       and issues with compliance.  There have always been


       issues with the short-acting beta agonists for the


       need to frequently dose; special issues with our


       children and whether or not they could be dosed in


       the schools; the difficulty in our blue-collar


       workers, of course, who need frequent dosing of


       their medications.


                 The short-acting beta agonists have, as I


       say, benefits and risks.  Overdosing of the


       short-acting beta agonists was associated with


       tremor, particularly with the peak.  There were


       changes in metabolism, hypokalemia and changes in


       glucose metabolism which may or may not be


       clinically significant but could be under certain


       circumstances.  There was also tachycardia


       associated with people using some higher doses or



       people who had more co-morbidity issues.  Then


       also, to inform us, we had very useful data from


       Saskatchewan looking at thee use of short-acting


       beta agonists in Canada.  These are combined data


       for formoterol and albuterol.


                 These are the deaths per 100,000 per year


       and the number of canisters.  We can see that as we


       start getting up in the number of canisters,


       especially win formoterol, one sees an increase in


       deaths due tot he short-acting beta agonists or


       associated--not necessarily due to but associated


       with the short-acting beta agonist class.  So this


       was, of course, a concern to all of us and is part


       of the recommendations presently in the guidelines.


                 We also had different side effect profiles


       of other medications available to us before the


       long-acting beta agonists and drugs we would have


       to consider today as substitutes.  First and


       foremost, theophylline, particularly the


       longer-acting forms.  Their safety profile is


       important to look at.  There was a narrow


       therapeutic window, as everyone knows, with these



       drugs.  There were very, very important drug


       interactions.  In fact, if we look at our elderly


       asthmatic population, commonly these patients would


       enter the hospital with use of an antibiotic or a


       medication for reflux causing a drug-drug


       interaction and making the patient theophylline


       toxic.  Furthermore, if we look at drug


       interactions in the hospital, there is a huge


       safety concern about medication errors, and


       what-have-you, and theophylline were always at the


       top of the list.


                 Other medications we had were oral beta


       agonists, both short-acting and long-acting and,


       again, much less of an efficacy profile as compared


       to the long-acting inhaled agents and a much


       greater safety risk with tachycardia, tremor and


       reflux.  Oral corticosteroids have to be used more


       and more when patients have more and more


       exacerbations and I don't have to review the


       laundry list of side effects that are well-known to


       everyone in the room.


                 Now, throughout the world we have--I have



       outlined in yellow here the G8 nations because of


       last week's meeting, but we can see the variation


       in prevalence of asthma symptoms as we get more


       industrialized societies.  We see a very large


       increase in the number of patients who suffer with


       airways disease.


                 On the other hand, there are facts that I


       find very, very consoling and comforting.  This is


       the death rate due to asthma in the United States


       going back to 1960 up to 2002.  These are the


       deaths per 100,000 so we can sort of get the rate


       that you are asking for.  Then we have the African


       Americans here and the Caucasian population here.


                 Short-acting beta agonists were introduced


       in the 1970s.  We had the inhaled corticosteroids


       introduced in the 1980s.  There have been enormous


       efforts in patient education, efforts by the expert


       panels of the National Institute of Health for


       guidelines and just generalized education programs,


       along with the introduction of effective controller


       medicines along with the long-acting beta agonists


       that seems to have led to a decline, although still



       very high, relatively higher in the African


       American population, but to the general United


       States population, from 5,400 to a little bit above


       4,000.  So, we are clearly doing something right.


       What the attribution would be here to the various


       things introduced is, of course, beyond my ability


       to say but, clearly, all these things together have


       helped us to improve the lives of our patients.


                 Now, what about the long-acting beta


       agonists?  Just briefly, you have seen an awful lot


       of this data this morning and, at the risk of being


       a little bit redundant, the first benefit, of


       course, is in patient symptoms.  These are better


       bronchodilators.  I think we would all agree that


       the data are overwhelming on this.  There is a


       reduction in the roller-coaster effect, and I will


       come back to that in a minute; better control


       because of the longer duration of effect and


       nocturnal symptoms.  Because of the control, we


       have less use of rescue medications.  We have


       improved morning lung function and also less


       diurnal lung function variability.  It doesn't



       completely eradicate it but it does minimize to


       some extent some of the early morning dipping that


       leads to waking up or even worse outcomes.  Also,


       equal important perhaps, it gives us protection


       against exercise-induced bronchospasm and that is


       the exercise of normal daily activities in normal


       life, and it is nice to have that kind of


       protection on board so you don't have to


       continuously dose yourself.


                 Looking at the formoterol data, Dr. Geba


       has shown us the safety data.  We saw that there


       are 3 pivotal trials that you have in your briefing


       documents.  There is superior improvement in the


       FEV                                    1 over placebo over the course of

12 hours.


       This duration of action is sustained for 12 weeks


       so there doesn't appear to be a signal that there


       is any tolerance or reduced efficacy.  There is a


       reduced need for nighttime rescue medicine, and the


       onset of action is similar to albuterol, as has


       been outlined.


                 Just again showing the similar 12-hour


       studies, this is the 040 and the 041 that you have



       seen a moment ago.  This is at week 12.  Here we


       see the 12-hour curve and this is the mean change


       in baseline FEV                                                        1.

Here is the short-acting


       albuterol and placebo, and here is the sustained


       effect of the long-acting beta agonist, in this


       case formoterol.  First of all, let me draw your


       attention to the pre-dose or trough.  Often we


       power clinical trials on long-acting beta agonists


       and this changes well over 10 percent here, around


       12 percent, and that usually can be transferred to


       as meaningful clinical improvements such as


       symptoms in the morning and what-have-you.  Over


       the course of the day you see the roller-coaster.


       That in itself means nothing but what this means


       here is that as these drugs flow off here our


       patients become symptomatic and have to disrupt


       their daily activity to take rescue albuterol.


                 Also one other thing I would like to show


       here is the peak effect.  Patients perceive change


       and when you have a nice plateau effect a lot of


       side effects such as tremor are much less


       perceptible to a patient.



                 Other outcomes besides bronchodilator is


       rescue medication.  These again are from 040 and


       041, the run-in for the 3 arms, formoterol,


       albuterol and placebo.  We see over the course of


       4, 8 and 12 weeks a nice decline in the rescue


       albuterol.  These parameters are less rigorously


       defined but we think the minimal clinical


       improvement in that parameter is 0.8 puffs per day


       ranging to 1.  So, it appears to be in the ballpark


       of something that means something to a patient and


       we could quantitate that.


                 Simply, formoterol reduces nocturnal


       symptom scores.  I am showing 040 and 041 which I


       believe are the adult studies run-in and over 12


       weeks and the companion study here.  We see a nice


       decline from about 0.5 to 0.1 in the nocturnal


       asthma symptom score parameter.


                 To end up, one of the studies that I take


       consolation from as a physician is the FACET study.


       Again, points are well taken here today, we are


       talking about deaths.  It is very difficult in


       asthma studies though to power our studies, as



       everyone in the room has heard over and over again,


       on death as the outcome.  Exacerbations are


       extremely important and the reason that I take a


       lot of comfort from this study is that it is a


       one-year study.  Also, the exacerbations are


       precisely defined, as Dr. Sorkness mentioned this




                 In this study we see--again in the


       mechanical function, 835 patients, 12 months--a


       marked improvement after the run-in.  The patients


       were in the run-in symptomatic on inhaled


       corticosteroids.  But this study gives us some look


       at what is the added value of adding a long-acting


       beta agonist--added clinical value--to inhaled




                 Here are the two budesonide arms.  I think


       there is no surprise that on the mechanical


       function we do see a change of 7 or 8 percent.  But


       I think one takes a better message away from


       looking at exacerbations.  First of all, if


       inflammation is ongoing and not being checked the


       patient is going to know that they are going to



       breakthrough with an exacerbation.  This is our


       best clinical surrogate for the so-called masking


       of inflammation, that is, the breakthrough of


       exacerbations.  In a study of one-year duration we


       have adequate duration to bring this signal out.


                 So, in this study, as Chris Sorkness


       pointed out this morning, we have an arm with 100


       BID of budesonide and with formoterol.  Then, the


       second arm is 400 BID with the addition of


       formoterol.  And, the exacerbations were described


       as mild with an increase in terbutaline, the rescue


       medicine, and severe, defined by a 30 percent


       change in peak flow and oral prednisone.  The


       decline in mild exacerbations was about 29 percent


       and 40 percent with the combination.  But with the


       higher dose, high dose budesonide reduced the


       exacerbation rate by 49 percent.  When one adds


       formoterol to it it went to 62 percent.  So, there


       is a net gain there and the actual clinical


       implication of that is that it appears to be


       significant.  We really haven't put a number on


       that yet but that, in my mind, is a very reassuring



       piece of evidence that supports the use of this


       class of drug.


                 Using the best evidence that one has, the


       expert panels--many of the members of which are in


       the room here--have come together, and you have


       seen this before over and over again, and have


       concluded that inhaled steroids and long-acting


       beta agonists have a complementary effect.  One can


       lower the dose of inhaled corticosteroids by using


       the combination, and not everyone responds to


       inhaled corticosteroids although a great majority




                 For more severe patients we would be using


       higher doses of inhaled corticosteroids and


       long-acting inhaled beta agonists and, hopefully,


       we will be able to avoid the use of prednisone and


       its very difficult side effect profile.


                 So to summarize, long-acting beta agonists


       have really become an established part of the


       current standard of asthma treatment in the United


       States and also internationally.  It is an integral


       part of internationally established guidelines



       using the best evidence that we have at our


       disposal at the present time.  It is well


       established that long-acting beta agonists have a


       place in the treatment regimen for asthma but must


       be conjunction with inhaled corticosteroids for


       those with moderate and severe persistent asthma.


                 Again, I don't think at the present time


       there is an alternative inhaled controller


       bronchodilator or one on the immediate horizon that


       is suitable for asthma.  So the LABAs, the


       long-acting beta agonists, have provided documented


       improvement in symptoms, airway function and


       quality of life and you have heard a great deal


       about that today.  For whatever reason, since the


       introduction of long-acting beta agonists,


       controllers and a national effort in education and


       guidelines, asthma hospitalizations and mortality


       have decreased, which is very reassuring at least


       to me and I think to many others.  Long-acting beta


       agonists in conjunction with inhaled steroids


       represent a medication category critical for


       optimal care of patients with moderate to severe



       asthma.  Thank you very much.


                        Questions by the Committee


                 DR. SWENSON:  The time now is open for


       questions to Novartis.  Dr. Schatz?


                 DR. SCHATZ:  Some of the things we are


       hearing suggest a possible disconnect between


       exacerbations, as has been studied and defined


       usually to include oral steroids and emergency


       departments or hospitalizations, and then death or


       near death.  So, I guess my question has to do with


       2307.  How were SAEs defined?  However, were


       asthma-related SAEs defined?


                 DR. FLOYD:  Dr. Geba?


                 DR. GEBA:  To answer this question I would


       like to bring up a slide that gives the definition


       of asthma-related and the definition of serious.


                 We used predefined asthma terms, MedDRA


       terms, MedDRA preferred terms.  Asthma-related AEs


       were defined as asthma, dyspnea, bronchospasm and


       chest discomfort, cough, wheezing, etc., acute


       respiratory failure and hypoxia.  For the


       definition of SAEs it was one of the above plus one



       of the following, death, life-threatening


       hospitalization, disability, congenital


       abnormalities--this is regulatory definition, and


       required intervention to prevent further


       impairment--standard definition.


                 DR. SCHATZ:  Just to follow-up,


       intervention could mean oral corticosteroids?


                 DR. GEBA:  Yes, it could be.  It could be