Witness Testimony

Mr. Chairman and Members of the Subcommittee, I am Dr. Lawrence Olanoff. I am Executive Vice President of Forest Laboratories and head of the Forest Research Institute. On behalf of Forest, I would like to thank the Subcommittee for the opportunity to discuss our views on the issues presented in this hearing. Among other things, I will discuss our new Clinical Trial Registry.

I am a medical doctor and I have trained as a clinical pharmacologist. I have chosen to devote my career to the development of human pharmaceuticals.

Depression - and in particular depression in children and young people - is a terrible and dangerous disease. It is terrible in that it inflicts terrible pain and it robs victims of the ability to enjoy their lives. It is dangerous because it also presents a substantial risk of suicide. This is an issue about which Forest, and its most senior executives, have painful and personal experience. The book, THE NOONDAY DEMON, authored by Andrew Solomon, is a seminal study of depression and his own depression and in that book Andrew describes his own near suicide. That book won the National Book Award and is dedicated to his father, who helped to bring him back from the brink of suicide. Andrew's book is dedicated to his father and the dedication reads "for my father who gave me life not once but twice". Andrew's father is our Chairman and Chief Executive Officer, Howard Solomon. In the case of the two most senior executives responsible for the launch and marketing of Celexa®, an antidepressant marketed by us, their child's depression took an even greater toll. Both lost young sons to suicide. One is still a member of our Board and has retired from Forest and now devotes his time to a foundation started by him to prevent suicide among young people.

So we at Forest know all too well that depression, without proper therapy, can be a devastating illness. I tell you about these personal experiences to emphasize our motivation at Forest. We strive to develop and provide the best possible treatment for this horrible disease. Our mission is to alleviate depression and thus help to prevent suicide.

Forest is a medium size pharmaceutical company that primarily markets drugs in the United States. Also, unlike the larger pharmaceutical companies who are members of PhRMA, which we are not, we do not conduct drug discovery activities in-house, but license all our products from other companies, usually European companies which sell the same products in their own markets. Our licensors often have already completed some clinical studies before we license the drug and continue to do studies after we license the drug. In the case of Celexa, for example, the drug was already approved in Europe for six years when we first licensed it. We have also, in the case of most of the drugs we have licensed, performed the additional necessary development activities and clinical trials in the United States to obtain FDA approval of these products.

The basis for a drug's evaluation are the clinical trials in which it is tested. A controlled clinical trial is a method for providing objective evidence about the safety and effectiveness of a drug. In a controlled clinical trial, subjects are divided into different test groups. Most commonly, one group will receive a placebo (a pill with no active drug), while another group will receive the test drug. We attempt to design these studies to assure, to the greatest extent possible, that the only thing that will cause a different result between the two groups is the drug itself. It is essential therefore that the patients in each study group be carefully balanced from the start in as many respects as possible to avoid distorted results. If, for example, one group included more patients with more severe illness, that might affect the results and make it difficult to know whether or not the drug itself was the cause of the study outcome or whether any observed differences arose simply because the study groups were different from the onset. Similar considerations apply in comparisons of the results across different clinical trials if there are different study designs and different patient populations. In addition, the terms that investigators use to describe adverse events and outcome measures may vary among different trials, particularly studies in different countries.

I want to make one additional point about clinical trials of depression and many other psychiatric diseases where the endpoints are not objective measurements like blood pressure but more subjective in nature like, for example, mood. It is well known and accepted in the scientific community that there is typically a relatively high proportion of placebo-treated patients who respond favorably (also called a high placebo response rate) in these studies, particularly because in the environment of a clinical trial, as distinct from clinical practice, drug therapy is accompanied by greater attention to the patient by the investigator and his staff, especially where this was not part of the patient's experience before entering the study. This means that some patients may improve from their condition at the start of the study even though they are not receiving active drug treatment. The high placebo response rate can make demonstration of a therapeutic effect for the test drug difficult when the overall response difference between test groups is compared by statistical analysis. Dr. March, the author of the NIH supported Treatment for Adolescents with Depression Study (TADS), involving Eli Lilly's Prozac and described in the recent JAMA publication, comments in regards to other SSRI/SNRI pediatric depression studies that, "In the 2 earlier fluoxetine studies, the placebo response rates were lower than placebo response rates seen in other pediatric antidepressant trials. Because the response to active drug was comparable, it was the placebo response rate that generally determined the effect size and hence whether a trial was positive or negative." Thus, in many cases studies of drugs intended for the treatment of depression fail to demonstrate a response: i.e., those studies are referred to as "negative" in the scientific literature, although they might be more logically termed "no-effect" studies. In fact, the vast majority of these no-effect studies do show some modest advantage of the test drug compared to the placebo treatment. It is just that the difference between the groups fails to meet the required statistical hurdle that confirms that the benefit observed did not simply occur by chance. For the reasons I stated above, the attainment of a positive study is especially notable and this explains why it may take more than one study to achieve a study with a positive outcome. Of the 15 placebo controlled studies in pediatric depression submitted to the FDA, only 3 were considered positive: two fluoxetine studies (which led to the approval of Prozac for pediatric depression) and the Forest sponsored U.S. study with Celexa which is discussed below.

It is clearly essential to determine whether drugs that are safe and effective for use in adults can also be appropriately used in children. This is an important inquiry, because children may be physiologically and psychologically different from adults and they therefore may react differently to drugs. The law therefore currently requires the conduct of pediatric studies where the indications sought for in adults may be applicable to the pediatric population.

Forest supports the goals of the pediatric study law and FDA's implementation of that law with respect to antidepressant drugs. If a drug should not be used in children, doctors should know that. Conversely, if a drug can help some children suffering from depression, it would be a tragedy if doctors and patients were discouraged from its use in that population, condemning those children and their families to unnecessary misery and to a possibly preventable risk of suicide. This places a heavy responsibility on the FDA as it has in so many areas. In our experience we have found the FDA to be unbiased and expert.

So far as we are aware, based on the data available to us and the complexity of the subject, we believe that FDA is trying to achieve the right result.

I want to emphasize that, because the FDA has not approved pediatric labeling for our products, Forest has always been scrupulous about not promoting the pediatric use of our antidepressant drugs, Celexa and Lexapro®. That is the law, and we follow it.

Prior to approval, companies are required to fully disclose the results of all studies related to the indication sought in the NDA to the FDA regardless of their outcome, and Forest has always done so. The FDA reviews those studies and decides what information is necessary in the package labeling which is the ultimate source of information for the physician. The FDA may approve a drug based on two positive studies even if there are negative or no-effect studies in the application, and it may or may not require mention of the no-effect studies in the label because, in large part, the scientific community has accepted that positive studies are generally more informative than no-effect studies. Companies frequently continue to develop drugs despite no-effect studies; the FDA approves drugs even when there are no-effect studies; and journals will often reject no-effect studies for publication.

That brings us to the question of publication and disclosure of the results of clinical trials, both those that show a positive effect and those that fail to detect a positive effect. Aside from the review by the FDA itself, perhaps the most objective review of a pharmaceutical company's clinical studies is the peer review system of prestigious medical journals. It is our experience that journals publish only a small portion of the studies conducted or submitted to their editors, and that they are usually interested in reporting positive studies and are often not interested in publishing studies which are not positive. Positive studies may be breakthrough studies; they are likely to be of greater interest to physicians. No-effect or negative studies, particularly if there are a number of known prior such studies in the same drug category, are often considered of less interest to their physician audience. That is certainly not always the case, but it is understandable that medical journals, like the media in general, want to publish what they believe their readers would be most interested in.

The next question relates to studies for new therapeutic uses completed after the FDA has initially approved the drug for a particular indication. There is no established procedure for disclosure of such study results. The general principle observed by Forest is that information that better enables physicians to treat their patients should be available to them.

Recently much public attention has been directed towards approaches to achieve the timely and full disclosure of all clinical study results for approved and unapproved uses of marketed products. Many have recognized that scientific publications alone cannot serve as the sole vehicle for this purpose for the reasons I have cited above.

Forest has focused on this complicated issue and announced this week the development of detailed procedures to assure that all results of Phase III and IV trials are reported in a Clinical Trial Registry. Forest's Clinical Trial Registry will contain the following information:

Ongoing Studies

Forest's Clinical Trial Registry will include a listing of Forest-sponsored ongoing phase III and phase IV clinical studies for all Forest drugs. In particular, when Forest initiates a Phase III or Phase IV clinical study, the number, title, start date and key objectives will be posted to the Clinical Trial Registry.

Completed Studies

For all phase III and phase IV Forest-sponsored studies relating to currently-marketed Forest products completed since January 1, 2000, Forest will by December 31, 2005 post summaries of the results of these studies on the Clinical Trial Registry. This will include summaries of clinical study reports for clinical studies of the use of Celexa and Lexapro by pediatric patients. These summaries will include results for the protocol-defined efficacy and safety outcomes, as well as a description of the trial design and methodology.

For all phase III studies relating to Forest products completed after today, Forest will post summaries of the results on the Clinical Trial Registry upon the commercial introduction of the product in the United States. For phase IV trials conducted for the approved indications completed after today, Forest will post summaries of the results within a year of study completion.

For studies submitted to scientific peer-reviewed journals whose policies do not permit disclosure of study results prior to publication in these journals, the clinical study summary will be posted at the time of publication. Also, Forest will post a summary on the Clinical Trial Registry of: (a) those Forest-sponsored phase I and phase II studies completed after January 1, 2000 for products which Forest currently markets, and (b) those Forest-sponsored studies completed prior to January 1, 2000 for products which Forest currently actively promotes, which provide additional important information for physicians and the care of patients.

In addition to our own activities in this regard, Forest will participate in and be guided by any industry, legislative, regulatory or medical association initiatives to facilitate this effort.

Now that I have provided the details of our recently announced Clinical Trial Registry, I want to comment on Forest's past practices in disclosing clinical trial results by referring to our three principal products. The first is Namenda, our recently approved drug for the treatment of moderate-to-severe Alzheimer's disease. The product was approved in October, 2003. We have released information on four placebo controlled studies relating to unapproved uses, in each case promptly after we received the study results as these results were judged to be material to investors. In the case of use of this drug in the treatment of patients with mild-to-moderate Alzheimer's disease, we disclosed that one study was positive and that two studies (one of which was performed by a European licensee) showed no effect. We hope to ultimately obtain FDA approval for Namenda in the treatment of mild-to-moderate Alzheimer's disease. The other study related to use of the active ingredient in Namenda in the treatment of neuropathic pain and it also failed to meet the FDA standard for approval even though an earlier study had shown promising results. Again, we promptly disclosed these results. We are continuing to study the drug for that indication and likewise ultimately hope to be able to meet FDA requirements for that indication.

The second drug is Celexa, which was approved for depression in adults in the U.S. in 1998 and which had been first approved in Europe as early as 1989. For this product there are two placebo controlled studies in pediatric-adolescent population. One was a study conducted by our licensor in Europe over a five year period which recruited patients from seven different countries and underwent various modifications in the test protocol over the course of the study in an attempt to improve its slow enrollment rate. We did not originate, design or monitor that study. This study included patients with characteristics that we and other sponsors would not use in our studies of depression and which we believe caused a substantial degree of patient variability and potentially important differences in disease severity between the active and placebo group from the very start of the study. Specifically, the European study allowed for the enrollment of patients without regards to a past history of hospitalization due to psychiatric illnesses, past history of suicide related events or a history of a failure to respond to other antidepressants.

Further, patients were allowed to enter into the study with other co-morbid psychiatric illnesses or on other concomitant psychotropic medications or receiving psychotherapy. This study was unique across the experience of placebo-controlled pediatric trials in that it allowed both hospitalized patients and outpatients to be enrolled. In fact, a third of the patients enrolled had a past history of suicide related events and the patients in the Celexa group had a higher rate of previous psychiatric hospitalizations and a greater number of the Celexa treated patients were hospitalized due to psychiatric illnesses at the start of the study compared to the placebo group. These design features and potential imbalances between groups make it very difficult to interpret any resultant differences in the incidence of relatively infrequent events such as suicidal ideation or behavior as related to a particular treatment assignment. Finally, this study did not demonstrate effectiveness of Celexa for pediatric use which we believe was due largely to a high placebo response rate, amounting to some 60% of the placebo treated patients.

The second trial was a well-controlled study, designed and monitored by Forest in the United States with investigators and trial centers determined by us, which did demonstrate the efficacy of Celexa for pediatric patients. In contrast to the European study, the U.S. Celexa study was more typical of other pediatric depression studies in that it enrolled only outpatients, did not allow patients to enter with a history of suicidal behavior or treatment resistance, did not allow for concurrent psychotherapy and was far more restrictive in the use of concomitant psychotropic drugs or the presence of comorbid psychiatric illnesses. We felt that this study was important to the medical community as it was the first and only positive study after a string of no-effect studies for all the other modern antidepressants except for two previous positive studies for fluoxetine, which has been approved for the treatment of pediatric depression. When we first unblinded these Celexa studies in 2001, we looked carefully at all the safety data combined across the two studies. We did not see in these results any evidence of a statistically significant or clinically relevant increase in suicide related events (SREs). We fully reported what we believed to be suicide related events under appropriate adverse event descriptive terms in our submission to the FDA in 2002 and the FDA in its review of our submission, while not granting approval for a pediatric indication, did conclude that there were no new safety issues identified in this population which would require labeling. When this issue was raised again in 2003 due to potential concerns over the SSRI/SNRI class as a whole, we reviewed our entire pediatric safety database for any SREs using the FDA's provided algorithm. Our conclusion and the conclusion of the FDA in early 2004, as reflected in the August 16, 2004 memo by Dr. Mosholder, of the FDA, was that the difference between the citalopram and placebo groups in the incidence of SREs was relatively small (risk ratio of approximately 1.4) and not statistically significant. The FDA conducted a new analysis in August based on a reclassification of SREs by experts at Columbia University. After their elimination of questionable cases, this reclassification reduced the number of SREs by some 40% for the citalopram group compared to our own earlier classification. This reclassification led to a new FDA calculated risk ratio of 1.37, also not statistically significant and the lowest of all the risk ratios among the SRI/SNRIs except for Prozac. After adding to this overall Celexa database, the safety experience from the Forest-sponsored and the only pediatric placebo-controlled trial with Lexapro (escitalopram, which is the therapeutically active enantiomer of citalopram), this risk ratio would be reduced to approximately 1.2. The Lexapro trial is relevant as it was conducted in the U.S. according to a protocol design very similar to that of the earlier U.S. pediatric trial for Celexa. When these two U.S. studies are taken together, and separate from the European Celexa trial, the risk of suicide related events is actually two-fold higher in the placebo group compared to these two related SSRIs; however, the numbers of events in these U.S. trials were too low to demonstrate any statistical differences.

As I have previously indicated, these two U.S. trials are in substantial contrast by design to the European trial where event rates were higher in both treatment groups. As stated by the FDA medical reviewer, Dr. Hammad in his review of August 16, 2004, in describing the U.S. and European Celexa trials, "These two Celexa trials varied in almost every aspect. The combination of the differences might have led to higher probability of having higher risk patients in trial 94404."

Forest conducted further analyses to attempt to better understand why certain patients in the Celexa trials experienced SREs. The analyses revealed that such patients generally experienced numerous antecedent psychosocial stress factors and as a group responded substantially less well to treatment, whether they were treated with placebo or active drug, compared to the patients who did not experience SREs.

It was these psychosocial factors and the lack of therapeutic response that appeared to better predict whether a given patient would experience an SRE, rather than their drug or placebo treatment assignment or any prior activation-like side effects.

Dr. March in his recent publication of the TADS results makes the same observation of the patients who experienced what he described as 'harm-related adverse events' which included suicide related adverse events. Dr. March states that, "Incident narratives indicate that irritability, agitation/ restlessness, and anxiety were not commonly reported in association with harm-related adverse events, suggesting that other factors, such as substance use and psychosocial stressors, may be more important in mediating the risk of harm-related adverse events."

Both Celexa studies were completed at virtually the same time, despite the fact that the European study was started four years before our study. In fact, the results of both studies were obtained shortly before we terminated virtually all promotion of Celexa for any use. We stopped promoting Celexa because we had obtained approval for Lexapro, a more potent antidepressant which we considered a superior product. However, the fact that we had a successful study demonstrating efficacy in a pediatric population, after there have been so many no-effect studies for so many other similar antidepressant products was important and something the study investigators felt should be made available to the medical profession. This study was therefore presented at several scientific meetings and accepted by and published in a prominent peer reviewed journal in June, 2004. The no-effect or negative European study results were not hidden by us and were available in several sources including a 2003 presentation at a prestigious meeting of psychiatrists specializing in the care of pediatric and adolescent patients, at which the safety and efficacy findings of both studies were described.

The final example is Lexapro, the antidepressant drug that Forest is currently promoting. We performed a clinical trial of this drug in a pediatric population and that trial failed to show a statistically significant therapeutic effect but did not demonstrate any safety issues for Lexapro in these pediatric patients. We promptly issued a press release disclosing the results of that study. Based on our overall analysis of the study results we are continuing to study Lexapro for that indication. That press release also discussed the positive and no-effect trials of Celexa.

We understand that there is great interest in the issue of disclosure of the results of clinical trials. With respect to Forest, we believe we have consistently acted appropriately and in compliance with all legal and regulatory requirements when informing physicians about out products. As I stated earlier, we are prepared to put into effect a publicly accessible clinical trial registry to facilitate the timely disclosure of our phase III and IV clinical study results for our marketed products.

We at Forest have a deep - and deeply personal - commitment to doing what is best for patients, and particularly children suffering from depression. Our mission is to help heal and treat young people. Forest will continue to search for more effective ways to inform physicians about clinical trial data on its products. I look forward to today's discussion of these important issues.

SUMMARY OF TESTIMONY OF LAWRENCE OLANOFF, M.D
EXECUTIVE VICE PRESIDENT OF FOREST LABORATORIES, INC.
BEFORE THE SUBCOMMITTEE ON OVERSIGHT AND
INVESTIGATIONS, COMMITTEE ON ENERGY AND COMMERCE
TUESDAY, JULY 20, 2004

• Depression -- particularly in young people -- is a terrible and dangerous disease that presents a risk of suicide. Forest and its most senior executives have personal and painful experience with the issue of depression, and it is Forest's mission to develop and provide the best possible treatment which will alleviate depression and prevent suicide.
• Controlled clinical trials provide the evidence by which the FDA judges whether drugs are safe and effective. Because some patients afflicted with depression may improve even when treated with a placebo, it can be difficult to demonstrate efficacy for an active drug in controlled trials of this disease. Many controlled trials of drugs for pediatric depression fail to show a positive response. The placebo-controlled study of Celexa conducted in the U.S. by Forest was one of only three trials of a total of 15 that clearly showed efficacy in the pediatric population for a modern antidepressant. Another trial of Celexa conducted by HLundbeck in Europe failed to show efficacy.
• When Forest has completed controlled clinical trials of a drug it is studying for a new indication, Forest submits all of the data, "no effect" trials as well as positive trials, to the FDA. Forest has submitted to the FDA the data of all studies of its drugs in the treatment of pediatric depression, whether the results detected a positive response or failed to do so.
• Forest has disclosed both positive and "no-effect" studies with respect to its principal products, including its antidepressants. Forest believes it has consistently acted appropriately and in compliance with all legal and regulatory requirements with regards to the disclosure of its clinical trial results. Once a drug has been approved for use by the FDA, however, there are no established procedures for the disclosure of further study results. The general principle observed is that information that better enables physicians to treat their patients should be available to them. To better achieve this objective, Forest will make publicly accessible the results of its phase III and IV trials of its marketed products, as well as the results of certain other studies, on a clinical trial registry. Forest will participate in, and will be guided by, any industry, legislative, regulatory and medical association initiatives to facilitate this effort.
• Forest supports the FDA's efforts to determine whether a particular SSRI should be used in the pediatric population. Forest's antidepressants, Celexa and Lexapro, are not approved for pediatric use and Forest has not promoted that use. Forest also believes that based on its own analysis, as well as the FDA's most recent reanalysis, of the available placebo controlled clinical trial data there is no evidence of a statistically significant or clinically important increased risk of suicidality in the pediatric population.
• Forest welcomes this Committee's inquiry into the issue of disclosure of the results of clinical trials and shares the goal of assuring that physicians have the best information possible to make their prescribing decisions, including the use of antidepressant medication in young people, with the objective of improved patient care.