Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 19
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary 4 Content of a Clinical Trial Registry This chapter describes the guiding principles for clinical trial registration and the key fields in a registry. For each topic, background material is followed by comments made by some of the speakers and participants at the June 27 workshop. GUIDING PRINCIPLES AND GOALS FOR CLINICAL TRIAL REGISTRIES Based on the premise that registries must provide the public with sufficient information to provide public health and safety benefits, the committee’s initial discussions suggested that the following guiding principles were desirable for a clinical trial registry: Be global in perspective. Offer access to the public at no charge. Be located on a single website or linked via a single portal. Be open to all prospective registrants. Be managed by a not-for-profit organization or trusted government agency. Have the capacity for electronic searches. Provide a mechanism to ensure the validity of the registration data. Have a process to ensure adherence to the registry standards. Avoid reducing the incentive to do clinical research, whether public or privately funded.
OCR for page 20
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary A clinical trial registry that follows these principles could be expected to meet the following goals: To provide patients and their health care providers with adequate and reliable information about clinical trials that may be enrolling patients. To provide health care providers, patients, and others with the results of a clinical trial once the trial is completed and the product is available for prescription. To link each clinical trial initiated with a reported outcome, thereby preventing selective or biased reporting of results. To meet the first three goals in a way that protects proprietary research data, as necessary, and preserves innovation. However, a clinical trial registry is NOT intended to replace the advice of a health care professional regarding benefits and risks nor is it intended to replace the comprehensive information on a product label as required by the relevant regulatory authorities. It is also not intended to replace peer-reviewed publication—although the veracity of journal publications could be better assured by the presence of a transparent and more comprehensive mandatory clinical trial registry. Furthermore, to meet its purposes, a registry should provide only objective, scientific information about the clinical trial and not promote a product. Comments on Guiding Principles and Goals for Clinical Trial Registries Marjorie Speers, Executive Director of the Association for the Accreditation of Human Research Protection Programs, began her workshop presentation about the needs of patients and the public by stating that, in the debate over registries, the purposes of such registries have been blurred. Are they intended merely to inform the public about an ongoing or proposed clinical trial? Are they intended to be vehicles to recruit individuals into clinical trials? Are they intended to assist physicians in treating patients? Are they intended to prevent the suppression of negative results? Are they intended to build the public’s trust in clinical trials? All of these purposes have been expressed at one time or another. Alan Breier, Vice President, Medical and Chief Medical Officer, Eli Lilly and Company, suggested that different platforms might be needed
OCR for page 21
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary to address the different purposes. He suggested that using a clinical trial registry as an enrollment tool is a separate aim that should be addressed elsewhere. Eli Lilly’s patient surveys indicate that people considering or enrolling in a clinical trial want to ask questions and engage in discussion with a live person. Jerome Yates, representing the American Cancer Society, stated that 90 percent of patients who query a clinical trials source need follow-up answers to their queries. Could the patient advocate community help to answer questions? Dr. Speers agreed there should be some type of help function for those who require assistance and some type of consultation service when the public has questions about a particular study. Alan Goldhammer, Associate Vice President, U.S. Regulatory Affairs, Pharmaceutical Research and Manufacturers of America (PhRMA), submitted this comment to the Institute of Medicine (IOM) website: “PhRMA supports the four goals that IOM has identified regarding the establishment and use of clinical trial registries…. The WHO [World Health Organization] document notes that ‘…one or more of data items 10, 13, 17, 19, 20 may be regarded as sensitive for competitive reasons by the sponsor who may wish to delay release of the information.’” He asked that this key point—“the need for confidentiality to preserve innovation”—be highlighted in the final IOM summary. Jeanne Ireland, Elizabeth Glaser Pediatric AIDS Foundation, commented that “only a mandatory system with a breadth of information will improve the public trust.” Catherine De Angelis, Editor-in-Chief of the Journal of the American Medical Association, noted that three things must happen with a clinical trial registry: (1) It must be managed by a nonprofit organization; (2) it must be able to support pertinent information; and (3) it must be available to anyone freely. Dr. Breier echoed those sentiments, stating that such a registry needs to be global, free, and publicly accessible. “A clinical trial done in Russia can be just as important to patients in the United States and vice versa—everybody counts. We live in a global world so having a global approach to this becomes also critical or we have significant gaps.” He expressed concern that writing summaries in patient-friendly language may edge the summaries toward being viewed as promotional. Michael Manganiello, representing the National Institutes of Health’s (NIH’s) Council of Public Representatives, asked if there would be an education campaign to help the registry reach its intended users.
OCR for page 22
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary Sharon Terry, Genetic Alliance, asked what kind of education campaign would be tied into the development of a new registry. How could the tremendous number of patient groups with expertise and branches in various states, plus provider groups, be tied in so that people are aware of what the registry is and use it themselves or for their patients? Hugh Tilson, Centers for Education and Research on Therapeutics, described the development of a clinical trial registry as a “huge social intervention, which therefore requires proper evaluation,” including cost-effectiveness analysis. Any registry approach developed needs to be pilot tested. He also suggested research to determine what the stakeholders want and what the impact on patient and other stakeholder behaviors and public health will be. He called for research on how to summarize a body of data and suggested that a new education agenda would include how doctors can communicate about the entire, evolving body of research, and how to educate the media. P.Pearl O’Rourke, Director of Human Research Affairs, Partners HealthCare System, Inc., noted that a trial is an active and alive entity, and raised questions about how the registry would incorporate planned and unplanned changes. Sometimes preliminary findings indicate a need for changes in the study design, she noted, and amendments are added and adverse events occur. “How are we going to accurately place these details into a registry? For example, my interest in enrolling in a trial may well be altered if 15 of the last 17 people had a big adverse event. How will early termination be handled? Particularly if it was terminated because of scientific misconduct, how are you going to document that in a registry?” Robert Temple, Director of the Office of Medical Policy in the Center for Drug Evaluation and Research, Food and Drug Administration (FDA), asked: “Do we mean for this to include trials that are poorly designed and ill controlled? We see trials all the time in which people on one drug are just switched to another drug, and FDA’s Division of Drug Marketing, Advertising, and Communications sends out scads of letters saying that’s no good, that’s not an adequate study. Do we want all those in there? I wouldn’t think so.”
OCR for page 23
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary TRIALS TO BE REGISTERED The International Conference on Harmonization differentiates exploratory trials from hypothesis-testing or confirmatory clinical trials as follows: Hypothesis-testing (confirmatory) clinical trials are well-controlled studies designed to provide meaningful results by examining prestated questions (hypotheses) using predefined statistically valid plans for data analysis, allowing solid conclusions to be drawn to support specific product claims. Pilot or exploratory studies are, by design, intended to provide preliminary information about a disease or condition, endpoints that might be measured to evaluate treatment efficacy, the profile of a possible new treatment, etc. These studies are performed to generate hypotheses and aid decision making for possible future product development. According to these definitions, some of Phase II and all of Phase III (registration) and Phase IV (post-marketing) would be considered hypothesis testing (confirmatory). In discussing which clinical trials should be included in a registry, the workshop participants gave particular attention to the subset of Phase II studies that are deemed “exploratory.” The three main proposals—those of the International Committee of Medical Journal Editors (ICMJE), the international pharmaceutical industry, and WHO—are broader than the requirement for ClinicalTrials.gov, which calls for posting of trials regarding serious and life-threatening conditions. These groups would require registration of all prospective, hypothesis-driven, interventional clinical trials (i.e., all confirmatory trials, not just those for serious or life-threatening illnesses and conditions) upon trial initiation. They all exclude Phase I trials. Comments on Trials to Be Registered Committee Chair Philip A.Pizzo and Sharon Terry, President and Chief Executive Officer (CEO) of the Genetic Alliance, a patient advocate group, called for inclusion of exploratory trials in registries, expressing concern that exploratory studies will be increasingly used to test hypotheses. As patient selection uses more sophisticated genetic profiling, the potential exists that smaller exploratory studies may cross the line into hypothesis testing. Miriam O’Day, Senior Director of Public Policy at the Alpha-1 Foundation, a patient advocate group, made a simi-
OCR for page 24
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary lar point, noting that exploratory studies may be especially important for genetic studies. Dr. Temple also suggested that some trials considered “exploratory” be included. Phase II trials are well-designed trials for the most part. “They may have multiple hypotheses and be exploratory in that sense, but they’re often used as part of the critical database.” William Vaughan sent a statement from Consumers Union to the IOM website: “It is essential that the goals, specifications, and endpoints of clinical trials be registered and made public.” He added, “Though in some cases it may not be necessary to register and disclose results from Phase I and some Phase II trials, we urge the Institute to err on the side of more public information, not less.” Industry representatives expressed concern that early exploratory studies should not be included for the following reasons: Early exploratory trials are designed only to set direction (i.e., to generate hypotheses) for possible future studies; they are not designed to provide definitive or confirmatory answers of safety or efficacy of drugs under development. Approximately 90 percent of compounds in exploratory trials fail in development. In many cases, exploratory trials involve small numbers of healthy volunteers and are of short duration; posting this information could leave patients with the misperception that more clinical trials are available for enrollment than in actuality. Dr. Speers called for including “at least advanced Phase II trials.” Dr. O’Rourke, noting the differences between Phase I and Phase IV trials and the desirability of a comprehensive registry, suggested requiring different data points for these different phases. Phases III and IV would have robust information, while Phase I might provide a more general description. “We should err on the side of inclusivity and register all trials, including exploratory Phase II trials,” stated Alfred Sandrock, Vice President of Medical Research-Neurology at Biogen Idec, Inc. He also suggested that the registry do its best to capture trials in the post-marketing setting, that is, the Phase IV trials. The FDA may not be fully aware of some of these trials, and some of the results may be hidden. He noted, however, that these personal views are shared by Burt Adelman, Executive Vice President of Development at Biogen Idec, Inc., but are
OCR for page 25
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary not entirely consistent with those of the Biotechnology Industry Organization (BIO). BIO has endorsed the concept of a clinical trial registry for marketed products, but has not supported a requirement for registering clinical trials involving drugs that are not available for commercial use. Specifically, James C.Mullen, President and CEO of Biogen Idec and chairman of the BIO Board of Directors, sent these comments: We support disclosure of all clinical trial information and results for marketed products. We believe that this is consistent with public health needs. For unapproved products we do not believe it is necessary to disclose trial information and results. Whilst some argue that these trials should be equally transparent, the public health argument is not as compelling when contrasted with the likelihood of introducing more information noise into the lay press and public and publicizing company know-how, strategies, and confidential information. On Eli Lilly’s clinical trial registry (Lillytrials.com), summaries of all the studies on the marketed drugs—Phases I through IV—are posted when the drug becomes available, so that every clinical trial conducted on a drug that goes to market will be posted, according to Dr. Breier. Harold Sox, Editor of Annals of Internal Medicine, explained the ICMJE position, which defines two types of trials. Clinically directive trials, which are intended to influence clinical policy and are typically large trials of agents that have already gone through preliminary testing, “should be registered—period, no argument.” Exploratory trials, which precede clinically directive trials, have been the subject of the most discussion between industry and the editors. The ICMJE excludes Phase I trials, which are designed to study major unknown toxicity or determine pharmacokinetics, but notes: “Between those two extremes are some clinical trials whose prespecified goal is to investigate the biology of disease or to provide preliminary data that may lead to larger, clinically directive trials.” Dr. Sox said the ICMJE has changed its position regarding those intermediate trials. Recognizing that requiring public registration might slow the forces that drive innovation, each journal editor will decide on a case-by-case basis about reviewing unregistered trials in this category. Authors whose trial is unregistered will have to convince the editor that they had a sound rationale when they decided not to register their trial. Editors plan to get together after 2 or 3 years of experience to come up with a body of “case law” that editors could apply more generally.
OCR for page 26
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary DELAYED DISCLOSURE OF INFORMATION AVAILABLE AT THE INITIATION OF TRIALS Fields 1 through 13 in the IOM background material (see Table 4–1) are also common to the ICMJE requirements, the voluntary commitment made by the pharmaceutical industry, and the WHO project. Fields 14 through 17 in Table 4–1 are the elements for which there are disagreements between the pharmaceutical industry’s proposed elements and the elements used by WHO and accepted by the ICMJE. Therefore, these elements were the focus of the discussion at the June 27 meeting. For the most part, the elements that needed further discussion corresponded with elements in the WHO list (Table 3–3) labeled “sensitive” for commercial reasons, except for one. WHO considered target sample size (Field 17) as a sensitive element, but that corresponds with Field 13 in the IOM list, which was not at issue. TABLE 4–1 Clinical Trial Registry Data Fields Discussed at IOM Meetings Information to be registered when a clinical trial is initiated Information to be registered for completed clinical trials # Clinical Trial Data Field Description—to be registered prior to the first patient visit in a study 1. Unique Trial Identification Number 2. Name of Sponsor 3. Brief Title 4. Trial Description in Lay Terminology 5. Trial Phase 6. Trial Type (e.g., interventional) 7. Trial Status (e.g., enrolling, completed) 8. Intervention Type (e.g., drug, vaccine) 9. Condition or Disease 10. Key Eligibility Criteria (including gender and age) 11. Location of Clinical Trial 12. Contact Information* 13. Estimated Target Number of Subjects
OCR for page 27
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary # Clinical Trial Data Field Description—to be registered prior to the first patient visit in a study Clinical Trial Data Fields Requiring Further Discussion 14. Hypothesis Statement: Statement of Intervention(s) and Comparison(s) Studied 15. Definition of Primary and Secondary Outcome Measures 16. Key Trial Dates: Registration Date, Trial Start Date (anticipated or actual), Projected Year of Trial Completion (=last patient, last visit) 17. Results a. Once a drug is marketed and commercially available in at least one country, summary results of all clinical trials other than exploratory will be posted, either by linking to a peer-reviewed publication or by providing the summary results in a common, nonpromotional format. b. Summary exploratory results will be posted if deemed to have significant medical importance. c. If results are published, a citation or link to the journal article will be posted in the registry and/or a summary of the results will be provided in a standardized format, such as ICH E3. *Some large-scale clinical trials have numerous investigators on different continents. In addition, many investigator sites are not equipped with the personnel to field numerous phone calls inquiring about a clinical trial. A possible solution is inclusion of a toll-free number that allows a central call line to direct patients and health care providers to the appropriate and most convenient clinical trial site. This central number also would allow the health care community to identify and contact the individual directly responsible for the conduct of the trial, if necessary. On certain occasions, the trial sponsor may consider some information in Data Fields 14 through 16 to be highly proprietary and consequently may want to delay public disclosure. Specifically: A hypothesis statement describes the nature of the trial, so that anyone reading it will know what is being tested in the trial and specifically which disease is of concern. For serious and life-threatening diseases, this information is already required and is typically posted in the “Trial Description” data field on ClinicalTrials.gov (Item 14). Industry has offered to register the intervention type (e.g., drug, vaccine), but on certain occasions, the sponsor may want to keep the intervention and/or comparator confidential (Item 14). The sponsor may want primary and secondary outcome measures to remain confidential in cases when such disclosure could put
OCR for page 28
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary an entire research program at risk of being unfairly copied by others who have not invested to develop the necessary expertise (see U.S. DHHS, 1998). The hypothesis and outcome measures in a clinical trial protocol describe the specifics of the sponsoring entity’s product development plan (Item 15). Generally the projected year of trial completion is acceptable; however, industry representatives indicated there may be circumstances in which this information is considered proprietary. For publicly held companies, stock market analysts carefully monitor the development of a product. Accordingly, projecting a year of clinical trial completion and subsequently missing that date—which can occur due to multiple, uncontrollable factors—could have a significant negative impact on a company’s shareholder value, especially for a small biotech company whose portfolio only consists of a compound or two (Item 16). The physicians working on the clinical trial also have full access to the hypothesis being tested, the intervention and comparison arms, the protocol-specific hypothesis statement, and a complete listing of primary and secondary outcome measures as part of the detailed clinical trial protocol. However, these physicians are required to keep the protocol information confidential as a result of a confidentiality agreement signed with the trial sponsor. The confidentiality agreement protects against unauthorized disclosure of the research that is being invested in by the clinical trial’s sponsor. Comments on Delayed Disclosure of Information Available at Trial Initiation “There is a legitimacy to confidentiality,” Dr. Breier stated. “Confidentiality spurs innovation, it gives it a competition that speeds things to market. If we remove all aspects of confidentiality of the innovation, the energy and some of the speed to market could get lost.” He gave examples of confidentiality that are embraced by the research community: peer review of scientific papers and NIH grant review. In both cases, the reviews are confidential to encourage the most thoughtful reviews of the research and its presentation. He provided a case example of the importance of confidentiality:
OCR for page 29
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary A drug called a selective estrogen receptor modulator is in Phase III testing; it has a 6-year Phase III plan. It will cost more than $100 million for the Phase III program, and there are multiple ways to design the clinical plan and different indications. Posting the full clinical plan at the inception, when the drug will not be on the market for over 6 years, will not benefit patients or prescribing physicians. The drug is not available. However, it will allow other companies to mimic the clinical plan and potentially be first to market. That could discourage innovation if all of the key aspects of this clinical plan were made available 6 years before the first prescription could be written, and I would argue that if we put the patient at the center, that’s not critical to the patient, but it is to the business model. Jeffrey Drazen, Editor-in-Chief of the New England Journal of Medicine, countered that competitors could not come up with a trial design that gave the answer in less time. “The competitor will always be behind.” Dr. Sox voiced concern that concealing the content of a field in a trial registry for 6 years or more is too long. People who are trying to summarize a body of evidence should be able to know about uncompleted trials so they can at least make a guess as to how an ongoing trial might affect the body of evidence of which it will eventually become a part. Dr. Breier responded that information would be available to important selected parties—journal editors should have access to the blinded data and the non-blinded data any time they are looking at a manuscript. Annetta Beauregard, from Eli Lilly and Company, clarified that for the majority of cases, all the fields—the intervention, the primary and secondary outcome measures—will be on the Web before the first patient is enrolled. Delayed disclosure will be sought in a minority of cases. Frank Rockhold, Senior Vice President and Director, Biomedical Data Sciences at GlaxoSmithKline (GSK) Pharmaceuticals Research and Development, said that if the purpose is to get as many trials as possible onto a registry, “don’t get hung up on the few trials that need to have their data masked until the end of the trial.” The FDA’s Robert Temple stated that any registry, if it is to facilitate understanding of what is planned and, when looking at the results, what was planned, needs to provide a complete description of the trial and any changes. “You really need to see the whole protocol, the statistical analysis plan, and any amendments to the protocol.”
OCR for page 30
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary Dr. Drazen noted that the data that are in contention—the drug or intervention that is being used, the outcome that is being measured—are major points. “In our discussions with a number of commercial sponsors of clinical trials, the idea has come up to place the details of clinical trials in a so-called ‘lockbox,’ which would be a way of assuring that the trials were carried out the way that they had been proposed, but that the detailed information wasn’t going to be widely available at the time the trial was either planned or completed,” he said. He asked the patient representatives in particular what they thought of this approach. The public/patient representatives were generally in favor of some limited form of data shielding in order to ensure continued development of innovative treatments. Ms. Terry said she thought “the lockbox concept is one that allows the commercial sponsor to feel safe as it evolves, whatever it has to go through in terms of drug development or device development, but also for people to get some information as that process goes. But since the information is banked somewhere, it also allows making sure that the trial was conducted in the way that it should have been.” She later added this qualifier: “I think the public will trust if the system is more healthy and more transparent and that a great deal has to remain outside the lockbox. And that ultimately it should be all outside, but I don’t think we can jump today to a tomorrow when everything is transparent.” Myrl Weinberg, Health Sciences Policy Board member and President of the National Health Council, suggested the word “lockbox” be replaced. “That is not a positive word that would make me feel particularly trusting. We really need to be careful from the beginning and then clarify exactly what would be held back and at what point would it be released.” Dr. Speers added, “There’s going to be a tradeoff. If the goal is to get good participation and to have carrots rather than sticks, which I support as well, then there’s got to be some support or recognition that some information is proprietary.” Patty Delaney, of the FDA’s Cancer Liaison Program, speaking as a former participant in a clinical trial, stated that trial sponsors are deluding themselves if they think they can keep their trial’s endpoints a secret. For example, patients entering a trial are told everything going on in that trial as part of the informed consent process. Therefore, there is no secret. The patient is free to tell anyone they want, including going to an Internet chat room to discuss the trial’s objective and endpoint. “In this day and age I’m going to go to the Internet. I’m going to go to the International Myeloma Foundation and say here’s the trial I entered, here’s what’s
OCR for page 31
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary going on, because people like me consider ourselves to be at the top of the pyramid of information and we want everybody to know about the trial. So that secret is blown at that point. Whether the trial is in the registry or not, everybody’s going to find out because patients are getting smarter and smarter.” Deborah Zarin, Director of ClinicalTrials.gov for the National Library of Medicine (NLM), stated that secondary endpoints are important. She used the example of the selective serotonin reuptake inhibitors (SSRIs) and suicidality. “If suicidality was always a secondary outcome measure, never a primary outcome measure, and you’re a systematic reviewer and you want to look at Paxil and suicidality but you had no idea how many trials ever looked at that, you’d be in the same conundrum that we’ve talked about. Now suppose you’re the parent of a child and you’re thinking of enrolling her in a study of SSRIs for depression, you might want to know if, in fact, someone had looked at the link between SSRIs and suicidality. What if you were the parent of a child who had committed suicide in such a trial? You might feel like you want someone to know about that, you want someone to have the opportunity to learn about that.” Dr. Sandrock commented that “a registry that does not include primary and key secondary outcome measures would ring hollow to me and probably to the public as well.” He continued: If a company chooses to employ a novel outcome in a Phase II or exploratory trial, that outcome could be listed in the registry in enough detail to give the reader an accurate view of what is being measured but not specific enough for others to emulate. For example, an outcome measure in a neurodegenerative disease trial could be listed as the 12-month change in regional brain atrophy as measured by magnetic resonance imaging (MRI). It does not detail how the MRIs are actually acquired, what segmentation parameters are being used, and how the data are being analyzed. A competitor would not readily be able to do its own trial with the same endpoint without that additional information. But a journal editor, a patient or a physician, will still have a good idea of what is being measured in such a trial. Phase III endpoints are typically standard endpoints such as survival, or are endpoints that have been extensively validated and sanctioned by regulatory agencies—hardly a trade secret. In fact I would argue that there are precious few secrets in this business. My belief is that
OCR for page 32
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary companies that participate fully in the registry can and will continue to find ways of being innovative. Sherry Marts, Vice President for Scientific Affairs at the Society for Women’s Health Research, suggested that a registry include a notation of whether subgroup analyses are being done. “From a patient perspective, if I as a woman or a minority am looking at three different trials for my condition and only one of them is going to break the data out by sex, that will be the one I sign up for. That’s a crucial bit of information that patients are going to need.” TIMING AND CONTENT FOR REPORTING RESULTS OF TRIALS To provide accountability as well as transparency, both WHO and the pharmaceutical industry call for results to be reported. WHO calls for results to be reported upon trial completion, and the Joint Position of the pharmaceutical industry limits results reporting to products approved for marketing in any country in the world. The ICMJE does not call for the posting of results. The June 27 discussion covered the function of results reporting, what the content of results reporting should be, and when results should be posted. Comments on the Timing and Content for Reporting Results of Trials The journal editors explained why the ICMJE did not call for describing the results of trials in a trial registry. Dr. Sox stated that they are more concerned with recording the existence of a trial and some basic data. He continued, “Seeing how many clinical trials are simply unworthy of publication after going through the peer review process, I have to worry about making publicly accessible the results of trials that have not gone through that process.” Dr. Drazen added that, if you know about a trial, you have the opportunity to query what happened with the trial. “We can’t deal with results reporting until we can get the trial existence part right.” The editors also expressed concern that showing the results of poorly done clinical trials is a disservice to the public. Posting results that have not gone through peer review may be dangerous.
OCR for page 33
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary Dr. Speers warned that data are not the same as information. Item 17, results of clinical trials, is most problematic because results are data, not information: I cannot imagine the type of results or the level of detail that could be included in a registry that would provide truly valuable information about clinical trials for the public. The proposal is to provide summary results. What does that mean? One suggestion is to link to a peer-reviewed publication. However, a single peer-reviewed publication will not put the results in a form where the public can easily interpret them. I ask the IOM to think very carefully about the purpose of providing results and to ask whether results can be provided in a way that would be truly meaningful to the public. She also stated that the registry should not get in the way of the doctor-patient relationship. Ms. Delaney stated that “patients should absolutely be given results of the trials they are in.” Ms. Terry went further to say that a comprehensive database could redefine the patient’s relationship with the system. “Patients can handle a great deal of information with the right systems in place to support that integration of data to information. Don’t keep information from them.” She also added that the publication of negative data or failed endpoints will improve research and clinical practice. Giving patients more rather than less information is the direction to take, Ms. Ireland said. “I’ve simply been amazed at the sophistication of patients, the amount that they know and understand and want to know about their own care and the care of their loved ones.” She added that results should be included, unless it could be proven that there would be a serious competitive disadvantage. “Negative results and toxicities need to be part of a registry,” said Dr. Pizzo. Ms. Terry made the same point, stating that knowing negative data and failed endpoints is helpful. Dr. Temple stated that a registry should provide results. “If they don’t do that, they miss a large fraction of the concern that’s driving interest in unpublished data, which is that you’re not telling me all the studies that there are. Now how exactly to do this properly is not a trivial question. Who writes it? It’s not peer reviewed and the FDA is worried about the potential for promotional reports. Nonetheless, if data don’t become available, one could ask what all this is for.”
OCR for page 34
Developing a National Registry of Pharmacologic and Biologic Clinical Trials: Workshop Summary Dr. O’Rourke, representing academic medical centers, expressed concern about databases that include data regarding trials open to enrollment and trial results. “These two types of databases have different goals, different audiences, and require different rules of operation. Realistic goals and rules for those two quite different types of information need to be looked at very seriously.” The most important issue, according to Dr. Breier, is ensuring that all relevant clinical trial results are available in a comprehensive, objective, and unbiased manner when a prescribing decision is made. “It’s an issue of WHEN, not IF. If a drug is in early development and it is not on the market and it is not available for a prescribing decision, the importance of that information is less so to the patient. The patient cares when the doctor has to make the prescribing decision and he wants all of the relevant information available in order to make the best prescribing decision.” Dr. Breier stated that results reporting is the way to address the file-drawer phenomenon, or failure to report negative results. GSK Pharmaceuticals has a registry that posts results once a product is approved, on the assumption that disclosing results on a register does not preclude publication, according to Dr. Rockhold. Before the end of the year there will be well over 1,000 trials in the GSK registry. Currently 28 products and more than 500 trials are posted. “I feel strongly that whatever the strategy is, it needs to include the results.” NLM’s Dr. Zarin noted that NIH is working on a proposal to enhance access to a wider range of information on NIH-supported clinical trials—with positive, neutral, or negative results. They see challenges, though, in reporting results, including verification of results and problems with misinterpretation.
Representative terms from entire chapter: