National Academies Press: OpenBook

Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary (2009)

Chapter: Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials

« Previous: Session 1A: Organization of the NCI Clinical Trials System
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 27
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 28
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 29
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 30
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 31
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 32
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 33
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 34
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 35
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 36
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 37
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 38
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 39
Suggested Citation:"Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials." Institute of Medicine. 2009. Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary. Washington, DC: The National Academies Press. doi: 10.17226/12535.
×
Page 40

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Session 1B: Lessons from Non-Cooperative Group Multi-Center Clinical Trials The next portion of the first session of the conference was devoted to lessons that can be learned from those who run multi-center clinical trials outside the Cooperative Group Program. The speakers for this portion of the workshop were a cardiologist, an industry scientist, and a community physician. Organization of Multi-Center Clinical Trials Dr. Robert Califf is a cardiologist, vice chancellor for clinical research, and director of the Duke University Translational Medicine Institute. He addressed organizational and operational issues that affect all clinical trials, not just those focused on cancer. He then discussed efforts that his university made to improve the efficiency of its clinical research efforts, and he made general recommendations for the clinical research enterprise at large. Dr. Califf began his talk by noting the complexity of clinical trials, but he pointed out that this complexity is to be expected given that a clini- cal trial is a transaction between the health care provider and the patient. “This by nature is a complicated and complex interaction, in which the investigator side of that equation has very split loyalties and obligations in terms of the requirement to function under a certain set of rules of the trial, and in trials that deal with sick people, also having a major obligation to put the welfare of the subject first,” he said. “This leads to an underestimation of the number of transactions that have to occur and overly ambitious expectations of how well this can occur. . . . We would 27

28 MULTI-CENTER PHASE III CLINICAL TRIALS be better off if we had more realistic expectations of what a clinical trial can accomplish in the first place.” In general, there are three models for clinical trials: the NIH model, the hybrid model, and the industry model. Dr. Califf noted that industry trials are regarded as being more efficient than publicly funded trials, while NIH trials are regarded as keeping the patients’ best interests as the top priority. He suggested that the primary task is to develop a function- ing hybrid model, and that lessons can be learned from both the NIH and the industry models. Dr. Califf concluded that there should be a balance of power, because clinical trials involve a set of human interactions in which there is no single right answer in regards to how best to run or oversee a trial. He also emphasized that there is a complex tradeoff between efficiency and ethics. “The degree to which you are willing to tolerate imperfection is a key decision that needs to be made,” he said. Dr. Califf noted that the coordinating centers for clinical research can be academic medical centers, which tend to be inefficient and have limited operational capabilities, or contract research organizations (CROs). The CROs essentially work as extensions of their sponsors, which typically are pharmaceutical companies (Shuchman, 2007). Nonprofit corporations or academic research organizations can also coordinate clinical research studies, and they aim to combine the efficiency of the CRO with an aca- demic or “public good” mission. He pointed out that while academic cen- ters do not do clinical trials to generate additional income, there are other benefits of doing such research, including enabling delivery of innovative therapies to patients, aiding in determining the best medical practices, and providing extra care to patients without additional cost. More fun- damentally, however, academic research centers conduct clinical research because research itself is a core mission of the institution, Dr. Califf said. Decision making in a clinical trial is a complex process, Dr. Califf pointed out, and although it is done by consensus within executive com- mittees, ultimately the sponsor “holds the trump card over withdrawing the funding.” Industry sponsors also hold the trump card when it comes to publication of the study’s findings, he added. “Our most prestigious academic centers, when they sign up to be a site in an industry clinical trial,” he said, “do not require that the results of the trial be published, routinely sign agreements to keep the plans for human experiment secret, and do not require a trial architecture that insures protection from sup- pression of negative results” (Davidoff et al., 2001; Schulman et al., 2002). What academic centers do require, he said, is the right to publish the data on their own patients in a trial. But if there are 100 centers participating in the trial, “it is statistical malpractice to publish the data from your own patients about the primary result,” Dr. Califf said. “Why don’t we put our foot down?” he asked. “The answer is because

LESSONS FROM NON-COOPERATIVE GROUP TRIALS 29 if I say no, the industry sponsor will just go to the next academic center that says yes. None of our physicians want to lose access to the cutting- edge drugs that industry has to offer.” As an example within his own field of cardiology, Dr. Califf pointed to a study that found that conflicts of interest went unreported in more than 80 percent of clinical trial results examined (Weinfurt et al., 2008). “We have real problems in terms of the architecture of trials and the results of contracts that we sign,” Dr. Califf stressed. One of these major problems is that many clinical trial results are not reported, especially if they are negative results. The registering of all clinical trials on ClinicalTrials.gov and the World Health Organization’s registry is helping in this regard, Dr. Califf noted.11 He added that aca- demic bureaucracy can also impede the reporting of clinical trial results. He then went on to discuss the costs of clinical trials and how to reduce them. Dr. Califf noted a recent conference and subsequent publica- tion on sensible clinical trial guidelines and recommendations to reduce bureaucracy in clinical trials (Sensible guidelines for the conduct of clinical trials, 2007; Eisenstein et al., 2008). At that meeting, which was attended by industry, government, and academic representatives, it was stressed that industry and others spent billions of dollars on clinical trial research (Moses et al., 2005). This suggests that there are probably sufficient funds for clinical research and that the problem is that those funds are being spent inefficiently. When the attendees calculated the costs of doing a specific clinical trial protocol by various methods of organization, they found massively different cost estimates for the same trial generating the same results that would pass FDA scrutiny (Figure 2). Dr. Califf reported on recent efforts by his own university to reduce the costs of its clinical trial operations. Duke University set up a task force that recently evaluated its clinical research to assess how to make it more efficient. As is true with all academic medical centers, the task force found that Duke’s clinical research was historically more of a “mom and pop” undertaking, according to Dr. Califf. “Each investigator was on his or her own, trying to make it without much organization or infrastruc- ture support,” Dr. Califf said. He described the core principles of the task force: first, a reaffirmation that clinical research is an important function of an academic medical center, and second, that efficiency in clinical trials is important. The task force identified a number of ways to make clini- cal research more efficient. One option was, for instance, to remove the responsibility for such research from individual clinical departments and place it within groups of clinical research investigators in a particular therapeutic area to form site-based research units that would be sup- ported by a clinical research support office. 11  See http://www.clinicaltrials.gov and http://www.who.int/ictrp/en.

30 MULTI-CENTER PHASE III CLINICAL TRIALS $450 $400 $350 2007 U.S. Millions $300 $250 $200 $150 $100 $50 $0 Full Cost Industry Streamlined More Industry Streamlined Total Coordinating Center Site Payments Other FIGURE 2 Patient trial costs. At the Sensible Guidelines for the Conduct of Clini- cal Trials Conference, representatives from academia, industry, and government agencies convened to discuss approaches to reduce bureaucracy in clinical trial design and conduct. Attendees reached massively different cost estimates for a specific clinical trial protocol that would generate the same results and still pass FDA scrutiny. Figure 2.eps SOURCE: Califf presentation (July 1, 2008) and Eisenstein, E. L., R. Collins, B. S. Cracknell, O. Podesta, E. D. Reid, P. Sandercock, Y. Shakhov, M. L. Terrin, M. A. Sellers, R. M. Califf, C. B. Granger, R. Diaz. 2008. Sensible approaches for reducing clinical trial costs. Clinical Trials 5(1):75–84, copyright © The Society for Clinical Trials. Reprinted by permission of SAGE. The research units function like small businesses embedded within the medical center, Dr. Califf said. Their responsibilities include acquir- ing economic approval for each planned and ongoing study, scientifically reviewing such studies, providing human resources for all non-faculty research staff as well as for orientation and oversight, and conducting the trial. The research units are led by a faculty member, the lead clini- cal research associate, and an administrator. The support office for these research units provides what Dr. Califf called the “rules and tools,” includ- ing information technologies and an understanding of the reimbursement rules that enables the office to effectively manage Medicare billing. Opera- tional oversight is done jointly by the medical school and the university health system.

LESSONS FROM NON-COOPERATIVE GROUP TRIALS 31 Dr. Califf concluded his talk by providing lessons and suggestions for improving the clinical research enterprise. In addition to publicly reporting start-up metrics, he suggested applying business practices to the organization of clinical trial operations. Such practices include struc- tural alignment, financial accountability, and appropriate rewards for site-based investigators. He also recommended developing more effective mechanisms for public–private partnerships that do not have conflicts of interest or compromise the academic mission. Other mid-term recommendations were to do “research on research” in order to develop evidence to support improved federal and global guidelines and to build capacity through reengineering the clinical enter- prise. Efforts in this regard have been started by the Clinical Trials Trans- formation Initiative.12 A more long-term goal is to develop a national learning system that would continuously record clinical practice data in electronic health records and disease registries. With such data, random- ized controlled trials could be conducted by inserting randomization into the data already being collected, Dr. Califf noted. For example, the Society of Thoracic Surgeons has a national database that keeps the records of about 80 percent of the patients who have cardiothoracic surgery in North America.13 Clinical trials can be run using this registry data without col- lecting much additional data, Dr. Califf said. Industry-sponsored multi-center clinical trials Next, Dr. Renzo Canetta of Bristol–Myers Squibb gave the industry perspective on industry-sponsored multi-center clinical trials, which, as he noted, have increasingly become multinational clinical trials. A survey of FDA-registered investigators reveals that although slightly more than half are from the United States, an increasing number are from European countries, Russia, India, and Argentina (Parexel International Corporation, 2008). Even more telling is the patient accrual to Bristol–Myers Squibb oncology pivotal trials between 1992 and 2007, which revealed a striking decline of accrual in the United States and an increase in accrual in other countries throughout the world (Table 1). Dr. Canetta noted that of the 10 oncology drugs Bristol–Myers Squibb introduced into the United States over the past 25 years, 5 were also approved for sale in other countries. The globalization of trials has been driven by the increasing international commercial interests of many pharmaceutical companies, he said, and by the difficulties that face patients and investigators who wish to participate in clinical trials in the United States. 12  See http://www.trialstransformation.org. 13  See http://www.sts.org/sections/stsnationaldatabase.

32 MULTI-CENTER PHASE III CLINICAL TRIALS TABLE 1  Accrual to Bristol–Myers Squibb Oncology Pivotal Trials % Pre-2000 % Post-2000 United States 70.8 40.5 Western Europe 18.6 26.8 Canada 5.8 9.5 Eastern Europe 2.9 4.7 Japan 1.6 1.3 Latin America <1.0 6.4 Asia/Pacific <1.0 5.8 Oceania <1.0 4.0 Africa <1.0 0.5 NOTE: Includes >20 trials, >20,000 patients. SOURCE: Canetta presentation (July 1, 2008). Dr. Califf pointed out during his presentation that the cost for each person in a clinical trial is three to four times as much in the United States as it is in India or China (Lustgarten, 2005). This greater cost is due, in part, to the added layers of bureaucracy linked to conducting clinical tri- als in the United States, and it is leading to an increase in overseas trials, he said. He also expressed concern about the “massive shift in clinical trial participation that is currently occurring away from the United States,” (Agres, 2005; Getz, 2005) and added, “What is particularly disturbing about recent trends is that when you look at the dropout of clinical inves- tigators in the United States compared to the dropout of the rest of the world, we are losing our most experienced clinical investigators. Very experienced clinical trialists are beginning to say ‘It is just not worth it any more—I am losing money, I am at risk of being vilified in my local newspaper for experimenting on people, and the regulatory risk is just too high.’” Later during the conference, Dr. Califf underlined this point by saying, “In the last six global clinical trials that we have been approached to coordinate by industry, the only major question industry [asked] us was how few patients do we have to put in the United States to be credible with the FDA. In other words, there is a mandate to not enroll patients in the United States because of the dynamics.” The globalization of clinical trials has had several consequences, including the need to fulfill numerous different local regulations in study activation, monitoring, and adverse event reporting, which results in the generation of extensive final clinical study reports that can be 1,000 pages long. Essentially, there is global auditing of the data generated. “Our data are the most scrutinized, regulated, audited, and [corrected] that you

LESSONS FROM NON-COOPERATIVE GROUP TRIALS 33 might dream of,” Dr. Canetta said. “We have been audited by the FDA, European health authorities, and the Japanese Ministry of Health.” There are complexities involved in integrating multiple study databases from several different countries, he added, and occasionally study materials and tools, such as patient-reported outcomes instruments, must be trans- lated into different languages. Dr. Canetta then gave a breakdown of the costs involved in running a multi-center industry-sponsored clinical trial. As others have noted, about 40 percent of the money and time spent on these trials is involved with the start-up phase of the trial. Start-up activities include negotiations with investigators and institutions, internal review by the sponsor, fulfillment of local regulations, IRB approval, and special protocol assessment. There can also be 34 or more internal reviews within Bristol–Myers Squibb, Dr. Canetta said. On average, across all therapeutic areas, it takes his com- pany eight months from the time that a protocol is conceived to the time that the first patient is treated in the trial. The company currently hopes to achieve five-month review times that might be accomplished by aligning review cycles so some are run in parallel rather than sequentially. Once a protocol is approved, there are numerous costs to industry involved in running a multi-center clinical trial, Dr. Canetta said, includ- ing funding for the personnel and infrastructure needed to run the trial; the fees of adjudication committees, IRBs, and data management centers; travel and meeting costs; and the costs of additional tests and procedures that are not reimbursed by insurers. The independent, blinded radio- logic or other data review done by adjudication committees is becoming common for trials with non-survival endpoints, Dr. Canetta said. This involves procuring the data or images—not all of which are digital—from other hospitals and then shipping them and storing them at a central location. “We are talking about patients that are on treatment for longer amounts of time. They basically have a life, and their life brings them to other places so procuring their X-rays and computed tomography (CT) scans costs money,” Dr. Canetta said. Other major costs for industry are those linked to producing and pro- viding the drug tested and the placebo or other comparators. Dr. Canetta noted that industry will provide an experimental drug to patients who take part in a study for as long as they wish to remain on the therapy, even after the conclusion of the study. Although this was not a major cost during the previous era of cytotoxic chemotherapy, it certainly is a signifi- cant cost in today’s era of chronic prolonged treatment, Dr. Canetta said. “What that means is that now we are still dosing free of charge patients who started receiving a drug in the Phase I trial started in November of 2003,” he said. The relabeling of a comparator, as is required by certain countries, or the masking of a placebo can also be costly, he added.

34 MULTI-CENTER PHASE III CLINICAL TRIALS Another type of increasing cost is that involved in sample process- ing and sample shipment, which in the current era of genomics and population pharmacokinetics is done on a global scale. There also is the cost of maintaining records for a number of years after the conclusion of the study or after the submission of the registration dossier. “This time changes from country to country,” Dr. Canetta noted. “It is not 5 years only because FDA says so. It is more because there are other countries that want more.” A final clinical trial cost Dr. Canetta mentioned is the cost of adver- tising for proactive recruiting of patients to trials. “We have been able to improve accrual to some of our trials by having advertisement campaigns about the existence of a particular trial in a particular geographical area,” he said. “That is something that in the past did not exist.” All these costs add up to a stunningly large figure of over $1 billion spent by industry on research and development for each new drug (new molecular entity) approved in the United States (DiMasi and Grabowski, 2007), Dr. Canetta said. He added that the costs linked to failed experi- mental drugs account for three-quarters of the cumulative drug develop- ment costs (DiMasi et al., 2003; Parexel International Corporation, 2008). These costs rise as an experimental drug passes through the various phases of clinical testing. Dr. Canetta concluded his talk by pointing out the irony of the public outcry for independent research that ignores the reality that industry- sponsored trials are the most regulated, intensively monitored, and scruti- nized experiments in biomedical research. He noted that it is in industry’s interest to continue to collaborate with capable investigators and organi- zations, wherever they are. The increasing globalization of the industry makes multinational collaborations attractive, he added, pointing out that the increasing need to identify and treat patients with specific genomic characteristics makes it necessary to access patients across a wider geo- graphic area. Dr. Canetta suggested mixed partnerships with cooperative groups whereby the cooperative group uses its own members and industry provides additional international investigators that industry takes the responsibility to monitor. “This has worked out very well in terms of the timeliness and the accrual. I think it is time that the NCI coopera- tive groups started thinking in these terms as well, because it is a real- ity,” Dr. Canetta said. Dr. Schilsky added during the discussion that his cooperative group is currently involved in such a mixed international partnership with Novartis for the testing of a leukemia drug within select genetic populations. “CALGB is the IND holder for the United States and essentially all of North America,” he said. “Novartis is the IND holder for the rest of the world. They have organized the extra U.S. sites and we

LESSONS FROM NON-COOPERATIVE GROUP TRIALS 35 have organized the North American sites. The protocol is activated and is working.” In the discussion following Dr. Canetta’s presentation, Dr. Califf pos- tulated that industry databases are “too clean and that a huge amount of money is spent on irrelevant study and that the NIH databases do just as well at getting the right answer.” Dr. Canetta responded by acknowledg- ing that industry probably does collect too much data, but he added that much of the data industry collects in excess of what would be collected by a cooperative group is used to satisfy the diverse regulatory demands of the various countries in which industry’s trials are run. In the same discussion Dr. Comis noted that cooperative group stud- ies are often used for supplemental New Drug Applications (sNDAs) or supplemental Biologic License Applications. “So in the United States at least, there is this symbiosis between what companies do and what the publicly funded system does,” he said. “It really benefits the whole nation and all cancer patients, so what would happen if our side went away?” Dr. Canetta responded that without cooperative group partnerships, the costs of industry doing investigations would increase and this would probably end up reducing the number of sNDAs that would be submit- ted. “But I think what will go away is probably more diffuse rapid access to experimental agents and getting physician investigators familiar with novel treatments,” Dr. Canetta said. “That will be really detrimental to the public interest.” Dr. Schilsky added that there probably would be fewer adjuvant therapy trials, fewer combined modality clinical trials, and fewer comparative effectiveness trials, “and lots of other things that the coopera- tive groups do that are important to patients,” he said. Private physician perspective on multi-center Clinical trials The next speaker, Dr. Alan Keller, chairman of clinical research at Cancer Care Associates in Tulsa, Oklahoma, addressed multi-center tri- als in the community and gave the perspective of a network of private physicians who participate in clinical trials. Dr. Keller began his talk by giving a number of reasons for private physician networks to participate in clinical research, including being able to provide the best care for their patients by accessing cutting-edge drugs, playing a proactive role in drug development, and getting good exposure for the group, which helps in patient and physician retention and local marketing. He added, however, “If anybody is in this to make money, they are in the wrong business, even on the private side.” Typically Phase I or investigator-initiated trials do not generate income for private groups. There is some potential for that

36 MULTI-CENTER PHASE III CLINICAL TRIALS in large Phase III registration trials, Dr. Keller said, but these larger trials are also more labor intensive. Part of the reason it is difficult to generate income from clinical trials is that the fixed costs for the infrastructure needed to develop and over- see the trial are high, but the capital investment is low. Patient accrual also is unpredictable and often unbalanced from study to study, and this affects the financial feasibility of running trials. Although some had hoped that enhanced information technology making clinical trial information accessible via the Internet would improve patient accrual, Dr. Keller said this has not been the case in his experience. He noted that the fixed costs involved with the development and regulatory and quality-assurance oversight of a clinical trial make up one-quarter of the costs of running a trial. Practice costs, study operations, and management account for about half the cost, and the remaining quarter is spent on direct research expenses and physician compensation, which in Dr. Keller’s group was $100 per patient. Most of the funding for privately run clinical trial net- works is from drug sponsors, although Dr. Keller noted that increasing numbers of participating investigators are members of CCOPs, which are funded by the NCI Division of Cancer Prevention through a competitive peer-review grant program. Dr. Keller noted that the basic infrastructure of private multi-center clinical trials is essentially the same as that of the cooperative groups (Box 4). Quality assurance of the trials is managed through frequent BOX 4 Multi-Center Trials: Private Group or Network Infrastructure • Leadership group • Committees o Monthly teleconferences o Meetings with industry • Administrative staff o Protocol writing o Central data management (investor initiated) o Data and Safety Monitoring Board o Finance/budget o Statistical support • Office space SOURCE: Keller presentation (July 1, 2008).

LESSONS FROM NON-COOPERATIVE GROUP TRIALS 37 audits by the drug industry or CROs. As is the case with cooperative groups, private research groups are subjected to numerous IRB reviews and a great deal of regulatory paperwork, he said. Dr. Keller then went on to describe what works and what does not work for private multi-center trials. On the list of what works he included pre-study involvement that allows investigators to participate in the study design and also initiation meetings, as long as they are done by teleconfer- ence or some other long-distance means so that investigators do not have to travel and lose a day of work to attend them. Dr. Keller was a strong advocate for central IRBs, although his group does not use the NCI central IRB. His group uses a private, commercial central IRB that meets frequently. “They help us from an economic stand- point in that in my practice we are in at least 17 different sites of service, so I do not have to do 17 different local IRBs,” Dr. Keller said. As for what does not work, Dr. Keller pointed to the excessive paper- work required by the FDA. “If we can cross-file drug INDs, why not cross- file investigators? How many 1572s14 and curriculum vitae does the FDA really need?” he asked, noting that when his group joins a trial with the pharmaceutical industry, it has to ship about four boxes of forms, many of which had been filled out for previous studies. To alleviate this exces- sive paperwork, Dr. Keller suggested setting up a national Web-based database for investigators that would contain all of an investigator’s relevant FDA forms and that could be accessed with a personal identifica- tion number. E-mail reminders could be sent every two years requesting online updating, and trials could be open only to eligible investigators who have up-to-date registrations in the system. Dr. Keller also suggested retooling financial disclosure requirements. Although he recognizes that such disclosure is important, it often is not relevant, as in the case of double-blind studies or studies in which an individual investigator does not manage the study data or contributes a small percentage of the cases and cannot block publication of negative data. He recommended requiring financial disclosure only for physicians who author the study report and contribute more than 10 percent of the patients in non-blinded studies or who own 5 percent or more of company stock. Financial disclosure should also be required of those who are on the data and safety monitoring board of a study. ”I do not think you need a financial disclosure in every instance,” Dr. Keller said. “It is just another piece of paper that goes in the box and gets sent in.” Dr. Keller was also critical of extensive auditing procedures. “You 14  The 1572 is a federal form and is the statement of the investigator that he will abide by the federal guidelines set forth in the Code of Federal Regulations for the use of drugs in an investigational setting. It is available at http://www.fda.gov.

38 MULTI-CENTER PHASE III CLINICAL TRIALS are in trouble when the number of appointments you have with auditors exceeds those that you have with your patients,” he said. In addition, he pointed to shortcomings in the reporting of SAEs. “I must spend maybe an hour a day as the [principal investigator] signing SAEs that come from Poland or somewhere else,” he said. These piecemeal reports do not put the SAEs within a meaningful context, Dr. Keller said. For example, the reports do not indicate on a global or a national multi-center scale how frequently the SAE is occurring within the trial. He was also critical of the frequently required follow-up burden of some clinical trials that requires the costly storage of records for long periods of time. Sites with insufficient accruals and low or non-accruing physicians also are problematic, Dr. Keller said. He presented an analysis of a group of about 800 physicians that found that about 60 percent of the physi- cians accrue less than one patient per year per study and about half of all research sites also have low accrual rates. Not only are low-accruing sites and physicians an economic drain on the system, Dr. Keller pointed out, but they also affect the quality of studies. “Poor accrual equals poor data,” he said, and he suggested closing sites that do not accrue enough patients. At the same time, Dr. Keller noted that as screening of patient vol- unteers moves from phenotype to genotype, the patient pool is getting smaller and spread out across a larger geographic area. “We need every- body to participate in clinical trials,” he said. “If we are looking for a target that is only in 20 percent of the patients and we only get 2 percent of eligible patients into clinical trials, it will take us forever to finish these trials. So we need to open this up across the country and make the system a lot simpler.” The increasing availability of more effective cancer drugs for second- and third-line indications for patients who did not respond or are no longer responding to initial treatment also limit patient accruals for experimental drug trials. Poor Medicare reimbursement for prescription drugs is also limit- ing patient participation in clinical trials and physician willingness to undertake these trials. As a result, many physician groups, including Dr. Keller’s practice, are limiting the number of their new patients on Medicare and are requiring those on Medicare who cannot afford their drug copayment to be treated in the hospital. Dr. Keller’s physician group recently decided that, due to this Medicare situation, it would no longer enroll Medicare patients in NCI-supported clinical trials. “We have cost- shifted for the very last time—I do not think there is anything left to shift,” Dr. Keller said. “We are under water, and the research was the first thing to fall here.” In the discussion after Dr. Keller’s presentation, Dr. Ronald Herberman, director of the University of Pittsburgh Cancer Institute, noted that his

LESSONS FROM NON-COOPERATIVE GROUP TRIALS 39 patients who are on Medicare fee-for-service plans do not have a bur- densome amount of medical bills. But in his area there are a number of patients who rely on Medicare HMO Advantage programs. “This has become virtually an absolute negative for any such patient to be able to afford to go into a clinical trial,” he said (Lin et al., 2008). “I believe that this is something that needs to be vigorously addressed at the national level,” he added. Another participant raised the question of how to deal with the inertia of private practitioners who are accruing a patient or less a year, given that “there is no carrot or stick to offer them.” Dr. Keller responded that they tackle this at the site level by explaining to investigators there that it is not financially feasible to keep the site open and then they close the site. In a later presentation, Dr. Laurence Baker, professor of internal medicine at the University of Michigan Medical School and chairman of SWOG, reiterated this approach, noting that this group’s policy is that a member in good standing with the cooperative groups must accrue 50 patients a year from its institution, not including patients accrued from affiliates under its supervision. “We also made it clear to our leaders, including committee chairs, that leadership in the group meant that there was accrual from your institution,” he said. “We spend a lot of time talk- ing about reducing the number of institutions and how we should police ourselves and reduce institutions that insufficiently participate.”

Next: Session 2: Barriers to Patient Recruitment and Physician Participation »
Multi-Center Phase III Clinical Trials and NCI Cooperative Groups: Workshop Summary Get This Book
×
Buy Paperback | $46.00 Buy Ebook | $36.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The NCI-sponsored cooperative groups have made important contributions to improving treatment for many types of cancer, including breast, ovarian, colorectal, and childhood cancers. Cooperative group research has been instrumental in establishing innovative treatments that improve outcomes and quality of life. Despite these successes, the Cooperative Group Program has faced a number of challenges that threaten its effectiveness.

To address this problem, the National Cancer Policy Forum (NCPF) convened a workshop titled "Multi-Center Phase III Clinical Trials and NCI Cooperative Groups" in Washington, DC, on July 1-2, 2008. The purpose of the workshop was to outline the challenges that the public clinical cancer research enterprise faces, and to identify possible solutions to these challenges.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!