Session 2:
Barriers to Patient Recruitment and Physician Participation

Following Dr. Keller’s presentation, there was a great deal of discussion concerning how to boost patient accrual and achieve greater physician participation in clinical research. Dr. Keller said, “In my experience, people either like doing clinical investigations or they hate it and will not do it. I’m not sure you can change that.” Dr. David Johnson, professor of medical and surgical oncology, director of the Division of Hematology/Oncology, and deputy director of the Vanderbilt-Ingram Cancer Center, suggested that more money should be spent on the doctors who are doing most of the accruing—the 20 percent who accrue 80 percent of the patient population. “Then,” he said, “patients will seek these individuals out, and it might also change the behavior of those physicians who find it burdensome to participate in the clinical trials process.” Dr. Keller agreed with this approach, as did Dr. Gwendolyn Fyfe, senior staff scientist in clinical hematology/oncology at Genentech. “It makes a lot of sense to pay sites that recruit a lot of patients more so that they can have excellent infrastructure,” she said. “It also makes sense to take community sites and make it easier for them by collecting less data. Do we really need to know where grade one toxicities happen at every site? Why not collect [just] deaths and SAEs at community sites and make it easy for them to participate in a trial?” Dr. Keller agreed and added, “There are virtually no incentives in this country for any doctor to enroll a patient on a clinical trial, and there are huge disincentives. Every portion of the clinical trials program has to look at how they can eliminate the disincentives that they contribute to the process.”



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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 41
Session 2: Barriers to Patient Recruitment and Physician Participation Following Dr. Keller’s presentation, there was a great deal of discus- sion concerning how to boost patient accrual and achieve greater physi- cian participation in clinical research. Dr. Keller said, “In my experience, people either like doing clinical investigations or they hate it and will not do it. I’m not sure you can change that.” Dr. David Johnson, professor of medical and surgical oncology, director of the Division of Hematology/ Oncology, and deputy director of the Vanderbilt-Ingram Cancer Center, suggested that more money should be spent on the doctors who are doing most of the accruing—the 20 percent who accrue 80 percent of the patient population. “Then,” he said, “patients will seek these individuals out, and it might also change the behavior of those physicians who find it burdensome to participate in the clinical trials process.” Dr. Keller agreed with this approach, as did Dr. Gwendolyn Fyfe, senior staff scientist in clinical hematology/oncology at Genentech. “It makes a lot of sense to pay sites that recruit a lot of patients more so that they can have excellent infrastructure,” she said. “It also makes sense to take community sites and make it easier for them by collecting less data. Do we really need to know where grade one toxicities happen at every site? Why not collect [just] deaths and SAEs at community sites and make it easy for them to participate in a trial?” Dr. Keller agreed and added, “There are virtually no incentives in this country for any doctor to enroll a patient on a clinical trial, and there are huge disincentives. Every portion of the clinical trials program has to look at how they can eliminate the disincentives that they contribute to the process.” 

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS Dr. Ruckdeschel countered that both his research on accrual to patient trials at the physician level and also data from Dr. Comis suggest that the high degree of variability in patient accrual is caused by individual investigator behavior (Albrecht et al., 2008; Comis et al., 2003; Lara et al., 2001). “It is the physicians and not the complexity of the IRB,” he said, “so if we are going to give people in institutions more, we need to get down to the physician level, because it really is individual physician behaviors that guide this, and much less the nuisance work that is part of that.” In response, Dr. Keller said, “In my practice it falls to just a couple of indi- viduals to interact with the IRB and do all that regulatory stuff. Therefore, it is a big deal.” Another participant suggested that the difference between clinicians and clinical researchers should be clearly defined and that aca- demic rewards or a cost structure that supports and encourages clinical research should be built into the system. Dr. Califf ended the discussion on an ominous note by adding that clinical research is now being subsumed by an economic measurement system that constrains the “impassioned doctor.” “He gets outvoted by his own practice partners because he is hurting the financial status of the practice,” he said. “You probably cannot fix this incrementally but rather need to do radical surgery, which is very risky. But the sense I am getting is that things are moving very fast in the wrong direction.” Continuing in the same vein as the previous discussion, the second session was devoted to understanding barriers to patient recruitment and to physician participation in clinical trials. Accrual of sufficient numbers of patients into clinical trials is a major barrier to the timely comple- tion of clinical research. Accrual is affected by both patient and provider attitudes about participation in clinical research. Several surveys have found that the single most important factor affecting accrual is whether a provider offers a clinical trial to the patient (Albrecht et al., 2008; Cox and McGarry, 2003). But many issues can affect patients’ decision mak- ing, including the informed consent process, unwillingness to receive a placebo treatment, and perception of personal benefit (Llewellyn-Thomas et al., 1991; Wright et al., 2004). Health care providers have to consider the time and resources that must be devoted to clinical trial participation and the liability of participation in the current regulatory climate. In addition, the recent acceleration in the development of new cancer treatments and medical technologies demands more clinical trials, compounding the chal- lenges that limit physician participation and patient recruitment. Beyond patient and provider attitudes, accrual is affected by strict eligibility crite- ria (George, 1996), and there is controversy on what criteria are necessary to obtain results applicable to the general population. Academic clinical researchers face additional challenges when making decisions about participation in cooperative group clinical trials. These tri-

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION als take a significant time commitment, but cooperative group trials are collaborative by nature, making it difficult for faculty whose primary research activity is participation in these clinical trials to succeed in the traditional academic system that rewards independent work. Deans in academic medical colleges must look after a host of career development issues for their academic research faculty, including teaching compe- tencies, committee service, research effectiveness, the ability to obtain research funding, and the tenure process (Hait, 2006). Dr. Mendelsohn opened the session by outlining his perception of the barriers to patient and physician participation in clinical trials. He cited issues in patient recruitment, including concerns about experimentation and randomization versus access to what is seen as the best treatment and also a failure to effectively communicate with patients about the tri- als. He also cited overly stringent eligibility requirements, which exclude many potential participants. As for the issue of what is preventing physi- cian participation in clinical trials, Dr. Mendelsohn agreed with previous speakers’ assessments that there is inadequate payment and frustration with excessive paperwork. He suggested that ethical concerns that place a current patient’s welfare before the need to gather knowledge for future patients and inadequate real-time prompting on protocols appropriate for patients both play a role as well. Finally, he stressed that another barrier to physician participation is a lack of recognition for their efforts, which impedes career advancement. “Publications of collaborative clinical trials have lots of middle authors15 who do not get much credit when promo- tion and tenure are being discussed,” he said. ACADEMIC CHALLENGES The session began with a panel discussion about academic chal- lenges to the effective conduct of clinical trials. The panelists included Dr. Laurence Baker of SWOG, Dr. Gordon Bernard from Vanderbilt Univer- sity, Dr. Michael Caligiuri of Ohio State University, and Dr. Alan Lichter of the American Society of Clinical Oncology (ASCO). Prior to the discus- sion, each panelist opened with a 10-minute overview of his perspective on the challenges. The first panelist, Laurence Baker, professor of internal medicine at the University of Michigan Medical School and chairman of SWOG, began by saying that the single most important barrier to a 15 Thereis a great deal of variability in the contributions made by authors of scientific publications. While the roles of the first and last authors are relatively standardized, the contributions made by the so-called middle authors can range from virtually nothing to equivalent to the first author’s. This ambiguity causes problems for tenure and promotions committees.

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS more successful cooperative group program was a lack of harmonization among cooperative groups, medical and cancer centers, and SPOREs. Instead of cooperation there is competition fueled by limited financial resources and a lack of the sort of communication that would foster more efficient alignment. “We say we are interested in cooperating, but we do not quite live up to that,” Dr. Baker said. “I do not think the system needs to be blown up. It does need to be reengineered pretty quickly, though.” Most major medical and cancer centers strive to be top-tier research organizations, he noted, but they recognize that grant support is insuf- ficient to run a research enterprise and that they have to rely also on philanthropy and patient care to underwrite their research costs. “Our cancer centers have become addicted to the revenue of clinical activities,” Dr. Baker said. “Clinical investigators are encouraged to increase patient care activities through increased utilization of chemotherapy, laboratory, imaging, radiation, and surgical services, so that revenue can be obtained, sustained, and grow in this research enterprise.” With this fiscal tightening has come a lessening of support for coop- erative group trials and, subsequently, insufficient patient accruals, Dr. Baker said. He gave an example of one medical center that joined SWOG in 1990 but resigned in 2003 because its executives claimed they could not afford to do cooperative group trials that only reimbursed $2,000 per patient. Another center has had four different principal investigators in the four years it has been part of SWOG. The previous principal investiga- tors said they resigned because the center director could not provide suf- ficient support for their research activities. The center has not reached the minimum number of patient accruals—50 patients—that SWOG set for their clinical trials. Another medical center, Dr. Baker said, had established priority criteria for the center that placed cooperative group trials fifth on the list and behind contract trials funded by pharmaceutical companies, even if the cooperative group trial was authored by one of its own faculty. This is in contrast to the experience of another cancer center and SWOG member, which had a center director and faculty committed to research. This cancer center substantially improved its low patient accrual rate after recognizing its problems and committing itself to rectifying them. The conclusion that can be drawn from these examples is that “money is necessary, but it is not sufficient,” Dr. Baker said. “We need to have cancer and medical center leadership that is committed to research.” He added that the goals for cancer centers, cooperative groups, and SPOREs need to be aligned regarding therapeutic clinical trials. One way to improve the system and the goals alignment would be to increase the NCI’s $2,000 cap on per-patient reimbursements, Dr. Baker suggested. He also recommended providing salary support to investiga- tors who design and conduct studies so as to permit a reduction of the

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION time they spend in the clinic. In addition, he suggested providing career development awards for new and mid-level faculty members doing out- standing clinical research. Unlike Dr. Mendelsohn, Dr. Baker did not think a lack of recognition of the participation in clinical trials stifles career promotion. Instead, he said, the problem is department chairs or center directors who want their clinicians to earn more clinical revenue. “We need to better recognize the downstream revenue created by these clinical investigations,” he said. Later in his talk, Dr. Michael Caligiuri, director and chief executive officer of the Ohio State University Comprehensive Cancer Center, con- curred. “The bottom line is that cancer does make money,” he said. “It is clear that the oncology product can be profitable and that you need a certain armamentarium to attract patients at an academic medical center.” That armamentarium includes clinical trials, he pointed out, so efforts in this regard should be appreciated by department chairs who are making salary and tenure decisions. “We have developed a transparent culture between what I will call the suits in our hospital doing the administration and the physicians who are doing the research. We’ve tried to embrace a culture where our administrators understand that protocol-driven research brings the kinds of patients we want into our cancer center. That has created a whole new perspective for our administrators in dealing with salaries, bonuses, etc., for clinical investigators.” Dr. Schilsky added that the U10 grant16 recognizes the clinical leader of a cooperative group, because that individual is listed as the principal investigator on an NCI-funded grant. “It gives that person credibility, respect, authority, infrastructure, and an ability to leverage local insti- tutional funds,” he said. He added, however, “That whole part of the process has just been progressively eroded, and I think it is putting the stability of our major institutions in jeopardy.” Dr. Baker pointed out that department chairs should recognize clinicians who receive a fraction of their salaries from federal funds that support their clinical research, just as bench scientists who receive federal grants for their research are recognized. Dr. Baker gave several suggestions for improving patient accrual to clinical trials, including eliminating unnecessary exclusions. An example of an unnecessary exclusion is when the patient has had prior treatment regimens; this was necessary in the cytotoxic chemotherapy era, when there was justifiable concern about the effect of several successive treat- 16 The U10 grant is the NIH Cooperative Clinical Research award mechanism. Under this cooperative agreement mechanism, the principal investigator retains the primary responsi- bility and dominant role for planning, directing, and executing the proposed project, with NIH staff being substantially involved as a partner with the principal investigator.

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS ment regimes on the bone marrow of the patient, Dr. Baker said, but this exclusion is no longer relevant today. Another exclusion that is no lon- ger relevant is having had a prior cancer. “I am one of 11 million cancer survivors, but we know that hundreds of thousands of us will develop a second cancer,” he said. “Don’t we deserve to be included in studies of potentially important new agents?” During the discussion, Dr. Johnson described additional barriers to patient accrual, including consent forms that require patients to pay for their clinical care if their insurance does not provide adequate reimbursement and also patients’ concerns about having to pay for their own medical care if the experimental treatment causes any injuries. Dr. Baker also questioned the wisdom and ethics of placebo-controlled trials for metastatic cancer. Although placebo-controlled studies are con- sidered a scientific way to prove the value of a new treatment, this design is always the lowest hurdle a pharmaceutical company must achieve for regulatory approval. “Certainly it makes no sense to have sham IVs and [pharmacokinetic] studies and multiple venipuctures,” he said. “That is what makes our patients upset. We would be better off recognizing that cancer clinical trials should primarily serve the needs of patients, not those of society.” Dr. Baker concluded his talk by saying, “We should recognize the strength of the U10 mechanism and consider expansion of it. Our U10 sites are the most important sites of our group for science, accrual, and leadership.” He noted that of SWOG’s current 48 studies, 44 have been authored by a member faculty from a U10 institution, and the average accrual from these institutions is more than 70 patients a year. Later in the discussion, Dr. Schilsky agreed: “When we look at the statistics on our U10 holders, all of whom are at major academic medical centers like SWOG, they account for essentially all of our committee leaders, and the majority of our protocol chairs and committee members, as well as pro- vide a substantial fraction of our patient accruals.” The second panelist, Dr. Gordon Bernard, professor in the Department of Medicine and assistant vice chancellor for research at the Vanderbilt University Medical Center, discussed some challenges faced in academic medical centers regarding physician participation and patient recruit- ment. He highlighted many of the topics that had already been discussed: the need for standardized definitions and metrics for evaluation across institutions, protocol design and Medicare reimbursement issues, the need to reduce the number of non-value-added steps, the need to make it easier to conduct steps in parallel, and the fact that too many trials harm an institution’s ability to conduct efficient clinical research. Dr. Bernard suggested that “The consent forms that come out of the NCI IRB are lengthy, complicated, wordy documents that are difficult

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION for my committees to swallow.” Dr. Stephen Grubbs, medical oncologist at the Helen F. Graham Cancer Center and principal investigator of the Delaware Christiana Care CCOP, agreed with this point during his pre- sentation. “My average consent form is now 30 to 35 pages long,” he said. “Anybody who thinks that my patients are reading 35 pages of a consent form is out of their mind. I go through those consent forms in summary, page by page with the patients. But I do not believe that they ever go home and read the whole thing.” Dr. Bernard suggested standardizing consent and contract language and having a “short form” approach to consent forms. He noted that the Association of American Medical Colleges has a working group currently trying to develop a short form that can be layered on top of a long, com- plicated consent form. The short form states in a few words what is going to happen to the patient, with links to the rest of the document for those who want more detail. He also suggested expedited study approval and review of consent language at secondary sites once an accredited IRB has given approval. “You would not need a full committee to review it—just the chair could look at it,” Dr. Bernard said. Finally, Dr. Bernard discussed his viewpoint on the challenges facing academic medical centers with respect to physician participation and patient recruitment in cooperative group clinical trials. He has had clini- cal investigators ask him to direct institutional funding to make up the deficit for the cooperative group trials, but it was not possible to justify that these trials should be given priority over other clinical research. Their clinical trials office, like others, rely on revenues generated for its survival, but it always operates on a deficit. Dr. Bernard also discussed academic rewards, and started by explaining that Vanderbilt has two categories of faculty: physician scientists, whose primary income is to come from aca- demic grants, and clinician educators, whose primary income is to come from seeing patients. While clinician educators do the patient recruit- ment, there is not a sufficient incentive for them to recruit because their evaluations do not depend on it. Dr. Bernard recommended more effort in addressing these issues, and he added that Vanderbilt is trying to find ways to allow young investigators to succeed. INSURANCE BARRIERS The third panelist, Dr. Michael Caligiuri, director and chief executive officer of the Ohio State University Comprehensive Cancer Center, said that his experience as a director of a comprehensive clinical cancer center led him to believe that lack of insurance reimbursement is an impediment to patient accruals (Lara et al., 2001; Mattel et al., 2004), estimating that such a lack of insurance prevented about one-quarter to one-third of the

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS patients at Ohio State University’s Comprehensive Cancer Center from participating in clinical trials. Dr. Caligiuri explained that commercial health insurers often refuse to pay for routine care costs associated with a clinical trial, even though those same costs would be reimbursed if the patient were not receiving experimental treatment. These costs are sub- stantial and include physician visits, blood work, and X-rays. A lack of such reimbursements affects many patients, Dr. Caligiuri said, because they reside in states where there are no laws that require commercial health insurers to pay. Even the 23 states that currently have clinical trial coverage laws do not necessarily require insurers to cover all cancer patients, such as patients on Phase I or II trials or those with employer self-insured plans,17 in which a large company self-insures its employees (NCSL, 2008). “So without a federal policy, cancer patients cannot be guaranteed that if they enroll in a potentially life-saving clinical trial there will be someone there to pay for the process,” Dr. Caligiuri said. “This is a huge impediment for most patients giving any further consid- eration of an experimental agent.” After a lobbying effort by Dr. Caligiuri and his colleagues, the Ohio legislature passed a bill, ultimately signed into law by Governor Ted Strickland, obligating health plans to pay for routine costs of care when a cancer patient enrolls in a clinical trial. 18 Dr. John Feldmann, medical director at the Moses Cone Regional Can- cer Center, noted later in his presentation that many large companies now self-insure their employees. Any self-insured plan that is offered as part of a benefits package is covered under Employee Retirement Insurance Security Act (ERISA) (Butler, 2000; DOL, 2008). This is the federal law that sets forth the minimal standards that must exist in any independent health plan; it currently does not require covering the routine care costs linked to clinical trials. Furthermore, this act has a section that preempts all state laws. “So if you have a self-insured plan, no state law about clini- cal trial coverage will be applicable,” Dr. Feldmann said. This is a major problem, given that in some areas, such as North Carolina, about half of patients now have self-insured plans offered by their employers. To overcome this impediment, Dr. Caligiuri suggested requiring 17 A plan offered by employers who directly assume the major cost of health in- surance for their employees. Some self-insured plans bear the entire risk. Other self- insured employers insure against large claims by purchasing stop-loss coverage. Some self-insured employers contract with insurance carriers or third-party administrators for claims processing and other administrative services; other self-insured plans are self-administered. 18 To amend sections .0 and .0 and to enact section .0 of the Revised Code to prohibit insurers, public employee benefit plans, and multiple employer welfare arrangements from excluding coverage for routine patient care administered as part of a cancer clinical trial, SB 186, 127th General Assembly of the state of Ohio.

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION ERISA plans to provide coverage for the routine costs of participating in clinical trials. This was pursued by two members of Congress, Rep- resentative Deborah Pryce and Senator Sherrod Brown, and led to the introduction of the Access to Cancer Clinical Trials Act of 2007 (H.R. 2676 and S. 2999).19 Dr. Caligiuri noted later in discussion that getting political action in this regard was relatively easy, because there is bipartisan support for ensuring that the routine costs of care linked to participating in clinical tri- als be reimbursed and because requiring this does not require a financial commitment on the part of the state or the federal government. “You have a paying customer who is not getting what is due,” he said, “so it was really easy to get support for it from both Republicans and Democrats.” He suggested fostering a grassroots effort by clinical trialists and patients to support the passage of the federal law. A workshop attendee noted that there are a number of cancer organi- zations, including ASCO, that have been working with Senators Edward Kennedy and Kay Bailey Hutchison on legislation that addresses coverage of routine care costs linked to cancer clinical trials as well as other cancer care concerns. “Given the congressional interest in some sort of compre- hensive cancer omnibus bill, there might be an opportunity there,” the participant said. The same discussant also noted that there is debate about whether CMS should codify its Medicare clinical trials coverage policy. Currently there is a national coverage decision to reimburse the routine care costs linked to clinical trials with therapeutic intent, but individual contractors with the Medicare program can opt not to cover this cost. Dr. Leslye Fitterman, epidemiologist in the Office of Clinical Standards and Quality at CMS, agreed that contractors at the local level have the author- ity to determine what is reasonable and necessary, and she said that this is why there is so much disparity in interpretations of therapeutic intent in terms of understanding what is actually needed and what is considered standard of care. That disparity could be rectified with a federal ruling in this regard. Dr. Ruckdeschel noted that in Michigan there is only one major health insurer—Blue Cross and Blue Shield—and it has a “Don’t ask, don’t tell” policy about reimbursing routine costs linked to clinical trials. “We do not see this as an issue,” he said, “and when we have discussed this in other forums, a lot of states have said they do not want to stir this up with 19Although the legislation was referred to committee, it was not acted on during the 110th Session of Congress. U.S. Congress, House of Representatives, 2007. Access to Cancer Clinical Trials Act of 2007. H.R. 2676. 110th Cong., 1st sess. (June 12, 2007). U.S. Congress, Senate, 2008. Access to Cancer Clinical Trials Act of 2008. S. 2999. 110th Cong., 2nd sess. (May 8, 2008).

OCR for page 41
0 MULTI-CENTER PHASE III CLINICAL TRIALS a piece of legislation because they have a similar modus operandi, where they do not tell the insurance company the patient is on the study, the insurance company does not ask, and things just proceed as if it were a normal bill.” But Dr. Feldmann cautioned that the “Don’t ask, don’t tell” policy carries certain risks: “If you have a supply drug and you bill for the infusion without an associated J-code because the drug has been sup- plied, the insurance company will detect the fact that the patient is on an experimental trial, and then everyone is at risk because the costs will be enormous at that point.” One participant pointed out that in his area of the northwestern United States, about 12 percent of patients on Phase III trials or Phase II clinical trials with therapeutic intent are denied reimbursement of rou- tine care costs. And Dr. Mendelsohn pointed out that there is even less reimbursement for patients in Phase I trials. “We have lots of patients that turn down a chance for a Phase I trial and take the standard old 5FU [5-fluorouracil] because the insurance company will pay for that,” he said. “That is a new phenomenon in the past year.” Dr. Bernard noted in his presentation that if protocols for Phase I and Phase II trials are rede- signed so that clinical response is the clear primary objective, Medicare might be more likely to pay for the costs linked to patient participation in the trials. “Of course a Phase I study explores a lot of safety issues and the like,” he said, “but I do not think you would give it to somebody if you did not think it had a prayer of affecting the tumor size or whatever it is that you are measuring as your surrogate marker of disease.” ACADEMIC RECOGNITION FOR CLINICAL TRIALISTS The final panelist, Dr. Allen Lichter, executive vice president and chief executive officer of ASCO, addressed the pitfalls of career advance- ment for the clinical trialist. A former medical school dean and clinical researcher who worked for many years in the Cooperative Group Pro- gram, Dr. Lichter has found that working in oncology clinical coopera- tive groups is frequently not well rewarded with academic recognition and advancement. He said that this is caused by a number of factors, including: • a lack of awareness by promotions committees of what such research entails; • the collaborative nature of the research, which makes it difficult to mark individual accomplishments; • the time factor involved in clinical research; and • the under-funding of much of this effort.

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION Clinical research is not well understood by academic promotion com- mittees, Dr. Lichter noted. There is no other specialty that has a clinical research infrastructure similar to oncology, he said. Furthermore, the work is done largely off-site with investigators from other institutions who are largely unknown to the committee, and many committees assume that most clinical trials are performed by industry, so they believe that there is not much academic productivity or thought that goes into it. Clinical trial research is also undervalued, Dr. Lichter added. “There is very little sense of the intellectual rigor and complexity that goes into trial design and protocol execution,” he said. “The assumption is, ‘This is Tylenol against aspirin for headaches and can be designed in about 10 minutes and carried out by anybody.’” There also is no sense of the intense time commitment involved nor an appreciation for the critical role these studies play in advancing the field, Dr. Lichter said. “We should probably bring in those 30-foot process maps [prepared by Dr. Dilts]20 and say this person deserves promotion, because they negotiate this every day.” In the discussion, Dr. Curran did note, however, that there is some prestige involved in having committee leadership appointments in cooperative groups and that this prestige can result in opportunities or promotions in the investigators’ own institutions. Another problem making it difficult to attain recognition and promo- tion from such collaborative work is that it is not independent. “We stand up as academic leaders and talk about the importance of team science,” Dr. Lichter said, “and then we sit down in promotions committees and ask what has this person done that is independent. I have said to the faculty when I was dean that if Abbott and Costello were trying to get tenured for comedy, we would turn them down because they did not have inde- pendent comedy routines. If we value team science, we have to figure out how to reward it.” As others had noted already, much of clinical research time spent by investigators is not funded, and division directors and department chairs are clamping down on uncompensated time. “To maintain salaries in most clinical departments today,” Dr. Lichter said, “there is a very much ‘eat what you kill’ salary structure. If you are not seeing patients and gener- ating revenues, you are not going to have your salary supported. Many promotions committees flip to the grant page and look for grant funding in your name, yet grant funding in the name of the investigator is either lacking or is as a co-investigator.” Another problem is that clinical trials can take years from conception to publication, “and in most academic medical centers today, the tenure clock is not calibrated to this time scale,” Dr. Lichter said. 20 See Dilts et al., 2006, 2008.

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS CCOP PERSPECTIVE ON PATIENT ACCRUAL The final session of the first day of the workshop began with Dr. Stephen Grubbs, medical oncologist at the Helen F. Graham Cancer Cen- ter and principal investigator of the Delaware Christiana Care CCOP, discussing the contributions, benefits, and challenges of the CCOP pro- gram. CCOPs contribute one-third of the total patient accrual to NCI treat- ment and prevention trials, Dr. Grubbs said, including a large number of early-stage patients, to which the community physicians have more access (NCI, 2003). Dr. Grubbs noted that only half of the participat- ing physicians are medical oncologists or hematologists—the others are primary care physicians, surgeons, radiation oncologists, or urologists. By participating on cooperative group committees and other NIH com- mittees, CCOP members also provide a community perspective in trial design. And because CCOP is a national network with members spread throughout the United States and Puerto Rico, it allows relocated patients to continue their participation in a clinical trial—something that proved invaluable, for example, when Hurricane Katrina forced many Gulf Coast residents to move. There are also benefits to the community physicians who participate in CCOP, including the opportunity to provide state-of-the-art cancer treatments to their patients and to design and participate in research studies. The physicians also benefit from the training of their research staff, investigator mentoring, and peer review that the cooperative groups provide, Dr. Grubbs said. He then went on to discuss what factors affect participation in clinical trials. Studies on a national scale show that the top two reasons that patients do not enroll in clinical trials are that physicians do not offer clinical trial information to the patients or that there is no protocol in which the patient can participate. Other factors high on the list are age and performance sta- tus requirements of the trial, the patient’s desire for standard therapy, and barriers that prevent patients in underserved communities from participat- ing (Albrecht et al., 2008; McCaskill-Stevens et al., 1999; Mills et al., 2006). When he conducted a study of barriers to patient accrual at his CCOP in Delaware, he found that three-quarters of the 1,000 patients studied were not eligible to participate in a clinical trial.21 Among those who were not eligible to participate in the clinical trial, there was no trial available for 54 percent, 20 percent could not participate because of poor performance status, 6 percent were denied participation because of a prior cancer his- tory, and 5 percent could not participate because of abnormal laboratory results. 21 S. S. Grubbs, Delaware Christiana Care CCOP, unpublished data, 2008.

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION Of the patients who were eligible for a trial, 38 percent enrolled and 62 percent declined to enroll in clinical trials. Two-thirds of those who declined did so at the advice of their physicians, many of whom thought the patients were too old to participate in a clinical trial. An age of more than 70 years was the biggest factor against enrollment in a clinical trial, followed by gender, with women less likely to participate than men (with the exception of participation in breast cancer treatment trials). “I think there is a silver lining in there,” Dr. Grubbs said. “We were successful in getting women on the breast cancer trials because of the publicity and patient advocacy that is going on in our country. When women get breast cancer, they already have it in their heads that clinical trials are a positive thing.” Other Delaware CCOP accrual barriers that Dr. Grubbs discussed are a lack of a dedicated recruitment effort for medical and non-medical oncology clinical investigators and infrastructure resource limits, which have become especially pressing with the declining CCOP budget. “Our budget to run the CCOP only covers about two-thirds of the cost of my hospital, which kicks in a third to cover it. That does not even include the in-kind giving that the private practices do,” Dr. Grubbs said. Other barriers are regulatory and documentation burdens, a lack of clinical research associate retention, and the insurance restrictions mentioned by other speakers. “Our problem in our state is the ERISA exemption,” Dr. Grubbs said. Another major barrier is the constraint that participating in clinical trials places on a physician’s time—a constraint that is exacerbated by the increasingly more complex trials and required consents, Dr. Grubbs said. “I give double bookings for my average patients on a complex trial like one with two targeted molecules for colon cancer chemotherapy,” he said. “It takes me that much time to sort through all the toxicities and do the dose adjustments for each drug.” Nontraditional physician investigators, such as radiologists and pathologists, also find their time overburdened conducting the tests and analyses required for clinical trials, and so they often resist doing them, especially since they are not paid to do these extra tests and procedures, Dr. Grubbs said. Dr. Grubbs suggested several ways to overcome some of these accrual barriers, including providing adequate funding for the CCOP infrastruc- ture and offering third-party reimbursement for oncology care linked to participating in a clinical trial. He also suggested developing more clinical trials for the majority patients for whom there is no appropriate protocol or who cannot participate in available protocols because of ineligibility. In addition, he suggested easing regulatory burdens and simplifying consent forms. Dr. Grubbs said he would like private insurers and CMS to recog-

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS nize the value of clinical trials and to reimburse the associated expenses more readily, and he stressed the need for public education about the importance of participating in clinical trials. He suggested that there be fellowship training tracks to train and encourage community clinical investigators and also ways to publicly recognize those community phy- sicians who regularly participate in oncology trials so the patients seek them out. He also suggested establishing a mentoring program for new and underachieving community research sites. To improve patient accrual in his CCOP, Dr. Grubbs said, the organi- zation recently started requiring a minimal annual patient accrual amount from its participating physicians as well as attendance at a cooperative group or research meeting at least every other year. He is encouraged by how physicians in his group have been responding to the new require- ments, he said. “I have got four physicians that have put two people on trial in the last five years that have each accrued three to four patients so far, and the requirement has only been around for six months,” he said. “But there will be gnashing of teeth at the end of the year, because some physicians are going to be stripped of their ability to participate in trials.” Dr. Grubbs also showed how CCOP participation has transformed the quality of cancer clinical care in Delaware. It spurred the development of one of the most advanced state cancer control programs in the United States today, he said. This program includes a state cancer consortium, a statewide colorectal screening program with funding for the uninsured, and funding for uninsured patients for up to two years of cancer control, including cancer clinical trial activity. CCOP participation has also raised the cancer standard of care in the state, which has resulted in a significant drop in cancer mortality in Delaware. Delaware, New York, and Maryland lead the most improved category in cancer mortality rates, with drops of more than 12 deaths per 100,000. In addition, Delaware’s cancer mortality is dropping at a rate double that of the national average (DCC, 2007). He attributes much of that drop in cancer rate to the CCOP and the clinical trials it runs in the state. Dr. Grubbs concluded by stressing the importance of the CCOP pro- gram. “CCOPs produce high quality and quantity clinical research and are the vehicles that elevate community quality cancer care,” he said. “The CCOP program is robust now but under the same multiple pressures experienced by academia and private practice medicine.” The next speaker, Dr. John Feldmann from the Moses Cone Regional Cancer Center in North Carolina, presented a perspective from a much smaller CCOP. He began his talk by noting that about 80 percent of patients still receive care in community practices. “It is clear that aca- demic centers alone cannot produce the number of patients we need for

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION timely completion of NCI studies,” he said. “And community physicians, not only through CCOP, but also independently, need to remain in the process somehow.” However, Dr. Feldmann noted a peculiar distribution of open protocols among participating CCOP sites in North Carolina. More than three-quarters of those sites have less than 10 protocols open, and nearly half the sites currently have no protocols open. “Obviously, there are a large number of sites doing a small or moderate amount of research throughout the state, which is a problem,” he said. He raised the question of whether these sites would do more research if they had more resources. Dr. Feldmann then discussed the pressing issues that are limiting the participation of patients and physicians in cooperative group clinical trials, including reimbursement and funding issues, regulatory burdens, competition from industry trials, and problems with trial publicity. He noted that, traditionally, a sizable portion of the revenue that supports community oncology practices is the profit margin made on chemother- apy. But because of recent changes in reimbursement by CMS, that margin has been shrinking, and some small practices that do not have enough volume to receive a discount on the drugs are actually having difficulty making any margin at all, Dr. Feldmann reported. Hospitals with a sig- nificant indigent load can purchase chemotherapy drugs at a reduced price because of a Health Resources and Services Administration (HRSA) program called 340B.22 So marginal payors are increasingly being referred to hospitals, Dr. Feldmann noted, as are Medicare recipients who cannot make their drug copayments. The recent 10 percent cut in Medicare reimbursements23 puts more financial pressures on private practices, fostering cuts in personnel and patient time that make it less likely they can afford to participate in clinical research. “Everyone in private practice is trying to cut personnel, and those personnel needed for good data collection or to deal with the regulatory burdens of clinical trials are going to be among the first to go,” Dr. Feldmann said. In addition, practitioners, who now have less time to spend with patients, are less likely to explain to patients their clinical trial options and to do consents; nor are they likely to have the time and willingness to fulfill other obligations of a primary investigator. One way to deal with this is for the private practitioner to work with local hospitals that can provide data management services. But this is difficult for rural 22 Seehttp://www.hrsa.gov/opa/introduction.htm. 23 The 10 percent Medicare payment reduction that went into effect July 1, 2008, was rescinded when both Houses of Congress overrode a July 15 presidential veto to maintain current funding levels for the rest of the year, while providing a 1.1 percent payment increase in 2009. Medicare Improvements for Patients and Providers Act of 00, P.L. 110-275, 110th Cong., 2d sess. (July 15, 2008).

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS practices to carry out, because the hospitals are often smaller and do not have the financial resources to provide this type of service, Dr. Feldmann said. As had been previously mentioned, Dr. Feldmann noted that NCI funds are insufficient to address the financial insufficiencies faced by CCOP physicians, which increasingly rely on hospital partners to sub- sidize clinical trial costs or on industry trials revenues. But hospitals are also facing increasing financial pressures, and Dr. Feldmann was skepti- cal that the new partnerships between the NCI and industry could solve these financial problems. Most of the industry trials involve INDs, and the trials are much more complex and require more time. “The payments may be threefold higher, but in our experience, the increased work has been out of proportion to that increase in reimbursement, and these trials have not improved the financial situation,” Dr. Feldmann said. Industry clinical trials also compete for the same limited pool of patients seen by community physicians, he added. Patients are more likely to accrue to industry trials, because they usually involve a Phase II component that does not require randomization. Other competitors for NCI cooperative group trials are industry-only networks. “These are being sold to smaller practices and even to some practices currently in CCOP as a way to do research in the office without as much financial risk,” Dr. Feldmann said. The network handles the regulatory burden centrally, and the design is more efficient for industry partners, because they can often contract directly with these networks and not have to con- tract with all the individual practices involved. But for financial reasons almost all of these networks now exclude NCI cooperative group trials, Dr. Feldmann said. He also expressed concern about the studies being done as part of CMS’s new Coverage with Evidence Development program.24 These stud- ies use CMS’s extensive database to determine the effectiveness of certain treatments and whether they should be covered by the insurer. “Our con- cern is that this will take the place of support of clinical research eventu- ally by CMS because it is a faster and more effective way of getting data,” Dr. Feldmann said. “From a research standpoint, this is probably not the ideal way to advance medical knowledge.” He then discussed clinical trials publicity. He noted that a Harris poll (Harris Interactive, 2001) showed that only 16 percent of cancer patients were aware of clinical trials, although they generally had a favorable impression of them. He also pointed out the difficulties in searching for appropriate clinical trials for patients. Although there are multiple trial searching sites, the information on them is often inaccurate, outdated, or 24 See http://www.cms.hhs.gov/CoverageGenInfo/03_CED.asp.

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION incomplete. In addition, many of the trials are not regionalized, with the result that “studies from 2,000 miles away are mixed in with those next door,” he said. “It can be very difficult for the patients to find them.” Some states, such as Georgia and North Carolina, are trying to regionalize their search engines and to make the clinical trials information more accurate and up to date, he noted, but more needs to be done in this regard. Dr. Feldmann concluded his presentation by stressing that commu- nity participation is essential for timely trial completion, but it is going to need increased outside support in the community, possibly via part- nerships between practices and hospitals. “Increasing industry trials at the expense of NCI trials is a huge threat,” he said, and much needs to be done to resolve the reimbursement and insurance issues that he discussed. “We are going to need some joint regional efforts, which may include expansion of the CCOP program or local partnerships with a state between practicing groups and local medical centers. Something is going to have to be done to shore up this effort.” PERSPECTIVE FROM A NATIONAL HEALTH CARE SYSTEM The next speaker, Dr. Richard Kaplan, associate director of the National Cancer Research Network (NCRN),25 discussed the United Kingdom’s recent successful initiative to boost patient participation in clinical trials. The United Kingdom has a national health care system called the National Health Service (NHS) for which all its citizens are eligible. The system includes the NCRN, which is a single national network for cooperative cancer clinical trials. This new network funds research nurses and data managers as well as the expertise of radiologists, pharmacists, patholo- gists, and other clinicians to a limited extent. The funding that was put in place for this was brand new money, and funded personnel were charged with doing research—they could not be diverted to meet the ordinary clinical loads of a very busy, overburdened system, Dr. Kaplan said. The data coordinating centers that are also managed under the NCRN are funded separately. The network also supports cancer steering commit- tees composed of scientific experts and consumer representatives. These committees oversee existing studies, consider new research questions, develop new proposals, and provide expert advice. The NCRN was established in 2001 with the goal of doubling patient participation in clinical trials within its first four years. It has exceeded this goal: within six years, patient accruals rose from 3 percent of the annual incidence of cancer to 13 percent. The NCRN recruits 28,000– 32,000 patients per year in treatment studies, plus another 30,000–40,000 See http://ncrn.org.uk. 25

OCR for page 41
0 MULTI-CENTER PHASE III CLINICAL TRIALS on screening or prevention trials, numbers roughly equal to the annual patient accruals in the NCI Cooperative Group Program, even though the United Kingdom’s population is about one-fifth the size of the U.S. popu- lation. The expansion in enrollment was seen more in the community hospitals than within academic medical centers, which already had good participation in clinical trials. The growth in randomized trials is leveling off, while the growth in nonrandomized trials continues to expand. This is in part because more genetic epidemiology studies are being done, Dr. Kaplan said, but it is also because the workforce put into place at NCRN’s start now has about the maximum number of large complex trials it can effectively handle. The main reason that the NCRN has been so successful in boosting patient accruals, Dr. Kaplan said, is the availability of increased funding dedicated to research staff involved in clinical trials. “The biggest driver of success was that new research nursing staff was put in place: if you get enough personnel out there, you can—up to a point—put more patients on clinical trials.” Although there might be some lessons that U.S. clinicians can learn from the NCRN experience, Dr. Kaplan noted a number of ways in which health care in the United Kingdom and the United States differ that can affect patient accrual to clinical trials. For example, people in the United Kingdom generally do not have access to innovative drugs or new uses of drugs outside of academic or industry-run clinical trials run within NHS. This restricted access makes government clinical trials more attractive to British patients and physicians than they are in the United States, where off-label use of drugs is prevalent. Dr. Kaplan claims that U.K. physicians are especially well motivated to support clinical trials as a way to be more evidence-based in their practices. “There is a genuine belief that it is their job to try to put patients on a study and to look for evidence of a treatment’s effectiveness,” Dr. Kaplan said. Patients in the United King- dom are often more accepting of being randomized in clinical trials than are those in the United States, he added. They typically follow clinician advice, even if the advice is to be randomized, and they infrequently seek multiple additional opinions. Another factor increasing patient and physician participation in clinical trials is the fact that in principle all extraneous standard health care costs linked to patients participating in clinical trials are automati- cally covered by the national health care system in the United Kingdom, although Dr. Kaplan noted that individual NHS Trusts (regional medical services) sometimes have insufficient funds to cover these extra costs, and access to a trial may be limited or capped. On the other hand, there is a strategic alignment of charity and government funding of cancer clinical research in the United Kingdom, the National Cancer Research Insti-

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION tute,26 with charities “very committed to the work of the cancer research network,” Dr. Kaplan said. If more resources are needed—for the data centers, for example—the Institute partners work out which amongst all the government agencies and funders can provide the cash or other sup- port necessary, he said. Dr. Kaplan added, however, that the NCRN may not be able to main- tain its momentum in boosting patient accruals to cancer clinical trials in the face of the trials nearing full capacity and the economy starting to falter, which makes increased resources unlikely. Furthermore, the increased burden of following patients on established trials will interfere with physicians’ ability to take on new ones. “The burden of following up on something close to 200,000 patients now is beginning to add up,” he said. Dr. Kaplan ended by noting that the NCRN is not a perfect system. Some of the most important studies are work-intensive, and with capacity now fully occupied, some local networks are declining to conduct them. Also, most of the studies are done on common cancers, with rare diseases being at a disadvantage in the system. In addition, accrual has been slow for studies that need to recruit from primary care or non-oncology clinics. There also need to be specific mechanisms of support for imaging, pathol- ogy, and pharmacy resources. New resources and new incentives have to be built in to fine tune the success so far. He also suggested that improved alignment of clinical trials internationally during the development phase would provide more complementary and synergistic research data and would prevent trials being duplicative except where that was desirable. PATIENT PERSPECTIVE The final speaker of the day was Ms. Deborah Collyar, cancer patient advocate and founder of the PAIR (Patient Advocates in Research) inter- national network, who spoke about patient perspectives regarding partici- pation in clinical trials. She began her discussion by debunking common myths perpetuated about clinical trials, including the idea that patients join clinical trials for selfless altruistic reasons. “I have not met anyone who enrolled in a clinical trial to help future patients,” she said. “They enroll based on the lottery concept. They hope for the best, for that win- ning ticket, but they realize it is probably not going to happen.” Instead of curing patients, clinical trials often simply offer them more time, Ms. Collyar pointed out. Thus patients have to decide if that extra time is more beneficial than the costs linked to pursuing the treatment. Those costs are not just financial but may also include pain or discomfort See http://www.ncri.org.uk. 26

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS beyond what the patient would experience with the cancer alone as well as impairment of lifestyle. Other factors that influence the decision to participate in a clinical trial include whether a patient has insurance and whether the patient has a support system. “It’s a life decision, not just a medical one,” Ms. Collyar said. She was critical of the informed consent process, noting that any pro- cedure done in a hospital or clinic often requires patients to sign a consent form. From the patients’ perspective, in order to receive the experimental therapy offered in a clinical trial they have to sign a consent form. “They do not understand the difference between what we are talking about in research versus a medical procedure,” she said. She added that another common patient misconception is that research is the equivalent of treat- ment. And even with all the regulations that are currently in place con- cerning informed consent, problems can still arise, she noted, offering as an example the case of Jesse Gelsinger, who died in 1999 from complica- tions related to receiving an experimental gene therapy (Gelsinger and Shamoo, 2008; Somia and Verma, 2000; Wirth and Yla-Herttuala, 2006). Patients also often erroneously assume that placebos are not given in can- cer trials, and this misconception needs to be honestly addressed. Another myth, this one on the part of health care workers, is that patients do not want to be told that they are dying, Ms. Collyar said. She argued that most patients do want to know they are dying so that they can focus on what is most important to them for the remainder of their lives. Another common misconception, according to Ms. Collyar, is that a lack of patient awareness about clinical trials is impeding their partici- pation in them. Patients do not pay attention to clinical trial awareness campaigns until they are afflicted by a condition for which there is not adequate treatment, she said. In addition, as others had pointed out, she noted that half of all patients are not eligible for clinical trials (Lara et al., 2001). “If we tell more people that clinical trials are the greatest thing since sliced bread, and they find out that they are not eligible for one, that creates a larger problem than what we have today.” She added that “low enrollment is not their fault, it’s our fault—we have to fix the system so people can get more involved in it.” Ms. Collyar offered several suggestions for improving clinical tri- als from a patient perspective. She suggested having informed consent templates that are written in plain language that patients can easily read and understand. Patients should also be given easy-to-read summaries of research results and be acknowledged for their valuable contribu- tions to research when they participate in clinical trials. “People want to know that their contributions are making a difference,” she said. Ms. Collyar also suggesting doing a better job helping patients not just with their medical treatment but with making decisions that fit their lives and

OCR for page 41
 PATIENT RECRUITMENT AND PHYSICIAN PARTICIPATION with making clinical trials a normal part of the decision-making process. Patients need help navigating their therapy and their clinical trial options, she said. Ms. Collyar was critical of the Privacy Rule27 developed under the Health Insurance Portability and Accountability Act (HIPAA), stating that the rule needs to work for the patients and not against them. HIPAA has created paperwork nightmares and made it difficult for patients to acquire their own records. HIPAA also can keep researchers from obtain- ing critical biospecimens and information that could further research. Often patients want access to their own biospecimens and information when new tests and treatments become available, yet the way in which institutions interpret the privacy rule or intellectual property agreements prevent that access. To avoid these constraints, Ms. Collyar recommended creating universal standards for data sharing. She also said that the NCI’s role in clinical trials should be to facili- tate them and to provide oversight but that the institute should not be involved in regulating clinical trials. She suggested that the NCI should sponsor a yearly public meeting between the NCI director and the clini- cal trial leaders and the cooperative group chairs; this meeting could be webcast and archived so it would be available and open to the public. At that meeting, people could explore what has been successful and why, as well as actions needed for improvement. Ms. Collyar suggested consulting patient advocates more often when determining study designs, consent forms, and other aspects of clinical trials. Unlike individual patients, patient advocates have a larger and more long-term view concerning the best ways to improve health care and research. “Patient advocates can help clinicians create and conduct better clinical trials that answer more patient-related questions faster,” she said. Ms. Collyar acknowledged that the lack of funding for clinical trials is a problem and suggested pooling money from many diverse sources, private as well as public, that could be available for publicly funded studies. “We have to be realistic about the fact that the government is not going to be able to support public clinical trials all by itself,” she said. “We have to build a feasible business plan on what we need and how we are going to fund it.” Ms. Collyar concluded her talk by saying, “We should not keep tweak- ing an antiquated system. What we are doing is just putting band-aids on a patient or a system that is bleeding out. We have to stop imitating our mentors and start living with and dealing with the world that our kids 27 National standards developed by the Department of Health and Human Services to protect the privacy of personal health information. See http://www.hhs.gov/ocr/hipaa.

OCR for page 41
 MULTI-CENTER PHASE III CLINICAL TRIALS and grandkids see.” She recommended having futurists and systems- oriented experts involved in plans for improving the clinical trial system and called for building an action plan that is actually implemented. “We have to stop just talking about these issues and start taking action,” she said. She concluded, “Whether we should get rid of the cooperative groups is not the appropriate question, but rather what can we do today to create a system that actually works.”