Clinical research and medicine have entered a time of great promise, but they also are faced with new challenges. The rapid acceleration of basic science, including advances in genomics and proteomics, are elucidating mechanisms of disease, yielding new methods to identify and potentially treat abnormalities, and effectively are transforming acute diseases into chronic diseases. These advances signify substantial progress in our national research endeavor; however, they simultaneously are reshaping not only the entire clinical research system but also our healthcare delivery system and the practice of clinical medicine. The pace and volume of medical discoveries and evolving clinical practice and corresponding policy require the development of new evidence in comparative effectiveness and outcomes, which recently have experienced tremendous increases in investment through the American Recovery and Reinvestment Act of 2009 and other substantial efforts.1, 2 Thus, as we proceed through this transformation and face new research and clinical practice demands, the question of how best to improve the translation of clinical research into clinical practice remains unanswered, and the substantial discovery-delivery gap remains.
The National Institutes of Health (NIH) established the NIH Roadmap Initiative (the Roadmap) to address these challenges and other needs in the scientific community.3 Through the Roadmap, as 1 means of restructuring its clinical research enterprise, the NIH is exploring practice-based research networks (PBRNs) to pursue the twin goals of accelerating science and facilitating the translation of research into practice.4-6
In terms of accelerating the science, PBRNs provide great promise in allowing access to both greater numbers and more heterogeneous samples of prospective research participants. For several reasons, both of these elements are increasingly important to accelerate the pace of research progress.4, 5 First, it has become more and more apparent that intervention effectiveness hinges on specific genotypes expressed in limited subsets of the population. This drives a need for research population heterogeneity to assure the inclusion of individuals who express each of the relevant genotypes. Second, this fuels the need for more clinical trials to examine different agents or approaches that may be effective in treating these multiple population subgroups. Opening more trials that test more agents creates a need for more individuals to enroll on those trials to allow their timely completion. Together with the long-held estimate that <1% of Americans seeking healthcare do so at academic medical centers (heretofore the locus of nearly all clinical trials),7, 8 filling these research needs will mandate the extension of mechanisms like PBRNs into the community and access to larger, more heterogeneous samples of prospective research participants.
Practice-based research networks also can facilitate the translation of research into practice by playing a central role in a critical 2-way flow of information.9, 10 First, PBRNs connect academic investigators to community practitioners who contribute “upstream” information to researchers regarding study design and implementation considerations, including insight into the clinical issues and the tacit, practice-based knowledge that exists in community-based practice settings. Second, PBRNs can facilitate the “downstream” dissemination and implementation of research results into community-based clinical practice by promoting a sense of trust and ownership among community providers who are first-hand participants in the whole process, enhancing providers' acceptance of research results and strengthening their commitment to acting on research findings. This level of participation also often encourages practice organizational changes that facilitate the research and allows participants access to more or earlier information compared with others who did not participate in the research. In this way, PBRNs contribute both evidence-based practice and practice-based evidence and help close the discovery-delivery gap.11, 12
Despite the promise of PBRNs and the substantial federal commitment to develop and support them in a wide range of disease areas,13-22 PBRNs face significant challenges with regard to implementation and sustainability.23-28 Major challenges include 1) developing and maintaining an infrastructure of clinicians and qualified staff, 2) maintaining adequate and consistent funding for research, 3) developing collaborations with local physicians who manage PBRN functions, and 4) instituting systems of accountability and efficiency,23, 28 all of which are greatly facilitated by 5) establishing a cultural context that values research and scientifically based practice.29 Recognizing these challenges is important; however, little practical guidance exists in the literature regarding principles for addressing them in establishing and operating effective PBRNs.
The National Cancer Institute (NCI) Community Clinical Oncology Program (CCOP) is a cancer-focused PBRN that, since its establishment in 1983, has grown to contribute approximately 33% to NCI clinical treatment trial enrollment and the overwhelming majority of its prevention and control trial enrollment (Fig. 1). Moreover, the CCOP has been integral not only to core accrual for NCI trials but also to the development of major trials, such as the Breast Cancer Prevention Trial (BCPT),30 the Study of Tamoxifen and Raloxifene (STAR),31 and the Selenium and Vitamin E Cancer Prevention Trial (SELECT),32 which was designed by a CCOP research base. In addition to providing 31% of the 13,388 BCPT participants, 33% of the 19,747 STAR participants, and 29% of the 35,534 SELECT participants (allowing SELECT to be completed well ahead of schedule), it is widely held that, without the enhanced infrastructure of the CCOP, these studies—most notably SELECT—could not have been started let alone completed.33 The current performance of the CCOP and its critical centrality to such research is the result of nearly 30 years' direct experience, continuous self-evaluation, adaptation, and learning from other NCI PBRNs and quality-enhancement efforts, including the Cooperative Group Outreach Program (CGOP) and the Community Hospital Oncology Program (CHOP).34, 35 To provide practical guidance for other new and developing PBRNs, in this review, we detail the organizational characteristics and strategies that have contributed to the CCOP successfully addressing the challenges of implementation and sustainability, meeting the needs of clinical research, and the translation of research into evidence-based practice. We draw from 3 decades of evaluation of the CCOP, CGOP, and CHOP programs,24-27, 34-43 and we expand on other recent work28 by discussing the current and evolving practical needs that face practice-based research.
Overview: The National Cancer Institute Community Clinical Oncology Program
The NCI has established a national clinical trials infrastructure through the cooperative groups, which are bodies of researchers, cancer centers, and community physicians that work together to conduct trials. The CCOP is a PBRN and a component of the NCI clinical trials program through which community physicians participate as full research members in NCI clinical trials (Table 1).44 The CCOP network funds community hospitals and physicians to participate in NCI-funded clinical trials that are designed by NCI cancer centers and clinical cooperative groups (collectively called “CCOP research bases”), commonly in conjunction with the CCOP. The community-based networks of hospitals and physicians (“CCOP organizations” or “local CCOPs”) and CCOP research bases are funded through peer-reviewed grants based on their productivity in developing, conducting, and accruing to clinical trials.
Table 1. Overview, the National Cancer Institute Community Clinical Oncology Program
NCI indicates National Cancer Institute; CCOP, Community Clinical Oncology Program; MBCCOP, Minority-Based Community Clinical Oncology Program.
The NCI Community Clinical Oncology Program
Established in 1983
Program funds through cooperative agreements (peer-reviewed grants)
Selected cancer centers and clinical cooperative groups (“CCOP research bases”) to design, develop and manage clinical trials; and
Community hospitals and physicians in local networks (“CCOP organizations”) to enroll patients onto cancer trials and disseminate study findings in the community setting
Included 47 CCOP organizations and 14 MBCCOP organizations located in 34 states and Puerto Rico in 2009 (NCI External Review Committee 200955)
61 CCOPs comprised of 390 hospitals and nearly 3470 community physicians in 2009 (NCI External Review Committee 200955)
Currently contributes approximately one-third of all enrollment to NCI trials, including 21% minorities(McCaskill-Stevens 2005,41 NCI External Review Committee 200955)
CCOP trials contribute to evidence-based national standards for cancer care
CCOP-affiliated physicians participate in the development of clinical trials (Helfrich 2007,25 McKinney 200627)
Disseminate national standards to community-based practices
The CCOP has a track record of translating findings from NCI-sponsored clinical trials into community practice (Table 2). It is the academic investigators within the CCOP research bases who primarily drive the science, but the CCOP physicians participate in the development of the trials, contribute substantial numbers of patients to the trials, and then translate that science into practice by modifying their subsequent practice patterns based on the results of those trials. In this way, the CCOP program has facilitated the translation of several research-based clinical innovations into practice. For example, epidermal growth factor receptor (EGFR) inhibitors have demonstrated efficacy in patients with metastatic colorectal cancer who have failed chemotherapy.45 The CCOP network brought a large, heterogeneous population of patients to the trials, facilitating the identification of Kras, a protein in the EGFR pathway, as a predictor of prospective treatment effectiveness.46 Because of the findings from that study, practicing oncologists and patients who are considering cetuximab today can ascertain the probability of treatment effectiveness prospectively and, thus, avoid unnecessary treatment risks, side effects, and their associated substantial costs. In addition to participating in the study (in terms of offering the trial to his eligible patients), a CCOP investigator participated as a coauthor of a report on the American Society of Clinical Oncology in guidance for screening patients,47 thus exemplifying the 2-way communication between research and practice in the CCOP network.
Table 2. Translating Bench to Bedside: Examples of Translation Facilitated by the Community Clinical Oncology Program
Clinical Tests and Interventions
CCOP Contribution to Study Accrual
CCOP indicates Community Clinical Oncology Program; Kras, v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog; EGFR, epidermal growth factor receptor; CALGB, Cancer and Leukemia Group B; PACCT1, Program for the Assessment of Clinical Cancer Tests; TAILORx, Trial Assigning Individualized Options for Treatment; NSABP; National Surgical Adjuvant Breast and Bowel Project; HER-2, human epidermal growth receptor 2; NCCTG, North Central Cancer Treatment Group; BCTP, Breast Cancer Prevention Trial; STAR, Study of Tamoxifen and Raloxifene.
Approximated or estimated contribution based on data to date; trials are ongoing or data are being finalized.
Avoiding ineffective treatment: Kras and EGFR inhibitors
Kras testing to guide introduction of chemotherapy (Karapetis 200846)
Only patients with a wild-type Kras benefit from EGFR inhibitors, eg, cetuximab; patients who will not benefit from cetuximab can be identified prior to treatment and avoid risks, toxic side effects, and substantial costs of this chemotherapy
Similar to that experience, the CCOP network facilitated the completion of studies in the development of OncoType DX, a genotype scoring system that evaluates patients' risk of cancer recurrence. Specimens and data from previous trials were used to develop and validate OncoType DX,48 and an ongoing NCI trial (Program for the Assessment of Clinical Cancer Tests or PACCT-1)49 will evaluate the need for chemotherapy in hormone-responsive patients who have OncoType DX scores that are difficult to interpret. For physicians and patients who are considering treatment, the findings from that study will better inform their balancing of the risk of recurrence with other considerations and whether the prospective chemotherapy treatment benefits will justify the risks.50 These 2 examples build on the enormous success of several foundational adjuvant breast cancer trials, a few of which are presented in Table 2, and demonstrate the centrality of the CCOP program in the evolving cancer research and clinical care environment.
Keys to Community Clinical Oncology Program Success and Overcoming Challenges
In today's evolving research environment, at least 4 major challenges must be addressed by PBRNs to continue medical progress that benefits patients, providers, and society (Table 3). Through a combination of experience, deliberation, and serendipity, the CCOP has put into place structures, policies, and practices that address these challenges and have contributed to its strong performance.28, 34, 35
Table 3. Community Clinical Oncology Program Principles for Successfully Addressing the Major Challenges Facing Practice-Based Research Networks
Develop and maintain infrastructure of clinicians and qualified staff
Maintain adequate and consistent funding for research
Facilitate collaborations between academic and community physicians
Flexible systems of accountability and efficiency
Key Principle 1: Building an infrastructure
Community-based physicians in a PBRN maintain the critical flow of patients into clinical research studies and collaborate with academically based researchers to test new ideas in the practice setting. In building a research infrastructure, physician interest is essential, but it often is not sufficient. Community-based physicians face demanding schedules and heavy workloads. A well-trained research staff, particularly research nurses, is critical to the successful integration of clinical research into community oncologists' practice.27, 42
Research nurses help facilitate and integrate clinical research into physicians' daily activities and ensure that research-related activities offer minimal disruption to clinical practice and care. Research nurses screen charts and flag them for patients who are eligible for trials, help physicians communicate with patients about clinical research opportunities, and provide assistance with other trial-related activities. Physicians and nurses in CCOP trials form a critical mass of healthcare providers who are comfortable with implementing and delivering the new treatment and care strategies that arise from these trials.42
To maintain high-quality research, mechanisms for the ongoing training and education of all staff need to be in place. The CCOP research bases provide training and educational opportunities twice a year for all affiliated academic and community-based physicians and research staff. At a local level, CCOP organizations also provide affiliated physicians, research nurses, and staff members with education and training sessions at which nationally known speakers present the latest clinical and research developments. Training includes information about protocol design for specific studies as well as trial-related skills, such as procedures for enrolling participants, collecting and submitting trial data, and reporting adverse events.27 Education and training activities vary across CCOP organizations, depending on how they choose to staff their operations.27, 42 Education includes continuing medical education sessions for scientific discussions specific to the ongoing clinical trials. Several research bases also offer mentoring programs for newer investigators and certification examinations for the Society of Clinical Research Associates. Training by the research bases and at the local CCOPs provides an unprecedented opportunity to educate clinicians who otherwise might not be exposed to clinical research, and such exposure also encourages the uptake of new treatment modalities.
Key Principle 2: Funding to empower local physicians
Maintaining adequate and consistent funding for research is a persistent challenge in a research enterprise. Practice-based research networks must direct funds to community-based physicians who deliver care and provide accrual for clinical trials. Although academic centers play a large role in a clinical trials program, the CCOP program uses peer-reviewed, cooperative agreements to award grant funding directly to local networks of physicians and hospitals rather than passing the money through academic medical centers or intermediaries, thus contributing a great degree of local flexibility to ensure the money is used to meet each site's specific needs. Providing grants directly to the sites allows the local organizations to plan for multiple years of funding (for stability); permits sites to plan and allocate staff, office infrastructure, and other resources according to the research needs; and brings recognition of NIH funding to a community site. At the same time, because the cooperative agreements allow each CCOP flexibility, they also permit substantial NCI management and oversight and provide a framework for regular performance monitoring and financial reporting.
The stability and flexibility conveyed through direct funding is augmented by the CCOP organizations affiliating with multiple research bases for a broad menu of trials that matches the research interests of the organization and the population it serves. CCOP productivity is measured by enrollment of patients to those trials in conjunction with data quality, and CCOP typically will have over 100 open trials. Budgetary flexibility allows each CCOP organization to manage its staff, time, and other resources to best support the trials it has selected to open, ensuring its ability to meet its accrual goals and otherwise be successful. In an example of how the funding structure empowers decision making by local communities, the physicians and community served by the Upstate Carolina CCOP in South Carolina expressed strong interest in the NCI-sponsored prostate cancer clinical trial, SELECT.51 Accordingly, they reallocated a large proportion of their resources and staffing to support enrolling patients onto that this trial, resulting in remarkably high accrual.
Each CCOP organization's funding is based on accrual to all trials it has open rather than any 1 specific trial, and these accrual targets are set annually based on the availability of trials and historic trends in the CCOP organization's patient accrual. Figure 2 illustrates the funding of the CCOP program and average per-CCOP funding and underscores the importance of ensuring a relatively consistent, predictable level of funding from 1 year to the next to enable CCOP organizations to maintain staffing and operations as trials open and close. Despite significant ebbs and flows in levels of federal funding for clinical research and myriad changes in the financing, structure, and delivery of healthcare,24 the cooperative agreement mechanism helps ensure the viability of this system.
Key Principle 3: Collaboration strengthens research and practice
Practice-based research networks help bridge the gap between academic research and community practice through clinical research that addresses an identified need and will produce usable results. For a 2-way communication system between academic research and community practice to be effective, clinicians must understand the importance of the research, and the research must be responsive to the needs of the clinicians.
The CCOP structure provides many opportunities for interaction between clinicians and academic researchers. CCOP research bases hold regular conferences during which academically based researchers and CCOP-affiliated physicians and research staff meet in scientific committees, attend plenary sessions, and participate in training sessions. These conferences provide a forum for learning about clinical research science, planned or ongoing trials, and clinical trial management. Community physicians acquire a greater understanding of the scientific relevance of particular questions and why certain protocol features may be important for scientific rigor while simultaneously providing input to the academic researchers, who learn how to design trials that are clinically relevant, feasible in community settings, and reflect the needs of the community itself. For example, minority physicians and those who serve minority men were persistent in communicating that African-American men were developing prostate cancer earlier than age 55 years, which, at the time, was a common age-eligibility criterion for many trials. By lowering the age-eligibility criteria for African-American men to 50 years, the recruitment of these men was substantially enhanced.33 The end result of incorporating input from community physicians is the ability to design scientifically rigorous trials that are feasible to perform in the community and also are more likely to result in uptake into practice, because participating physicians have a venue in which to become familiar with the drugs and procedures tested during the trial, observe their benefits for patients, and develop ways to mitigate possible adverse effects.
Key Principle 4: Flexibility in operations of organizations
Individual CCOPs have flexibility regarding their structure and operating procedures that builds on the stability provided by predictable, direct funding and the flexibility provided by selecting which clinical trials to open from a large menu of options. This approach optimizes the skills and interests of local clinicians as they work within the constraints of provider organizations and patient populations. CCOP performance guidelines require that organizations meet overall accrual targets, implement protection for human participants, and maintain the standards for data quality set by their affiliated research bases; however, addressing the reality that these provider organizations and populations vary widely in their characteristics, priorities, and norms, CCOPs have wide latitude in how they meet those guidelines. There are no requirements regarding a specific organizational structure or size, number of open trials, or even accrual targets for individual trials; rather, CCOPs largely self-design a system that works for them. NCI program staff review each organization annually for overall performance and provide technical assistance as needed, CCOP research bases periodically review data from each CCOP to assure that it is meeting its quality standards, and each CCOP is peer reviewed every 3 to 5 years when grants are opened for competition.
This flexibility in strategies for meeting performance standards provides CCOP organizations with considerable discretion in how they manage their operations. Like the Upstate CCOP in South Carolina, CCOP organizations participate only in those trials that suit the structure, function, needs, and interests of the affiliated providers and provider organizations. Some CCOP organizations operate in a centralized manner from a single office, whereas others use a more decentralized model with an administrative core that coordinates research nurses who are used individually by different participating physician practices. CCOP organizations also may participate in nonhealthcare settings, such as health information booths at grocery store entrances or civic events, to distribute information and enroll participants in cancer prevention trials. Many CCOP organizations use outreach workers to increase trial access for minority populations. These workers staff information booths at minority-focused health events, make presentations in minority churches, and engage in “academic detailing” in community-based organizations that serve minority populations.27
The flexibility of the CCOP program flexibility is exemplified through the Minority-Based CCOP (MBCCOP) program, which evolved from the recognition that, historically, most cancer care for minorities has been at academic institutions. Thus, academic centers subsequently were allowed to be part of the MBCCOP program in addition to the traditional community-based programs. These sites serve as excellent venues for examining how well new agents, complex trial designs, or new technologies can be disseminated and implemented (or not) in special populations. These sites typically have outreach services that inform the MBCCOP of referring physicians' level of acceptance for a specific clinical/scientific advancement. Despite being comprised of only 13 local MBCCOP sites (compared with the other 47 CCOPs), the MBCCOP program allows the inclusion of a substantial population of racial and ethnic minorities in studies, as demonstrated in Figure 3, and facilitates their access to practice-based evidence and evidence-based practice.
Historically, implementation and dissemination of new healthcare innovations have been slow. For example, although it was known as early as 1601 that providing citrus to sailors would prevent scurvy, it was not until 1795 that the British Navy ordered the inclusion of citrus fruits as part of the rations on all navy ships.52 More recently, a study of physician awareness and use of the latest hypertension prevention, diagnosis, and treatment guidelines from the NIH reported that <50% of physicians followed guidelines for use of the latest prevention, diagnostic, and treatment guidelines regarding hypertension; and 40% were not even aware of those guidelines53 despite data indicating that >50% of the 50 million Americans with high blood pressure are not being treated to goal, although their physicians understand that high blood pressure is the most preventable cause of heart attack and stroke.54 There are similar examples of lack of translation in almost all fields of medicine.
The CCOP network is an example of how PBRNs can be mechanisms to effectively address the challenges facing clinical research and practice; continue medical progress that benefits patients, providers, and society; and more effectively translate research into practice. Organizational and funding flexibility, together with meaningful yet adaptable performance standards, allow CCOPs to vary in terms of size, location, provider composition, and the types of trials in which each site chooses to participate. These principles have enabled it to be a vehicle both for including community oncologists as full partners in the NCI clinical trials network and for giving those oncologists a mechanism to learn and use state-of-the-art treatments in the context of national standards. In an environment of ever-changing local, state, and federal regulatory burdens that often unintentionally constrict healthcare and clinical research, the firmly established internal processes of CCOPs enable the consistent and appropriate provision of resources for sites to meet those requirements. These processes also allow for the identification and engagement of competent and knowledgeable young investigators and various support personnel for research careers within this PBRN framework.
Challenges will always face the development and sustainability of CCOPs and other PBRNs and their continual adaptation. For example, recent advances in technology and clinical research have resulted in the development of new drugs and methods of treatment that can be increasingly complicated to use, let alone systematically implement, as a part of a new, high-quality standard of care in community practice. Accordingly, special attention needs to be given to providing incentives to bring young clinician-investigators and research nurses into the PBRN arena with their expertise in emerging technologies and practice skills. Practice-based research networks can provide a structured forum to observe and first use healthcare innovations, thus facilitating their initial adoption and their subsequent systematic incorporation into practice.
Looking forward, even the CCOP network will find value in revisiting these principles in the context of its continual evolution and adaptation to the changing landscape of cancer research. Just as clinical cancer care is moving toward individualized therapies and personalized medicine, so too is cancer clinical research. Clinical science is advancing to where antecedent knowledge of individual biologic variation is a factor that drives not only outcomes but also treatment selection and even randomization within studies. CCOP practices will have to adapt to these and other changes to maintain their ability to stay ahead of the curve in research. The CCOP program, as it seeks to remain in the vanguard of state-of-the-art translational science and charts its course after a recent program evaluation,55 will benefit from revisiting these guiding principles and grounding itself in them to help it successfully move into the future.
The CCOP structure has provided a framework for involving community patients and practitioners in clinical research while simultaneously expediting the translation of clinical research results into practice. With ever-growing emphasis on efficiency and, more recently, new pressure to assess differential treatment effectiveness in the broader population, PBRNs in other disease areas may find the CCOP experience useful for developing links between community-based service-delivery organizations and academic centers. This is critical for meeting the challenge of translating research into practice.
We thank Dr. Victor Vogel and the peer reviewers for their thoughtful comments, and we acknowledge the support of our Division of Cancer Prevention colleagues Ms. Lindy Wong and Drs. Leslie Ford and Peter Greenwald.