Crowley, William F. Jr. MD; Thier, Samuel O. MD
Five years ago, we outlined problems facing clinical research in academic health centers and proposed several solutions.1 We then instituted a comprehensive program at the Massachusetts General Hospital targeting each of those problems. The present article is a five-year update (1996–2000), in which we outline the progress made intramurally and report the extramural changes in the environment for clinical research that continually reshape its future.
OVERALL VISION AND PROGRAMMATIC CONSIDERATIONS
The Clinical Research Program (CRP) of the Massachusetts General Hospital (MGH) was established in 1996 with the broad goals of improving the quality and increasing the quantity of clinical research within our institution (see Figure 1). The CRP also assumed the overall responsibilities of balancing our institutional investments in clinical investigation between translational research, clinical trials, and outcomes/epidemiology. As described in our initial communication,1 we envisioned clinical research in each of these areas as frequently requiring a different workforce, training, motivation, and resources.
Within each of these subdisciplines of clinical research, the individual programs mounted by the CRP have been modified by several considerations. First, we delineated our future vision for each of these types of clinical research. Next, we assessed our institution's overall strength within each area. Finally, we kept in mind the ever-changing extramural environment. Consequently, we have made selective investments in each of the following three areas.
This type of clinical research focuses on the bidirectional transmission of new information between bench and bedside. Given the National Institute of Health's (NIH's) heavy past investments in basic research as well as our institution's longstanding traditions and commitment to applying basic knowledge to the care of our patients, translation investigation seemed a logical area of focus. It appears to us that its future is inextricably tied to the information forthcoming from the sequencing of the human genome and thus will be increasingly genetic and phenotypic in focus. In addition, the increasing convergence of computers, lasers, and the miniaturization of surgical and radiologic therapeutics will also have a major impact upon this form of clinical investigation, making local and circumscribed interventions into disease processes increasing less invasive and less debilitating. Furthermore, the need for potential local gene therapy for some illnesses is likely once the central issue of vector development has been addressed.
The MGH has a strong base of fundamental research, with 72% of its FY98's NIH budget ($134 million) supporting basic research. This base provides a solid foundation available for subsequent translation of research into patient care, only a small percentage of which is suited for transfer at any given moment. Additionally, extramural forces are increasingly driving clinical research in a translational direction. Nowhere is this trend more apparent than with the recent final sequence data from the human genome. This achievement appears to us to make inevitable an enormous demand for extensive translational research, particularly of the phenotyping variety. The impact of this demand is likely to be felt differentially in those academic health centers (AHCs) that have successfully assembled integrated patient care networks, defined their populations well, and developed cadres of well-trained clinical investigators. The Partners Health-care system's network,* with its approximately 1,000 primary care physicians and 1.2 million patients, thus provides an increasingly important context and resource for critical studies of genetic disorders. Taken together, this constellation of resources makes a focus on translational research quite logical for our system.
The future infrastructure of translational research will depend upon assembling a broad platform of genomic capabilities. This type of investigation requires building and integrating basic science laboratories' capabilities in genomics with the research directions of clinical investigators. This potentially natural development of partnerships between basic and clinical investigators is likely to be catalyzed by the establishment of new core laboratories in genomics, proteomics, extensive relational databases providing phenotypic information, integrated educational effort in genetics, and the development of central repositories for tissue and DNA samples with standardized access. These resources should catalyze new and highly promising synergies between basic and clinical research communities. In fact, the development of such partnerships has already begun, driven in large part by certain astute basic scientists' recognition of the value of information derived from study of naturally occurring disease models. Investing in such partnerships is necessary to set the stage for a generation of translational investigators to participate in a remarkable decade of research following the availability of the sequence of the human genome. Development of these capabilities is likely to require establishment of several strategic partnerships with both biotechnology and pharmaceutical companies in addition to building internal capabilities. In each of the circumstances with external alliances, however, decisions must be made as to tradeoffs versus enhanced capabilities for AHCs whose missions intrinsically differ from those of industry.
This next phase of genetic research should be a particularly important opportunity for all AHCs because of their unique juxtaposition of well-characterized patient populations, longitudinal medical records, and experienced clinical investigators. This unique constellation is endemic to AHCs in our nation's health care delivery system. Stated another way, if these unusual entities (i.e., AHCs) did not exist, we would probably have to construct them now to maximize our leveraging of information from sequencing the human genome.
Given these largely historical circumstances, by far the most important aspect of setting the stage for these broadened translational efforts is likely to be an increased educational effort in genetics. Most current clinicians and clinical investigators within AHCs were trained prior to the molecular and genetic revolutions. In addition, patient populations and the general public will also require considerable education and reassurance in this area. For patients, such reassurance will be likely to require legislation eliminating discrimination on the basis of genetic and medical information in the workplace and in health insurance, i.e., a federal initiative.
In response to these imperatives, the CRP has integrated a strong molecular genetics focus into its translational research courses, establishing specific didactic efforts in molecular genetics to equip clinical investigators with broader skills in this important area. The MGH has also established the Genomics Core Laboratory, headed by the MGH's director of the Department of Neurogenetics, with a staff that includes a nationally known geneticist and educator (Dr. Bruce Korf). This laboratory is also part of the Partners Healthcare System and has the mandate of establishing a broad clinical, research, and educational effort in genetics across the MGH network. In addition, we are currently searching for a director of genetics for our system and are dedicating considerable future resources to that effort, that person's support, and the work that person will direct.
To initiate these efforts and gain experience with these technologies before substantially investing in them, the CRP has initiated a series of strategic collaborations with various biotechnology partners. In the spring of 1999, the Partners Healthcare system signed a five-year joint research agreement with Millennium Pharmaceuticals (MP). This research collaboration provides our clinical and basic investigators access to MP's core laboratory facilities in genomics, mutational screening, human genetics, proteomics, and bioinformatics. In the spring of 2000, the Partners Healthcare system signed a second collaboration with deCODE Genetics granting us use of their information technologies, anonymous joint access to their Icelandic health care and genealogy databases, and collaborative research programs to map genes and loci identified in the more homogeneous Icelandic population in our more heterogeneous Partners Healthcare system patients. Each of these academic-industrial collaborations is guided by a joint steering committee comprising equal numbers of senior members from each organization. This committee serves an overall umbrella function catalyzing this interface collaboration, for which we have high expectations.
Such academic-biotechnology collaborations are likely to represent a way of leveraging the cost, time, and personnel involved in conducting research in AHCs in this arena, given the enormous resources that venture capital and biotechnology companies have devoted to building their genomic capabilities in the past five years. Few academic centers currently have the resources that can match those of industry in this area. Conversely, to succeed in this field, corporations require access to well-defined populations and clinical investigators. In addition, the rapidly developing technologies in this area are undergoing remarkable increases in efficiency and decreases in costs. Thus, over time the types of investments that AHCs will need to make should be increasingly clear and less costly. Until that time, collaborations with industry seem the wisest way to gain initial experience with this rapidly advancing technologic interface and defer capital expenditures. However, while innovative and important, such collaborations are not without risk; hence our efforts to establish supervisory boards to monitor their impacts upon our respective organizations.
Involvement in large clinical trials (Phase II and Phase III) represents an asymmetrical investment across the Partners Healthcare system. While clinical trials are generally not a major focus at the MGH, our sister institution, the Brigham and Women's Hospital, is a traditional leader in these activities. The more typical interest of MGH investigators (and Brigham and Women's as well) is the Phase I or early Phase II trials, which establish important biologic insights into the particular disease process being investigated. Thus, the MGH's current infrastructure, culture, and role models for the conduct of large clinical trials need development, with a few notable exceptions (our psychiatry department is already well prepared in these areas). More typically our trial efforts are small in numbers of patients enrolled, lower in cost, and more frequently involve junior faculty members. These realities dictated that considerable time and effort need to be expended to establish an industrial relations unit within the CRP to present our broad biologic capabilities to various biotechnology and pharmaceutical companies.
The larger pharmaceutical companies are focused not only on the accuracy but also on the speed and cost of conducting large-scale clinical trials for FDA approval. The biotechnology industry, on the other hand, is more typically interested in feasibility studies, use of surrogate markers, and assistance in earlier evaluation of the biologic potential and therapeutic strategies for agents in the development process—in many ways a better fit for the typical AHC agenda. There has been a recent rapid growth of biotechnology companies in Massachusetts (250 companies, 25,000 employees, and $2.5 billion in annual sales), and a clear and productive synergy exists between the goals of these small biotechnology companies and the MGH. These partnerships have also been key to the general improvement in the MGH's clinical trial revenues that has occurred since the CRP was established (outlined later in this paragraph). However, clinical trial activities have significant educational startup requirements for our investigators. Since the CRP began, this educational need has been reflected in the numbers of registrants for CRP courses:
▪ 450 registrants when the program started in 1996,
▪ almost three times as many in 1997 (1,122),
▪ 1,367 registrants in 1998, and
▪ 1,672 in 1999.
Clinical trial activities also entail significant infrastructure startup requirements for investigators (see Figure 1). Our industrial relations group, led by a former director of a contract research organization, provides marketing, scientific and administrative interfaces, budgeting, and contractual and institutional review board (IRB) preparation support for investigators. These collective results, as well as other institutional efforts to streamline IRB and corporate sponsorship and licensing offices, have synergistically resulted in significant increases in clinical trial activities. Our total clinical trial revenues for the three years before the CRP were relatively flat, being approximately equal to the $6 million total annual costs for each of those years; revenues more than doubled in the three years after the CRP, to a current $11–13 million in the last year two years. This growth is even more impressive because it incorporates simultaneous losses of several high-volume trialists from our faculty to other academic centers during the same period. Considering the small sizes of many of the new trials and their corresponding budgets, this large increment in clinical trial revenues has required the establishment of significant infrastructure services to support these efforts (see Figure 1).
Establishment of an industrial relations unit, like the establishment of some of the units defined in Figure 1, is a front-loaded investment wherein the early stages of developing enhanced delivery capability, educating clinical investigators about the benefits of involvement in clinical trials, and marketing their results are successful only if sustained over time and accompanied by improved delivery capabilities. However, training each new investigator in clinical trials has an amplification effect with each passing year of faculty retention, as these individuals serve as role models for subsequent investigators. Consequently, we anticipate further growth in this area, space permitting.
Finally, the Harvard Medical School has established a major initiative in clinical trials, the Harvard-wide Clinical Research Initiative (HCRI). This program will hopefully provide the infrastructure and coordination for all large-scale industrial trial activities occurring across all of the school's affiliated hospitals. The intramural investments made by the establishment of an industrial relations unit within the CRP assure that the MGH can be an effective partner in this bold initiative.
Another traditional strength of our institution has been a strong outcomes research group. Recent cost-reduction pressures from health maintenance organizations and decreased governmental reimbursements have turned all AHCs' attention to the incorporation of outcomes studies into their various operations-improvement and disease-management efforts. The CRP has utilized its interdisciplinary perspective, institutional convening power, and information technology resources to blend these mandates for operations improvement with the outcomes research strengths of our General Medicine Division.
These collaborations are now providing an impetus to improve our understanding of the disease-management process and develop a collaborative framework that represents a true new opportunity for clinical research. In addition, pharmaceutical companies increasingly understand that they are no longer involved merely in drug-target identification, pharmaceutical development, and marketing. The more forward-looking of these companies have already developed surprisingly sophisticated disease-management teams and hired major academic leaders in outcomes research. Similar to the rapid increase in the sophistication of clinical trials within the pharmaceutical industry over the last two decades, disease-management expertise is anticipated to grow within these organizations in the near future. While initially justified to improve the sales of pharmaceuticals, it is likely that there will be a strong convergence of commercial interests with those efforts of AHCs to streamline disease management and lower their costs. This synergy should prove a powerful force for changing the landscape of outcomes research for those AHCs poised to reap the benefits of this convergence. Hence, the CRP has made a large investment in this area.
Finally, seeking novel educational partnerships for these disease-management efforts is a logical outcome of such initiatives, particularly when it becomes possible to demonstrate that their use leads to decreases in cost of caring for patients with various diseases. These partnerships are a model of marrying clinical research with several intramural clinical-improvement initiatives to turn AHCs into self-learning organizations.
ORGANIZATION OF THE CLINICAL RESEARCH PROGRAM
With the above vision of the future of clinical research as well as our investments in each specific area, the CRP strategically deployed itself into five distinct units, as indicated in Figure 1 and described below.
The Administrative Core
The present director of the CRP is a mid-career scientist with a background in translational research. He continues to direct an active clinical and laboratory research program in neuroendocrinology and reproduction that comprises ten faculty and 40 support personnel. It is our bias that the leadership of clinical research requires active clinical investigators who understand the needs and resources required to sustain a contemporary research program. Thus, all CRP faculty appointments have been drawn from the ranks of active investigators willing to devote significant (10–15%) effort to the CRP in addition to the director's 50% commitment. Such part-time faculty participation necessitates a strong commitment by its administrative core.
The present chief administrative officer was a senior administrator of a specialty academic center for ten years before assuming the position and has a strong financial background. Her role is to oversee all of the personnel, financial, and administrative interfaces with the institution and public policy issues relating to clinical research. Importantly, the CRP reports directly to the president of the MGH; hence the establishment of a new administrative interface required a senior administrative figure to integrate this new organizational structure into our institutional culture.
The Industrial Relations and Clinical Trials Unit
The Industrial Relations and Clinical Trials Unit is at present supervised by a former director of a contract research organization and has three full-time employees focusing on the facilitation of industrial relations and clinical trials. This staff educates potential principal investigators (PIs) in how to prepare an IRB submission, negotiate industrial contracts, construct industrial budgets, hire properly trained study co-ordinators, and interface with our Corporate Sponsored Research and Licensing Office. They also represent the marketing and project management arm of the CRP that is frequently lacking in academic/industrial interfaces. Most sponsors wish to call one central number (i.e., a “front desk” for clinical research) to help them access the complexities of an academic health center, understand the contractual process, facilitate IRB approval, and streamline contract negotiations. In addition, the project-management functions of the CRP's industrial support unit provide the necessary impetus to keep all of these regulatory issues on track for the busy PI, a service that is doubly important for the investigator naïve about these processes.
The Clinical Research Support Office
Investigators naïve about the clinical research process are frequently overwhelmed by the prospects of identifying a research mentor, limiting a project to a testable hypothesis, obtaining funding, and then meeting each of the numerous regulatory requirements—all at the same time. Relatively simple matters such as submitting an initial IRB proposal or obtaining an Investigation of a New Drug license from the Food and Drug Administration can represent apparently insurmountable institutional barriers to new clinical investigators without assistance. The Clinical Research Support Office (CRSO) provides support in such efforts for both new clinical investigators just completing their fellowships and established basic scientists wishing to translate their basic research into applications for patient care. Therefore, this office emphasizes one-on-one interactions with investigators. The director not only helps to identify funding sources, but also performs personalized critical reviews of study designs and grant applications to optimize chances for success. The latter service is in greatest demand. This office has also mapped the pathways for submission of a clinical research proposal that allows investigators to anticipate and plan for the administrative hurdles of the approval process. Most of these maps represent relatively simple algorithms and are available on our Web site, 〈http://crnet.mgh.harvard.edu〉. However, there is no substitute for the personal element of “hand holding” for walking investigators through the process.
In addition, this office provides limited preliminary statistical support for investigators lacking such capabilities. While our General Clinical Research Center (GCRC) offers statistical support to NIH grantees, many institutional investigators are ineligible to use this GCRC service, since they lack NIH grant support. IN such cases, the CRSO makes available a statistician's time free of charge to starting investigators. The specific statistical consultant chosen for this key role was selected to be able to understand medical questions and explain statistical concepts in easy-to-understand terms for the uninitiated investigator. In addition, the service is structured to provide prompt attention to the investigator's need. As part of this start-up service, the investigator also receives an estimate of his or her ongoing statistical needs as well as a budget estimate and justification for the use of this service, for inclusion in the investigator's subsequent grant submission. At our institution, this service can be provided with a 50% effort of the statistician.
The final role of the CRSO is to provide role modeling and mentoring for young investigators. The present director of the CRSO is an excellent role model for young investigators initiating careers in clinical investigation; she also directs the Department of Medicine's clinical research training track. In this program, two to four medical interns each year are accepted and trained as clinical investigators during their house-officer years. One of our current institutional problems is the small number of senior clinical investigators, so there simply is not an adequate supply of senior mentors. Thus, the mentoring functions of the Clinical Research Program in general and the CRSO in particular serve a critical role in sustaining young investigators in this difficult career path in the absence of a multitude of well-established senior role models. In the final analysis, this may be one of the most important long-term impacts of our program.
The Education Unit
The number of institutional investigators participating in clinical research declined throughout the 1980s and 1990s; so, too, did the broad general repository of knowledge about clinical research across our institution. Therefore, the Education Unit was established to initiate several academic enrichment programs focusing on the clinical research process across our institution. This unit has taken two broad educational approaches to this institutional problem of lack of clinical investigators, reflecting two different views—an epidemiologic and an etiologic one.
The epidemiology of our current problem demonstrates a depletion of sufficient institutional knowledge about clinical investigation at all levels of the clinical research process from investigators to study coordinators through nurses and administration. Consequently, the Education Unit targeted broad didactic programs for each of these constituencies. The goal of each was to increase general institutional knowledge of clinical research via survey and general lecture courses. These general overviews have proven enormously successful and account for much of the growth in the number of registrants for CRP courses, indicated earlier.
On the other hand, the cause of our present institutional circumstance is a simple lack of senior role models for clinical investigators. Clinical investigators, because of their need for clinical and subspecialty training, generally enter the research arena late and depart early. A career in clinical investigation cannot commence until all clinical and specialty training has occurred. Compounding this late arrival on the research scene is the other fact that to be successful, clinical investigators must operate effectively in the interfaces between patient care, research, and administration. Consequently, they often depart early from the clinical research arena, being called to a variety of administrative and/or industrial posts, generally with excellent success. Thus, this combination of late entry and early departure makes their investigative-career “half-lives” relatively short.
However, we believe that only a few new clinical investigators in each year of trainees are needed to repopulate the clinical research landscape of academic health centers. Of course, these individuals must be trained with much greater intensity. The Education Unit has thus developed a series of intensive courses in topical areas we believe to be important and consistent with our vision of the future in clinical investigation as outlined above. These include courses in molecular genetics, clinical trials, and physiologic investigation. Each is typically targeted to a small number of participants (20–30 per course), is quite intensive, involves high faculty—student ratios, and is targeted at developing a small cadre of extremely well-trained clinical investigators. With the recent awarding of an NIH K30 Program to Harvard Medical School, many of these intensive programs have been expanded to fit into our planned Harvard-Medical-School—wide master's-degree program in clinical research. However, exploration of the feasibility and success of many of these master's-level courses has clearly been poineered within the Education Unit of the CRP.
Our first and perhaps most important educational effort was the establishment of an educational program for study coordinators. We viewed this as the quickest way to increase the quality of our clinical-trial efforts. These all-day seminars are offered three to four times yearly and provide basic training in the requirements and conduct of clinical trials. The lecture sessions are supplemented by monthly luncheon seminars administered for and by the study coordinators that reinforce and expand upon these efforts. Taken together, these efforts have helped to establish a sense of institutional community among these coordinators and resulted in a handbook that is useful for both them and their clinical investigators. In addition, we have now begun to work with our human relations department to develop career ladders for allied clinical research support personnel. Thus, focusing on the education of our study coordinators appeared to us to be the most rapid way of increasing the mean level of institutional knowledge and execution of clinical trials.
Disease Management and Epidemiology Unit
Given the growing importance of disease management as a form of outcomes research as well as its ability to synergize with several other institutional mandates as outlined above, the Clinical Research Program established a separate unit to initiate and coordinate these efforts. At present the unit is led by a senior professor of medicine whose research focus provides the catalytic and project-management functions for the MGH's diabetes disease-management initiative. It also serves as an interface with the Clinical Effectiveness Program at the Harvard School of Public Health, where its unit chief is a principal course director of the Harvard School of Public Health's summer program in clinical effectiveness, in which most outcomes investigators train.
Information Technology Unit
Central to any contemporary effort in clinical research is the management of information. As a result, the CRP established its Information Technology Unit, headed by the associate director of the MGH's Laboratory of Computer Science. This unit serves several roles. It has established a clinical research Web site, CRNET (〈http://crnet.mgh.harvard.edu〉), the first institutional information resource dedicated to clinical research. All clinical trials available to patients, their clinicians, and investigators across the full Partners Healthcare system are listed at this Web site, along with their hypotheses, goals, and eligibility criteria according to an organ-system—specific index. An active e-mail list of over 700 investigators who have submitted IRB proposals within the past three years has been established. This list provides frequent updates (two to three each week) about funding opportunities, educational seminars, information relevant to clinical investigators, minutes of the MGH's new Clinical Research Council meetings (also constituted by the CRP), and progress reports of the CRP's annual activities, made available via the Web site.
Perhaps the most important function the Information Technology Unit has served has been its critical contribution to our computer-based disease-management efforts. Given the apparently contradictory mandates of saving our busy clinicians' time while simultaneously instituting novel disease-management programs on their office desktops, this unit has done an outstanding job.
We have described the progress made during the first five years of the Clinical Research Program at the Massachusetts General Hospital. The most impressive outcome of those years may be that the institutional tide of clinical research has begun to turn. This program, with a current annual budget of $1.3 million/year and 14 full-time-equivalent employees, represents an institutional commitment we believe is essential for academic centers wishing to maintain balanced portfolios in clinical investigation in the future. It is likely that only a minority of AHCs will be able to afford broad programs such as this, as all are operating in a dangerous fiscal climate.
In this tightening fiscal environment, we believe that AHCs will rediscover that they are practically unique among the spectrum of health care organizations in their ability to train clinical investigators and generate new therapies for the future. These missions cannot be accomplished in their fullest measure within managed care organizations, whose primary emphasis must be on clinical services and cost containment. They are thus precious missions of AHCs, worth defending, and will require an investment if AHCs are to maintain their innovative position in our contemporary health care scene.