PANEL DISCUSSION II—THE ROLE OF ACADEMIC MEDICAL CENTERS IN TRANSLATING RESEARCH INTO CLINICAL PRACTICE
INTRODUCTION—DR. MICHAEL CALIGIURI
Dr. Caligiuri introduced the panel members.
DR. RONALD B. HERBERMAN
Background
Dr. Herberman is currently the Director of the University of Pittsburgh Cancer Institute, Associate Vice Chancellor for Cancer Research, Hillman Professor of Oncology, and a Professor of Medicine and Pathology. Most recently, he has assumed the position of Director of the University of Pittsburgh Medical Center (UPMC) Cancer Centers, a prestigious network of the UPMC Health System.
Key Points
-
In order to address the challenge of effective dissemination of research advances, the University of Pittsburgh Cancer Institute and other NCI-designated Comprehensive Cancer Centers and academic medical centers have formed multidisciplinary groups that focus on a particular cancer type or a modality of treatment. The aim of these groups is to stimulate discussions, interactions, and, hopefully, collaborations across a spectrum of experts—from basic cancer researchers to clinical oncologists. Although this facilitation of translational research is critically important, it is also essential to achieve advances that occur beyond initial clinical applications and disseminate them into widespread practice.
-
It is also important to foster effective interactions between academic medical centers, particularly cancer centers, and industry; NCI has been playing an important role in the academic public-private partnership program launched last year. This is a challenging area and one that merits substantially more funding support.
-
The biggest challenge to dissemination lies in moving promising applications out of the academic cancer center and into the community. In response, the University of Pittsburgh Cancer Institute has been developing a hub-and-spoke model in which the Cancer Center is closely integrated and coordinated with a wide community network in the region and is responsible for central oversight and personnel training. The Cancer Center then becomes responsible for what goes on in the regional community network. The model is still in the early stages, but the indications so far are promising. More clinical research studies are being implemented in the community and more community oncologists are becoming actively involved in clinical research; in turn, those oncologists are making an impact on some of the clinical trials to increase their relevance to the problems of cancer in the community, which are often different from those in the academic center. Critical to building trust with the community oncologists is ensuring that patients who can be treated in the community are, indeed, treated in the community. Community clinical oncology representatives are included in the Cancer Center leadership.
-
This kind of model does require resources. NCI, through two funded grants, is helping the Institute implement this model. Also, there has been development of a process by which the revenue from the clinical practice of oncology is generating increased efficiency, which then generates revenues that can be put back into support for infrastructure and staffing.
-
Communication is very important, and the challenge requires creative approaches. Videoconferencing has proven important in facilitating effective connections between the Institute and the community it serves. Hand-held PDAs containing information about all open clinical trials and eligibility criteria are being evaluated for use by practicing clinical oncologists.
-
No single cancer center has all of the relevant expertise and capabilities. The Institute has been setting up collaborations among different cancer centers, particularly those within its region. One collaboration, for example, is a consortium to evaluate biomarkers and their usefulness in predicting tumor development and clinical responses to therapy.
-
Bioinformatics is aiding effective communication. NCI’s recent funding for the Cancer Biomedical Informatics Grid (caBIG) is very promising; cancer centers around the country will be able to more effectively communicate with each other, receive full clinical annotation of tumor specimens, and determine whether particular biomarkers, for example, are useful for the intended purpose.
-
Tumor registries merit more widespread evaluation and implementation. Registries should be built upon and converted from incidence registries to outcomes registries, which will better enable researchers to evaluate best practices and deviations from those practices.
DR. STEPHEN M. PRESCOTT
Background
Dr. Prescott is the Executive Director of the Huntsman Cancer Institute. He was a founder and the Director of the Eccles Program in Human and Molecular Biology and Genetics at the University of Utah. Currently, he is a Professor of Internal Medicine and holds the H. A. and Edna Benning Presidential Endowed Chair. Dr. Prescott received his M.D. degree from Baylor College of Medicine.
Key Points
-
The Huntsman Cancer Institute serves the largest geographical region of any NCI-designated Cancer Center. The region’s population is unique in that it is home to the largest concentration of Native Americans. The region also contains Utah, which was founded by the Church of Jesus Christ of Latter-day Saints (Mormons). Because of the genealogical focus of the Church, it has been possible to link much of the Mormon population in a database that when linked with the Utah cancer registry, provides a thorough description of the population’s cancer incidence and outcomes. This allows researchers to look at genetic predisposition and other factors that influence the likelihood of cancer, patterns of response to treatment, and likelihood of adverse events and outcomes. More than 400 colon cancer deaths have been prevented in the last 7 years using information from these databases.
-
The Utah database represents a completely European-derived population. If there are alleles that predominately affect other ethnic groups, this database will not be helpful. However, genetics investigators have discovered more than 40 disease-causing genes that have been determined to be broadly applicable to all populations, including Asian-and African-derived populations. However, it is difficult to share this database and still protect privacy as mandated by HIPAA. Understanding barriers created by HIPAA and other guidelines and exploring ways to overcome those barriers is very important.
-
From a policy point of view, prevention is paramount. Understanding the genetic basis of cancer and how that information can be used to design novel biomarkers, in vivo imaging, and other approaches will be essential in developing an effective prevention program.
-
There are a number of barriers to dissemination. First, when focusing on prevention, the wrong individuals are often considered the “front line.” Instead of concentrating on high-end, specialized systems, such as cancer centers and academic medical centers, prevention begins with general internists and family practitioners, as they are the doctors doing health maintenance and screening—i.e., the “front line.” Second, there are institutional barriers in many organizations, particularly academic medical centers. Academic institutions can make improvements by overhauling reward systems that highlight individual achievement over transdisciplinary teamwork. Leaders of academic centers need to build teams that span the spectrum of specialties and reward shared victories and achievements. An additional barrier is over the NIH/NCI grant system, which rewards principal investigators over inter-or transdisciplinary teams.
-
There are also barriers to creating effective partnerships, especially those with the commercial sector. Partnerships tend to be based on contractual negotiations: the researcher has a deliverable, and the company has a responsibility; once the researcher has delivered his or her product, there is no more information sharing.
-
Long-term support from the Federal Government and/or the pharmaceutical industry will be needed to develop new technologies and drugs.
DR. EDDIE REED
Background
Dr. Reed is the Director of the Mary Babb Randolph Cancer Center at the Robert C. Byrd Health Sciences Center, West Virginia University (WVU). His research interests are focused on gynecologic and genitourinary malignancies. Dr. Reed is also interested in viral carcinogenesis. The primary focus of his studies has been high-risk human papillomavirus (hrHPV) and its association with cervical cancer and cancers of the head and neck. Dr. Reed’s laboratory recently completed an almost 800-person surveillance study involving 33 sites across the state.
Key Points
-
West Virginia University serves a population that is unique in several respects. Nationally, the state of West Virginia ranks 51st in median household income, is second in median age of its residents, has the fifth highest poverty rate, and is the second most rural state. Although there are several modern metropolitan areas within West Virginia, the majority of the state is rural and has scattered resources to deal with mounting health problems and other issues.
-
Two things are being done well and deserve special emphasis; both of these require continued and expanded support. The first is the Centers for Disease Control and Prevention’s Breast and Cervical Cancer Early Detection Program (BCCEDP), which seeks to provide screening to women who are uninsured or underinsured. The state of West Virginia provides funds to assist individuals who are diagnosed with cancer through this screening program. West Virginia University has initiated a study of the cost-effectiveness of the program.
-
There are currently several problems with the Program. First, it is not funded as well as it should be to reach the families it could easily serve. The Program should be strongly supported, and consideration should be given to expanding this program to include other major malignancies for which data suggest screening would profoundly impact the health of the population. In those malignancies for which data are incomplete, such a program could be used as a platform to conduct important clinical studies to answer critical questions about screening and its effectiveness. The application of this approach to a broader range of malignancies will be needed to make a major impact on the cancer burden in West Virginia. Also, WVU is conducting research regarding the cost effectiveness of the BCCEDP in order to demonstrate the benefits of screening to citizens and insurance companies.
-
Second, the program and others like it represent unexploited opportunities to gain important new knowledge about cancer causality in real time and space. Most cancer prevention knowledge is developed using animal models, which have severe limitations in terms of applicability to the human situation. Supporting active research linking service (as occurs in the BCCEDP) to peer-reviewed research on important clinical questions that can be easily addressed in this setting is an opportunity that should be taken advantage of.
-
The second thing being done right is the NCI plan to expand the number and breadth of NCI-designated Cancer Centers. The expansion of the NCI Cancer Center network is essential to the dissemination of state-of-the-art clinical care and research to all American citizens. However, limiting expansion to just one institution per year is a major mistake. Denial of NCI designation to institutions that meet the rigorous peer-review criteria is directly contradictory to the goals of reducing the nation’s cancer burden.
DR. HAROLD C. SOX, JR.
Background
Dr. Sox is the Editor of the Annals of Internal Medicine. He also chairs the Medicare Coverage Advisory Committee of the Centers for Medicare and Medicaid Services (CMS) and serves on the Report Review Committee of the National Research Council. Dr. Sox is President of the American College of Physicians–American Society of Internal Medicine. He received his M.D. from the Harvard Medical School.
Key Points
-
There are currently three barriers to successfully reducing the risk of cancer. The first is infrequency of the disease in the screened population; healthy people are by definition the targets of screening programs, and they seldom experience the target disease.
-
The second barrier is that prevention is inefficient because disease is infrequent in screened populations or populations undergoing chemoprevention; most patients who undergo screening or chemoprevention do not have the target disease and will never get it. Therefore, many people are exposed to the intervention and its possible harms in order to benefit a single individual. Yet another barrier is that the rates of performing preventive services are too low.
-
Some of the problems with cancer prevention occur because relatively poor technology exists for targeting people at high risk and reducing their risk. Worse, the cancer community does not consistently deliver agreed-upon important services.
-
There are ways to overcome these problems. The first is to reduce the harms of screening programs. These occur because so few screened patients have or will get the target disease. Most screened patients can only suffer the harms of screening, including worrying that they have cancer and undergoing invasive, painful tests to be sure that they do not. Improving screening tests to reduce the frequency of false positive results would mean fewer people experiencing several weeks of worry followed by a negative biopsy. More accurate screening tests and safer, cheaper follow-up tests after positive screenings are needed. Academic medical centers can play a leading role in the research on this topic.
-
The second way to overcome these problems is to improve the ability to identify patients who are at the highest risk for the target disease and for the complications of follow-up tests and treatment. Academic medical centers typically are the sites of the epidemiologic studies that allow researchers to stratify patients by their risk of cancer.
-
The ability to deliver screening and risk-reduction services to patients should also be improved. One way is to develop electronic medical record systems for small office practices, where most care occurs. These systems can remind office staff when a patient is due for preventive services, detect when a patient has actually had the service, and connect with national data systems for tracking screening performance. Another way is to create incentives for practices to offer preventive services to a high proportion of eligible patients.
-
Academic medical centers can also conduct more research on the behavioral factors that motivate physicians and patients to follow suggested guidelines regarding prevention. For example, in the smoking intervention arena, research has shown that the most important factor in influencing patients to follow through on tobacco cessation programs was a strong statement from their physicians about the dangers of tobacco. The cancer community needs to make it clearer to physicians how important they are in their patients’ decisions to stop smoking. Also, patients need to be motivated by understanding that quitting tobacco use really can make a difference.
-
Knowing a person’s genetic endowment, as in the case of the Mormons in Utah, is one way to identify people who are at higher risk; this is supported by what is known about breast cancer. It would be easier to motivate people, for example, to stick with smoking programs if research showed that they had a genetic polymorphism that made them more susceptible to the effects of cigarette smoking.
-
Identifying and targeting high-risk people can have consequences: In breast cancer, for instance, only a small proportion of women who are at risk actually carry the genes associated with high risk. If prevention efforts focused only on women at high risk, many people at average or slightly below average risk would not be screened, even though their risk was high enough to justify screening.
-
Personal knowledge of patients’ histories, risk factors, and habits is a huge advantage that primary care physicians have; this helps them do their part to reduce the risk of cancer. However, physicians simply may not remember to do preventive health services in the office setting. Electronic medical records can do the job of reminding them.
-
Research shows the most important factor in increasing cancer screening rates in primary care populations is scheduling a visit solely for the purpose of prevention. Primary care physicians should be more systematic about scheduling preventive service visits, and Federal agencies should find a way to reimburse them for visits devoted entirely to disease prevention and health promotion.
THE ROLE OF ACADEMIC MEDICAL CENTERS IN MOVING RESEARCH INTO COMMUNITIES—DISCUSSION
Key Points
-
It is important to develop novel approaches to link academic cancer centers with the communities and regions they serve. This is one of the criteria for designation as an NCI Comprehensive Cancer Center, but it is minimally funded, at best.
-
The most effective mode of cancer treatment is prevention. However, the prevention approaches that target high-risk individuals require access to medical records on a large scale. Legal barriers, including HIPAA, hinder such access-based approaches.
-
Electronic medical records are the key to improving delivery of preventive health services to primary care patients and in making primary care physicians accountable for their performance. Every primary care physician should have an electronic medical record capable of supporting preventive health services. However, NIH has rules about the handling and security of electronic records that seriously impede use of electronic technology. Researchers should meet with Government agencies to decide how to use electronic records while still safeguarding privacy.
-
Academic medical centers need to contribute to the research that shows that prevention reduces bottom-line costs.
-
The University of Pittsburgh Cancer Institute is able to provide high-quality radiation therapy in the community. Studies conducted by the major Blue Cross insurance provider in Western Pennsylvania indicate that the farther away a community is from the urban and academic settings, the less compliance there is with recommendations for radiation therapy. Using the cancer center/community integration model, the Institute has made available not only good-quality radiation therapy, but even intensity-modulated radiation therapy (IMRT). This therapy is as available 100 miles from Pittsburgh as at the Institute itself by employing the right type of high-speed communication lines conducive to centralized radiation physics planning, which is then sent out into the community. The result is that the same IMRT can be delivered in the community as at the Center.
-
NCI mandates that the Comprehensive Cancer Centers develop programs that generally consist of groups of individuals with both basic and clinical prevention expertise. The Ohio State University Comprehensive Cancer Center has found that when a Ph.D. scientist and a physician-scientist are partnered in leading those programs, the amount of translational research doubles. Appointing people who understand both basic and clinical oncology benefits patients.
-
Another facet of communication that should be explored is that among cancer patients, the community, and physicians. This communication should emphasize the psychosocial and spiritual as well as the physical burdens of cancer. There are many common links among cancers, yet they are considered to some degree to be different diseases. The stages that patients go through are much the same.
