| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Special Report |
1 From the Biomedical Imaging Program, National Cancer Institute, Executive Plaza North, Rm 800, 6130 Executive Blvd, Rockville, MD 20892-7440. Received December 21, 1999; accepted December 23. Sponsored by the National Institutes of Health Bioengineering Consortium, American Institute for Medical and Biologic Engineering, and the Radiological Society of North America. Address correspondence to the author.
Index terms: Radiology and radiologists, research • Special Reports
 |
Introduction |
|---|
 TOP Introduction BIOMEDICAL IMAGING SYMPOSIUM...RECOMMENDATIONSCONCLUSIONS FOR BIOMEDICAL... |
|---|
Symposium speakers and audience participants developed the following recommendations to achieve these goals.
|
BIOMEDICAL IMAGING SYMPOSIUM REPORT |
|---|
|
TOPIntroductionBIOMEDICAL IMAGING SYMPOSIUM...RECOMMENDATIONSCONCLUSIONS FOR BIOMEDICAL... |
|---|
The objectives of the symposium were to (a) identify the most important challenges and opportunities in biomedical imaging science, (b) develop strategies for integrating imaging science with biologic and medical research, (c) provide a forum for interdisciplinary imaging scientists to discuss the vision and future of biomedical imaging, and (d) recommend areas of future investment to NIH.
Three general topics were addressed. These were imaging at the cellular and molecular levels; imaging in the early detection, staging, and recurrence of disease; and imaging in therapy for various diseases. Speakers and audience members provided the NIH with recommendations regarding the use of imaging sciences in medicine, mechanisms to enhance research and development of imaging technology, and the imperative for preparing a future generation of imaging scientists.
Imaging at the Cellular and Molecular Levels
Symposium speakers described recent advances in imaging at the cellular and molecular levels in two broad categories. Optical coherence tomography, ultrasonography (US), and magnetic resonance (MR) imaging were discussed as examples of cellular imaging technologies in which increased spatial resolution has given us new windows into the anatomy, cell structure, and histopathology of living tissues. These imaging modalities are noninvasive or minimally invasive and can be used to perform serial "virtual biopsies" or obtain "snapshots" of target organs and tissues in living animals over time. Positron emission tomography (PET), MR imaging, fluorescence spectroscopy, and nuclear medicine techniques are examples of the detection methodologies used with molecular probes and contrast agents and typify the powerful new field of molecular imaging. Molecular probes are often signal-producing agents linked to drugs or proteins and are designed to be specific for a particular molecule or biologic process. A probe may also be introduced as a gene that codes for an optically active protein or contrast agent–binding receptor when expressed. An example would be fluorine 18 fluoroethylspiperone, a PET molecular imaging probe that binds to the dopamine D2 receptor. A PET image would show only those tissues where the D2 receptor is found on the cell surface. Another example is a caged gadolinium molecule, which changes conformation and becomes an MR imaging contrast agent in the presence of calcium.
It is very exciting that along with the deeper understanding of disease afforded us by molecular biology and genetic research comes the potential to visualize these processes with molecular imaging. Molecular information has profoundly affected our approach to diagnosing and treating disease, and many diseases are being redefined in terms of the their characteristic genetic or molecular abnormalities. Likewise, new therapies are designed to specifically target the abnormal gene or phenotypic pathway. There is a concomitant need for noninvasive or minimally invasive imaging procedures that will provide the information to make these molecular diagnoses or track the effects of these targeted therapies at the molecular level in patients. Molecular imaging techniques are already used in isolated cells, tissues, and animal models and are being developed for use in humans.
The science of molecular imaging needs further development as a research tool in small animals and in humans, and it needs to be refined for routine clinical use. This will require close collaboration between basic scientists who make the molecular discoveries and the imaging scientists who can create useful imaging procedures. In addition to substantial improvements in the spatial, contrast, and temporal resolution of imaging device sensors, chemists will need to create the diagnostic imaging probes that will amplify the molecular signals and add chemical specificity to the detected events. Team science is essential for these advances to occur. A challenge for NIH is to facilitate and support the multidisciplinary efforts needed to bridge these gaps.
Imaging in the Early Detection of Disease
The outcome of therapy often depends on early detection in a symptomatic patient; fast diagnosis; and accurate, careful staging of the disease. Imaging methods such as MR imaging, US, conventional radiography, and computed tomography (CT) have become standard tests in disease detection and patient care. New clinical modalities with improved sensitivity are on the horizon: These include PET and functional MR imaging of the brain for neurologic disease; MR imaging of cardiac function; contrast material–enhanced MR imaging for breast and other cancers; and high spatial resolution three-dimensional US for breast cancer, prenatal examination, and heart function. While imaging capabilities are becoming more sensitive, thus enabling detection of a greater percentage of lesions than ever before, they need to provide more specific biologic characterization to improve our fundamental understanding of disease and provide a diagnosis more specific to the development, selection, and evaluation of therapies.
Public health is improved by population-based screening for disease, but these screens must be fast, inexpensive, easy to administer to very large numbers of people, and reliant on technology that can be widely distributed. They must also result in few false-positive and few false-negative diagnoses. Low specificity of diagnosis is a major problem in current early detection methods, both for symptomatic patients and for population-based screening. For example, new technology that overcomes the low specificity of current screening methods for cervical, breast, prostate, colon, lung, and skin cancer would substantially lower the cost of diagnosis and therapy. Likewise, the incidence of death and impairment could be reduced if patients who have had a stroke or myocardial infarction could be evaluated in the emergency room quickly and accurately for their likely response to existing therapies.
To enhance our ability to detect disease at its earliest stages, scientists must determine how to define disease better in terms of the specific molecular and genetic abnormalities underlying the symptoms. Novel imaging technologies are needed to pinpoint signifying events that mark disease onset and define its biologic characteristics. Scientists also must determine the most effective types of imaging modalities for different diseases, including diseases with very low incidence and prevalence which are often difficult to detect. Improved quantification of imaging test results is also important for standardization and comparison purposes, allowing us to better monitor therapy or disease progression.
NIH could identify the diseases for which patients would benefit from increased specificity of early detection. These diseases would represent areas of particular opportunity for imaging technology development, implementation, and dissemination. NIH could initiate discussions with manufacturers, researchers, and clinicians to develop these priorities and to alert and educate the research community about these high-priority areas.
Imaging in Therapy
Several decades of advances in the imaging sciences have demonstrated that imaging has become an important, if not critical, factor in the care of patients. We are now witnessing a far-reaching change in image-guided therapy. This "revolution" is based on synergistic interactions among scientists from many disciplines. In the symposium, image-guided therapy was explored from the point of view of the surgeon, the radiologist, the researcher, the bioinformatics specialist, and the bioengineer. Imaging in therapy encompasses an enormous variety of technologies. These include use of multiple imaging modalities for diagnosis and position of the lesion in three-dimensional space; real-time imaging of anatomy, physiology (eg, blood flow, neuron activation), and function during surgery; video imaging in laparoscopic surgery; image-guided placement of catheters and other devices; and surgical computer-aided diagnosis and distance medicine. There are new interventional imaging techniques such as thermal coagulation by means of a focused ultrasound beam. Molecular and other imaging modalities can be used to monitor drug delivery and action, gene expression, or metabolism. Serial imaging studies are used to monitor response to therapy for tumors. Developments in hardware and software in CT, MR imaging, US, PET, and single photon emission CT, or SPECT, in concert with clinician-directed applications have made this possible.
The various sciences underlying this field must be rigorously investigated to maximize the benefits that improved imaging modalities can bring to delivery of health care. Within academic health centers, however, image-guided and computer-assisted intervention does not have a clearly defined structure or home. This makes it difficult to establish a targeted grant applicant pool or criteria for research and clinical training. Therefore, the challenge is to develop a multidisciplinary approach, similar to existing centers of excellence, that will break down traditional barriers, foster collaborative interactions, and harness these technologies for hypothesis-driven research.
|
RECOMMENDATIONS |
|---|
|
TOPIntroductionBIOMEDICAL IMAGING SYMPOSIUM...RECOMMENDATIONSCONCLUSIONS FOR BIOMEDICAL... |
|---|
To achieve these goals, it must be recognized that imaging science encompasses a variety of disciplines including medicine, biology, physics, chemistry, engineering, and bioinformatics. Continued success depends on highly trained, multidisciplinary teams, as well as adequate resources for development and testing. Biomedical imaging therefore provides an ideal opportunity for productive collaboration between government, private industry, health care providers, and academia.
Multidisciplinary Research Programs
An important challenge is to encourage and facilitate the establishment of multidisciplinary research and development programs in biomedical imaging, with specific emphasis on molecular imaging or image-guided therapy.
Multidisciplinary programs are essential so that discoveries being made at the molecular and submolecular levels be translated into new imaging methods for disease detection, diagnosis, and therapy. Recent program announcements such as "Bioengineering Research Partnerships" (http://grants.nih.gov/grants/guide/pa-files/PAS-99-010.html) and "Bioengineering Research Grants" (http://grants.nih.gov/grants/guide/pa-files/PAR-99-009.html) indicate that NIH recognizes the value of multidisciplinary research that includes a strong bioengineering component. Current funding mechanisms include collaborative Interactive Research Project Grants that use the R01 mechanism (http://grants.nih.gov/grants/guide/pa-files/PA-96-001.html) and R24 grants (Resource-related Research Projects). Additional programs are needed to encourage collaborative projects between biologists, chemists, physicists, pharmacologists, computer scientists, bioengineers, and clinicians.
A challenge for NIH initial review groups, or IRGs, is to facilitate multidisciplinary participation in cellular and molecular imaging research. Relevant review panels should include experts in the various physical and biologic scientific disciplines that compose the field of biomedical imaging. Another challenge for NIH is to become receptive to applications in which a single project is led by multiple equal principal investigators with complementary expertise.
The National Cancer Institute recently released a request for applications for "In Vivo Cellular and Molecular Imaging Centers" (http://grants.nih.gov/grants/guide/rfa-files/RFA-CA-99-004.html). Similar programs focused on molecular imaging are needed for diseases other than cancer. Centers of excellence focused on image-guided diagnosis and therapy are also needed. These centers should be multidisciplinary, have multidepartmental support, and should combine onsite technology development with strong basic science and system engineering. Partnerships with industry are also desirable. The centers for image-guided diagnosis and therapy should develop a multifaceted scientific program that might include research in basic development of medical imaging technology, computer science, targeted delivery systems for therapeutics, endoscopic and endovascular tools, microsensors, or molecular imaging probes.
Imaging Technology, Probes, and Contrast Agents
Recent discoveries present a substantial opportunity to foster the development of new biomedical imaging technologies and the molecular probes and contrast agents that are the tools for linking imaging to specific biologic processes.
Many of the important advances in biomedical imaging over the past decades (eg, CT, US, MR imaging, PET) originated from investigators with roots in disciplines such as physics, chemistry, engineering, and other areas that have not been well represented in the mainstream of medical research. Among investigators who apply basic science to develop new imaging technology, there is a perception that this field of research is not well understood by funding bodies at the NIH and that the level of support is not commensurate with the level of impact that new imaging technology has had on research and medical care. A challenge for NIH is to fund early technical development of basic and generic imaging technologies, before specific disease or organ-oriented applications are proved. Programs that will attract the necessary experts in chemistry, physics, molecular biology, pharmacology, and bioengineering to collaborate in these efforts are needed.
A related opportunity for NIH is to develop programs and review mechanisms that will enable engineers, physicists, and chemists in the imaging sciences to participate more successfully in the open, peer-reviewed NIH system. This would allow them to fund their careers by the R01 mechanism, just as the biologic scientists have been able to do for many years. A challenge will be to dispel the notion that research directed at development of technology is intrinsically less meritorious than studies of biology or pathophysiology.
Molecular probes and contrast agents are the tools that link an imaging device to specific biologic processes of disease. These agents serve to selectively "light up" a specific region, tissue, lesion, or cell because of some unique aspect of its particular biology. Future success in molecular imaging will proceed directly from a timely investment in the design of these reagents. Funding will be required not only for research and development, but also to train people capable of designing and developing novel imaging probes. Molecular probes should be specific for biologic processes such as gene expression at the level of transcription or translation, signal transduction (cell surface receptors), enzyme action or other metabolic processes, blood flow, or drug action. Of particular importance are molecular imaging probes that are minimally invasive and allow tissues in animals and people to be monitored in vivo, and in real time. These will be used for disease detection and diagnosis and, ultimately, for guiding treatment. They will serve as tools in basic research and drug development, as well as in the design of trials in preclinical models of disease and in patients.
Supporting molecular imaging probe design and synthesis with use of such approaches as combinatorial and parallel chemistry and high-throughput chemical screening technologies is an area of important opportunity. Likewise, the development of in vivo molecular imaging technologies for screening molecular imaging probes and drugs in genetically engineered mouse models of human disease are worthy areas for investment.
Cooperative efforts between universities, other not-for-profit organizations such as the Department of Energy and the National Science Foundation, and the pharmaceutical industry in developing molecular imaging probes should be encouraged. These partnerships should be especially useful in drug studies and other forms of clinical research. Imaging of drug biodistribution will facilitate drug discovery, validation, and efficacy and the approval process. Molecular probes may help elucidate any changes in the biologic systems targeted by drugs. It is hoped that molecular imaging probes will be used as surrogate markers to help assess the outcome in clinical trials of new therapies that span the spectrum from gene therapy to conventional drugs.
Support for the preparatory work required to obtain Investigational New Drug, or IND, approval for probes and contrast agents is needed. Molecular imaging agents are likely to be highly specific for a particular application and may therefore not have a large market potential. Industry may have limited motivation to support the necessary work involved in Investigational New Drug preparation. If this support is not provided, most of these potential agents will never reach clinical trials.
Education and Training
New training programs in molecular imaging are needed to create a generation of scientists for whom the principles of imaging, physics, bioengineering, molecular biology, physiology, pharmacology, and pathophysiology form an intellectual continuum.
Traditional scientific training programs are designed to teach traditional disciplines. New training and educational paradigms are needed to prepare scientists for the type of interdisciplinary research required for success in the imaging sciences. These scientists will need chemistry, physics, molecular biology, pharmacology, and bioengineering so that they can understand the principles of medical imaging, probe targeting and development, tracer methodologies, normal physiology, and the process of disease. The need for scientists with this type of integrated education is expected to grow along with the demand that molecular and cellular imaging take its place in the research toolbox of cutting-edge biologists.
New initiatives in training at the interface of molecular imaging, chemistry, physics, pharmacology, integrative physiology, and bioengineering must necessarily attract training faculty with a commitment to this integration. Needed are interdisciplinary graduate programs that provide the student with a continuum from the basic imaging physics, through the chemistry of molecular probes, to cellular and molecular principles, and extending into disease processes. The course content, laboratory experience, and research milieu of such programs must cross traditional disciplinary boundaries. Such training should be made available to students from the predoctoral to postdoctoral levels. These programs should also have the resources to provide postdoctoral training for qualified young researchers who have degrees in the traditional biologic sciences, chemistry, or biomedical engineering disciplines. Finally, the fastest way to stimulate new research in this field is to provide for the continuing education of senior faculty who already command research resources, in order to bring to their current programs these new technologies and concepts at the interface of imaging and biology. Fellowships and sabbatical support for immersion experiences and for intensive didactic exposure should be made available to senior faculty.
Proposals for new interdisciplinary bioimaging training programs should be reviewed for merit by senior scientists from the various disciplines, to ensure that curricula encompass all the relevant scientific disciplines within the field of biomedical imaging. It is important that these reviewers understand and work within the multidisciplinary philosophy.
Clinical Trials and Informatics
Clinical studies, with careful attention to integration of informatics, are needed to assess biomedical imaging technologies and to advance biomedical imaging research.
Clinical trials of new imaging technologies and new imaging probes are essential and require considerable resources. There is a challenge for government and private entities to collaborate on funding and conducting the necessary clinical trials of biomedical advances. Many clinical trials are performed for reasons other than testing imaging modalities, but use imaging as part of the trial. These present an opportunity and a challenge to ensure that the biomedical imaging used in the trial is state-of-the-art technology. Furthermore, there is a challenge to NIH to provide leadership in establishing standards for recording image data and to employ these standards in all its clinical research.
Informatics, an essential component of biomedical imaging, concerns the collection and processing of imaging data for use in research and medicine. It also refers to the establishment and management of large databases of imaging information and to the process of extracting information from them. Rapid evolution of medical procedures in general and medical imaging in particular has led to a gap between the mass of clinical information a clinician must synthesize and the technology available for integrating this deluge of information into a coherent, readily interpretable picture. Today, an unparalleled opportunity to fill this gap arises from inexpensive powerful computers and networks, the transfer of image analysis methods from military and space agencies to civilian use, and the increasing availability of digitally stored clinical imaging data.
Most extant imaging databases have been developed by individual research organizations. Data are stored in incompatible formats, and images from one cannot be compared with those from another. No widely accepted formats or standards have evolved for collection and storage of three-dimensional data. Therefore, a system to standardize imaging information is greatly needed, along with national databases to archive, organize, and retrieve data. These resources will be invaluable for the validation of new diagnostic technologies so that they can become clinically useful.
The following areas of opportunity were identified by speakers at the symposium:
1. Interdisciplinary collaborations are needed to develop new informatics-based systems and models that integrate clinical and biomedical imaging data to support medical decision making.
2. Databases are needed that contain both biomedical images and pathology reports or other outcome and clinical data. These should comprise an adequate number and distribution of cases to address institute priorities. Databases would be used, for example, in testing and validation of disease-specific, computer-aided image analysis methodologies. This effort should include standards for acquiring data, specifications for data compression and storage, standards for nomenclature, and standards for database structure so that investigators can work across multiple databases.
3. Computer-aided image analysis methodologies are needed for increasing the specificity of current clinical imaging methods.
4. Biostatistics components are needed in all image analysis programs to better quantify the results of imaging tests.
5. Informatics and computer science could be integrated into funding priorities, program announcements, and other aspects of research, education, and training.
There is an opportunity for NIH to collaborate with other agencies or exert leadership in any or all of the above areas. NIH could hire staff or use external advisors with expertise in systems design, process engineering, and informatics at all levels of the organization, from review groups through institute policy and planning. Potential research methods could be evaluated against such design criteria in order to make good choices to use in expensive animal and clinical testing phases. NIH could also make such expertise available to clinical trial groups planning to test biomedical imaging or engineering objectives and foster interdisciplinary collaborations that bring systems expertise from engineering and informatics together with the appropriate expertise in biomedical imaging, medicine, biology, chemistry, pharmacology, and/or technology development.
Relationship between NIH, FDA, HCFA, and Industry
Greater cooperation among NIH, FDA, HCFA, and industry (both large and small businesses) would improve the speed with which new imaging technologies, probes, and contrast agents can be transferred into clinical practice.
We are currently seeing the rapid development of new molecular probes and contrast agents that allow biologic processes to be imaged. Many such probes are now ready to be tested in phase I and II clinical trials. These compounds differ from pharmaceuticals in that they are designed to have no effect on the biologic process under investigation. Also, many pharmaceuticals are given over sustained periods of time, whereas molecular probes are administered a few times at most. It would be an impediment to clinical research if these new molecules were required to go through the same regulatory requirements used to assess pharmaceuticals. The same standards of safety, but not the same process, should be applied to both probe molecules and pharmaceuticals (with adequate consideration of the extremely small amount of the molecular probe that is usually administered). On the other hand, different criteria must be used to assess the efficacy of probes versus drugs. Most important, these probes must be proved to enable accurate assessment of the biologic process under study or the location, amount, and state of its specific recognition site.
Symposium speakers encouraged NIH to continue its communication with FDA and HCFA aimed at streamlining existing procedures that affect transfer of molecular imaging probes and other imaging technologies into clinical practice. These agencies could facilitate this process by developing efficient and effective ways for technology to be evaluated, approved, or discarded. Policies for physician and hospital reimbursement must be in place before approved technology can move fully into the clinical setting. Clinical implementation is a particularly difficult process for technologies that employ both an instrument and a biologic imaging probe because both have to be reviewed and evaluated. Communication among NIH, FDA, HCFA, and industry, such as that recently promoted by the National Cancer Institute, could lead to streamlined technology assessment and minimize redundancy in the approval processes at FDA and HCFA.
|
CONCLUSIONS FOR BIOMEDICAL IMAGING RESEARCH |
|---|
|
TOPIntroductionBIOMEDICAL IMAGING SYMPOSIUM...RECOMMENDATIONSCONCLUSIONS FOR BIOMEDICAL... |
|---|
|
Footnotes |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. J. Kelley, R. J. Davidson, and D. L. Nelson An Imaging Roadmap for Biology Education: From Nanoparticles to Whole Organisms CBE Life Sci Educ, June 1, 2008; 7(2): 202 - 209. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Dzik-Jurasz The development and application of functional nuclear magnetic resonance to in vivo therapeutic anticancer research: 2002 Sir Godfrey Hounsfield lecture delivered at the President's Day, Manchester Br. J. Radiol., April 1, 2004; 77(916): 296 - 307. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A S K Dzik-Jurasz Molecular imaging in vivo: an introduction Br. J. Radiol., December 1, 2003; 76(suppl_2): S98 - S109. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
