En Español | Plain Text Version | Skip Navigation

You are here:  Home > Find People > Groups at NINDS > Brain Tumor Progress Review Group

Imaging

Skip secondary menu

Co-chairs: Ronald G. Blasberg, M.D., and John C. Mazziotta, M.D., Ph.D.

Participants:

Ramon Gilberto Gonzalez

Fred Hochberg

Kenneth A. Krohn

Kathleen R. Lamborn

Sarah J. Nelson

 Edward A. Neuwelt

William M. Pardridge

Bruce R. Rosen

Gail Segal

Arthur W. Toga

STATEMENT OF THE PROBLEM

The past 60 years have seen a progression in the field of imaging from the early use of cerebral angiography and pneumoencephalography to the development of early radionuclide techniques and the advent of X-ray computed tomography (CT) in the 1970s. The current era has seen the implementation of magnetic resonance imaging (MRI) in all its permutations (structural, functional [fMRI], perfusion, diffusion, and spectroscopy), single-photon emission computed tomography (SPECT), positron emission tomography (PET), and intraoperative ultrasound. These techniques, now part of the clinical mainstream, are used individually or in combination to better understand the basic mechanisms, pathophysiology, and clinical features of brain tumors and their responses to therapy. They are also used to make therapy safer and more accurate, ultimately improving the quality and duration of life for patients.

Functional imaging--the visualization of physiologic, cellular, or molecular processes in living tissue--now provides insights into tumor blood flow, glucose or oxygen metabolism, and many other hemodynamic, physiologic, and biochemical processes. Such approaches may provide a means to identify molecular structures or receptors that cover the surface of a tumor and to help predict its natural history and response to certain treatments. Attempts are already being developed to use such strategies to examine gene expression, a strategy that will improve detection, staging, treatment selection, treatment monitoring, and prognosis. Imaging techniques are also now being linked to surgical and radiation therapies. Pre- and intraoperative imaging methods using CT, MRI, PET, and SPECT, as well as intraoperative methods such as radioisotope probes, optical imaging, and intraoperative MRI, are all finding a place in planning surgical or radiation therapy for patients and in sampling or targeting tissue for biopsy. These technologies also have a role in avoiding critical brain areas when destructive lesions or surgical resections are planned. They may also be important in evaluating the plasticity of normal brain tissue after such procedures, either during development in pediatric brain tumors or in adults.

The development of cellular- and molecular-based imaging will provide many new opportunities to assess brain tumors at a molecular level in animal and clinical models, as well as the ability to monitor gene therapy. As imaging techniques continue to evolve, it will be possible to visualize and quantitate changes as cells transform from normal to precancerous to cancerous. It may one day be possible to evaluate at-risk patients earlier in cancer pathogenesis, perhaps before a tumor becomes malignant. It is anticipated that, with the information obtained from the use of such imaging techniques, it will be possible to visualize the actual molecular signatures of cancer in vivo. The ability to detect fundamental changes associated with a tumor cell will thus vastly improve our ability to detect and stage tumors, select appropriate treatments, monitor the effectiveness of a treatment, and determine prognosis.

For example, patients may be selected for a particular drug therapy on the basis of imaging before drug administration. A drug's effect on specific protein interactions, signal transduction, or metabolic pathways could be measured, thereby providing new endpoints for monitoring drug response. In all probability, nomograms of response could be created for populations receiving therapies. Clinicians would benefit from quantitative methods for the identification of "partial response" and "complete response." These would serve as endpoints to replace survival in clinical trials.

In imaging, as elsewhere in cancer research, animal models of cancer are making it possible to perform certain kinds of studies that are difficult, if not impossible, to perform in humans because of practical or ethical considerations. A distinct advantage of noninvasive imaging in animal models of cancer is the ability to perform repetitive, noninvasive observations of the biologic processes underlying cancer growth and development without sacrificing the animal. The development of small-animal imaging devices, which can produce serial images of experimental brain tumors in small animals and incorporate all of the functional strategies just described, should provide a powerful new tool for experimental studies of brain tumor behavior and response to treatments. Further development of targeted contrast agents, ligands, and imaging probes also need to be supported because they will provide better in vivo elucidation of the key metabolic pathways and specific cell cycle functions that become altered in cancer.

It is also clear that better use can be made of existing data derived from imaging techniques. Combining in vivo phenotypic information about tumor characteristics with ex vivo analysis at genetic, cellular, and chemical levels can provide better correlations among these variables and the patterns seen in images obtained in vivo. Four-dimensional (spatial and temporal) data analysis of these characteristics should provide new insights into the natural history of tumor growth, patterns of spread, and responses to therapy. When optimized into probabilistic data structures that account for variance, not only in normal brain structure but also in tumor behavior, these approaches should provide new and useful tools for optimizing clinical trials. Image analysis techniques that integrate information across modalities, spatial and temporal scales, subjects, trials, and species will require the development of new algorithms; an emphasis on neuroinformatics for the incorporation of images into clinical trial databases; and the incorporation of genetic, demographic, and clinical data sets.

CHALLENGES AND QUESTIONS

The goals of the National Cancer Institute's plan for Fiscal Year 2001 with regard to imaging are quite appropriate to the challenges associated with imaging brain tumors. These goals include the following:

• Develop and validate imaging technologies, probes, and radiocontrast agents that have the sensitivity to detect precancerous abnormalities and very small cancers.

• Develop imaging techniques that identify the biological properties of precancerous or cancerous cells that will predict clinical course and response to interventions.

• Develop minimally invasive imaging technologies that can be used in interventions and assessment of treatment outcomes.

• Foster interactions and collaboration among imaging scientists and basic biologists, chemists, and physicists to help advance imaging research.

• Create infrastructure to advance research in developing, assessing, and validating new imaging tools, techniques, and assessment methodologies.

Each of the goals listed above is associated with a set of challenges and questions. The major challenge for the imaging community is to accurately measure tumor burden and function (phenotype). This goal will be pursued by using different imaging technologies, probes, and paradigms (see Exhibit 1) in order to better characterize brain tumors before, during, and after treatment. A further challenge will be validation of the new imaging paradigms through animal experiments, in vivo versus ex vivo (molecular) analysis, and clinical correlations.

One goal of surrogate marker imaging is differentiating the cellular and molecular characteristics of tumor from those of normal brain tissue. Another goal is to image the biology (molecular biology) of brain tumors with new techniques and probes. A final goal is to arrive at non-lethal endpoints for the assessment of treatment and the natural course of tumor growth. This can be done with the development of four-dimensional nomograms specific for tumor and age of patient and with spatial-temporal measurements, using a multimodality, multispectral approach.

Image Outcome and Monitoring Interventions

Imaging has a role in monitoring both treatment progress and outcomes in addition to drug toxicity. Available radiographic endpoints are inadequate as treatment markers and endpoints. Efforts must be made to achieve realistic criteria ("cut points") for partial and complete response to therapy and to define criteria for assessing toxicity to white matter, cortex, and ventricular structures. In the case of treatment, there are several aspects under consideration, each with its own challenges. Drug effectiveness should be assessed within the context of tissue concentration (labeled drugs, etc.), drug delivery of small versus large molecules (i.e., blood-brain barrier, blood-tumor barrier), and biological effect (function) of the tumor.

Image-guided strategies include the following:

• Preoperative and intraoperative planning (e.g., image-guided stereotactic biopsy and resection using PET, MR, fMRI, magnetic resonance spectroscopy [MRS], and optical intrinsic signal [OIS] imaging)

• Radiation therapy (e.g., image-guided stereotactic radiosurgery using PET, MR, fMRI, MRS, and OIS)

• Development of multimodal image registration (e.g., MR, fMRI, MRS, OIS, and PET)

• Development and availability of improved instrumentation (e.g., high-field human MR systems) and hybrid imaging devices (e.g., combined CT-PET or MRI-PET tomographs)

Assessment of Treatment Toxicity

Radiation and chemotherapy can have toxic effects not only for the tumor they are intended to treat but for brain function as well. Toxicity to brain white matter has received little attention in the past, and little is known regarding the mechanism (e.g., demyelination, axonopathy, edema) of this toxicity and its temporal course. Toxicity can also be gauged by imaging the vasculature of normal brain and tumor. Similarly, toxicity indices can be created for gray matter function and plasticity in treated children.

Transgene and Molecular-Based Therapies

Imaging at the molecular level (assuming a homogeneous region of interest) is on the horizon. The task will be to generate reporter gene constructs that can be imaged as markers for transgene delivery (e.g., viral vectors) and for markers of change in specific protein interactions, signal transduction, or metabolic pathways. These protein interactions and the specific steps in signaling pathways can be targeted by specific anti-tumor drugs, and drug efficacy assessments can be made by noninvasive imaging of the specific pathway. In the future, patients may be selected for therapy on the basis of imaging before drug administration. Response could be monitored by measuring changes in specific protein interactions, signal transduction, or metabolic pathways. In this way, new endpoints for monitoring drug response could be developed.

Development of Databases, Informatics, Standards, and Software Tools

The NCI has created brain tumor study groups offering Phase I and II studies that provide the basis for intergroup Phase III trials. Examples include the nationwide approaches to brain lymphoma and oligodendroglioma. It is important to translate the above-described imaging advances to serve clinical trials. Systems must be provided for sharing, accessing, archiving, and integrating information across all data types. Databases provided for these clinical trials must contain real-time imaging displays as well as quantitative data (e.g., volume, spectral ratios, diffusion ratios, and normalized cerebral blood flow [CBF] and volume [CBV] information). The field of informatics is growing rapidly in the biological sciences and can make a major impact in the area of brain tumor research. Unlike other organs in the body, the brain has a distinct and important architecture. Both the type and location of brain tumors are therefore important in understanding their causes, growth patterns, and response to therapy.

A logical framework for integrating information about these lesions would be to use the anatomical structure of the brain itself as the framework for an atlas that would store information about all patients, whether studied in scientific protocols or undergoing conventional clinical treatments. Imaging can provide this architecture for databases and atlases. It will be important to develop the tools and informatics methods to integrate imaging studies across modalities, spatial-temporal scales, subjects, trials, and species. Such databases and atlases should be four-dimensional (three in space and one in time, where the latter variable can be the age of the subject as well as the time course of tumor growth and treatment). Because brain anatomy is highly variable among individuals in a population, it is also advisable that such atlases be probabilistic in nature, thereby providing distribution estimates for the locations of regions and tumors. These four-dimensional, probabilistic, image-based databases can be linked to other data sets, including clinical, molecular, histological, and therapeutic variables.

To accomplish such an endeavor, it will be important to develop tools to normalize image acquisition across sites and to develop a standardized core image analysis and feature-extraction pipeline for the quantitative processing of imaging studies from all modalities. For example, a nationwide trial for the therapy of brain lymphoma in 600 patients can provide seminal data on the rate of response, relationship between drug dose and volumetric diminution of tumor, changes in spectra of magnetic resonance, and diffusion parameters. The clinical trials format is the ideal mechanism by which reliance on individual data sets can be replaced while providing integration across laboratories and experimental trials. This important, final recommendation is critical and, based on experiences in other fields, will require appropriate funding to provide the force for data integration. It will be necessary to create data standards (e.g., voxel size, diffusion weighted imaging (DWI) software sharing) and communication pathways and brain atlases at nationwide meetings.

RESEARCH AND SCIENTIFIC PRIORITIES

The following research priorities were identified for brain tumor imaging and are further detailed in Table 1:

Priority 1: Develop and validate new imaging markers and techniques to facilitate the spatial-temporal assessment of brain tumors. These will be applied to animal models, human phase I trials, and phase II-III clinical trials. (See Exhibit I for details of new probes, markers, and imaging techniques based on current imaging technology.)

Priority 2: Develop imaging techniques for predicting outcome and for planning and monitoring interventions in patients with brain tumors.

Priority 3: Develop databases, standards, and software tools that integrate demographic, clinical, and imaging information in a form that can be used to identify characteristics that are critical for managing brain tumors and tailoring therapy to individual patients.

RESOURCES NEEDED

Resources needed to address these challenges include the following:

• Continuation and expansion of the imaging research programs described in Exhibit II

• Development of a "shared" and "accepted" informatics and database infrastructure

• Meetings to facilitate relations among industry, academics, regulatory, and funding agents

• Research and training funds to support the scientific priorities

Table 1. Scientific priorities for imaging in the study of brain tumors

I. Develop and validate imaging markers for the spatial-temporal assessment of brain tumors for use in human research, clinical trials, animal models, and patient care.

A. Assess tumor burden as non-lethal endpoints for treatment assessment and natural course.

B. Develop tumor- and age-specific spatial-temporal nomograms (four dimensional).

C. Multimodality, multispectral approach

D. Validation via:

1. Clinical correlation

2. Molecular correlations

3. In vivo vs. ex vivo analysis

E. Validated for: 1. Monitoring treatment

2. Targeting biopsies

F. Differentiating tumor from normal brain 1. Preoperative planning

2. Intraoperative planning

G. Angiogenesis and tumor phenotype imaging

II. Image outcome and monitor therapy in patients with brain tumors.

A. Treatment 1. Drugs a. Tissue concentration--labeled drugs, etc.

b. Drug delivery--blood-brain barrier, blood-tumor barrier

c. Small vs. large molecule

2. Image-guided strategies a. Preoperative and intraoperative planning

b. Hybrid devices

c. Small-animal imaging

3. Molecular a. Reporter gene imaging i. Transcription of endogenous genes

ii. Post-transcriptional modulation of mRNA

iii. Protein interactions

b. Genotypes--antisense imaging strategies 4. Radiation a. Tumor

b. Brain

c. Vasculature

B. Toxicity 1. White matter a. Mechanism (demyelination, axonopathy, edema, etc.) as function of age and treatment plan

b. Temporal course

2. Vascular a. Mechanisms

b. Temporal course

3. Gray matter a. Mechanisms

b. Temporal course

c. Plasticity

III. Develop databases, informatics, standards, and software tools to:

A. Integrate across modalities, spatial-temporal scales, subjects, trials, and species

B. Develop four-dimensional probabilistic image databases linked to databases of clinical, molecular, histological, and other variables

C. Develop tools to normalize across image acquisition sites and to standardize a core image analysis and feature extraction pipeline (e.g., post-processing, distribution, etc.)

D. Alter the sociology of data sharing

EXHIBIT I:

New Probes/Markers and Imaging Techniques Based on Current Imaging Technology

I. MRI

A. Surrogate markers: 1. Methods need to be developed for choosing among many options; for example, using MRI alone, one can assess tumor volume, hemodynamics (CBF, CBV, BBB permeability), tissue water diffusion, and blood O2, as well as proton and 31P-MRS. At present, it is impossible to perform all these imaging motifs/measurements well, and there is little standardization of techniques across sites.

2. An overall concern is how to pick and choose among the different MRI motifs (let alone nuclear, CT, and other imaging techniques), select targets for validation, and compare studies by using the different imaging methods in the study of different patient groups with different endpoints.

3. Priorities and challenges might include both technology development (needed to facilitate improved quantification, especially for MRS and fMRI) and basic clinical validation studies (although the challenge above holds true here).

4. Ways need to be found to to standardize acquisition across multicenter trials with industrial collaboration.

B. Toxicity: 1. There is an apparent clinical need for functional cognitive studies. Because other sessions, especially the Radiation Therapy session, identified the same need, this call should be heeded.

2. fMRI can be combined with anatomical, biological, and functional assessment of white matter changes. fMRI/DWI tomography would be one priority, and the other imaging modalities described below would also be involved.

3. Assessment of vascular toxicity/BBBs in this category will overlap with surrogate markers, especially for anti-angiogenic treatments. Hemodynamics is a near-term goal for MRI.

C. Imaging Therapeutic Effect:

1. It is difficult to precisely define the need for quantitative in vivo pharmacodynamics within the tumor. This issue seems largely relegated to nuclear imaging, but MRS is likely to play some role as well. The area of pharmacodynamics represents an opportunity for industrial collaboration (e.g., by major pharmaceutical corporations) in a national effort to radiolabel all prospective therapeutic drugs.

2. Gene expression presents a "grand challenge." MRI/MRS is likely to play an important role in this new field of "molecular imaging"; it has the right elements and will be developed further in the decade ahead.

II. MRS has been a tool for examining the alterations in cellular metabolites associated with carcinogenesis and treatment response in animal and cell systems for more than 20 years. Recent developments in technology have allowed MRS to be increasingly routinely applied to patients with brain tumors in research and clinical studies. There are three major areas for development: A. Single voxel water suppressed proton MRS is the most widely available MRS technique for clinical applications. It has been shown to assist in characterizing tumor type and grade, distinguishing tumor from other mass lesions, and determining whether changes in lesion morphology correspond to tumor recurrence or treatment-induced necrosis. Challenges to this technology include the following: 1. Choosing the most appropriate region of the lesion to study 2. Developing databases of in vivo spectral characteristics and using pattern recognition techniques to identify fingerprints that are predictive of histology

3. Validating the in vivo findings by correlating them with results obtained by ex vivo analysis of molecular markers, histology, and nuclear magnetic resonance (NMR) spectroscopy of excised tissue

4. Educating radiologists and oncologists concerning the most appropriate applications of the technology and the interpretation of the biological significance of clinical MRS data

B. Single-voxel proton MRS techniques can only interrogate regions that are thought to be suspicious from morphological or other physiological criteria. Proton MRSI can be used to map out spatial heterogeneity in both the lesion and surrounding tissue for studies in animal tumor models and patients. Applications include directing tissue biopsies, planning surgical resection or other focal therapies, defining the extent of disease, and evaluating response to therapy. Challenges to this technology are as follows: 1. The need for improved data acquisition techniques for optimized shimming, more robust water and lipid suppression, more accurately tailoring the excitation to the region of interest, and improving signal-to-noise ratio by using either higher field magnets or more sensitive, custom-designed radiofrequency coils

2. Developing post-processing methods for displaying imaging and spectral data, registering serial data from follow-up examinations, and deriving quantitative indices describing the metabolic changes within the lesion and surrounding normal tissue that can be used for treatment planning and assessing response to therapy

3. Validation of the technology as a tool for routine clinical evaluation of brain tumor patients in both single and multi-institutional settings

C. Multi-nuclear MRS: At present, the applications of 31P, 13C, and 19F MRS are limited by low sensitivity and mainly involve cell and animal model systems. In cases where specific drug therapies have a signature that can be detected using one of these methodologies, there is promise for noninvasive monitoring of drug delivery and function. More generally, it is possible to measure pH, cellular bioenergetics, and phospholipid metabolism. Challenges for this technology include the following: 1. Obtaining 13C-labeled drugs at a price that is economic for routine basic and clinical research studies

2. Availability of high field human MRI scanners in order to obtain in vivo 13C, 19F, and 31P data at adequate signal-to-noise and spatial resolution

III. Recent developments in CT merit its consideration as a relatively inexpensive and widely available method for assessing physiological responses to brain tumor therapy. The development of the helical scanning technique and, most recently, multi-detector technology permits the measurement of CBF at very high spatial resolution with high precision and accuracy. This technology is moving rapidly and is likely to become the predominant CT technique in the course of the next few years. This capability and widespread availability presents an opportunity for inexpensive and widely available functional imaging of brain tumors. It is expected that high-resolution, high-precision CBF and vascular permeability imaging will be most amenable for assessment of anti-angiogenic therapies. It is possible, perhaps likely, that it will be useful for evaluating a wide variety of novel therapies. For these reasons, it is prudent to encourage explorations in the use of this technology in the evaluation of brain tumors.

IV. Nuclear (PET, SPECT, gamma camera; special topics that require further development)

A. Cell proliferation (assessment within 5 years) 1. Established probe: 11C-TdR

2. Developing probes: 18F-3'FLT, 124I-IUdR, 76Br-FbrAU

B. Angiogenesis (validation of assays for monitoring anti-angiogenesis therapy, assessment within 5 years)

1. Blood flow (15O-water, 99mTc-sestamibi, 201Tl-thallium, 133Xe-saline, etc.)

2. Blood volume (15O- or 11C-carbon monoxide-labeled erythrocytes [RBCs], or 99mTc-RBCs)

3. Capillary permeability (82Rb-rubidium, 68Ga-DTPA, 68Ga-transferrin, 18F-, 123I-, 131I-, 124I- or 99mTc- labeled albumin)

4. Oxygen metabolism (15O-oxygen)

C. Hypoxia (assessment within 5-7 years) 1. Established probe: 18F-fluoromisonidazole

2. Developing probes: 61Cu- or 64Cu-ATSM, 18F-EF1, 18F-EF5, others

D. Transporter up-regulation (assessment within 5-7 years) 1. Amino acid transporters (11C-methionine, 18F-FET, 18F-FACBC, etc.)

2. Nucleoside transporters (11C-FMAU, etc.)

3. Choline transporter (18F-fluorocholine)

4. Glucose transporter (11C-3OMG, 18F-FDG)

5. Other substrates

E. Cell surface receptors/antigens (endothelial cells and tumor cells; assessments over next decade) 1. Transferrin receptor (67Ga-transferrin, 111In-DTPA transferrin chelate, etc.)

2. EGF receptor (radiolabeled antibody or peptide)

3. Benzodiazepine receptor (iodinated-PK11195)

4. Other cell surface receptors/antigens (e.g., Flt1 and Flk1/KDR receptors for VEGF)

F. Cell matrix antigens (assessments over next decade) 1. Integrins (RGD- and other radiolabeled peptides G. Molecular imaging strategies/issues (assessments over next decade) 1. Reporter gene imaging (indirect assay; principle established)

2. Enzymatic amplification vs. receptor binding and internalization

3. Oligonucleotide/aptamer (direct binding/assay of mRNA/proteins; to be established)

4. Probe/contrast agent delivery issues (small molecules vs. macromolecules)

H. Molecular imaging specifics (assessments over next decade) 1. Endogenous gene expression (transcriptional activation/depression)

2. Post-transcriptional modulation/stabilization of mRNA

3. Protein-protein interactions of specific steps in selected signal transduction pathways

I. Gene therapy (assessments over next decade) 1. Vector delivery and transgene expression (reporter transgene imaging established)

2. Trafficking and targeting of genetically modified T cells

V. Optical intrinsic signal (OIS) imaging: A. The development of more rapid, sensitive, and efficient optical instrumentation with multispectral capabilities, in order to observe the etiology of various functionally active cell types in normal and pathogenic tissues

B. Imaging systems compatible with interoperative MR environments

C. The ability to examine molecular concomitant to intraoperative optical measurements using probes and other markers

D. Compatibility between other intraoperative instrumentation and optical acquisition and display, such as microscopes, stereotactic localizers, etc.

E. Co-localization of multimodality displays, including optical, pre- and intraoperative imaging, tomographic, and projection data

F. The ability to combine intrinsic and tracer-based optical images

G. New optical contrast agents to identify specific aspects of brain and brain tumor biochemistry. To a large extent these agents can be constructed as modifications of the nuclear probes.

EXHIBIT II:
NCI Imaging Research Priorities

Cellular and Molecular Imaging in Cancer: The Goals

Over the past several years, the National Cancer Institute (NCI) has been keenly aware of the potential power of imaging techniques and molecular imaging in particular. The Biomedical Imaging Program (BIP) (http://cancer.gov/bip/default.htm) of the Division of Cancer Treatment and Diagnosis is responsible for the extramural grant portfolio and programs related to oncologic imaging. Imaging has been identified as an area of "Extraordinary Opportunity" in the past several "NCI Bypass Budgets". The NCI Bypass Budget is a public document produced annually by NCI to identify for the Administration and Congress those scientific priorities on which the budget appropriation will be spent. The imaging-related goals of the NCI include:

i. Develop and validate imaging technologies and agents (e.g., probes, radiocontrast agents) that have the sensitivity to detect precancerous abnormalities or very small cancers.

ii. Develop imaging techniques that identify the biological properties of precancerous or cancerous cells that will predict clinical course and response to interventions.

iii. Develop minimally invasive imaging technologies that can be used in interventions and in assessing treatment outcomes.

iv. Foster interaction and collaboration among imaging scientists and basic biologists, chemists, and physicists to help advance imaging research.

v. Create infrastructures to advance research in developing, assessing, and validating new imaging tools, techniques, and assessment methodologies. The NCI has already made significant progress in the past several years toward reaching these goals with the introduction of various programs and initiatives:

• NCI has awarded three grants to support In-Vivo Cellular and Molecular Imaging Centers (ICMIC) (http://grants.nih.gov/grants/guide/rfa-files/RFA-CA-99-004.html). The ICMIC grants will facilitate interaction among scientists from a variety of fields to conduct multidisciplinary research on cellular and molecular imaging. The integration of this breadth of expertise is still in its early stages.

• The NCI has also funded nine pre-ICMIC planning grants (http://grants.nih.gov/grants/guide/rfa-files/RFA-CA-99-002.html) . The pre-ICMIC planning grants provide time and funds for investigators and institutions to prepare themselves, organizationally and scientifically, to establish an ICMIC.

• Small animal models, particularly genetically engineered mice, are powerful discovery tools, but we have yet to capitalize fully on their potential in cancer research. NCI has funded five Small Animal Imaging Resource Programs (SAIRP) (http://grants.nih.gov/grants/guide/rfa-files/RFA-CA-98-023.html). This initiative supports activities to develop and apply a wide variety of imaging modalities that focus on functional, quantitative imaging. Quantitating image data for small animals will lead the way to quantitative methods that can be applied in humans. An additional five SAIRPs will be funded in fiscal year 2001.

• New drug discovery programs are producing an increasing number of molecules for investigation, in turn stimulating a need for research that integrates imaging techniques into preclinical and clinical studies to assess newly developed therapeutic agents. NCI has set aside funding for the development and application of labeled therapeutic agents as compounds for imaging studies and imaging agents that serve as metabolic markers of response to newly developed therapeutic agents. The Development of Clinical Imaging Drugs and Enhancers (DCIDE)(http://cancer.gov/bip/dcide.htm) program will facilitate the development of novel imaging agents in preclinical development. A detailed overview of the newly approved DCIDE program will be presented at a future date in Academic Radiology.

Cellular and Molecular Imaging in Cancer: Meeting the Goals

To ensure that the initially defined goals for cellular and molecular imaging are met and completed in future years, the NCI has set forth in the 2001 Bypass Budget specific priorities and initiatives. These include:

1. Accelerate development of clinically useful technologies for detecting malignant and precancerous cells and for visualizing their functional characteristics. • Expand the number of In-Vivo Cellular and Molecular Imaging Centers (ICMIC) in 2001, 2003, and 2004.

• Expand the Small Animal Imaging Resources Program (SAIRP) to improve access to researchers testing new approaches to diagnosis, treatment, and prevention in animal models of cancer. NCI will foster collaborations between this program and the Mouse Models of Human Cancers Consortium (MMHCC).

• Support multidisciplinary centers of expertise to develop optical technologies and perform clinical feasibility tests of instruments able to visualize epithelial tissue at risk for common cancers and recognize the optical signatures of precancerous abnormalities. This often involves molecularly oriented techniques.

2. Develop, synthesize, validate, and distribute to the research community novel imaging agents.

• Expand a program similar to NCI's Rapid Access to Intervention Development (RAID) initiative (which is designed to accelerate the movement of novel interventions from the laboratory to the clinic) specifically for imaging agent development. The Development of Clinical Imaging Drugs and Enhancers (DCIDE) program (http://cancer.gov/bip/dcide.htm) will facilitate and promote preclinical development and validation of important imaging agents and ligands. NCI will, on a competitive basis, synthesize, test, and distribute probes that image the physiological and functional status of tumor tissue in the human body. The DCIDE program will be described in detail in a future issue of Academic Radiology.

• Establish a publicly available database of agents available to the research community, together with information on their properties.

3. Expand and improve clinical studies of molecularly based imaging modalities and image-guided interventions.

4. Integrate molecular and functional imaging technologies into drug development and early clinical trials (http://cancer.gov/bip/concepts.htm - c4).

• Support the development of in-vivo and molecular clinical imaging research tools for assessing the biological effect of cancer drugs on their intended target or pathway.

• With this continued investment in the future of imaging research, it will soon be possible to apply the techniques developed to image novel molecular targets, specific genetic pathways, signal transduction, cell cycle alterations, angiogenesis, apoptosis, and numerous other biologically relevant processes known to occur in cancer in routine clinical practice.

Cellular and Molecular Imaging in Cancer: The State of the Art

It is likely that the NCI goals and visions of cellular and molecular imaging in cancer research and patient care will be met. It is gratifying to note that the power of imaging with PET, nuclear medicine techniques, MRS, ultrasound, CT, optical imaging, and other techniques is being recognized and these techniques are becoming available in routine clinical practice. These modalities will allow for the molecular, functional, biochemical, and physiologic assessment of important aspects of malignancy. Many of these imaging techniques are already beginning to show their potential power in the management of the patient with cancer. With the continued advancements that are hoped to occur, imaging will assume a critical and essential role in the basic scientific understanding, diagnosis, staging, and monitoring of cancer.

In addition to the initiatives already in place and listed above, the resources required to address the challenges and opportunities for using imaging technologies in the study of brain tumor development and their treatment will require targeted funding in each of the areas detailed above. In addition, it is recommended that funds be provided for training in the areas of image analysis, tracer development, image methods development, and informatics systems. Funding should also be developed in a way that mandates integration of data across modalities, spatial-temporal domains, subjects, trials and species. This will require appropriate funding for the development of atlases and databases in which such complex and large-scale data sets (from patient outcomes to gene chip arrays) can be organized, archived, accessed, and distributed. Funding agencies should also provide opportunities for investigators to meet across disciplines to develop standards for communicating such information and to alter the sociological outlook of investigators from one of isolation and territoriality to integration and sharing. Finally, specific emphasis should be placed on funding programs that develop tools to standardize and normalize image acquisition across investigational sites and to develop core, quantitative analysis and feature extraction algorithms for the post-processing and distribution of imaging studies acquired from all imaging modalities in patients with brain tumors.

Last updated May 03, 2007