SPECIAL EDITION
Clinical Skills

By Merlina Trevino

"The University is a nursery of scientific research and of mental education, a place for cultivation of ideals for students as well as for teachers." -Wilhelm Roentgen

Over the course of five years, radiology residents have their hands full learning the details of CT, MR, ultrasound, and other established imaging modalities. Researchers around the country, however, are making the case that it's time to start preparing residents for functional and molecular imaging as well.

To a certain extent, both functional and molecular imaging are already part of routine clinical practice in terms of radionuclide-based exams such as PET and SPECT (Figure 1). But radiology's anticipated migration to functional imaging involves agents far more subtle than FDG.

"A new generation of probes and approaches for PET will likely impact routine clinical practice in the next four to six years. A new generation of molecules very specific for certain cellular events is already starting in clinical trials," said Dr. Sanjiv (Sam) Gambhir, director of the Crump Institute for Molecular Imaging at the University of California, Los Angeles. "Engineered antibody fragments known as minibodies and diabodies targeted against various cell antigens, such as carcinoembryonic antigen, are already starting trials, with many more to follow."

DEFINING MOMENT

Debate continues over how molecular and functional imaging should be defined. Hybrid scans and other types of imaging may render the terms less meaningful (Figures 2 and 3).

"Biomedical imaging in the postgenomic era should provide more than morphologic information," said Dr. King Li, associate director of the Radiology and Imaging Sciences Clinical Center at the National Institutes of Health. "Any measurement of events occurring at the cellular or molecular level obtained in vivo from biologic systems in a spatially and temporally resolved manner should be included in the broad definition of functional/molecular imaging. If we use this broad definition, all imaging modalities can be included."

The functional imaging practiced today-PET staging of lung cancer or evaluation of degenerative brain disease, for example-is still far removed from the more complex, detailed types of studies that Gambhir and others predict will change the field. In anticipation, research facilities across the U.S. are readying their training programs for a field that will reflect a much stronger focus on function and on the molecular processes of disease.

"Clearly, we're seeing the shift at Johns Hopkins," said Dr. Richard Wahl, director of nuclear medicine at Hopkins and vice chair for technology and business development for radiology. "Even meetings that have been anatomic in focus in the past have moved toward functional and molecular imaging. Medical students have become more familiar with the molecular biology of common diseases. The curriculum has to link anatomy and molecular etiology. With molecular tests, you can understand what causes false positives and false negatives. This will help radiologists to better interpret their scans."

COMING TO A CENTER NEAR YOU

At the national level, the NIH's imaging sciences department is funding an imaging scientist training program that helps researchers tie the newest imaging techniques to tissue analysis via functional genomics and proteomics. Imaging that blends the structural and functional has taken a top priority at the new National Institute of Biomedical Imaging and Bioengineering. The institute's first research grants, announced in April, include a $1.4 million project at Yale University to develop MR functional and spectroscopic imaging techniques to study and treat neocortical epilepsy, and a $330,000 grant to the University of California, San Francisco Cardiovascular Research Institute to develop new optical methods for imaging cellular architecture and dynamics.

The National Cancer Institute, meanwhile, has funded in vivo cellular and molecular imaging centers (ICMICs), which enable schools to bring in outside speakers on the subject of molecular imaging or set up Web sites (see accompanying article). The center for molecular imaging research at Massachusetts General Hospital, one of the first to receive a NCI ICMIC grant, has produced numerous seminars and lectures focused on different aspects of molecular imaging.

"Attendance has grown with the exponential increase in interest in the field," said Dr. Umar Mahmood, an assistant professor of radiology at Harvard Medical School. "The center has formed an angiogenesis club so radiologists in the entire Boston area can meet to discuss what approaches are most successful. Radiology residents have also spent dedicated time at the center to learn molecular imaging skills."

UCLA, another ICMIC, earmarked funds for minisabbaticals to provide hands-on research and didactic training in the interdisciplinary aspects of molecular imaging, according to Harvey Herschman, Ph.D., a professor of biological chemistry. Because functional imaging and molecular imaging encompass topics outside of traditional radiological training, some institutions are finding that grants not specific to radiology provide grounds for additional training in these new techniques.

"The ICMIC grant allows us to fund postdoc fellows as well as research on new techniques and technologies," said Dr. David Piwnica-Worms, director of the molecular imaging center at Washington University in St. Louis. "We're able to fund a lecture series and symposia and bring in outside speakers to give seminars."

Learning opportunities are not confined to fellowships and grants for established researchers. Residency programs are grappling with whether, how, and when to add functional imaging to their curricula.

"Right now, conventional radiological training doesn't engage critical thinking about the molecular mechanisms of disease or about molecular therapeutics," Piwnica-Worms said. "Radiologists, as generalists of sorts, might know which new drug is used with which disease, but they may not remember the details of how that drug worked at the cellular and biochemical level. An understanding of the molecular and biochemical mechanism of disease needs to be better incorporated."

Others caution that it may be premature to shift gears when functional and molecular imaging have yet to prove they can live up to the lofty expectations set for them.

"What is theoretically possible may never see the light of day," Wahl said. "For diagnostic agents, the development cost can be as much $20 million to $25 million. General agents like FDG that are not absolutely specific have an advantage. Because there is a willingness to pay more for drug therapies, it may make more sense to link diagnostic with therapeutic agents. Developing drugs is extremely expensive, and that is the limiting factor."

The interlude of uncertainty may give imagers a chance to prepare for coming changes (Figure 4), Piwnica-Worms said.

"In the future, radiologists will have to understand the details of the pharmacological agents and the targets, where these fit in the biochemical milieu of the disease process. And they will have to interpret all of this information in terms of pathophysiology," he said. "We won't be looking at fuzzy edges, we'll be looking at receptor density or the rate constant of the enzyme or transporter. Molecular imaging may be a more sophisticated and judicious use of image-enhancing agents."