It wasn't so long ago that radiologists could be aptly described as the hard hats of the medical specialties--clinicians technologically and visually adept in manipulating sound and light rays that provide intimate peeks into the inner structure of the human body. As medical students, would-be radiologists probably could have afforded to perhaps snooze a little bit through biology and biochemistry classes. But gross anatomy was another story, requiring rapt attention.
"The radiologists I know are the most pragmatic of specialists," said Steven Larson, chief of the nuclear medicine service and director of the Laurent and Alberta Gerschel PET Center at Memorial Sloan-Kettering Cancer Center in New York. "They are not so interested in theory, but image analysis" which relies on a detailed understanding of anatomy. "It's kind of an engineering mentality," he said.
Throw those old assumptions away. Radiology is undergoing radical change thanks to a burgeoning new field called molecular imaging, a technology for peeking into cells for early traces of disease.
Decades in the making, as nuclear medicine grows ever more sophisticated, the evolution is now becoming a revolution, spurred by the mapping of the human genome and the development of new biology-based imaging technologies, particularly positron emission tomography, or PET.
The commercial introduction of PET/computed tomography four years ago marked the official union of biology and anatomy at the altar of radiology (Nov. 27, 2000, p. 48). Now wielding the new piece of hardware from their ever-expanding toolbox, radiologists are poised to move into uncharted areas, both diagnostic and therapeutic.
Molecular imaging potentially opens a variety of new doors and opportunities for radiology, bringing radiologists out of the insular environment of radiology departments and into other clinical areas such as pathology, surgery and oncology. The field will almost certainly change the way radiologists will have to be trained, and it offers limitless business opportunities for entrepreneurial hospitals.
But it also threatens new turf wars between radiology and other specialties such as oncology and nuclear medicine and a more traditional foe, cardiology (Dec. 1, 2003, p. 28).
"We're talking about a whole new way of thinking in terms of diagnosing and treating diseases that are common," said David Rollo, chief medical officer of nuclear medicine at Philips Medical Systems. "With almost every new breakthrough in healthcare, we've found that there is a resistance to the acceptance of new technologies in part because of its impact on a physician's usual method of practice but also because of a lack of education of both physicians and patients."
Bringing therapy to the doorstep of radiology will create tensions, though tensions often spawn progress, Larson said.
"Radiologists in the future will identify cancer and treat it. We will see an expansion of the interventionist's role," Larson said. "I think people will be referred for diagnosis and therapy to radiologists, but a lot depends on the choices made now about the preparation of the next generation of imagers. A lot will be done with closer interaction."
Riding the wave
Larson should know. He's been riding the wave for three decades. One of the world's foremost experts in targeted radiotherapy and molecular imaging, he is being honored with the Outstanding Researcher Award at this year's annual meeting of the Radiological Society of North America, set for Nov. 28 to Dec. 3 in Chicago. In 1983, he was recruited to the National Institutes of Health to establish a state-of-the-art PET center for researchers. Larson is being recognized by the RSNA in part for his many contributions to the advancement of PET as a clinical tool for oncology.
Larson defines molecular imaging simply as the imaging of key molecules important to the disease process. In cancer, cells go bad because of gene mutations, caused by a variety of factors, some known and many unknown. Researchers "have begun to image these processes of molecular change in cancer cells," investigating how genes are expressed, he said. For example, at Memorial Sloan-Kettering, Larson and other researchers from the departments of radiology and medicine and the cell biology program have found a way, using PET, to image a growth factor sometimes found in breast cancer cells called HER2. The amount of HER2 in a breast cancer cell is key in targeting which drugs will most effectively shrink or destroy the tumor. The method allows doctors to image the immediate effect of drugs on the tumor and it does so noninvasively. The results of Larson's research group study were published in the May 9 issue of Nature Biotechnology.
For all kinds of cancers and other diseases, the ability to directly measure a patient's immediate response to a drug regimen could save weeks of needless treatment. A similar process could also be indispensable in the development of new drugs, shaving months or even years off misdirected research and development prior to Food and Drug Administration approval.
"In the future we think that molecular medicine is going to be the way in which we treat human diseases by targeting the molecule that is at fault," Larson said. "The future of diagnostic imaging is going to be using this biologic imaging to determine when a patient needs a drug, whether it is doing what it is supposed to do and it also will be important in the development of new drugs."
At Ohio State University Medical Center in Columbus, researchers are about 1â years into a pilot study using molecular imaging techniques to diagnose melanoma of the eye. The study aims to develop a method for imaging the genetic expression of a tumor, mapping out information about an optical lesion's blood supply, blood flow and the leakiness of the vessels, said Michael Knopp, chairman and professor of radiology of the medical school's department of radiology. So far 10 patients have participated in the study. The most current findings will be presented at the RSNA meeting, he said.
"Radiology has been focused on morphology and structure," Knopp said. "It is not that we are abandoning it, but we're adding additional functional and molecular information to our repertoire, which we use for our noninvasive assessment. Radiology needs to expand to detect not only that something is wrong, but what is it" that is wrong.
Michael Phelps, co-inventor of the PET scanner in the early 1970s and director of the Crump Institute for Molecular Imaging and the Institute for Molecular Medicine at the UCLA School of Medicine in Los Angeles, will deliver the New Horizons Lecture at RSNA on "Molecular Imaging: From Nanotechnology to Patients." Phelps said a brave new world called systems biology is redefining the concept of disease.
"Disease is a reprogramming of the cells to gain and lose function and do harm to us. So we have to understand how they do that and develop new molecular diagnostics and therapeutics," Phelps said. "Patients don't get cancer; they develop cancer--and over many years." In systems biology, he said, "We don't believe in breast cancer, prostate cancer or Alzheimer's disease. We believe that cells are being reprogrammed, and there's a continuum of cell types as they progress so that one drug won't work. It means you must stage the disease out." Rather than aiming to develop "silver bullet" blockbuster drugs, the new approach calls for a diversity of drugs for each therapeutic situation. Like Larson, Phelps said identifying which patient populations will respond to which drugs will be a bonanza for clinicians as well as for the biotechnology companies developing drugs.
He might be prejudiced, but the tool of choice for molecular imaging will be PET, Phelps said. "When I invented PET in 1973 and 1974, we had no interest in clinical medicine. Our goal was to use PET to understand the biology of the disease," he said. When it was brought into clinical practice by the giant diagnostic imaging companies in the late 1980s, PET arrived with the advantage of all those years of research, he said.
PET is a nuclear-based technology that when used in conjunction with injectable radiopharmaceuticals, offers intimate glimpses of molecular function. When combined with computed tomography, or CT--an X-ray based method for imaging the body's anatomy--PET images have the added benefit of structural context. The fusion of PET/CT also speeds up the examination time of PET imaging alone by as much as three times.
This year about 1 million PET procedures will be performed in the U.S. with a growth rate of as much as 50% per year, Phelps said. Industry revenue will be about $1 billion, and PET/CT now accounts for 85% of all PET scanner sales. GE Healthcare reports it has sold 350 PET/CT scanners worldwide since 2001 and the company has nearly stopped selling stand-alone PET units. PET/CT units range in price from $2.7 million to $3.1 million. The price of CT units range from $700,000 for a four-slice scanner to $2 million for a 64-slice machine, according to GE. GoldSeal, the GE unit that sells refurbished equipment, has a stand-alone PET for sale, but there hasn't been a buyer for one in two years, a GE spokesman said.
An early phase
PET/CT marks only an early phase in terms of integrating radiology with other specialties. Phelps notes that the hybrid scanner essentially brought radiologists, the purveyors of CT, to loggerheads with nuclear medicine specialists, the purveyors of PET. "Over time we will get to the point where all of them read the PET/CT image and it will be done as one procedure," Phelps said. "And what will you call that specialty? I don't know. There are opportunities and there are problems."
For academic medical centers, molecular imaging offers a gold mine of business opportunities, particularly in the development of radiopharmaceuticals and contrasting agents targeted to specific molecules, said David Piwnica-Worms, director of the molecular imaging center and professor of radiology and molecular biology and pharmacology at the Washington University School of Medicine in St. Louis. "The interest in molecular imaging is much more focused on biochemistry and events in vivo (in the body)," Piwnica-Worms said. "Part of the field is directed at those injectable agents that can help (imaging devices) detect those molecular events."
At Washington University, for example, researchers are employing molecular imaging to investigate multidrug resistance in chemotherapy. They are developing radiopharmaceuticals that would allow them to image the process by which certain genes inappropriately usher chemo drugs out of the cell, rendering the patient drug-resistant. Such crucial information would guide therapeutic decisions, avoiding costly and useless rounds of chemotherapy.
"The biggest opportunities in molecular imaging for clinical avenues and business opportunities are in the diagnostic agent development," Piwnica-Worms said. "It means more types and broader arrays of sophisticated probes that ... will be available to inject in patients to interrogate cells about molecular events."
The medical technology companies have long been positioning themselves for a future that is commonly called personalized medicine. Most notably, General Electric Co.'s $9.5 billion acquisition of Amersham BioSciences in April married the biochemical expertise of the life sciences and imaging agents company with the engineering expertise of the former General Electric Medical Systems, creating GE Healthcare. The new company will ring up $14 billion in sales this year and has earmarked $1 billion for research and development--all designed to give GE a prominent stake in the arena of molecular imaging.
Taking a different tack, last year Siemens signed a cooperative agreement with the Center for Molecular Imaging Research at Massachusetts General Hospital, Boston, that provides for Siemens employees to work on-site to collaborate on developing imaging concepts. The center--under the leadership of Director Ralph Weissleder--was established in 1994 to further the field of molecular imaging and test imaging applications in the laboratory and in clinical trials.
Philips also "is taking a very aggressive approach to molecular imaging because we believe it is the future of healthcare," Rollo said. Philips has created a separate business line called molecular imaging that is focused on identifying companies developing new molecular probes and collaborating with them. "We are the imaging guys," so Philips is constantly upgrading its scanners, including single-photon emission computed tomography, or SPECT, to ease the processing of images, Rollo said.
Though molecular imaging is still in its infancy, there are some practical applications already in the clinic, Rollo said. Molecular imaging has two components--diagnosis and therapy--and in both cases a molecular probe in the form of an injectable radiopharmaceutical is used. In its diagnosis component, the probe will target the disease like a heat-seeking missile, facilitating the imaging of the disease process with a scanner. For the therapeutic component, the molecular probe that is injected will carry a payload of drugs to destroy the disease. Diagnostic agents on the market outnumber therapeutic agents by about a 9-to-2 margin, Rollo said.
One prominent therapeutic agent is Zevalin, marketed and distributed in the U.S. by Biogen Idec and approved by the FDA in 2002 as the first radioimmunotherapy. It is used for the treatment of 60% of the patients with non-Hodgkin's lymphoma who do not respond to traditional therapy. Zevalin seems to work on about 80% of those patients and half them will go into total remission, he said.
"The point of all of this is that the holy grail for molecular imaging is to ultimately be the diagnosis and treatment of choice for various diseases," Rollo said. He argues that in addition to its clinical promise, there is a compelling business case for molecular imaging. Because it is done mostly on an outpatient basis, targeted therapy offers fewer complications. In many cases, surgery would be avoided. As an example he notes that conventional treatment--surgery and radiation--for non-Hodgkin's lymphoma averages $60,000 per case. Using Zevalin on an outpatient bases costs about $25,000, he estimated, and can spare the time and cost of standard treatment for the 60% of the patients who would not respond to conventional therapy anyway.
"This could eliminate a lot of what may eventually be considered unnecessary and even inappropriate procedures," Rollo said. Because of all this, molecular imaging also promises to improve the quality of life.
"The train is moving and the FDA has created an accelerated approval process so many (drugs) that used to take a long time, the FDA has streamlined with a combination technol-ogy section specializing in molecular imaging applications and approval," Rollo said. "So a lot of stuff that wasn't here two years ago is here now."
Similarly, Siemens is convinced that molecular imaging has "huge potential in enabling early diagnosis of disease," said Mohammad Naraghi, senior vice president of business development at Siemens Medical Solutions. Early diagnosis will improve patients' quality of life and also wield a "huge impact in terms of cost," simply by reducing the expenses that accrue when a disease runs its course unhampered. An array of modalities--PET, SPECT, magnetic resonance imaging and optics--a promising imaging technology based on fluorescence that is still under development--are all being tweaked to play key roles in the new arena. Naraghi would not disclose what Siemens is investing in molecular imaging, but he said that clearly in the future there will be big growth in PET/CT and SPECT/CT scanners.
Of the big three imaging companies, however, GE has made the most public investment in molecular imaging through its acquisition of Amersham. That, along with several other life sciences businesses GE acquired four years ago, represents a change in focus from reactive to proactive medicine, said Alexander Tokman, general manager of global molecular imaging and radiopharmacy for GE. "The goal is to understand who is predisposed to which diseases and when they can expect to acquire it, and the third question is what to do about it--what is the best therapy?" he said. "So we've aligned ourselves around this new model of personalized medicine. ... To do this, we extended our expertise into nontraditional areas such as chemistry and molecular biology."
GE is actively involved in three areas of personalized medicine: molecular imaging, molecular diagnostics and bioinformatics, but not therapeutics, Tokman said. Though radiology departments were GE's traditional partners, the Amersham acquisition has allowed the company to expand its customer base to researchers and pharmaceutical companies to provide tools for basic research and drug development.
"We're aligning ourselves around personalized medicine and we're putting a strong bet that hospitals will be doing so as well. There's a tremendous amount of money from the government to expand molecular imaging programs around the country," Tokman said. "Every hospital now is going to be measured by how they are proactively evolving around the molecular imaging equation."
There are plenty of business opportunities for hospitals as well, he said. "Hospitals with nuclear medicine departments have some of the most profitable areas in hospitals," Tokman said. "Surgery is one of the highest revenue generators for hospitals, but as more and more noninvasive diagnostic imaging tests and targeted agents are introduced, there will be less need for surgery and intervention. That's the holy grail of personalized medicine."
What do you think?
Write us with your comments. Via e-mail, it?s [email protected]; by fax, 312-280-3183.