Many established PET sites are approaching capacity and are taking on additional systems to keep up with demand

-- by Roger Lindahl

Routine clinical use of nuclear imaging's premier modality is on the rise following decades of effort to expand its use beyond applications in research laboratories. An informal survey by Diagnostic Imaging of PET providers indicates that clinical use of PET in the U.S. is growing substantially, fueled by a sharp increase in oncology referrals.

Many existing PET sites are reaching volume capacity and installing an additional unit. Several have seen competitors purchase or plan purchases of dedicated PET systems in their regions, when not restricted by state certificate-of-need regulations.

Mobile PET services are also beginning to grow with the assistance of emerging radiopharmacy delivery networks. PET isotope delivery has allowed sites without cyclotrons to gain timely access to fluorine-18 fluorodeoxyglucose (FDG), the workhorse PET radiotracer.

An improved reimbursement climate for PET over the past two years and mounting evidence of its unique clinical advantages in cancer imaging are drawing in a growing number of referrals around the U.S. Concurrent advances in PET technology have reduced operating costs and cut exam times while offering better image quality and sensitivity.

"There is no question that PET will fly. It is already flying," said Dr. Johannes Czernin, director of nuclear medicine at the University of California, Los Angeles. "We train radiologists, nuclear medicine physicians, and technologists who want to do PET. Last year, we trained a total of 15 people in PET. This year, we have already trained 50 people (as of March)."

Some of the growth in FDG-PET imaging has been spurred by the introduction of hybrid dual-head nuclear medicine cameras, which can perform both single-photon emission computed tomography and PET (dual-photon) imaging using coincidence detection techniques.

HYBRID CAMERAS

There are considerably more hybrid PET/SPECT systems in service than dedicated PET units. About 80 dedicated PET scanners are currently installed in the U.S., a fraction of the 300 to 350 coincidence cameras in use, according to Dr. R. Edward Coleman, nuclear medicine director at Duke University. Five mobile PET systems are also in operation, he said.

PET providers differ on how they expect the hybrid and dedicated PET systems to evolve. Some believe that practices will shift to dedicated PET once they reach the necessary patient volume to justify use of a higher performance dedicated system. Others see the performance difference narrowing between the hybrid and dedicated systems and most of the PET work going to the more prevalent coincidence imaging cameras.

While PET is taking some procedural volume from CT scanning in oncologic imaging, chances are that PET and CT will settle into a complementary relationship. Ideally, this diagnostic relationship will evolve into the use of a single imaging system combining the anatomical strengths of CT and the functional diagnostic power of PET.

"This (hybrid PET/CT) would make reading PET scans drastically easier. It also should improve throughput, since the attenuation correction of PET scans would be done using CT data and some mathematical manipulations to correct for difference in photon energy. This would be a much faster transmission scan than what we currently do with germanium sources," said Dr. Barry A. Siegel, director of nuclear medicine at the Mallinckrodt Institute in St. Louis.

A hybrid PET/CT scanner would also facilitate the fusion of anatomical and functional PET images into a single image, he said.

RISING VOLUME

Business is said to be booming at most clinical PET sites. At the same time, PET instrumentation has become easier to operate and image acquisition processing faster. Whole-body scans that took 90 minutes to perform a year ago are now done in 40 minutes, Coleman said. Despite faster throughput, many sites are reaching and exceeding their patient capacity.

"Our growth (at Duke) has been tremendous," Coleman said. "From 1997 to 1998, we doubled (patient volume) and doubled again from 1998 to 1999. We have one PET scanner and we start it at 6:30 in the morning and go until 8 or 10 at night. We will do over 2000 clinical PET scans this year. We can't do any more."

PET volume at the University of Washington in Seattle has also steadily increased over the past three to four years, according to Dr. David A. Mankoff, an assistant professor of radiology. The university operates its PET system until 9 or 10 p.m. to meet demand. The University of Washington runs the only dedicated PET system in the Pacific Northwest, including British Columbia, although this may change soon, Mankoff said. Another dedicated system is expected to begin operation nearby in about six months.

Enthusiasm for clinical PET grew dramatically with the approval of Medicare reimbursement for several oncologic imaging indications by the Health Care Financing Administration, Coleman said. The first two decisions, in January 1998, approved payment for PET imaging of solitary pulmonary nodules and lung cancer staging. Then, last July, HCFA approved PET imaging of colorectal cancer, melanoma, and lymphoma.

"This has made the interest in clinical PET grow tremendously," he said. "The number of individuals interested in doing clinical PET is increasing daily."

PET providers at West Virginia University noticed a definite pickup in referrals after the HCFA reimbursement decisions, said Dr. Raymond R. Raylman, research director at the Center for Advanced Imaging. The center's volume rose from around 20 patients a week to between 30 and 35 a week. When the university's PET system was installed, the state declared that this would be the only fixed PET site in West Virginia. Now, WVU as well as two or three private companies are investigating mobile PET opportunities, he said.

"For whatever reason, mobile PET in West Virginia has become a super-hot topic," Raylman said.

Private insurance payers have also been reimbursing PET, almost to the level of standard nuclear medicine, Siegel said. The PET payment collection rate at Mallinckrodt is only slightly below that of nuclear medicine overall.

"Eighty percent of private insurers cover for whole-body PET imaging," Czernin said. "This is not only a Medicare issue. Many cancer patients are younger than Medicare age."

Even so, providers are hoping that more widespread payment coverage of PET applications by HCFA will boost usage further, he said. "We expect broad (Medicare) coverage for all PET applications, probably this year," Czernin said.

But reimbursement can be a two-way street. Providers should be aware that PET payments might be trimmed in the future.

"There is a definite potential danger of (PET) reimbursement getting less," said Dr. Martin Sandler, interim chairman of radiology at Vanderbilt University. "The outpatient DRGs (diagnosis-related group Medicare payments) are at risk of going down."

Vanderbilt operates both a dedicated PET system and a hybrid coincidence detection camera. The performance of coincidence PET imaging technology is steadily improving, while the price of dedicated PET has not dropped substantially, Sandler said. For these reasons, many sites will continue to rely on the hybrid PET/SPECT technology.

"We distribute over 200 doses of FDG out of Vanderbilt a month," Sandler said. "I would say the majority of the doses go to patients of coincidence cameras. Most institutions, in order to afford a PET camera, have to consistently do three to four patients every day. That is a difficult volume to sustain for a medium-sized hospital."

Apart from independent cyclotron operations, five or six radiopharmaceutical firms are setting up networks for PET radioisotope distribution, Czernin said. This trend should bring the price of FDG down, helping smaller sites break even sooner.

The cost of operating a PET site has also been reduced by refinements in cyclotron technology. Newer cyclotrons are simple enough that they no longer require a dedicated physicist to supervise radioisotope production.

"It isn't any harder to run a PET scanner at this point than an MR," Mankoff said.

ONCOLOGY DOMINATION

While improved reimbursement for PET imaging is one reason usage is rising, the primary driver of referrals is clinical need. Specifically, FDG-PET imaging of tumors has been shown to provide utility in diagnosis and staging not found in competing modalities.

"The incremental value of PET over other techniques (for non-oncologic imaging) is not as large as the incremental value of PET in oncologic diagnosis," Coleman said.

That is why 90% to 95% of clinical PET work currently is in oncology, he said.

The other two major applications for PET are neurology, including the imaging of epilepsy and dementia, and cardiology, which includes the diagnosis of coronary artery disease and myocardial viability. Neurological use of PET has been slow to develop partly because there have been no therapies available for degenerative disorders like Alzheimer's and Parkinson's disease, said Dr. J. James Frost, a professor of radiology and neuroscience at Johns Hopkins University. That situation may be changing, as drug companies are on the track of neuroprotective therapies.

"Once those (therapies) are developed, then suddenly you have a great need for detecting neurodegenerative diseases very early in their clinical evolution so you can institute therapy," he said. "Imaging would have a very important role (in this)," Frost said.

Neurological PET could have a formidable competitor, however, in functional MRI. Although the utility of fMRI is still being investigated, it shows considerable potential in the brain, said Christopher Chabris, Ph.D., a research fellow at Harvard University.

"You can do a lot more (with fMRI) in mapping the functions of the brain precisely in terms of resolution and location," he said. "People can do a variety of mental tasks in one session. You can localize more functions than you can with something like PET. For these applications, a combination of functional MRI and anatomical MRI is the big leader."

While oncology dominates PET services overall, there are a number of centers that successfully focus on the heart. These sites bring in business despite the less favorable reimbursement climate for cardiac applications, said Dr. K. Lance Gould, a professor of cardiology at the University of Texas, Houston.

"PET can be the basis for a fundamentally different approach to managing heart disease," he said. "It is a very sensitive marker of changes in perfusion capacity that can reflect diffuse coronary artery disease before there are any flow-limiting stenoses."

The ability to provide very early diagnoses of coronary artery disease potential with PET enables physicians to better manage at-risk patients with drugs and dietary and lifestyle changes before more invasive therapies are needed, Gould said.

This approach has proved so attractive to individual patients that UT's cardiology PET scanner has been kept busy largely with patients who pay directly, bypassing insurance payers, he said. Gould's PET practice is booked six weeks in advance.

It is in oncology, however?in the detection and tracking of very small malignant tumors?that FDG PET makes a dramatic difference in the management of patient care.

"What is really driving the business is the outcry from the referring oncologists, surgeons, medical oncologists, and radiation oncologists," Mankoff said.

Increased reimbursement has been helpful in making more referring physicians aware of the clinical value of PET, he said. The PET imaging volume at the University of Washington has grown dramatically without the need for active marketing. PET's oncologic advantages have been spreading by word of mouth among referring physicians.

"Referring physicians are not stupid. They refer patients if they think a test can provide additional information that has an impact on management," Czernin said. "That is what is happening."

In a survey of referring physicians responsible for approximately 600 patients sent to the UCLA PET facility, there was a 30% to 40% change in management due to the PET findings, he said.

The relative utility of PET in oncologic imaging varies among the different types of cancer. Lymphoma, in particular, has benefited from the use of PET.

"One of biggest problems in lymphoma is that, when you get a response to chemotherapy or radiation therapy, the mass will shrink in size. (It is then difficult to determine) whether the patient has gone into remission or the disease is still active," Sandler said. "It is hard to tell just by looking at the mass. Now, we can do a noninvasive test that is exceptionally accurate in predicting whether the patient is in remission or not."

Superior tumor diagnosis using FDG PET may be taking some procedural volume away from CT, MRI, and biopsies, although all remain useful and complementary to PET.

"The enormous value of FDG PET is really in tumor imaging," Frost said. "That is where there was a preexisting need because CT and MR had limited sensitivity and specificity."

PET can replace biopsies in some patients while showing where to biopsy in others, Coleman said. CT is also being replaced in functions like the assessment of response to therapy.

"The major modality on which PET has had an impact is body CT," he said. "Most of the PET imaging we are doing is for oncology. Most of the patients had previously been followed using CT scans. Now a lot are being followed with PET." 

<< BACK | BACK TO TOP