According to the National Cancer Institute, 8.2 million people were afflicted by cancer in 1998, with 1.2 million of those listed as new cases that year. The same year, 564,800 people died of cancer. By comparison, the National Heart, Lung, and Blood Institute reports 7 million coronary heart disease cases with approximately 500,000 annual deaths. The National Institute of Allergy and Infectious Diseases states that less than 1 million people in the United States had AIDS in 1998, and 40,000 of those were new cases that year. Conclusion: Cancer by many measures is Public Health Enemy No. 1.

Among the tools necessary to combat the disease, few have proved as effective at diagnosing and staging malignancies than perfusion single photon emission computed tomography (SPECT) and — increasingly — positron emission tomography (PET).

Edward Coleman, MD, vice chairman of the Department of Radiology at Duke University Medical Center in Durham, NC, observes that the use of SPECT in oncology work is growing at rates of around 10% annually, and estimates that 100,000 oncologic PET scans will be performed nationally by year’s end, and 200,000 by the end of 2001.

“These modalities have had important places in nuclear medicine for some time, but they are now emerging with well-defined and essential roles in oncology,” Coleman asserts.


SPECT, Coleman says, is a widely available imaging technology. Most US hospitals that provide nuclear medicine have SPECT capability, he notes.

PET, on the other hand, is not as broadly deployed. Ruth Tesar, immediate-past president of the Institute for Clinical PET, estimates that perhaps 400 hospitals and freestanding imaging centers currently offer some form of PET — and of those, only about 100 sites are equipped with dedicated, high-end scanners. Tesar predicts that PET in cancer may soon account for as much as 30% of all nuclear medicine procedures.

“I don’t think PET will supplant one type of procedure for another, but will simply add to the total number of nuclear medicine procedures,” she says.

The appeal of both SPECT and PET in oncology is their ability to diagnose and stage a variety of malignancies including lymphoma, neuroendocrine tumors, breast cancer, brain tumors, colorectal cancer, and many others — especially those that can escape detection by other imaging modalities.

“It is probably easier to list the tumors they don’t identify than it is to list those they do,” suggests Val Lowe, MD, incoming president of the Institute for Clinical PET and PET specialist at the Mayo Clinic in Rochester, Minn.

While both SPECT and PET employ isotopic tracers to detect malignancies, there are some points of differentiation. “SPECT utilizes a number of different radiopharmaceuticals, each for uptake in a specific type of tumor,” Coleman says. “There are gallium 67 for lymphoma, indium 111 octreotide for neuroendocrine tumors, and thallium 201 for breast cancer, to mention some of the most common.”

By comparison, there is one primary radiopharmaceutical used in PET: fluorine 18-labled deoxyglucose (FDG-PET)

“Ninety-five percent of the FDG PET examinations are for oncology,” says Tesar, who also is the executive director of Northern California PET, a freestanding PET-dedicated imaging center in Sacramento, Calif. “The reason for the predominance of FDG in oncologic PET is that it can broadly distinguish metabolic activity in tissue. Since tumors are more metabolically active than the surrounding tissue, FDG zeroes right in on the problem site for the majority of tumor types.”

SPECT’s technological limitations — vulnerability to noise and scatter, plus problems with attenuation — are such that it does not portray malignancies as well as is possible with PET, Coleman indicates.


Obtaining needed radiopharmaceuticals is an issue that must be surmounted, unless the enterprise owns a cyclotron and can produce all that it needs on-site, as does the Mayo Clinic in Rochester, where Lowe is on staff. In March, Mayo acquired a cyclotron for ensuring the enterprise a reliable stream of FDG.

“The ability to make our own FDG allowed us to increase the number of PET scans being performed each week,” Lowe says. “Previously, we were doing scans only 2 days a week. Once we were making FDG on-site, we increased to a full week schedule.”

Mayo had confined its PET scan days to two in order to leave itself room later in the same week for rescheduling procedures canceled due to delays in the arrival of FDG shipments.

“FDG distribution can be spotty if your site is in a part of the country — such as Minnesota — where weather and other variables can affect your ability to get FDG on a timely basis,” Lowe explains. “If the FDG we requested from our supplier did not arrive here when it was supposed to, we had the flexibility to reschedule the patients to the following day. But the demand for PET scans was rising and we had to increase the number of days for doing those procedures. Only by getting our hands directly on the production of FDG could we hope to accommodate an expanded schedule.”

Smaller centers unable to afford the cost of a cyclotron have no choice but to rely on outside FDG production sources. That was the reality faced by Tesar’s PET imaging center. However, she was able to justify the acquisition of a cyclotron for in-house FDG production by partnering with a commercial FDG supplier and selling radiopharmaceuticals to other PET operators in the immediate vicinity.

Tesar — for some 15 years an administrator of hospital-based nuclear-medicine departments — entered the PET arena after calculating the modality’s market potential.

“I realized that PET was an effective way to arrive at an oncology diagnosis,” she explains. “The market potential plainly was obvious. PET benefitted payors and patients alike by helping avoid unnecessary surgeries and extra procedures.”


PET makes good business sense because Medicare and other third-party payors decided to reimburse appropriately for it.

“In January 1998, the Health Care Financing Administration (HCFA) issued its first policy for covering oncologic PET, the imaging of solitary pulmonary nodules and staging lung cancer,” Coleman says. “That made a profound positive impact on the number of studies we did. Then, in July 1999, HCFA added policies for covering colorectal cancer, lymphoma, and melanoma.”

Coleman says Duke expects by the end of this year to log approximately 3,000 oncologic PET scans — about 13 per day — for the 12 months of 2000.

“We have one scanner, and each whole-body PET takes about an hour to complete,” he says, “This keeps us working from 7 AM until 8 PM.”

For those labors, Duke can expect to receive from payors roughly $2,200 per patient scanned.

Tesar calls the compensation “very adequate; it covers your fully allocated costs.”

A significant part of those costs is the equipment. Sources list the price of a dedicated PET scanner as anywhere from $800,000 to $2 million. The most expensive PET systems are those that deliver the highest resolution, Coleman says. These, he notes, are capable of very accurately characterizing lung modules as small as 5mm.

“There are other dedicated scanners that are slightly lower resolution, but less expensive,” he says. “These seem to be quite acceptable for clinical PET imaging. One of these is a fixed-ring sodium iodide-based system. Another is a half-ring detector that rotates. Another major group of instruments would be the hybrid — or camera-based — PET systems. These offer lower image resolution than the other types. However, they are improving rapidly and now several manufacturers are coming out with CT scanners that are attached to these hybrids — the result is a dramatic improvement in image quality. Some of these are now being evaluated.”


In addition to new equipment, research is proceeding apace to introduce new types of radiopharmaceuticals that will be even more oncologic specific.

“The limitation of FDG is that it is fairly nonspecific about the type of tumor it shows us,” Tesar says. “We need more specificity. We want to know if that lung tumor we are looking at is just a lung tumor or some other type of tumor — perhaps a metastasizing breast cancer. And we need to be able to get more specific for the less metabolically active tumors.”

Tesar thinks it likely such radiopharmaceutical innovations will reach the market sooner rather than later, thanks in part to the Food and Drug Administration Modernization and Accountability Act. Signed into law in 1997, this measure — among other things — gave the FDA time to work with PET researchers, suppliers, and operators on developing a final rule for regulating FDG, and two more years after that to implement that rule. But the law also provides for a less bureaucratic process of developing and putting new tracers into wide service.

Already, signs of innovation are evident in many quarters, including Duke University.

“We are very excited about a new radiopharmaceutical that we have developed here at Duke,” Coleman says. “It is F-18-labeled fluorocoline. We have developed this primarily as a prostate cancer imaging agent. FDG is not as accurate in prostate cancer as it is in other malignancies, so we felt we needed something better. Carbon 11 choline had been used in Japan for imaging prostate cancer. Those studies looked good. But carbon 11 has the limitation of a 20-minute half-life. We were able to put fluorine-18 — which has a 110-minute half-life — onto the choline and have had very good results from it.”

As Coleman sees it, the future of both SPECT and PET should be quite bright.

“It is difficult to know what will be the next set of federally approved oncologic applications for SPECT and PET, whether we are looking at it from the purview of the FDA or HCFA,” he concedes. “I can say, though, that, here at Duke, we are planning on all of the oncologic applications being approved. In July, we sent to documentation HCFA as part of our request for broad coverage from Medicare. We have submitted a document that supports the broad coverage of almost all of the major malignancies, plus dementia, epilepsy, and myocardial viability. The documentation includes data on more than 18,000 patients to support our contention that we should obtain broad coverage. I think that after they see the wealth and extent of our data that show the sensitivity, the specificity, the cost-effectiveness, the changes in disease management, they will have a solid appreciation for the power of PET in particular and its utilization in a wide variety of malignancies and other conditions that will lead to broad coverage.”

Tesar predicts that a HCFA move to adopt broader coverage of PET will further stimulate utilization, but that reimbursements will eventually recede.

“I suspect that within the next 3 years, HCFA will initiate a review of PET reimbursement rates,” she says. “At that point in time, they will almost surely discover a decline in PET operator costs and this will lead to reimbursements being adjusted downward. I don’t anticipate a sharp decline in the rates, but I do imagine it will decrease at least a bit.”

For PET operators, there will be other challenges as well, she warns. “Unfortunately, PET is not a slam-dunk business opportunity,” Tesar says. “Importantly, you have to learn where PET fits in on the clinical pathway. There are certain tumors that we have not done the full education process on among community physicians. Breast cancer is one of these — we still need to work with our community physicians and the patients to figure out the best utilization for PET. Granted, we have looked at staging and diagnosis pretty thoroughly in oncologic PET and we have a lot of data to support that. Still, we need to gather more data in the monitoring of therapy and looking at how we can intervene in terms of therapy.”

Lowe, speaking in his capacity as incoming president of the Institute for Clinical PET, says that the organization, during the coming year and possibly beyond, will seek to develop and implement educational programs that will equip physicians new to the technology to properly read the images.

“PET scans are very unlike any other radiologic examination,” Lowe says. “They also are very much different from most nuclear medicine examinations. They require that physicians do extra training to become competent with them.”

At the same time, the Institute will be exploring ways to maintain a position for PET at the forefront of molecular imaging.

“With the mapping of the genome now completed, the possibilities for molecular imaging are awesome,” he says. “We want to make sure PET plays a leading role in that regard now and in the years ahead. We think it will. And certainly it will if its reception as a tool for diagnosing and staging cancer is any indication.”

Rich Smith is a contributing writer for Decisions in Axis Imaging News.