Ronald Nutt and David Townsend were building a prototype positron emission tomography (PET) scanner at the University of Geneva (Switzerland) in 1993 when a physician whose name both men have since forgotten stuck his head in the room. Not really speaking to anyone, the stranger lobbed a remark toward the large, open space between two detectors in the middle of the embryonic PET machine.
"You should put a CT in there," he said.
Nutt and Townsend mulled over that casual suggestion for two years. Then, with Townsend transplanted at the University of Pittsburgh, they set to work to try to integrate the two radically different imaging technologies--PET and computed tomography--all in an effort to come up with a better one.
"The concept of having CT with PET was not the breakthrough," says Nutt, a founder, senior vice president and technology director of Knoxville, Tenn.-based CTI, a supplier of products and services for PET. "How to do it has been the most significant thing."
Little could that doctor in Geneva have known his off-hand comment would eventually meld two key fronts in the medical arms race. With HCFA on the cusp of broadening its Medicare coverage of PET technology, competition is rife among nuclear medicine companies that vie to best one another in a slow-burning but ready-to-burst-open market.
The union of CT and PET equals more than the sum of the parts, according to those who are aggressively building the new PET-CT scanners. CT is an X-ray-based method for imaging the body's anatomy; PET is a nuclear technology offering increasingly intimate glimpses of molecular function. When the two are combined, developers boast, the fusion offers more precise imaging than PET or CT alone and at twice the speed of PET.
By pinpointing the location of malignant tumors and providing clearer molecular images--all at a brisker pace--PET-CT will revolutionize both the practice and business of cancer treatment, they say--not to mention that it will weld together two distinct and sometimes at-odds specialities: nuclear medicine and radiology.
Although clinical data remain sparse, one claim is that as many as 30% of cancer patients who are scanned by these devices will see management of their cancer changed--sometimes by avoiding costly and invasive surgery. In 1999, about 351,800 patients worldwide received PET scans, and in 2000, some 453,000 are expected to receive them, according to Frost & Sullivan, a San Jose, Calif.-based market consulting firm.
The contestants. Three companies, one of which was acquired by another just last week, are unveiling versions of PET-CT scanners this week at the 86th Scientific Assembly and Annual Meeting of the Radiological Society of North America in Chicago. All told, hospitals at RSNA, on the prowl to lift their cancer care up a notch, will have four different PET-CT scanners from which to choose, each varying considerably in price and technology.
CTI PET Systems (CPS), a joint venture of Siemens Medical Systems, Iselin, N.J., and CTI, will showcase the Biograph, the end result of Nutt and Townsend's serendipitous moment in Geneva. Development of the Biograph was helped along by a $2.5 million grant from the National Cancer Institute. CPS, which commands a 68% share of the dedicated PET scanner market, according to Frost & Sullivan, expects to list the Biograph at $2.3 million. The federal Food and Drug Administration approved the Biograph earlier this month.
At RSNA, SMV (also known as Sopha Medical Vision), a nuclear medicine company with U.S. headquarters in Twinsburg, Ohio, and worldwide headquarters in Buc, France, will introduce POSiTRACE, a PET-CT scanner, which received FDA approval in September. The concept, discussed at last year's RSNA conference, represents SMV's first venture into PET imaging.
Meanwhile, in an illustration of just how fast-changing the PET industry has become, SMV was acquired by GE Medical Systems, Milwaukee, on Nov. 17 for an undisclosed price. GE officials boasted in a company press release that the combination of SMV's PET-CT product and GE's fledgling line will allow the giant to expand into even more market segments.
A somewhat lower-end model than CPS' offering, POSiTRACE will list at $1.5 million, according to Lonnie Mixon, vice president of worldwide marketing for SMV. Mixon says that SMV believes any 200-bed facility will be able to justify the price. SMV will be ready to launch a full commercial release of its PET-CT scanner at RSNA.
CPS, which now has the capacity to build three Biographs at a time, will take longer to release its product, shipping out a second-generation test version of its model to 437-bed Memorial Sloan-Kettering Cancer Center in New York in early spring. Five other large research institutions plan to get beta models before CPS expects to begin shipping in quantity by the end of 2001, Nutt says.
Meanwhile, GE Medical Systems has been furiously building its own high-end prototype in recent months, says Beth Klein, GE's vice president and global general manager for functional and molecular imaging. Klein says the project was in direct response to a market request for such a scanner. It's been kept under wraps at Rambam Medical Center in Haifa, Israel, where patient studies are under way.
GE, which already has a 20% share of the PET market by Frost & Sullivan's accounting, will be ready to roll out the machine as a work in progress at RSNA and will ship out "a few" beta systems in the first half of next year, pending FDA approval, Klein says. Expectations are that the still-unnamed scanner will be commercially available sometime next year at an average list price of $2.2 million, she adds.
Although GE lags behind CPS in developing high-end PET-CT, GE was the first to combine the two technologies. The GE Millennium VG Hawkeye was approved by the FDA in September 1999 and released commercially at the June meeting of the Society of Nuclear Medicine in St. Louis.
With a list price of about $700,000, the Hawkeye was designed to be affordable by combining a hybrid nuclear medicine camera that is not exactly a true PET with a low-grade CT scanner. Klein says more than 50 such systems will be shipped worldwide by the end of the year, and GE expects to double that shipment next year.
Quiet revolution.Even without CT's synergistic contribution, PET by itself has fostered a quiet revolution, particularly in cancer care. The expensive technology--whose price tag can easily go into seven figures--builds its information from a radioactive form of sugar, called FDG (fluoro-deoxy-d-glucose).
Glucose is a critical energy source for many cell types, particularly cancer cells, which require a lot of energy to fuel their growth. PET works off the assumption that malignancies will metabolize sugar faster than normal cells. The PET camera images this metabolism, determining where high or low glucose consumption is occurring.
As a result, PET can characterize the spread of a tumor in ways that a conventional CT anatomical scan cannot. Most important, PET zeros in on tumors more readily than CT: It could take three months or longer to detect a change in a lesion with CT, while the same change would be noted in six weeks or less with PET.
PET lights up tumors "almost like a beacon," Klein notes. Although PET still can't quite give doctors a noninvasive peek into the inner workings of single cells, "that is the direction (PET) is headed," says Homer Macapinlac, M.D., clinical director at Memorial Sloan-Kettering's PET facility.
CT's contribution to the equation is twofold. For one, it offers a key anatomical locating map, pinpointing the tumor. Second, CT plays a crucial role in filtering out extraneous radiation or "noise," called attenuation, that is one of PET's intrinsic downsides, according to Jonathan Frey, a product manager for CPS. All PET scanners have ways of correcting the attenuation, but CT can correct it in significantly less time.
Memorial Sloan-Kettering is planning on taking that time savings to the bank. The cancer hospital, which purchased its first PET scanner five years ago, has reached the maximum potential of the current device, says Patricia Soto, administrator of the Department of Radiology. In 1996, Sloan-Kettering logged 539 PET scans, which at top speed take about an hour and are scheduled during a 12-hour workday. By 1997, 971 PET scans were performed. This year the cancer hospital expects to scan 2,400 patients with the one machine, Soto says.
Adolf Pagliarulo, Sloan-Kettering's manager of nuclear medicine, is relishing the Biograph's arrival. Squeezing out every available second in the schedule, Sloan-Kettering still has a four-week backlog of patients waiting for PET scans, he says. With the Biograph, Pagliarulo expects Sloan-Kettering patients can be on and off the scanner in 30 minutes-double the capacity of the dedicated PET device.
For the time being at least, Pagliarulo says Sloan-Kettering will bill for PET-CT scans as if they were PET scans.
"My gut feeling is, in the future all high-end PET scanners will have CT attenuation correction," Pagliarulo says. "We've pushed the envelope of PET to the maximum. We believe this will be an additional tool to extend the horizon."
The melding of the two technologies won't come without bugs, Pagliarulo cautions. At the least, there will be licensing issues regarding the technical staff that administers the scans. At smaller facilities, where radiology and nuclear medicine are segregated specialties, the issues will be even stickier.
Evaluating the products. Carolyn Meltzer, M.D., medical director of the University of Pittsburgh PET facility, has been evaluating the Biograph prototype for about a year. The Pittsburgh study is focused on areas of the body that are difficult to image with CT or PET alone, such as head and neck and gynecological areas, she says. Surgery and radiation tend to distort the tissue in these spots, and there are no "anatomic landmarks," she says.
PET-CT imaging also may obviate "second look" surgeries for monitoring the success of cancer treatments, Meltzer adds.
Based on 200 cases representing a wide variety of cancers, Meltzer says, "having the information together led to more accurate diagnosis and better localization of the tumor" in several cases. PET-CT also holds promise for planning radiation therapy by helping radiologists aim their weapons more precisely, Meltzer says.
Klein of GE boasts that the unnamed high-end GE version betters the CPS scanner in its use of GE's top-of-the-line multislice CT scanner. That CT scanner is six times faster than the Biograph's, cutting into the study time even more, she claims. It also has been developed so that owners of GE's top-of-the-line dedicated PET scanner can upgrade to PET-CT for the cost of the multislice CT device, about $1.5 million, Klein adds.
SMV's version won't detect tumors quite as small as those the Biograph or the GE PET-CT can detect, but it's good enough for "real-world, clinical, day-to-day, whole-body oncology," maintains Mixon, the marketing VP. POSiTRACE may miss a 3-millimeter throat lesion--about the size of a period made by a No. 2 pencil--but SMV is "certain" it will detect 8-millimeter lesions.
Considering the nearly $1 million price differential between the POSiTRACE and higher-end PET-CT scanners, "that's a lot of money for something that tiny," Mixon says.
Similarly, GE's low-priced Hawkeye doesn't have the precision of the Biograph. Nor does it shave off the precious minutes of scanning time. In fact, the low-grade PET camera on the Hawkeye has missed some tumors picked up by a better quality, dedicated PET scanner, says R. Edward Coleman, M.D., a professor of radiology and the director of nuclear medicine at Duke University Medical Center in Durham, N.C.
Nevertheless, the Hawkeye still maintains an edge in its ability to map tumors, says Coleman, who is evaluating the Hawkeye system for GE.
Based on about 100 patients who have been scanned with PET-CT at Vanderbilt University Medical Center in Nashville, the Hawkeye has improved the accuracy of cancer staging--a system of evaluating tumor growth--about 30% of the time, says Martin Sandler, M.D., chairman of radiology and a developer of the machine. Sandler says he and his colleagues are just now exploring its possibilities in planning cancer therapy.
"In cancer (treatment) this is one of the biggest developments in many years," Sandler says. "CT changed the whole way we practice medicine; it brought anatomy to the bedside. This will bring physiology to the bedside."
Sandler gives Hawkeye-style PET-CT technology three to five years before it penetrates hospitals of all sizes.
Modest PET reception. Without the extra push of CT, PET has to date received a lukewarm welcome from the marketplace. Part of the problem has been the cost of purchasing, operating and staffing it; part has been payers' reluctance to reimburse for it, according to Frost & Sullivan. The tricky, short-lived radiopharmaceuticals on which PET relies also have daunted many hospitals.
In marketing terms, PET until now has been relegated pretty much to academic research facilities, says Scott Balsters, product manager of diagnostic imaging for Irving, Texas-based Novation, the group purchasing arm of University HealthSystem Consortium and VHA. So PET-CT has not yet appeared on most hospitals' radar screens.
"I still think regular PET will provide the majority of sales for a number of years," Balsters says. "It is going to take some time for (PET-CT) to evolve to the point where everyone is really satisfied with it and the price will start coming down."
But where money takes second place to being the first on the block to have the technology, PET-CT already reigns.
"We decided to go ahead (with PET-CT) because we believe it can help a lot of patients, and it will be a good capital investment when payers get around to paying for it," says John Gunn, executive vice president of Memorial Sloan-Kettering. He adds: "I think, quite honestly, with new technology, that process is much quicker than it used to be."
The dedicated PET market is indeed opening up. PET started coming into the mainstream in July 1999 when HCFA agreed to reimburse PET scans for five types of cancers: lung, colorectal, lymphoma, melanoma and the staging of certain lung cancers.
Now HCFA, which reimburses about $2,000 for each PET scan (depending on region), is considering broadening coverage to 19 different cancers. An executive committee took testimony Nov. 7, and a final determination by HCFA is promised by Dec. 15.
Michael Phelps, chairman of the department of molecular and medical pharmacology at the University of California at Los Angeles School of Medicine, says he and other members of the Washington-based Institute for Clinical PET presented a study of 18,000 patients at the hearing; the study demonstrated that PET was more accurate than conventional CT imaging 9% to 43% of the time, and that it changed the management of the given patients 15% to 50% of the time. The executive committee overwhelmingly agreed with the finding.
Expanded coverage of PET by HCFA will undoubtedly unleash the market, encouraging more hospitals to invest in the technology. At a time when radiology departments with aging equipment are struggling with how to best invest their capital dollars, many are already scrutinizing the pros and cons of dedicated PET scanners, says Robert Maier, president and chief executive officer of Regents Health Resources, a Brentwood, Tenn., radiology development and management firm.
"PET, I know, is on the wish list of a lot of hospitals," Maier says.
But if hospitals are considering PET, especially for cancer care, then it's a small leap to PET-CT, the developers argue.
"I'm very aggressive about what the technology will do, and I foresee down the road that PET will only come with CT," GE's Klein says.