May 5 , 2008
By Beth W. Orenstein
Vol. 9 No. 9 P. 14
Most clinicians know that imaging results from fluorodeoxyglucose (FDG)-PET scans can change treatment decisions for more than one third of their cancer patients, regardless of the type of cancer they have. Now, data collected by the National Oncologic PET Registry (NOPR) and published in the Journal of Clinical Oncology in late March confirm it.
An analysis of data collected on nearly 23,000 patients at more than 1,200 PET facilities nationwide from May 2006 to May 2007 found that FDG-PET prompted changes in treatment decisions in 36.5% of oncology cases, says Barry A. Siegel, MD, a cochair of the NOPR Working Group and chief of the division of nuclear medicine at the Mallinckrodt Institute of Radiology in St. Louis.
Currently, the Centers for Medicare & Medicaid Services (CMS) covers payments for FDG-PET scans for only nine cancer types where data has proven its effectiveness in treatment planning and management. The NOPR was developed to assess how FDG-PET affects care decisions and was open to patients with any of the more than 200 cancer types.
The initial data analysis showed that FDG-PET is effective in diagnosing and evaluating disease and capable of dramatically changing the course of patient care with cancers other than those on the CMS “PET-approved” list.
Based on the results, Siegel says, “Medicare should strongly consider opening up its coverage to include diagnosis, staging, and restaging for all cancers.”
Confirming Earlier Work
Peter S. Conti, MD, PhD, a professor of radiology, pharmacy, and biomedical engineering at the University of Southern California in Los Angeles and former president of the Society of Nuclear Medicine (SNM), agrees: “The study confirms what we have known since the mid-1990s—that FDG-PET is useful for imaging a wide variety of cancer patients and that it alters patient management. It affects patients’ lives and, therefore, it is appropriate to expand approval for reimbursement.”
It is impossible to predict what the CMS will decide, Siegel says, but the day after the data was made public, representatives from the NOPR sent a letter to the CMS formally asking it to expand FDG-PET coverage for the diagnosis, staging, and restaging of all cancers that are not currently covered by Medicare.
The CMS said it will review all the data and comments collected during a 30-day period that ends May 10 and will issue a decision. The CMS announced its intent to support a PET registry in January 2005. The registry was launched in compliance with the CMS’ Coverage with Evidence Development policy, which is to collect data through a clinical registry to help it make coverage determination decisions.
Siegel says it will take months for the CMS to reach a conclusion. “But at least the official request has found its way to Baltimore,” he says. Meanwhile, the registry will remain open, and additional analysis of the data will be conducted.
The initial analysis does not break down the results by cancer types, and that is something the researchers plan to do, Siegel says. They also plan to look at the impact of FDG-PET on treatment monitoring for all cancer types and publish those results. Siegel says these data are still “maturing.”
PET with FDG is used in oncology because cancer cells metabolize glucose more rapidly than normal tissues. Between 1998 and 2005, the CMS agreed to reimburse for PET use with nine cancer types. Included on the CMS’ approved list are cancers of the head and neck, esophagus, lung, thyroid, breast, cervix, and colon, as well as malignant melanoma and lymphoma. “Medicare was paying for PET for the nine cancer types and for a mixture of indications—diagnosis, staging, restaging for recurrent tumors, and treatment monitoring. “However, it wasn’t uniform across the board,” Siegel notes. For example, he says, until the registry, breast cancer was the only cancer that Medicare reimbursed for PET for treatment monitoring. (A complete list is available at www.cancerpetregistry.org.)
Medicare agreed to reimburse for PET used for all types of cancers and indications provided by the clinicians who participated in the registry.
Registry Data Published
Sponsored by the Academy of Molecular Imaging in Los Angeles and managed by the American College of Radiology (ACR) and the ACR Imaging Network in Philadelphia, the registry opened May 8, 2006.
To participate, clinicians were asked to fill out prescan and postscan forms. The prescan form asked the indication for the study, the cancer type if known and the suspected stage, the patient’s performance status, whether the referring physician would be the treating physician, and the planned treatment if PET was not available. Once the scan was completed, the PET facility had to upload the report to the database. The referring physicians also had to complete a form tailored to the study they had done. All the questions on the form were designed to assess whether the planned management was altered in light of the PET results.
The data were analyzed to see whether the PET scans changed treatment decisions—from treatment to nontreatment or vice versa, and from curative to palliative or vice versa. Treatment included surgery, chemotherapy or radiation alone or in combination, and nontreatment, including watching, noninvasive imaging, biopsy, or supportive care.
The researchers characterized the changes as major or minor. A switch in treatment type, such as from surgery to chemotherapy, was considered major even if the goal of the treatment remained the same. Adding or deleting treatments was considered minor as long as one type of treatment did not change after the scan. Finally, the researchers scored the treatment prescan and postscan plans as to whether they increased, decreased, or remained unchanged.
“We did some filtering, and we ended up with almost 23,000 patients in the study,” Siegel says. “We included all indications for PET except for treatment monitoring. We decided to analyze those data separately, and we’re close to finishing with that and expect to submit it to another journal fairly soon.”
Any change in treatment plans, whether more or less, was counted and included in the results. “We assumed that FDG-PET would only change management of the disease 15% to 20% of the time because the registry included uncommon cancers,” Siegel says. “Many of them weren’t in the literature. Many of them are not quite as FDG avid as cancers for which PET is already approved. So we said, ‘Well, maybe PET won’t be as effective as it has previously shown to be in lung and colorectal and lymphoma.’ But it turned out that it is.”
Virtually every other study that has reported on a percent change in management measured in a similar fashion has found that the frequency of change is somewhere between 25% and 40%, “and so our number is right on target,” Siegel says.
Nearly 80% of the PET centers in the United States participated in the study. “We had almost 1,600 PET facilities signed up in the registry and for this first paper, we had close to 1,200 of them submit at least one patient,” Siegel says. “So we basically looked at things way down in the trenches.”
In an accompanying editorial in the Journal of Clinical Oncology, Steven M. Larson, MD, chief of nuclear medicine service at Memorial Sloan-Kettering Cancer Center in New York City, says that because of the participation, the data can be considered representative. “The findings are relevant,” he writes, “and not biased in the way a single-institution study might be because of possible patient sampling biases, equipment, or experience of the PET readers.”
Conti says that unlike some newer technologies, PET has never been unduly overrated and has proven over time that it can play a significant role in cancer diagnosis and treatment. “It’s interesting because often when a new technology is introduced, early literature results might suggest accuracies approaching 100%. And then it kind of drifts down with experience over time and it settles at some level of impact. However, I think the initial PET data back in the early to mid-1990s was reasonably tempered, and now, if you go back historically, it’s very reflective of what we’re seeing almost two decades later, now in a wider variety of cancers,” he explains.
Siegel says the registry will continue until the CMS makes a decision, with the outcomes being fairly broad. The CMS could decide one of at least three ways: “It could say, ‘OK, we’ve made our decision and the registry is done. We’ll cover the other cancers.’ Or it could say, ‘The registry is done, and we’re not going to cover any other cancers or indications.’ Or it could say, ‘The registry continues, and we’re going to cover some cancers for some indications but not the rest, and you can continue to collect data for the rest.’”
Siegel says clinicians are hoping that the CMS will say the data are convincing and “we’re going to cover all cancers for at least these indications.”
Robert W. Atcher, PhD, president-elect of the SNM and program manager for the department of health and human services at the Los Alamos National Laboratory in New Mexico, says the timing of the PET registry results is ideal. “We are on the brink of starting what everybody refers to as personalized medicine, which is really individualized medicine. At this point in time, we can’t do the genomics of every single patient. It’s too expensive and too time consuming for what the yield would be,” he says. “But with PET, we can certainly identify certain probes that are going to give us information about metabolism or other function of the tumor cells that we’re not going to get from other imaging modalities. The information from PET will allow us to refine the care that we give allowing us to use a phrase we hear often these days that I like: ‘We can use the right drug in the right patient at the right time.’”
Some drugs are being developed that may only help a fraction of patients with a given cancer, Atcher says. Insurers could say that the population isn’t large enough for them to approve the drug for reimbursement. “But if we can do an imaging study for $1,000 or $1,500 and determine that this patient would benefit because they have the proper receptor or there is something else about their current status that indicates this would be a valuable intervention, then that’s going to really revolutionize the care of those cancer patients.”
PET scanning could save costs, Atcher says. Today, he explains, the physician prescribes a treatment and then sometimes has to wait months to determine whether it was effective. “With PET, we can make that determination a lot quicker and continue treatment if it is proving to be effective or change to a more effective treatment before spending too much time and money on a treatment that is not working.”
— Beth W. Orenstein is a freelance medical writer and frequent contributor to Radiology Today. She writes from her home in Northampton, Pa.