Panel pans clinical trials justifying PET coverage

October 6, 2008

PET imaging proponents reacted with dismay in August when a Centers for Medicare and Medicaid Services advisory panel found data for nine conditionally approved cancer indications of FDG-PET were too ambivalent to support CMS coverage.

The nine-member Medicare Evidence Development and Coverage Advisory Committee (MedCAC) heard testimony from the Canadian Evidence-Based Practice Center in Edmonton, AB, and the American National Oncologic PET Registry on Aug. 20 in Woodland, MD.

The MedCAC session was convened to respond to an application in March to grant routine payment for PET to diagnose, stage, and restage soft-tissue sarcomas and brain, cervical, ovarian, pancreatic, small cell lung, and testicular cancers. The application also asked Medicare to reimburse providers for PET prescribed to evaluate suspected recurrence of these conditions.

MedCAC deliberations explored whether enough positive evidence had been gathered to improve physician decision making or patient outcomes for these applications. Members of the American College of Radiology, American Society for Therapeutic Radiology and Oncology, Academy of Molecular Imaging, and SNM, as well as patient advocates, also testified.

The panel included MedCAC chair Dr. Saty Satya-Murti, a health policy consultant; Dr. Steve E. Phurrough, director of CMS's coverage and analysis group; a consumer advocate; a hematologist/oncologist; a radiation oncologist; a prescription management company executive; an American Cancer Society official; and guest panel member Dr. Richard L. Wahl, a professor of radiology and director of nuclear medicine and PET at Johns Hopkins University Medical Center.

With the exception of Wahl, these members cast generally negative votes regarding the clinical value of FDG-PET for indications evaluated by the Canadian metastudy and the NOPR. Satya-Murti and Phurrough did not vote because of their association with CMS.

Panel members were equivocal in votes concerning whether PET data could be applied broadly to other cancers, nonresearch PET facilities, and the general Medicare population. Their highest ratings were assigned to the diagnosis, restaging, and monitoring of cancers of the kidney, ovaries, and pancreas, and staging, restaging, and monitoring of cervical cancer. FDG-PET for the diagnosis of bladder, brain, cervical, prostate, and testicular cancer generally received low ratings. Panelists raised questions about the statistical significance of the Edmonton group's data. The quality of evidence was deemed poor to moderate.

"I was a little surprised that the vote was so low," Wahl told Diagnostic Imaging. "Particularly in some of the diseases where there was concordance between the NOPR data and the Canadian healthcare analysis data."

The MedCAC panel vote reflected the fact that the published literature is small. A U.S. National Library of Medicine/PubMed search identified 9546 citations since 1984 for PET and cancer, compared with 65,914 for MRI and cancer.

"There are missing pieces of evidence," said Dr. Barry A. Siegel, director of the nuclear medicine division at Washington University's Mallinckrodt Institute of Radiology in St. Louis.

The panel's members should have reviewed all the evidence in aggregate, along with biological data, instead of performing a limited technology assessment, he said.

The inherent limitations of registries, such as NOPR, for measuring clinical efficacy have been part of the problem. NOPR was created as a compromise between PET supporters and CMS. As a reflection of CMS's Coverage with Evidence Development policy, NOPR opened the door to reimbursement for new cancer-related PET applications while generating data about their effect on clinical decision making. The registry was established in 2006. First-year results published earlier this year were positive.

But CMS may not get all the answers it needs for a coverage decision based on NOPR findings, Siegel said. Scientists have known all along that available data for rare cancers are not very robust, which led many to support the registry. There was an agreement, however, that the registry was only one part of the puzzle.

"Medicare always said that they were going to make their decision in light of all the evidence," Siegel said.

The MedCAC panel did not recommend for or against Medicare reimbursement for the procedures or formally consider the future role of the NOPR, which was established to examine the effect of FDG-PET on clinical management of the nine conditionally approved indications at the August meeting. The committee's assessment will be forwarded for further review to CMS, which will publish a proposed decision memorandum concerning reimbursement for FDG-PET indications on Jan. 10, 2009. A 30-day comment period will follow. The final coverage decision is expected in April 2009.

"CMS has to weigh national coverage needs versus their realistic budgetary constraints," Wahl said. "But if they incorporate more of the biological considerations, they may come to broader coverage decisions."

-By H.A. Abella