Diagnostic Imaging
May 2003

NUCLEAR MEDICINE

PET reimbursement awaits pivotal year in Washington

At least a dozen rare cancer indications are being considered for FDG-PET coverage under federal guidelines

Advocates expect that this will be one of the most significant years for PET reimbursement policies since federal officials first approved FDG-PET coverage in 1998. Now that many of the most common cancers, including colorectal, lung, and lymphoma, are approved for Medicare coverage, the PET community must persuade the Centers for Medicare and Medicaid Services to adopt a different evaluation process for low-incidence cancers such as testicular and thyroid.

At least a dozen rare cancer indications are being considered for FDG-PET coverage this year. The difficulty arises because there are no large patient population studies of these cancers to fill the pages of peer-reviewed journals.

Several years ago, the CMS, then known as the Health Care Financing Administration, adopted a policy of evidence-based approval. The agency looks to the published literature and other expert sources to determine whether a modality improves patient management and is not merely a duplicate test. For FDG-PET, the government wants evidence that it has some stand-alone value; for example, that it could substitute for CT and/or MR follow-up or is as accurate as another modality for a particular indication but less expensive.

Proponents strived for many years to prove PET's efficacy, particularly in oncology. Medicare coverage came slowly and piecemeal, starting with a nod in 1998 for several lung cancer indications. The following year, officials granted coverage for specific indications of colorectal cancer, Hodgkin's and non-Hodgkin's lymphoma, and melanoma. Other oncologic indications were subsequently approved, as were myocardial viability following an inconclusive SPECT exam and refractory seizures (presurgical evaluation only).

Then in late 2000, federal officials recognized FDG-PET as a biologic modality and expanded coverage for six cancers to include diagnosis, staging, and restaging. PET practitioners saw this as a breakthrough and hoped that blanket coverage for all oncologic indications would be approved.

"Cancer is a disease of cells gone awry. There is a commonality to all cancers, and that's why FDG works," said Dr. R. Edward Coleman, director of nuclear medicine and vice chair of radiology at Duke University. "It doesn't seem right that we have to submit a request to CMS for every cancer."

The CMS, however, has continued to review FDG-PET on an indication-by-indication basis, frustrating those anxious to obtain coverage for low-incidence cancers. In late February, the CMS decided to group together five such cancers-brain, testicular, small-cell lung, cervical, and pancreatic-for evaluation, even though they were submitted separately. Initially, the PET community saw this as a sign that the evidence-based bar might be lowered for low-incidence cancers. But Coleman isn't so sure. The CMS has forwarded the request for the five rare cancers to the Agency for Healthcare Research and Quality, seeking more supportive evidence. This is the same path the government followed for the higher incidence cancers.

"We had hoped the government had determined a new mechanism for evaluating low-incidence cancers. But it's not clear it will evaluate these any differently than the higher incidence cancers," Coleman said.

No other imaging modality has to go through this tedious indication-by-indication process, said Ruth Tesar, executive director of the Northern California PET Imaging Center.

"This process is frustrating for referring physicians, who need to remember this detail every time they think about ordering a PET scan for their Medicare patients. It is generally not as difficult for the private patient, so this places the referring physician in a difficult place, knowing there are potentially two standards of care," she said.

IN THE CMS PIPELINE

For years, the CMS has been considering reimbursement for several rare cancers, including thyroid (submitted June 2001) and soft-tissue sarcoma (October 2001). Myocardial perfusion using the tracer nitrogen-13 ammonia was submitted in May 2002. The CMS referred these three requests to its Medicare Coverage Advisory Committee, asking that it recommend a review process for rare conditions. An answer for all three requests was expected March 1 but had not been issued as of late March.

Six additional low-incidence cancer requests were submitted to the CMS during February and March: ovarian, gastrointestinal stromal tumors, multiple myeloma, pediatric tumors, advanced prostate cancer, and fluorine-18 bone scans. A response to these requests could come as early as September 2003 or as late as February 2005. Coleman has submitted a coverage request for Merkel cell tumor, an extremely rare cancer that behaves like malignant melanoma. He is following several patients with these tumors at Duke.

Last year, the CMS dealt a blow to PET imaging when it rejected coverage for Alzheimer's disease. A new request was submitted in February, a sort of repackaging of the evidence for clarity, with indications limited to a more specific patient population, Coleman said.

"It's reasonable to require good evidence for the value of any procedure," said Dr. Ken McKusick, chair of the nuclear medicine ambulatory payment classification (APC) task force. "It's our responsibility to do the research and publish papers that demonstrate the modality's validity."

PET PAYMENT MAZE

McKusick, a retired nuclear medicine physician who practiced at Massachusetts General Hospital, concentrates on coding issues. Although CPT 78810 applies to all tumor imaging with PET, the CMS continues to assign G-codes for specific indications. G-codes are temporary designations that the agency uses until it collects sufficient cost data about a new technology. The CMS has tried for years to take PET out of the new-technology category, a move that would reduce reimbursement. But PET proponents have argued that the agency's cost data are flawed and have so far staved off such a move. Many PET practitioners say it is inevitable-and desirable-to assign APC codes to FDG-PET procedures. The challenge is to ensure that Medicare payment levels accurately reflect the actual cost of performing these studies, including the cost of the FDG.

The CMS employs a complicated formula to arrive at the Hospital Outpatient Prospective Payment System (HOPPS) technical component payment. This year, the technical payment level for whole-body FDG-PET procedures is $1767.64 ($1375 for PET, $392.64 for FDG). Many PET advocates believe that the CMS arrived at this payment level by crunching bad numbers. When it began collecting PET cost data a few years ago, for example, a number of academic centers had their own cyclotrons and their scanners were fully depreciated. Community-based hospitals that installed PET within the last several years have dramatically different expenses that may not be sufficiently represented in the CMS formula. Furthermore, PET proponents contend that early hospital cost reports contained many errors because the technology was so new.

Since the CMS sets the hospital outpatient technical component payment level, providers can do little but gather accurate cost data and petition for change. Nonhospital technical payments under the physician fee schedule are another matter. Once based on relative value units, they are now determined by the Medicare carriers and could change every month. Current payment levels across the country vary widely, ranging from $1600 to $2700.

The more savvy private payers are clearer about what they'll pay for certain PET procedures. They will use historical payment data as a reference or will establish a baseline technical component payment. Providers must be savvy as well; they should be willing to negotiate a reasonable reimbursement rate with each private payer. It is also critical to negotiate the clinical coverage criteria in addition to the payment level.

The CMS publishes its proposed rule changes for the payment schedule every summer. For the past two years, it has proposed moving PET from the new-technology category, which could cost providers nearly $300 in technical component reimbursement. The PET community has responded to the CMS during the 60-day comment period in support of leaving the PET technical component at the current level. One compelling argument to support reversal of the proposed rule change has been the lack of accurate PET claims data. It's anybody's guess what CMS intends to do this year and whether the PET will remain in the new-technology category.

Accumulation of a substantial amount of accurate charge and cost data from hospitals throughout the country may be critical to maintain reasonable HOPPS technical component reimbursement for PET in 2004. The local Medicare carrier payment levels must also be addressed for 2004. PET supporters must establish a national payment database as a reference guide for future payment policies.

PET and PET/CT procedures are becoming standard practice tools for many physicians. It is important for all PET providers to understand the processes for expanding coverage guidelines and establishing technical component payments for hospital and nonhospital providers.

Ms. Halliday is vice president of ImageMed Group, a company involved with management, consulting, and ownership in medical imaging. She can be reached at sue.halliday@imagemedgroup.com.

---

Red tape blocks path to payment

PET breast coverage decision sat for two years in pipeline

Generally, it takes 18 to 24 months for the Centers for Medicare and Medicaid Services to grant Medicare coverage for a submitted indication. During that time, PET proponents interact with the CMS to ensure that it has and understands the most current data. Following is the history of a request that took more than two years to complete the approval process.

June 2000: PET community requests broad coverage for all oncological indications, heart disease, and neurological disorders.

Dec. 15, 2000: CMS grants broader coverage for certain indications but rejects it for breast cancer. It will reevaluate breast cancer request and announce decision Feb. 28, 2001.

March 7, 2001: Breast cancer request referred for technology assessment. CMS expects to receive a report by the end of May and to present the issue to Medicare Coverage Advisory Committee (MCAC) diagnostic imaging panel on June 19, 2001.

June 19, 2001: MCAC's recommendation referred to MCAC executive committee for review.

Dec. 17, 2001: MCAC executive committee minutes received Nov. 30. New due date: Jan. 20, 2002.

Feb. 5, 2002: Decision memorandum is complete. Awaiting final clearance. New due date: Feb. 28, 2002.

Feb. 27, 2002: CMS grants Medicare coverage for certain breast cancer indications.

Oct. 1, 2002: Actual coverage becomes effective, almost two and a half years after initial request.

---

Rare cancers need new process

Dearth of studies challenges reimbursement criteria

The government bases its approval of Medicare coverage primarily on peer-reviewed studies that demonstrate a modality's cost-effectiveness and positive impact on patient management. Unfortunately for the PET community, many rare cancers do not have the volume of patient studies needed to satisfy the evidence-based criteria required by the Centers for Medicare and Medicaid Services. This year, the CMS will consider FDG-PET coverage for at least a dozen cancers of low incidence, and no one is quite sure how it will evaluate them. Following is a list of six of the cancers and the American Cancer Society's estimates for the number of cases in the U.S. in 2002:

Testicular-7500

Soft-tissue sarcoma-8300

Cervical-13,000

Brain-17,000

Ovarian-23,300

Thyroid-20,700