Algorithms dominated the discussion during the Centers for Medicare and Medicaid Services’ (CMS) annual lab pricing meeting, held at the agency’s headquarters in Batltimore on July 10—or more precisely, multianalyte algorithm assays, better known as MAAAs.
Algorithmic predictions are the linchpin of highly sophisticated molecular tests that can predict a patient’s genetic propensity for a specific condition or how they might react to a dosage of medication. Oftentimes, a cocktail of tests with the faintest clinical relation to one another is required to make an MAAA-based assay function effectively.
And the laboratory sector, under pressure from both providers and payers to move from a volume to a value-based business model, wants MAAA testing to be made available to the largest swath of patients as soon as possible.
OVA1 and the MAAA Value Proposition
One of the tests discussed during the meeting, OVA1, is a bold example of the value proposition behind MAAA testing. It uses five different immunoassays to determine the malignancy of an ovarian tumor prior to surgery to remove it—dictating whether a gynecologic oncologist should perform the procedure in order to prevent a rupture and upstaging of the malignancy rather than a less skillful general surgeon.
Currently, most doctors rely on CA125, a 30-year-old blood test that correctly predicts ovarian tumor malignancy only about 50 percent of the time, according to Donald Munroe, chief scientific officer for Vermillion, the Austin, Texas-based company that developed OVA1. Although CA125 is among the assay components in OVA1, the overall test’s accuracy in predicting malignancies is 96 percent, Munroe claimed.
Just about one-third of ovarian cancer patients receive the appropriate standard of care for removing an ovarian tumor—due in part to the inaccuracy of CA125 and shortages of gynecologic surgeons, according to Munroe. However, the occurrence of a malignant mass is relatively rare—about 22,000 malignancies are detected among the 300,000 ovarian tumors removed from patients in the United States each year. Those that receive the standard of care had higher five-year survival rates.
“Our goal is a bold one—to make regrettable first surgeries by nonexperts a thing of the past,” Munroe said. “We see this as a major breakthrough for patients, providers, and payers alike.”
AMA, CMS at Odds
The American Medical Association adopted MAAA as a new Current Procedural Terminology category for 2013, and there are nine different MAAA tests included in the CPT codes. However, CMS has yet to take an official position on MAAA-based tests, primarily because the agency has not determined if the algorithmic formulas used to knit an array of assays together as in the OVA1 example provide a specific therapeutic benefit.
CMS has not yet separately priced these codes and is currently instructing providers to continue to bill for MAAA tests using existing Healthcare Common Procedure Coding System codes.
Although the agency said it requires further study before it creates a separate payment methodology for MAAA tests, it was urged by a variety of speakers to create it as soon as possible, citing the growing use of such assays.
“CMS should recognize MAAAs as a class of tests,” said Eric Zimmerman, a partner with the law firm of McDermott Will & Emery, representing the Coalition for 21st Century Medicine, a lobbying group.
“The algorithm creates the value for the test,” said Chandra M. Branham, vice president of payment and health care delivery policy with the Advanced Medical Technology Association (AdvaMed).
Zimmerman said if CMS eventually embraces MAAA testing, gap-filling would be the preferred methodology for pricing, with crosswalking used when it makes more sense. Such decisions could also be deferred to the Medicare administrative contractors, or MACs.
“How we set payment amounts is up to you,” Zimmerman said.
Currently, Novitas is the only MAC that has set a payment rate for OVA1—$516. The Department of Defense has set its own payment rate of $650. Munroe recommended a national gap-fill rate be set for OVA1. He noted that the cost of all the tests combined is slightly less than $500 but that crosswalking for the test made no sense.
“The test aligns perfectly with the intent of gap-fill—to provide actionable information for doctors,” Munroe said.
From Code Stack to Gap Fill
Ken Song, M.D., chief executive officer of Ariosa Diagnostics in San Jose, Calif., urged the use of gap-fill for his company’s Harmony test, which detects fetal chromosomal abnormalities such as Down syndrome.
Although Harmony is as diagnostically accurate as amniocentesis, since it is a noninvasive assay, it is much safer. Song noted that up to 1 percent of fetuses tested using that older method wind becoming nonviable as a result—up to 2,000 a year.
“It’s a clear value proposition,” Song said. He added that most commercial payers have issued positive coverage for Harmony, and that several thousand tests have been performed since it received approval two years ago.
Song recommended gap-fill to price Harmony, primarily to keep it affordable. Under code-stacking, the test costs about $1,500. But Harmony is priced at about $795 for commercial payers.
“When you look at this relative to all the other input out there, it is well within the range of other tests covered by CMS,” Song said.
For Other Tests, Crosswalk Preferred
Not everyone in attendance at the meeting wanted to use gap-filling, a sometimes controversial method that is currently being used by MACs to price more than 100 new Tier 1 and Tier 2 molecular pathology codes on the Clinical Laboratory Fee Schedule (CLFS).
Paul Radensky, M.D., a partner in the law firm of McDermott Will & Emery and a physician, recommended that the Galectin-3 test for congestive heart failure be reconsidered and crosswalked to the ostecalcin test, CPT code 83937, which pays $41.03. It is currently crosswalked to the code for a nonspecific immunoassay, CPT code 83520. McDermott Will & Emery is the legal counsel for BG Medicine, the Waltham, Mass.-based company that manufactures and distributes the test.
According to a survey of laboratories that perform Galectin-3, the median cost of performing the test is $43.47, while 82520 pays $17.80.
“If it’s not clinically appropriate, don’t cover it,” Radensky said, but keeping the test at too low a price will punish laboratories, keep valuable information from clinicians, and stymie future research and development for testing.
Radensky’s request was also supported by AdvaMed’s Branham, who also urged CMS officials to provide greater transparency regarding the rationale behind some of their payment decisions. The agency had been heavily lobbied by the laboratory sector regarding the crosswalking of Galectin-3 to osteocalcin.
“When CMS decisions are not closely aligned with comments they should provide a more detailed response to the rationale and how it was released,” Branham said.
The CMS will announce its pricing decisions on the 2014 CLFS in September and will gather further comments at that time.