NEW YORK (GenomeWeb) – The US Centers for Medicare & Medicaid Services on Friday reined in a national coverage proposal for next-generation sequencing oncology panels that had the lab industry worried that they'd have to successfully take their tests through US Food and Drug Administration's review if they hoped to secure Medicare coverage and had cancer patients vexed that the government payor was limiting their access to precision medicine.
"The final national coverage determination has alleviated the primary concern of NGS lab-developed tests not being covered for Medicare patients," said Chirag Desai, director of health system genomics at BioReference Laboratories. He's reassured that at least for the near-term patient access to NGS cancer panels won't change much because labs can still seek local coverage through Medicare contractors if their tests don't have FDA's blessing.
CMS stated in the final policy that it will provide national coverage for FDA approved or cleared NGS companion diagnostic panels that have FDA-approved or cleared indications, when the test report specifies treatment options for a patient's cancer. The agency will cover such tests for Medicare beneficiaries who have recurrent, relapsed, refractory, metastatic, or advanced stages III or IV cancer. Patients can get coverage for testing one time using the same NGS for a primary cancer diagnosis and get repeat testing for a new primary cancer diagnosis.
Meanwhile, labs that choose not to seek FDA review for their NGS companion diagnostic cancer panels are still free to seek local coverage determinations through CMS' Medicare Administrative Contractors. Importantly, though, CMS noted that MACs may determine coverage "only when" NGS tests meet specific criteria, such as patients having advanced cancer.
Some industry observers were relieved to see that CMS had narrowed the final coverage provisions so dramatically compared to the draft version. In the draft NCD, CMS had proposed to nationally cover CDx indications on NGS cancer panels with premarket approval, and coverage with evidence development parameters for two scenarios — for FDA-approved or cleared tests intended for guiding patient care but without CDx status when patient outcomes are collected in a prospective registry, and for lab-developed tests (LDTs) without the FDA's blessing when they're part of a National Cancer Institute National Clinical Trial Network study.
Patients had to have recurrent, metastatic, or advanced stage IV cancer and not have had prior testing using the same test. When tests don't meet these criteria, or when they are performed on patients who don't meet the agency's provisions, CMS had proposed "non-coverage of NGS as a diagnostic laboratory test."
This non-coverage language spurred the lab community, hospitals, and patient advocacy groups to lobby Congress against the draft NCD. CMS received more than 300 comments on the draft NCD, the majority from academic medical centers, but also from patients, patient advocates, hospitals, labs, and industry organizations. According to the agency's review, around 41 percent did not support the draft proposal, 30 percent supported it with modifications, and 30 percent requested changes in the final version.
CMS tried to mollify stakeholders' concerns by stating that NGS cancer panels with premarket approval and 510(k) clearance as a CDx will have national coverage; by ensuring that NGS cancer panels without FDA's OK still had a path to local coverage through MACs; and by broadening its definition of advanced cancer.
"The adjustments made have resulted in a final coverage determination that is beneficial to oncology patients, their caregivers, and laboratories," said Laura Housman, CEO of Access Solutions Consulting, which specializes in helping companies commercialize companion diagnostics. The College of American Pathologists, one of the professional groups that had aired their criticisms of the draft NCD on Capitol Hill, lauded that government payor for addressing some of the concerns raised by CAP and other stakeholders. The American Clinical Laboratory Association, which also had serious concerns about the limited coverage proposed for non-FDA cleared/approved LDTs, said it is still reviewing the final NCD and gathering input from members.
Meanwhile, reimbursement experts and industry observers taking a closer look at CMS' 60-page NCD questioned the extent to which the final policy will really expand access to NGS cancer panels. Even though the agency is now willing to provide national coverage to NGS tests with FDA-approved and cleared CDx indications, the FDA has very rarely used the 510(k) clearance pathway for companion diagnostics. And since CMS has left it up to MACs to cover NGS cancer panels that haven't gone through the FDA, contractors have leeway to issue coverage criteria, including requiring FDA approval.
Others lamented CMS entirely discarding the CED aspects of the draft NCD that if crafted with more stakeholder input could have helped create a national framework for gathering evidence on whether NGS-guided precision oncology has broad utility in cancer.
"We have gone from one extreme to the other," said Bruce Quinn, a nationally recognized Medicare expert who consults diagnostics labs. "The draft NCD would have created an enormously burdensome, and not very well thought out coverage with evidence development system. However, the final NCD does nothing to help track even basic, relatively inexpensive information on what genes were found, what drug chosen, and if there was a drug response."
The NCD also raises questions about which CPT codes labs will use to bill Medicare for covered NGS tests and whether commercial payors will fall in line with CMS' policy outside of Medicare Advantage plans. "The big questions are still around pricing, coding, and payment and commercial payer response," said Charles Mathews, principal at consulting firm ClearView Healthcare Partners.
Battle for the market
CMS initiated this NCD last year after reviewing Foundation Medicine's FoundationOne CDx in parallel with the FDA. Given how narrowly CMS had proposed to cover non-FDA-approved/cleared LDTs in the draft guidance, stakeholders had written to the agency that its policy would create steep barriers for the majority of NGS panels on the market and allow Foundation to monopolize the oncology space with its test.
Market analysts continued to interpret the final NCD as being exceptionally positive for Foundation after discussions with the company. "FoundationOne CDx is the first and only NGS assay that presently meets the requirements of the policy, enabling national coverage for all solid tumors," the company said in a statement.
William Blair Analyst Amanda Murphy and other analysts claimed in their notes to investors that Foundation has verified with CMS that the final NCD grants it national coverage not only for the test's FDA-approved CDx indications, but for tumor profiling across any solid tumor. Murphy reasons this is so because FDA approved not only the test's CDx indications across five tumor types and 17 targeted therapies, but the entire 324-gene panel that also analyzes microsatellite instability and a relatively newer genomic signature, called tumor mutational burden.
Medicare experts GenomeWeb spoke to disagreed with this take, because CMS makes such a point to emphasize that for national coverage the test must have an "FDA approved or cleared indication for use in that patient’s cancer." And Foundation's MSI and tumor mutational burden signatures are not listed among companion diagnostic indications in FDA-approved labeling.
Furthermore, FDA's authorization of Memorial Sloan Kettering Cancer Center's 468-gene tumor profiling test demonstrates that the agency distinguishes biomarkers on NGS panels based on risk and indication. The FDA did not indicate the MSK-IMPACT test as a companion test, required for the safe and effective use of a drug, but as an NGS tumor profiling panel that can be used with other data to inform patient care. Alongside that authorization, the FDA issued a three-tier risk framework for biomarkers assessed by tumor profiling tests, in which the highest level of risk is for CDx-indicated markers.
Cowen analyst Doug Schenkel originally wrote in a note to investors on Friday that the final NCD doesn't give as much of a competitive advantage as the draft did. But today, after discussions with Foundation, issued another note that the final policy presents a much better scenario for the company by covering all cancers. "The lingering question appears to be whether or not CMS enforces the full reading of FDA labeling," he added.
This issue will ultimately turn on how coding and payment shake out, observed ClearView Healthcare Partners' Mathews. "On the one hand, I could see them agreeing to pay something near the [several-thousand dollar] full list price for FoundationOne CDx. Then it won't matter if the non-[FDA] approved elements are covered or not because they are being paid for," he said. "On the other hand, I could see CMS trying to delineate the specific value of the approved parts versus other content of the assay. That could get tricky."
Foundation's claim that FoundationOne CDx is "the first and only NGS assay that presently meets the requirements of the policy," is also worth questioning, since three other marketed NGS cancer panels — FoundationFocus CDxBRCA, Illumina's Praxis Extended RAS Panel, and Thermo Fisher Scientific's Oncomine Dx Target Test — have FDA-approved CDx indications in advanced cancer patients.
In a statement announcing the final NCD, CMS stated, "Tests that gain FDA approval or clearance as an in vitro companion diagnostic will automatically receive full coverage under this final NCD, provided other coverage criteria are also met."
Subsequently, Thermo issued a statement that CMS' NCD also means national coverage for its Oncomine test. The FDA last year approved Oncomine, which gauges 23 genes, but also has CDx indications for assessing EGFR mutations, ROS1 rearrangements, and BRAF mutations in advanced non-small cell lung cancer patients most likely to benefit from three treatment regimens.
CMS also clarified that the NCD isn't specific to tissue-based NGS cancer panels, which means that NGS liquid biopsy tests that have FDA-approved or -cleared CDx indications are also eligible for coverage. When liquid-based, multi-gene sequencing panel tests don't have FDA approval or coverage, it's up to MACs to decide coverage.
"The NCD does not limit coverage to how to prepare a sample for performing a diagnostic laboratory test using NGS," CMS said in response to stakeholder comments, noting that the agency included four studies on liquid biopsies as part of the evidence underlying the final policy.
Some market analysts also interpreted this language as a positive for companies like Guardant Health and Natera, which market liquid biopsy cancer tests. Guardant said in a statement that it has a pending local coverage decision in advanced lung cancer for its Guardant360 assay and has previously said it also intends to submit its test for FDA premarket approval by year end.
The MACs and other uncertainties
While industry players largely appreciated CMS clarifying that MACs could evaluate non-FDA approved/cleared LDTs for coverage in advanced patients, the language raises questions about why the agency needed to make such a detailed statement in this regard.
"One of the fundamental changes between the draft and final NCDs was elimination of the non-coverage provisions, so I think this new language is mostly a clarification," said reimbursement expert Girish Putcha, who is director of laboratory science at Medicare contractor Palmetto and chief medical officer at Freenome, a company developing a noninvasive screening technology. His comments for this article aren't on behalf of these organizations.
However, Putcha also noticed that CMS now plainly states that MACs may determine coverage "only when" the test and the patient satisfy the coverage criteria listed in that section of the NCD. "Does this mean that MACs cannot cover any LDTs or FDA-cleared or approved IVDs that are not companion diagnostics that do not satisfy all these criteria?" he wondered. "At least as intriguing and controversial is whether the approach that CMS has taken with this NCD gives MACs the leeway to ask for FDA clearance or approval as a condition of coverage."
Others in industry also recognized the possibility that MACs might take CMS' lead in requiring NGS cancer panels have FDA-cleared or -approved CDx indications for coverage.
The language in the final NCD certainly suggests that CMS is giving MACs the latitude to decide if FDA approval or coverage is important to them for all other NGS applications, observed Mathews. And when it comes to NGS companion diagnostics in advanced cancer, he expects that some MACs might tell labs to take their tests through FDA review for coverage.
"We are hopeful that the MACs will not initiate changes in coverage for NGS tumor testing without stakeholder input so as to prevent negative impacts to patient care," said BioReference Laboratories' Desai.
The final NCD also doesn't address how MACs should deal with off-label prescribing based on results from NGS tumor panels, especially now that Foundation claims it has coverage for its test across all solid tumors, not just the FDA-approved CDx indications.
"There is significant data on biomarkers with proven clinical utility that have gone through FDA [and] prospective clinical trials," Desai said, but noted that for the majority of biomarkers on large NGS panels labs haven't yet shown that a precision oncology approach improves patient outcomes through a clinical trial. Because the final NCD doesn't advance a mechanism for tracking outcomes compared to a controlled study group, one outcome will be an increase in off-label prescribing, he predicted.
"We may learn little if anything from the off-label prescribing that will likely accompany this NCD," Putcha agreed.
Foundation's report to patients will identify whether they have mutations in more than 300 genes, as well as microsatellite instability and a tumor mutational burden score, and note treatments based on patients' genomic profile that are not FDA-approved for their specific cancers. Vincent Miller, chief medical officer at Foundation, has previously said that there isn't much risk of increased off-label prescribing because such practices are unlikely to be covered under Medicare.
However, many more patients will now learn their off-label options from NGS test reports. "Nobody knows how MACs will handle all the patients with all the long lists of genes for long lists of off-label drugs," Quinn said.
At the same time, Desai felt that the coverage with evidence development requirements that CMS had proposed in the draft NCD were "unreasonable and unrealistic" and would have been impossible for US labs to implement.
Quinn had pointed out, for example, that CMS's requirements for conducting RECIST imaging to track patient outcomes would have been particularly cost prohibitive. However, he and others had also appealed for CMS to engage stakeholders to craft more reasonable CED provisions.
"We should still have a national dialog on how all the patients, genes tests, and given drugs can be tracked in some basic, not overly burdensome, directionally correct way that would be very informative," said Quinn. He had suggested that it would have been relatively simple and cheap for CMS to track Part B and Part D drugs alongside information on genetic test results and patients' cancer type.
In issuing the final NCD, CMS noted that stakeholders felt the CED mechanism was not necessary because labs are already developing or have developed the evidence to demonstrate NSG cancer tests improve health outcomes. "We strongly encourage continuing these studies and publishing the results of these studies especially on the endpoints of overall survival, progression-free survival, objective response, and patient reported outcomes relevant to the quality of life for Medicare beneficiaries," CMS said. The agency noted these data are important for patients to make informed treatment decisions and for the dissemination of new technology into the healthcare system.
Mathews observed that perhaps labs were developing this evidence because they thought they needed it for coverage. "I, for one, was hopeful that the original version of the policy would be a forcing mechanism that would make the industry align efforts to show the true value of using NGS in oncology patients," he said, adding that eventually there might be blowback "when somebody asks what the return on investment on all of this activity has been and nobody is organized enough to articulate it."
Ultimately, CMS has left it up to industry to voluntarily and transparently collect and share clinical utility and cost-effectiveness data on NGS cancer panels, which may or may not happen, in Putcha's view. "The elimination of any CED requirement suggests to me that our collective commitment to this wonderful learning healthcare system we hear so much about these days is perhaps less than advertised," he said.