NEW YORK (GenomeWeb) – In a draft local coverage determination issued last month, Palmetto's MolDx program proposed coverage for comprehensive genomic profiling (CGP) of advanced non-small cell lung cancer patients whose tumors have turned up negative for well-known EGFR mutations or EML4-ALK translocations.
According to the draft LCD, Medicare will cover genomic profiling for metastatic NSCLC patients who are non-smokers or fomer light smokers, defined as people who have a 15 or fewer pack-year history (see clarification note at the end of the article). Additionally, patients must be negative for EGFR or EML4-ALK translocations, based on a US Food and Drug Administration-approved companion diagnostic or a lab test. Medicare will not pay for CGP testing to gauge only germline mutations, which are inherited and occur in every cell in the body. "Alterations detected by CGP, if positive, may allow individuals to be treated with a targeted therapy for which they were previously ineligible," Palmetto said in the draft proposal.
Palmetto defined CGPs as "a single test that does not distinguish between somatic and germline alterations," and can detect base pair substitutions (ie. SNPs), insertions and deletions (up to 30-40 base pairs), copy number variations (ploidy < 4 with copy number ≥ 8), and translocations.
Next-generation sequencing tests can gauge four classes of genomic alterations, but the Medicare contractor said it would cover non-NGS-based tests as long as they could also assess these markers and perform on par with CGPs. Studies have shown CGPs to have an analytical sensitivity of between 95 percent and 99 percent for the four classes of alterations, a positive predictive value of greater than 99 percent, and inter-batch precision of 96.4 percent.
The draft LCD, which stakeholders can comment on between Feb. 10 and March 27, is being upheld by many in the industry as an innovative coverage policy that takes into account some of the challenges healthcare providers face in delivering molecularly guided personalized care to cancer patients. Still, the draft policy wouldn't cover testing for all lung cancer patients who might benefit from molecular characterization of their tumors
In lung cancer, obtaining sufficient tissue samples through invasive biopsies is difficult and often limits the molecular analysis that can be run. After a pathologist has analyzed a patient's biopsy sample to determine the kind of lung cancer he or she has (ie. NSCLC or small cell lung cancer), there is very little or no tissue left for further genomic testing. As such, profiling the tumor at one time for multiple molecular characteristics would be the best use of limited resources.
At the Personalized Medicine World Conference last week, Laura Housman, chief commercial officer of MolecularHealth, noted that there are a number of labs currently marketing NGS tests that can pick up the range of markers specified in the draft LCD. Furthermore, "we are beginning to see payor engagement around tumor panel coverage and payment," she said, highlighting Palmetto's recent draft LCD on CGPs for advanced NSCLC.
MolecularHealth will likely benefit from expanded payor coverage of CGPs, since last year the firm began offering testing on TreatmentMAP – an NGS-based diagnostic that gauges 500 cancer-linked genes and can help guide treatment strategies. Similarly, Foundation Medicine, which markets an NGS test called FoundationOne that analyzes around 200 cancer-linked genes, lauded the draft LCD.
"We believe this first step [by Palmetto's MolDx] is an important one towards broader coverage of expanded indications as evidence continues to evolve," the company said in a statement. "Although the insurance coverage process in the US for new molecular diagnostic approaches is not straightforward or easy, we are very excited about this important development in setting the stage for increased access to CGPs and supporting our efforts to obtain broad reimbursement for FoundationOne."
While Palmetto is not the Medicare contractor that determines coverage for Foundation Medicine's test, determinations by the MolDx program are often followed and emulated by other Medicare contractors around the country. Foundation said it believes that the FoundationOne test meets the conditions for coverage under Palmetto's policy and that its contractor, National Government Services, will likely adopt the same criteria.
MolecularHealth's Housman agreed that while the draft LCD is a "great step forward" from having no coverage of CGPs at all, the terms of coverage don't include a portion of the NSCLC population, and under certain clinical scenarios the policy may not align with how medicine is currently being practiced.
She noted that there are 222,000 lung cancer patients in the US, 90 percent of whom have NSCLC. In April 2013, the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology released guidelines recommending that all patients with advance lung adenocarcinomas should be tested for alterations in EGFR and ALK genes. The same bodies announced last month that they are considering updating these guidelines to consider whether the latest clinical evidence supports recommending testing for additional lung cancer mutations in ROS1, MET, ERBB2, RET, and NTRK1, as well as for resistance markers.
Within the overall NSCLC population, 15 percent and 4 percent of patients have tumors characterized by EGFR and ALK, respectively. However, under the terms of the draft LCD, only patients outside this group would be eligible for CGP testing as a covered benefit. Despite the latest guidelines recommending testing all lung adenocarcinoma patients for EGFR and ALK markers, currently only 50 percent of the NSCLC population is receiving such testing, Housman said. Then, further subtracting out the smoker population, approximately 26,000 never-smokers or light smokers would be eligible for coverage.
In the draft LCD, Palmetto cited sequencing studies that have shown that never-smokers or light smokers with lung adenocarcinomas have a higher frequency of EGFR mutations than smokers, and that mutations show up in particularly high rates among non-smoking Asian women. The rate for ALK fusions is similarly more frequent in these subpopulations.
Within this light or never-smokers' group, CGP would only be covered for patients with metastatic disease. "So, at the end of the day … under 24 percent of diagnosed NSCLC patients in an eligible population, negative for EGFR and/or EML4-ALK biomarker testing, will be able to have a CGP test that will be covered and paid for by MolDx," Housman said. "There is no free lunch in the world, in life, and in CMS."
Furthermore, Palmetto's draft guidelines may not align with FDA-approved labeling for targeted lung cancer drugs, she pointed out during her talk. For example, in the case of the EGFR inhibitor Tarceva (erlotinib), NSCLC patients could be prescribed the drug regardless of their smoking history. However, the label does tell doctors that smokers may clear the drug faster than non-smokers, and so smokers may require dose adjustments.
"The erlotinib drug label recognizes the different levels of drug absorption between smokers and never smokers," Housman observed. "So, there is proof [of drug efficacy] for all NSCLC patients."
The draft LCD asks labs to verify that patients have previously tested negative for EGFR and ALK translocations before testing them on CGP. However, "it's about a 50/50 shot that you get the prior report in," Housman said at the meeting. "And if you do, then what's the chance of that being accurate," since results between labs can vary. Citing these examples, Housman said that payment and regulatory guidelines for genomic tests need to address how treatment is given in the real world.
Labs providing CGP under the policy also have a number of items that they will have to report to CMS every six months, such as the number of patients they've tested, specifying which patients had or didn't have EGFR or ALK alterations by NGS-based testing, as well as what other mutations were found. When labs identify patients with EGFR or ALK markers by CGP, they will have to report to CMS the number of these patients, their response to treatments, and how sustained their responses were.
For situations where a CGP platform identifies a patient to have an ALK or EGFR alteration, labs must inform CMS of the specific non-CGP test that previously identified these patients as marker-negative. Conceivably, this will enable CMS to track the performance of other diagnostics. In these situations, labs must report to CMS whether the mutations picked up by CGP occurred outside the genes gauged by the non-NGS lab test or companion diagnostic, and whether the mutations are insertions, deletions, duplications, or translocations.
In addition to a variety of lab-developed tests, there are three FDA approved companion diagnostics that gauge EGFR mutations and ALK translocations – Roche's Cobas EGFR Mutation Test for predicting best responders to Tarceva; Qiagen's Therascreen EGFR RGQ PCR Kit for predicting response to the EGFR inhibitor Gilotrif (afatinib); and Abbott Molecular's Vysis ALK Break Apart FISH Probe Kit for predicting response to Xalkori (crizotinib).
In the draft LCD, Palmetto noted that although there aren't many studies comparing CGP and non-CGP diagnostic approaches head-to-head for gauging these mutations, "CDx or LDT sequencing techniques may miss deleterious EGFR mutations and ALK translocations." The genomic alterations might show up outside the region being sequenced or be the types of alterations that aren't picked up by the test, such as insertions or deletions, copy number alterations, or translocations.
By comparison, NGS can "offer the possibility of not only increased analytical sensitivity but also the ability to detect a broader range of genomic alterations than existing CDx and LDT techniques," Palmetto noted. In the LCD, Palmetto cited one study involving non-smokers or light smokers who were tested previously by a non-NGS test for specific genes, and tested them by NGS-based cancer genomic profiling. In that trial, CGP identified 7 percent more patients with EGFR mutations compared to previous non-NGS tests, and 6 percent more ALK translocations compared to analysis by florescence in situ hybridization testing.
"Although some of the EGFR-mutated malignancies found by NGS are less likely to respond to available EGFR tyrosine kinase inhibitors (e.g. exon 20 insertions), others, such as complex double mutations and exon 18 mutations (which are typically undetectable with so-called "hotspot" panels), are likely to benefit from targeted therapy," Palmetto noted. Similarly, the two patients in the study who were negative for ALK translocations by FISH analysis and identified by CGP, "would likely benefit from treatment with crizotinib," Palmetto said.
This article has been updated to clarify that draft coverage is for non-smokers or former light smokers with a 15 or fewer pack-year history. This breaks down to mean that a former light smoker is a person who has smoked one pack of cigarettes per day for 15 years; a half-pack per day for 30 years; or two packs per day for 7.5 years.