Response to Comments: MolDX: Next-Generation Sequencing for Solid Tumors

Thank you for the opportunity to comment on Palmetto’s proposed coverage policy for Next Generation Sequencing (NGS) for Solid Tumors (DL38045).

The Association for Molecular Pathology (AMP) is an international medical and professional association representing approximately 2,300 physicians, doctoral scientists, and medical technologists who perform, or are involved with, laboratory testing based on knowledge derived from molecular biology, genetics, and genomics. Membership includes professionals from the government, academic medicine, private and hospital-based clinical laboratories, and the in vitro diagnostics industry.

As the world’s largest organization of board-certified pathologists and the leading provider of laboratory accreditation and proficiency testing programs, the College of American Pathologists (CAP) serves patients, pathologists, and the public by fostering and advocating for excellence in the practice of pathology and laboratory medicine worldwide. 

We are submitting a joint comment letter because both our organizations are fully aligned in our views regarding this draft Local Coverage Determination (dLCD). We appreciate Palmetto’s willingness to provide limited coverage for NGS-based testing of solid tumors in cancer patients as part of emerging efforts to both improve detection of cancer and help guide treatment decisions and we respectfully ask that you consider our following recommendations.

1. Criteria for Coverage

A. dLCD statement: The Criteria for Coverage section states that all the following must be present for coverage eligibility: 

• As per NCD 90.2, this test is reasonable and necessary when:
  • the patient has either:
    • Recurrent cancer
    • Relapsed cancer
    • Refractory cancer
    • Metastatic cancer
    • Advanced cancer (stages III or IV)
  • AND has not been previously tested by the same test with the same primary diagnosis
  • AND is seeking further treatment
• The test has satisfactorily completed a TA by MolDX for the stated indications of the test
• The assay performed includes at least the minimum genes and genomic positions required for the identification of all FDA-approved therapies with a companion diagnostic biomarker for its intended use that can be reasonably detected by the test. Because these genes and variants will change as the literature and drug indications evolve, they are listed separately in an associated Coverage Article, as well as in the MolDX TA forms. 

Comment: There are multiple clinical scenarios whereby repeat testing of the same cancer is necessary, typically as the cancer evolves to evade front-line targeted therapy. Specifically, relapsed, recurrent and metastatic cancers under pressure by treatment often show losses and gains in mutations compared to the primary tumor, rendering them, in effect, new cancers.

In its Decision Memo for Next Generation Sequencing (NGS) for Medicare beneficiaries with advanced cancer (CAG-00450N), dated March 16, 2018, CMS states that repeat testing is allowed under certain circumstances. Any lab diagnostic tests using NGS that are FDA approved/cleared as a companion diagnostic are nationally covered (i.e., no contractor discretion) under this NCD, and coverage determinations for the rest of the diagnostic lab tests using NGS will be made by Medicare Administrative Contractors. If the patient has not been diagnosed with a new cancer, diagnostic lab testing using NGS is coverable but only if when a different diagnostic lab test is furnished from what was furnished previously (emphasis added).”

Recommendation: Palmetto should consider coverage for repeat NGS testing when the subsequent post- treatment cancer is recurrent, relapsed, treatment-refractory or metastatic.

B. dLCD statement: The second sub-bullet under “reasonable and necessary” Criteria for Coverage states: “AND has not been previously tested by the same test with the same primary diagnosis”.

Comment: In the Criteria for Coverage section under “Situations in which a test should not be used or when coverage is denied”, the proposed policy states that “the test in question will not be covered if another CGP test was performed on the same tumor specimen (specimen obtained on the same date of service).

Recommendation: To provide for clarity and uniformity, we recommend that the second sub-bullet under “reasonable and necessary” criteria for coverage be amended as follows:

“and has not been previously tested on the same tumor specimen with the same primary diagnosis on the same date of service”.

Alternatively, an additional bullet could be added to “Criteria for Coverage” that states:

“Repeat NGS testing on the same patient for the same primary diagnosis may be reasonable and necessary when performed on a different date of service.”

C. dLCD statement: The third bullet point under Criteria for Coverage states, “The assay performed includes at least the minimum genes and genomic positions required for the identification of all (emphasis added) FDA- approved therapies with a companion diagnostic biomarker for its intended use that can be reasonably detected by the test. Because these genes and variants will change as the literature and drug indications evolve, they are listed separately in an associated Coverage Article, as well as in the MolDX TA forms.”

Comment: The companion diagnostic space is rapidly evolving and under Palmetto’s proposed policy CLIA labs would be required to update their assay frequently (re-validate) or adopt a broader “future-proof” assay.

Clinical laboratories often want to take a modular approach to assay adoption. Having multiple assays that address companion diagnostic biomarkers allow labs to “pick-and-choose” based on relevant clinical indications and/or serialize testing based on prevalence, with the goal to maximize tissue availability and reduce testing cost per patient.

Recommendation: We recommend that Palmetto substitute the word “all” with “clinically relevant”.

3. Technical Assessment (TA) Checklist (M00151, V5)

TA statement: The first two questions under the “Test Details Checklist/Questionnaire” section of the Technical Assessment (TA) document ask: 

1. Does this test result in a report/information that is limited to providing patient genetic/genomic information and ancillary data that are not proprietary, utilizing methodologies for which Clinical Validity (CV) and Clinical Utility (CU) are well established in the literature? 
2. Is this a test based on novel/proprietary technology or algorithms, and/or provides a result based on such technology or algorithms? If yes, Clinical Validity and Clinical Utility must be described.

Comment: Questions #1 and #2 appear to be an attempt to clearly distinguish NGS-based tests that have proven clinical utility (CU) and clinical validity (CV) from those that do not. Most lab-developed NGS-based procedures, to the contrary, have some components of CV/CU that arise from literature-supported evidence and other components of CV/CU that are supported by unpublished novel or proprietary lab-specific algorithms or data. Most lab-developed NGS-based procedures therefore fall somewhere in the middle of the two options offered in the TA.

Additionally, while Palmetto has taken efforts to streamline its technical assessment checklist the document still appears to be overly burdensome for test applicants and lacking in transparency. For example, the checkboxes do not include information about the way in which the yes/no responses could impact and inform Palmetto’s ultimate coverage decision.

Recommendation: We recommend that Palmetto increase transparency regarding its evaluation and response to the questions and other information required under the TA document that is used by MolDX to evaluate test coverage. Specifically, we recommend that the technical assessment document further clarify: 

• the ramifications (to coverage) of choosing a “yes” versus “no” response to each question;
• there is seemingly no “middle ground” between “proprietary” and “non-proprietary” and whether clinical utility/clinical validity have already been established in the literature. The answer is almost always somewhere in between.

ICD-10 Codes

The ICD-10 diagnosis codes may not always be granular enough to accurately distinguish relapsed, refractory, and/or recurrent cancers as compared to the initial pre-treatment diagnosis code. We recommend the addition of all diagnosis codes that distinguish these cancers for solid tumors.

We would sincerely like to thank both CAP and AMP for their engagement on these issues and for their thoughtful considerations of this policy. We will address the issues by topic and specific points as necessary below.

1. Criteria for coverage for “repeat testing”. We agree that the current NCD language allows for testing of a tumor more than once provided that the same test is not performed. The Criteria for Coverage section includes language from the NCD (referred as NCD 90.2) that explicitly states that a beneficiary cannot be tested for the same primary indication using “the same test” that was previously performed. We believe this language is consistent with the proposed use case presented. We furthermore do not wish to misrepresent the language of the NCD, nor can the LCD supersede the NCD.

We concur that there are instances where repeat testing may be warranted. To fulfill the criteria that a test be both reasonable and necessary, we seek in this policy to further restrict coverage to prevent unnecessary testing of the same sample wherein no additional value is likely to be derived from testing, even if it is by a different test (specified here as a CGP). To this end, we included language restricting testing only in repeat testing in specimens “on the same date of service”. We believe the existing language is already compatible with the comments presented here and thus do not see a reason to alter the language further on these points. We do not believe these issues to be contradictory- the first statement is the condition for MAC discretion noted by the NCD, the second is the condition for coverage created in this policy.

2. Requirement for coverage of essential biomarkers. Thank you for the feedback. We understand that the clinical science is rapidly evolving and that the word “all” in the policy may be problematic as it conveys the idea that any test must include relevant clinical biomarkers for all indications. The intent here was to ensure that a test have all necessary biomarkers for the intended use of the test. We agree that replacing the word “all” with “clinically relevant” maintains and clarifies this intent. We will change the language of the policy as described here. It is also understood that some biomarkers may be necessary for clinical management that are not associated with companion diagnostics. We have additionally added language here that they may also be required. Regarding re-validations, we expect labs to revalidate their tests as often as reasonably possible to ensure their tests are clinically useful and whenever changes to the test are made. We will seek to work with the lab community when we feel changes are warranted to the list of required biomarker for coverage and provide sufficient time to allow labs to make the required changes.

3. Checklist M0151 comments. Thank you for the feedback on the updated TA process and forms. We understand that this process is imperfect and it will continue to evolve based on provider feedback as we seek to both simplify and improve the process. It should also be understood that these are not part of the LCD and changes to these forms will be an ongoing and collaborative effort. We seek to make the language both as clear and precise as possible.

1. Q1- CV/CU determination. The question presented here (Q1) is NOT intended to identify NGS tests that have proven Clinical Utility (CU) or Clinical Validity (CV). Instead, they are intended to inform the lab as to the requirement to complete form M00116, which requires providers to explain the CV/CU of their test. The intent of these questions is to reduce provider burden by identifying situations where this is not necessary. For clarity, it is understood that some internal lab processes may be proprietary for any given test yet not be relevant to the requirement of describing CU or CV of the test. The focus here is on the test result as stated in Q1. As an example, if a test reports mutations in cancer such as in EGFR, then the CV/CU of that finding is established and will not be specific to that lab or test; we rely on a chain of evidence to maintain that CV/CU has already been established for such a test result. It is likely irrelevant that an intermediary step in the process is proprietary- a proper validation (AV) will identify if the lab is identifying these mutations (with known CV/CU) accurately. According to this current form, Q1, if your test result is limited to genetic data based on accepted methodologies you do not have to fill out form M00116 and skip to Q3. If the result is a proprietary result (like an algorithm), then you do. Having proprietary components internal to the test is NOT listed as a requirement for either category. CV and CU evaluations are necessary when the test is measuring something new, wherein a chain of previously published evidence cannot be utilized. Typically this means the value of that measurement is not established as it relates to disease (CV), or when the usefulness/actionability of the finding is not accepted and must be demonstrated (CU). If a test identifies EGFR and/or other mutations but reports an algorithm value (instead of the mutation itself that has understood validity and utility), then the CV of that value must be established. Alternately, if a test performs gene expression, and that expression is compared to a proprietary data set that is not available or standardized, then the CV of the expression data must be established. Lastly, if the test is measuring something novel, then the utility of that biomarker must be established. All these scenarios would require form M00116.
2. Checkboxes and impact on coverage. The checkboxes in form M00151 do not impact coverage decisions, which are based on what is both reasonable and necessary in distinct benefit categories. They are there to help MolDX quickly and effectively understand what you claim your test is doing, which is very useful during the TA process. In turn, this helps reduce the processing time of TAs. These data may affect pricing for CGP tests.
3. Please let us know if you still feel this language lacks clarity. 

4. ICD-10 codes. Thank you for this feedback. Billing and coding information will be placed in an associated Article (per PIM Chapter 13), which will allow us to refine and improve the coding information. We welcome comments regarding what codes may be more appropriate for this policy.

The Coalition comprises many of the world’s most innovative diagnostic technology companies, clinical laboratories, physicians, venture capital companies, and patient advocacy groups. Coalition members have developed diagnostics that make personalized medicine possible. By helping providers understand the molecular nature of disease, these tests allow clinicians and patients to make individualized patient management decisions by assessing their risk of developing cancer, identifying potentially effective targeted therapies, and/or predicting the likelihood of cancer recurrence (among other intended uses).

Given the Coalition’s mission to facilitate development and commercialization of innovative diagnostics to inform important patient management decisions, we have a keen interest in the draft LCD, which would impact the nature and extent of Medicare coverage for next generation sequencing (NGS)-based tests for solid tumors furnished by certain Coalition members to Medicare beneficiaries in Jurisdictions J and M.

The Coalition is pleased that the MolDx program is proposing to operationalize the permissive local coverage under the National Coverage Determination for Next Generation Sequencing in Advanced Cancer through an LCD approach that will be more efficient for both providers and the Medicare program. We agree that, under the NCD and consistent with clinical evidence, laboratories should not be required to prove the clinical validity and clinical utility of biomarkers identified through comprehensive genomic sequencing profiles where those biomarkers have been established to select treatments for FDA- approved companion diagnostics or where the biomarkers are established in patient management algorithms under National Comprehensive Cancer Network (NCCN) or other well-established clinical guidelines in oncology. Since the usefulness of these markers is well-established, it is unnecessary for each laboratory seeking coverage for a comprehensive genomic profile test to reproduce the same evidence dossier of the clinical validity and clinical utility of these markers.

At the same time, these biomarkers are validated and useful in patient management only if the performance characteristics of the particular test have been established as reasonable and necessary by each laboratory seeking coverage. Such performance cannot be assumed or inferred from CLIA- certification and state licensure—especially across the broad range of solid tumors potential included under the draft LCD. Therefore, we support the requirement for rigorous analytical validation and submission of an evidence dossier establishing such validation for individual tests to be eligible for coverage under the draft LCD.

In summary, the Coalition supports the proposed Medicare coverage framework set forth in the draft LCD and its associated forms, provided a test will be covered only to the extent that its developer individually establishes that test’s analytical validity for the range of cancer types for which coverage is sought.

The Coalition appreciates the opportunity to provide comments on the draft LCD.

We would like to thank the Coalition for its supportive comments of this policy.

Invitae is the fastest-growing clinical genetics company in the United States. Our mission is to bring comprehensive genetic information into mainstream medicine to improve healthcare for patients. We are dedicated to making genetic information affordable and accessible to those who can benefit from it. Invitae's tests rely primarily on next generation sequencing-based genetic technology for hereditary cancer, neurology, cardiology, pediatrics, metabolic disorders, and reproductive health.

Intention of the LCD

We believe the intention of the LCD is to offer clarity and uniform rules for all labs which offer somatic tumor testing in MolDx jurisdictions.

Invitae plans to begin offering accurate, rapid, and state-of-the art somatic tumor testing beginning in (Q1 2020). Therefore, this LCD is of high interest to us and we hope that MolDx finalizes it quickly when the Palmetto and Noridian comment periods close. Our test will support FFPE tissue samples and will be used for solid tumor profiling. The test was designed based on FDA-approved drug targets and biomarkers and investigational targets and pathways for a complete molecular profile. Invitae will be measuring SNVs and Indels, CNVs, Fusions, MSI, TMB, and potentially other markers. Our validation plan was developed to meet New York State approval requirements and will follow appropriate MolDx processes.

We believe that there are three main ways in which the LCD needs to be clarified. These involve:

1. The binary definition of hotspot versus comprehensive genomic profiling (CGP),
2. The requirements for an “indication” in light of the rapidly evolving roster of FDA-approved CDx and CDx indications, and
3. Resolve confusion in CMS’s use of “primary cancer.”We also recommend that the LCD is the ideal place to provide clarity in how MolDx will define “primary tumor,” since the NCD uses this term without any clear definition.

We also recommend that the LCD is the ideal place to provide clarity in how MolDx will define “primary tumor,” since the NCD uses this term without any clear definition.

Defining Hotspot and CGP Testing

The LCD states that:

Two types of tests are considered for coverage, “Hot-spot” tests and comprehensive genomic profile tests (CGP). The definition of these terms, in addition to appropriate coding information is located in Coverage Articles associated with this LCD. These tests can detect any combination of the previously described variant types, but in general, Hot-spot tests are limited to SNVs and small INDELs, whereas CGPs can detect those variants in addition to CNAs, larger INDELs, gene fusions/translocations, and be used to calculate MSI status and TMB.

Unfortunately, no Article appears to be available for review and no Article is listed as an attachment to the online draft LCD (see underscore above).

Unfortunately, no Article appears to be available for review and no Article is listed as an attachment to the online draft LCD (see underscore above).

However, the LCD does appear to provide a very short definition:

                           Seq (SNV)         Small Indel       Large Indel/CN       Fusion      MSI-TMB

"Hot spot"              X                         X                         

"CGP"                    X                         X                         X                       X             X

Key Request: 

• MolDx should publish or circulate its full proposed definition of “hot spot” and “CGP” during the comment period, making it available for public comment.
  • Otherwise a pivotal part of the LCD, one of the most important parts, won’t be under public review.
• Without MolDx listing the definition of key terms, it is hard to comment on the LCD as a whole.

Defining “Current FDA CDx” for an “Indication”

The LCD states that,

“The assay performed includes at least the minimum genes and genomic positions required for the identification of all FDA approved therapies with a CDx biomarker for its intended use that can be ‘reasonably detected’ by the test. Because these genes and variants will change as the literature and drug indications evolve, they are listed separately in an associated Coverage Article as well as in the MolDx TA forms.”

Four Key Questions regarding “Indication:

(i) It is not clear what MolDx means by “indication.”

It may mean tumor type (e.g. MolDx will look for the FDA CDx genes in lung cancer, in colorectal cancer, in ovarian cancer, etc.). Is this correct? For example, if a CGP test is to be approved for coverage in the indication for lung cancer and also in the indication for colon cancer, then the test would need to meet the “all FDA CDx biomarkers” rule in these two cancers? 

(ii) In some cases, there are both germline and somatic mutation tests approved by FDA – will MolDx require both?

The “standard” of matching the available FDA CDx tests is unclear, especially when there are multiple FDA-approved tests. 

Even large tests will be unlikely to provide every potential fusion, and tests will likely calculate derived values like TMB somewhat differently.

We recognize that the requirements for an “indication” are limited to just the current FDA-approved CDx’s for that indication. However, these FDA standards are not always uniform. For example, Roche offers a FDA-approved KRAS test with just a few point mutations, while Illumina newly offers an enlarged RAS panel test with >50 mutation sites. Which one will MolDx use as the standard for new entrants like ourselves?

(iv) How will updates for expanded FDA CDx be implemented?

For example, will MolDx update its listing quarterly, with a period of several quarters of safe harbor while a covered test continues to be covered? Will new genes added to panels need to receive new review by MolDx as part of the TA process? Remember that labs also are meeting regulatory standards, such as submission of updated tests to New York State review before the test can be offered to patients in New York State. All of this requires a realistic appreciation of timelines.

NCD Controls Coverage by “Primary Diagnosis” – How Will MolDx Implement This?

The NCD requires that an NGS test may be used only if the patient has not “been previously tested using the same test for the same primary diagnosis of cancer” but does allow repeat testing “when a new primary cancer diagnosis is made by the treating physician.”

This raises several areas of confusion. If a patient on treatment develops new primary metastases, is that a new primary diagnosis of cancer? What guidance does the treating physician follow? After treatment, the tumor evolves and is not the same cancer, and different metastatic sites will vary. In addition, a metastasis is not usually called “a primary cancer.” The purpose of additional testing is to define new mutations that implicate the need for new and different treatments, not simply to confirm prior testing. In addition, in a significant number of patients, especially those with genetic risk loading, new cancers may well be new primaries as well. We appreciate clarification in the final LCD.

Thank you for your comments and feedback. Please refer to the current version of Article A56518, which has been available for public viewing since May 22, 2019- before this draft policy was publicly available. This should clarify a lot of the questions about definitions. Regarding indicated uses of the test, these must fall within the scope for MAC discretion identified by the NCD.

1. Regarding the term “indication”, this is not synonymous with disease type, but a condition or set of conditions wherein the test is indicated. NCD 90.2 sets forth some of the conditions, but the mere presence of a disease does not always indicate the use of a test.
2. This policy is for somatic testing only. However, germline information may be inferred from somatic testing with variable degrees of certainty. For example, knowing that a possible germline predisposition mutation is present in somatic tissue may require subsequent confirmatory testing. Similarly, knowing that a suspected germline variant is absent in somatic tissue may preclude unnecessary germline testing.
3. Specific genes and/or variants required for testing are listed in MolDX TA forms. These are separate from the policy and we urge you and others to provide feedback on the required genes and variants set in these forms. We will use subject matter expertise in identifying where to place cut-offs for what is reasonable as well as necessary.
4. It is not known at this time what the schedule for updating the requirements will be. However, we will work with the provider community to reduce undue burden by identifying how much advance notice will be necessary and ensuring there is sufficient notice and time between updates to minimize disruptions to lab operations.
5. Primary cancer definition: Please refer to Article A57902 for clarification of when a metastatic or recurrent tumor could be considered a primary.

F1CDx is a next generation sequencing (NGS) based in vitro diagnostic device for detection of substitutions, insertion and deletion alterations and copy number alterations in 324 genes and select gene rearrangements, as well as genomic signatures including microsatellite instability and tumor mutational burden using DNA isolated from formalin-fixed paraffin embedded tumor tissue specimens. The test is intended as a companion diagnostic to identify patients who may benefit from treatment with the targeted therapies listed in the product’s labeling in accordance with the approved therapeutic product labeling. Additionally, F1CDx is intended to provide tumor mutation profiling to be used by qualified health care professionals in accordance with professional guidelines in oncology for cancer patients with solid malignant neoplasms. The F1CDx test is a single-site assay performed at Foundation Medicine, Inc. Medicare covers F1CDx for beneficiaries nationwide, including patients in Jurisdictions J and M.¹

We recognize that the National Coverage Determination for Next Generation Sequencing in Advanced Cancer (CAG-00450N) permits local coverage through an LCD approach and that the MolDX program, through these Draft LCDs, is proposing to operationalize coverage in a way that will be more efficient for both providers and the Medicare program.

We would like to stress that biomarkers are validated and useful in patient management only if the performance characteristics of the particular test have been well established as reasonable and necessary by each laboratory seeking coverage. Such performance cannot be assumed or inferred from CLIA-certification and state licensure—especially across the broad range of diseases potentially included under the Draft LCDs. Therefore, we strongly support the requirement for rigorous analytical validation and submission of an evidence dossier showing such validation in order for individual tests to be eligible for coverage under the LCD.

Under the Draft LCD, a clinical laboratory seeking coverage for an NGS-based cancer test that is not nationally covered would be required to submit documentation allowing MolDX to complete a technical assessment (TA). The TA would be based on information provided by the test’s developer on certain forms available on the MolDX website (e.g., M00119 “Analytical Validity – Performance Specifications for Comprehensive Genomic Profiling Checklist: Somatic and Germline”, M00153 “Analytical Validity, Clinical Validation Summary Worksheet, NGS Solid Tumors, and M00154 “Analytical Validation and Clinical Validation Summary Worksheet, Myeloid Malignancies”). On these forms, MolDX requests analytical validation data, including information establishing the sensitivity, specificity, positive percent agreement, negative percent agreement, variant-level reproducibility, and reported limit of detection. However, the TA is not transparent about how the information provided on these forms will be evaluated to determine coverage of each NGS-based comprehensive genomic profile test or if there are minimum requirements that must be met. Additionally, we cannot underscore how important it is to understand and evaluate the bioinformatics processes the laboratory is using to interpret the sequencing data and generate the reports.

We appreciate the recognition by the MolDX program that not all NGS-based assays are equal. To that end, we request that the coverage framework set forth in the Draft LCDs and associated forms, ensure that a test will be covered only to the extent that its developer individually establishes that test’s analytical validity for the range of cancer types for which coverage is sought. Additionally, we request that the coverage framework evaluation criteria be made transparent in the TA documents.

The ability of any individual assay to accurately and reliably identify patients with such variants is, by definition, specific to the assay being performed. For example, our analytic validation of F1CDx for use in patients with solid malignant neoplasms was based, at least in part, on a retrospective analysis of 80,715 specimens from 43 unique tissue types.² It would only be appropriate to grant a new NGS-based assay comparable coverage if the new test’s developer is similarly able to establish that specific assay’s analytic performance across multiple tissue types. At the very least, and if not already the case, the TA threshold should be set at a level consistent with New York State Department of Health review standards as outlined by the U.S. Food and Drug Administration.³ 

¹ See Decision Memo for Next Generation Sequencing (NGS) for Medicare Beneficiaries with Advanced Cancer (CAG-00450N) (Mar. 16, 2018).

² See Summary of Safety and Effectiveness Data (SSED) – FoundationOne CDx™ (Nov. 30, 2017).

³ https://www.accessdata.fda.gov/cdrh_docs/reviews/DEN170058.pdf

We would like to thank you for the supportive comments on the policy. We agree that tests must demonstrate Analytical Validity for coverage for the indicated uses of the test. Criteria are established by a review of the published literature, standards set by regulatory agencies, as well as subject matter experts and advocating medical associations. Regarding the current TA forms, as discussed above this will be an ongoing process outside of the formal LCD process where we welcome feedback regarding these forms. We seek to both reduce provider burden as well as increase transparency in this process.

On behalf of Illumina, a leading developer and manufacturer of next generation sequencing (NGS) tools for both research and clinical use, we appreciate the opportunity to provide comments on this proposed LCD (DL38045) for Medicare beneficiaries with advanced solid tumors.

First, we would like to commend the Palmetto GBA MolDx program and express our support for the proposed LCD, which we believe will facilitate patient access to high quality molecular testing.

We agree with MolDx that, based on peer-reviewed published clinical evidence and the CMS National Coverage Determination for Next Generation Sequencing in Advanced Cancer (NCD 90.2), the clinical validity and clinical utility of biomarkers identified through comprehensive genomic profiling (CGP) have been established to aid in the selection of treatments. As stated in the proposed LCD: “NCD 90.2 confirms these tests to be both reasonable and necessary in Medicare beneficiaries.” As such, this proposed LCD will streamline the effort required for laboratories to seek coverage for a CGP test by eliminating the need to summarize and report evidence regarding the clinical validity and clinical utility of CGP assays.

We have a few minor suggestions and areas for clarification to share with MolDx.

Section: Coverage Indications, Limitations, and/or Medical Necessity

Comment

• Suggest adding “FDA-cleared” to the first sentence. “This policy describes and clarifies coverage for Lab-Developed Tests (LDTs), FDA-cleared, and FDA-approved clinical laboratory tests utilizing Next-Generation Sequencing (NGS) in cancer as allowable…”

Section: Analytical Validity and Clinical Utility

Comment

• Suggest adding “FDA-cleared” to the first sentence of the 2^nd “Labs seeking coverage for LDTs, FDA-cleared, or FDA-approved tests…”
• Suggest adding clarification regarding TA requirements for CGP assays that are already FDA-approved, FDA-cleared, or New York State-approved.

Section: Criteria for Coverage

Comment

• One of the criteria (per NCD 90.2) is that the patient “has not been previously tested by the same test with the same primary diagnosis.” However, the associated coverage article (A54795) states that coverage of CGP “precludes the use of any other molecular testing on that specimen.” Suggest clarifying the coverage article to state that CGP testing will be non-covered only if “the same test” has been performed (as stated in the LCD) as opposed to “any other molecular testing” (as currently stated in the coverage article).
• In the Criteria for Coverage section, the proposed LCD states that: “Because these genes and variants will change as the literature and drug indications evolve, they are listed separately in an associated Coverage Article.” However, genes and variants are not currently listed in the coverage article. Will they be added?

Section: Analysis of Evidence

Comment

• At the bottom of this section, the proposed LCD states: “Given the abundant literature on genetic and genomic testing in cancer diagnosis and care, this contractor feels strongly that NGS methodology for testing is appropriate for use in Medicare Beneficiaries. However, given the variability for what information tests can provide, additional information must be submitted by providers to ensure the contractor A) understands what test is being performed; B) Why it is being performed; C) If the test is both necessary and reasonable for cancer care for its intended use.” Suggest clarifying specific requirements for this statement. For example:
  • Is this required for each patient?
  • For (A): does the Z-code suffice?
  • For (B): does the ICD-10 code suffice?
  • For (C): How will this be evaluated? Per the statement in the Summary of Evidence section, a CGP test that passes TA assessment would be considered reasonable and necessary if used according to the coverage criteria, correct?

Thank you for the positive feedback on the policy and the suggested edits. We have incorporated many of the suggested edits into the policy. Those that were not incorporated are discussed below:

Analysis of evidence- Your questions are important but these are claims processing issues that are not specific to this policy. In general, any molecular test utilizing human DNA sequencing must have registered with DEX and have a Z-identifier code, and claims processing requires ICD-10 codes compliant with the policy that are listed in an associated coverage and billing Article. Technical assessments are described elsewhere; please go to our website for more information and forms for the TA process. If a test is reviewed and satisfactorily completes the TA process, and a claim is submitted with the proper ICD-10 code, CPT code, and Z-identifier the claim will process.

Please accept our comments related to draft LCD #38045 (MolDX: Next-Generation Sequencing for Solid Tumors), specifically “assay performed includes at least the minimum genes and genomic positions required for the identification of all FDA-approved therapies with a companion diagnostic biomarker for its intended use that can be reasonably detected by the test.”

We believe the “all” criteria should be removed.

The companion diagnostic space is rapidly evolving, and DL#38045 asks CLIA labs to either update their assay frequently (revalidate) or adopt a broader “future-proof” assay.

Clinical laboratories often wish to take a modular approach to assay adoption. Having multiple assays that address companion diagnostic biomarkers allow labs to “pick-and-choose” based on indication and/or serialize testing based on prevalence, with the goal to maximize tissue availability and reduce testing cost per patient.

A recent example to highlight the above rationale is NTRK. NTRK fusions were recently approved as a companion diagnostic biomarker for VIKTRAKVI® (larotrectinib) across all solid tumor cancers.

However, many next generation sequencing (NGS) assays – used for established companion diagnostic biomarkers like EGFR, BRAF, ALK, etc. – do not include NTRK fusion detection.

Some laboratories have elected to either use non-NGS methods (e.g. IHC) or send-out samples to external laboratories for NTRK fusion detection, rather than adopt a new NGS assay that does include NTRK fusions.

While there are merits for using a broader NGS assay, laboratories that decide to use a different approach should not be penalized for maintaining their existing assay, providing it continues to report (all FDA-approved therapies) and demonstrate clinical utility by detecting other established companion diagnostic biomarkers.

Thank you for your feedback and comments. I will only address questions that were not previously addressed in other comments. Regarding modular approaches to assay adoption, please note that this is not prohibited under this policy regarding CGP tests. There is no stipulation prohibiting non-NGS technologies as a component of a test. This is actually a fairly common practice, particularly when CGP tests are by their nature “undefined” with available CPT codes. Your lab could choose to include FISH for NTRK fusion detection as part of your CGP test if NTRK fusions are required for coverage. If/once you incorporate NTRK translocation detection into your NGS assay, please submit validation documents to us and inform us of the change. Please note that validation materials submitted as part of the TA must include all components of the test.

On behalf of Thermo Fisher Scientific’s Clinical Next-Generation Sequencing Division, we thank you for the opportunity to provide comments on the proposed MolDX Local Coverage Determination (Next-Generation Sequencing for Solid Tumors (DL38045)).

Thermo Fisher Scientific Inc. is the world leader in serving science, with revenues of more than $24 billion and approximately 70,000 employees globally. Our mission is to enable our customers to make the world healthier, cleaner and safer. We help our customers accelerate life sciences research, solve complex analytical challenges, improve patient diagnostics, deliver medicines to market and increase laboratory productivity. Through our premier brands – Thermo Scientific, Applied Biosystems, Invitrogen, Fisher Scientific and Unity Lab Services – we offer an unmatched combination of innovative technologies, purchasing convenience and comprehensive services. For more information, please visit www.thermofisher.com

Our Clinical Next-Generation Sequencing Division supports molecular diagnostics by creating and manufacturing a portfolio of assays that enhance performance in applications which range from translational research to therapy selection in patient care of Medicare beneficiaries.

Tests include molecular profiling of solid and myeloid tumor indications, liquid biopsy, and immuno-oncology. We manufacture multi-biomarker targeted assays designed for cancer research, enabling next-generation sequencing analysis of multiple biomarker types: fusions, insertion/deletions (indels), single nucleotide variants, and copy number variations. Of note is Oncomine Focus: a targeted, multi-biomarker assay that enables detection of hotspots, SNVs, indels, CNVs, and gene fusions from DNA and RNA in a single workflow. This comprehensive assay design is further enhanced with workflow automation, delivering insights quickly and helping accelerate research towards that next companion diagnostic or therapy in the future.

Thermo Fisher applauds the MolDX program’s proposal to update how NGS testing is covered, coded, and reimbursed today. We wish to submit the following comments:

Panel Design Requirements

Our primary concern is this language “assay performed includes at least the minimum genes and genomic positions required for the identification of all FDA-approved therapies with a companion diagnostic biomarker for its intended use that can be reasonably detected by the test.” We believe the “all” criteria should be removed.

The companion diagnostic space is rapidly evolving, and DL#38045 asks CLIA labs to either update their assay frequently (revalidate) or adopt a broad “future-proof” assay. These scenarios place a heavy burden on labs, in terms of both financial investment and man hours.

Clinical laboratories often wish to take a modular approach to assay adoption. Having multiple assays that address companion diagnostic biomarkers allow labs to “pick-and- choose” based on indication and/or serialize testing based on prevalence, with the goal to maximize tissue availability and reduce testing cost per patient.

While there are merits for using a broad NGS assay, laboratories that decide to use a different approach should not be penalized for maintaining their existing assay, providing it continues to report (all FDA-approved therapies) and demonstrate clinical utility by detecting other established companion diagnostic biomarkers.

We suggest the following alternatives:

1. Allow “all” content to be satisfied by multiple tests; this would be confirmed/updated through specific Z-code reporting, limiting administrative complexities of coverage, coding, and pricing

2. At minimum, add language to final LCD that affords time for CLIA labs and approved assays to expand tested/reported biomarkers – allowing for proper reaction to future test content expansions

From tissue-sparing techniques that generate a patient report in over 90% of tested samples, to a 10-day turnaround time, to multiplex results which yield dozens of answers to inform therapy selection: NGS testing is critical to oncology patient management. We applaud the thoughtful development of DL38045 – but are concerned that MolDX’s failure to address the items noted above may negatively impact beneficiary access to high quality testing.

Thank you for your comments. We believe all comments herein have been previously addressed in other comments.

Coding:

• CPT codes listed in the draft are currently limited to 81445 and 81479.
• While it is understood the LCD does not address hematologic malignancies, the definition of CPT 81455 includes solid organ or hematolymphoid neoplasms (see below).
• Therefore, should 81455 also be included in the LCD for any NGS solid organ panel test that has 51 or more genes? Even if these panels are generally considered to be “not medically necessary,” a statement to that affect added to the LCD would be very helpful.

Targeted genomic sequence analysis panel, solid organ or hematolymphoid neoplasm, DNA analysis and RNA analysis when performed, 51 or greater genes (eg, ALK, BRAF, CDKN2A, CEBPA, DNMT3A, EGFR, ERBB2, EZH2, FLT3, IDH1, IDH2, JAK2, KIT, KRAS, MLL, NPM1, NRAS, MET, NOTCH1, PDGFRA, PDGFRB, PGR, PIK3CA, PTEN, RET), interrogation for sequence variants and copy number variants or rearrangements, if performed

Approved Tests:

• This Draft states a test must “satisfactorily complete a TA by MolDX.”
• As a Medicare Advantage plan, we are required to implement and provide coverage consistent with the local contractors LCDs and LCAs, yet we don’t have access to the Change Healthcare database to know if a given test has completed the TA process.
• Thus, it is unknown how we would know if a test has been approved by MolDX or not.
• Will this be provided in some format, or will it be added to the LCD over time?
• Strict time frames to complete prior authorization requests make submitting a request to MolDX to determine if a test has been approved or not unrealistic, and not ideal.
• It would be useful to know what tests have completed this process, and either been determined eligible for coverage, or non-covered.

Criteria Question:

• The Criteria states, “…AND has not been previously tested by the same test with the same primary diagnosis.”
• Is this exclusionary criteria item specific to the exact same panel test?
  • Or would this apply to the same gene(s) that were targeted in the original test, such as if the referring physician used a different panel, by a different laboratory, but looking for the same gene variants tested for originally?

Thank you for your comments. We will address your comments below.

Coding: This policy is for coverage of lab-developed tests or FDA-cleared tests defined in article A56518 as either Targeted Panels or CGPs. Although not part of the policy itself, the billing and coding Article identifies the appropriate CPT code for Targeted Panels as 81445 or the NOC code 81479 for CGP tests as we do not believe these to be sufficiently described by an existing CPT code. CGP tests may have more than 51 genes sequenced as a component of the test. MolDX-associated MACs have the capability to adjudicate claims based on a Z-identifier code, which allows us to know what test is being performed and utilize the NOC code with specificity.

Approved tests: Anyone can create a public account on the DEX Diagnostic Exchange website and look up tests or companies that perform tests. Included in the public information is payor coverage information. Additionally, we will be creating an article to accompany this policy that will refer to all the tests that successfully complete the TA.

Criteria Question: This is language from the NCD and refers to the same test. Regarding LCDs, the test would be specific to the laboratory performing and validating their test.