Response to Comments: MolDX: Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia

The following comment were submitted to Palmetto GBA, CGS, WPS and Noridian and were received from multiple stakeholders:

1. Clarity on the type of molecular test being covered.
2. Request for addition of recently published clinical guidelines that describe the use of nonendoscopic cell-collection devices as an alternative to upper endoscopy.
3. Request for removal of numerical cut off expectations for sensitivity and specificity.
4. Clarification of criteria for tests that identify dysplastic vs. nondysplastic disease
5. Recommendation to modify the non-coverage language to coverage

1. Clarity on the type of molecular test being covered.

Requirements for DEX registration as a component of the MolDX program are limited to DNA and RNA based tests at this time. Other tests may fall under this policy but are not within the scope of MolDX.

2. Request for addition of recently published clinical guidelines that describe the use of nonendoscopic cell-collection devices as an alternative to upper endoscopy.

We agree with the addition of recently published practice guidelines to the policy. Updated ACG and AGA clinical practice guidelines have been referenced where applicable.

3. Request for removal of numerical cut off expectations for sensitivity and specificity.

The policy describes an “ideal” test as highly sensitive and highly specific, a goal for any diagnostic assay. It does not state those percentages as specific coverage criteria requirements. However, we understand that the perception of numerical cut offs for a service that has not yet been fully established and does not take into account the prevalence of the disorder or comparison to current practices may be confusing. Therefore, (>95%), as false negatives may miss patients with dysplasia and (>95%) meaning fewer patients without dysplasia would require follow-up procedures have been removed from the rationale for determination.

4. Clarification of criteria for tests that identify dysplastic vs. nondysplastic disease.

Based on published guidelines, treatment and surveillance protocols differ between patients with dysplastic disease and those without. For example, the ACG recommends endoscopic eradication therapy (EET) in patients with HGD, suggests EET or surveillance for those with LGD, and surveillance for those with NDBE. The intention of coverage criteria #5 was not to exclude tests designed to identify patients with NDBE but to distinguish between dysplastic and nondysplastic disease. Therefore, we have modified the language of coverage criteria #5 as follows:

The test identifies patients with dysplastic disease that may benefit from endoscopic treatment or surveillance or patients with nondysplastic disease who may benefit from surveillance.

5. Recommendation to modify the non-coverage language to coverage.

We recognize that the field is rapidly changing, and several non-endoscopic molecular tests are currently in use or development. However, upper endoscopy remains the gold standard for the diagnosis of BE. Per both updated ACG and ACA clinical guidelines the quality of evidence for non-endoscopic molecular testing is considered very low with a conditional strength of evidence. Additional trials are warranted to demonstrate that this service is reasonable and necessary. The policy will remain as non-covered until such time.

The following comment was submitted to Palmetto GBA, CGS, WPS, and Noridian:

On behalf of Medtronic, thank you for the opportunity to provide comments on the Molecular Diagnostics (MolDX) Proposed Local Coverage Determination for Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia. Medtronic is the world's leading medical technology company, specializing in minimally invasive therapies that provide innovative solutions that improve patient outcomes and expand access to care for gastrointestinal disease and cancers. We are committed to the continual research and development necessary to produce high-quality products and to support innovative solutions that improve patients' lives.

Medtronic appreciates the effort that went into the development of the Proposed LCD and recognizes the importance of increasing patient access to diagnostic testing that have the ability to detect and prevent esophageal adenocarcinoma (EAC) at earlier, curable stages. However, we have concerns regarding proposed non-coverage for non-endoscopic biomarker tests for screening and surveillance of Barrett’s Esophagus (BE), including Cytosponge™, and as such, we respectfully request you consider the comments outlined in this letter.

During the last five decades, the incidence of EAC has been rising. Survival rates following diagnosis are less than 20% after five years.¹ BE affects at least 5.6% of the U.S. population² and is the greatest risk factor for EAC, increasing the risk by 10-55 times that of the general population.³ BE is defined by the presence of specialized intestinal metaplasia of the tubular esophagus. The development of BE is believed to be a reparative response to reflux-induced damage to the native squamous epithelium, with subsequent replacement of metaplastic intestinalized epithelium, BE. Metaplastic BE is associated with an increased cellular proliferation and turnover that may result in progression to dysplasia.⁴ BE is a well-established premalignant condition for EAC, and therefore efforts to prevent and detect pre-cancerous lesions are critical to improving patient outcomes.

The Cytosponge™ cell collection device is minimally invasive and enables the sampling of surface esophageal cells in patients who require screening for esophageal pathologies such as Barrett’s Esophagus. Cytosponge™ consists of a 30mm compressed spherical sponge within a clear, vegetable-based capsule attached to a retainer card by a silicone braided polyester suture. The patient ingests the capsule with water. Then, after the capsule has dissolved in the stomach, it is retracted through the esophagus, collecting a specimen from the distal esophagus. Cytosponge™ can be administered without the use of sedation in a physician office, yields adequate samples for cytological and histological assessment and is well tolerated by patients.⁵

While endoscopy has been the standard of care in the diagnosis and management of premalignant conditions of the esophagus, physicians in real-world practice recognize the need for an effective and efficient non-invasive office-based screening test to detect BE. Non-endoscopic devices and biomarker tests can provide an additional screening method to prevent esophageal cancer or to diagnose it at an early stage. Low-cost cell collection devices such as Cytosponge™ also enable improved access to patient populations that have historically had challenges in accessing quality GI endoscopic care pathways due to socioeconomic and or geographical restrictions. We believe Cytosponge™ could unlock the opportunity to detect cancer earlier and produce improved outcomes if it is diffused effectively across all populations where BE is emergent and prevalent.

Updated guidelines for the diagnosis and management of Barrett’s Esophagus published on April 2, 2022 from the American College of Gastroenterology (ACG) suggest that a swallowable, nonendoscopic capsule sponge device combined with a biomarker is an acceptable alternative to endoscopy for screening for BE in patients with chronic reflux symptoms and other risk factors, directly supporting the use of Cytosponge™ as an alternative to endoscopy for screening of BE in defined populations.⁶ Medtronic requests the proposed MolDX LCD to be updated to reflect these revised ACG guidelines with broadened screening modalities to include a swallowable, non- endoscopic capsule sponge device combined with a biomarker as an acceptable alternative to endoscopy for screening BE in those with chronic reflux symptoms and other risk factors.

Clinical references

Januszewicz et al lists four prospective studies utilizing Cytosponge™ to detect BE.⁷ The largest of these, BEST2, a case-control study with 1110 patients, combined cytological assessment of the Cytosponge™ sample with TFF3 immunohistochemistry and demonstrated a sensitivity of 80% for detecting BE of any length, and 87% for circumferential BE segments measuring >3cm; and a specificity of 92% compared to endoscopy and biopsy.⁵ In a more recent US-based multicenter prospective cohort study with 191 patients, Shaheen et al reported 99.5% of patients successfully swallowed the device with overall sample adequacy of 91%. Overall sensitivity, specificity, and accuracy of the assay with TFF3 staining were 76%, 77%, and 76% and increased to 86%, 77%, and 82% for patients with ≥ 3cm BE.⁸

In requirements for a new diagnostic test you mention that the test would ideally be highly sensitive (>95%), as false negatives may miss patients with dysplasia, and highly specific (>95%), meaning fewer patients without dysplasia would require follow-up procedures. However, Paterson et al comments that Cytosponge cell collection coupled with TFF3 immunohistochemistry performance from BEST2 is comparable to values quoted for the sensitivity and specificity in meta-analyses of established screening tests; 79% and 95% respectively for distal colorectal adenocarcinomas using the fecal occult blood test and 71% and 95% respectively for proximal tumours; 94% and 88% respectively for human papilloma virus testing and 84% and 88% respectively for cervical smear cytology.⁹

From a safety perspective, a meta-analysis of pooled data from five prospective trials including 2,418 individuals and 2,672 Cytosponge™ procedures, has shown that over 96% of patients were able to successfully swallow the Cytosponge™ with only two adverse events and a median acceptability higher than endoscopy without sedation.⁷ Finally, as mentioned above, the processing of the Cytosponge™ specimen is a two-step process with H&E staining, followed by biomarker assessment with TFF3 immunohistochemistry (IHC). Shaheen et al⁸ defined an adequate sample as ≥ 1 columnar cell by H&E staining, and only submitted adequate samples for TFF3 IHC, which may be a good quality control to consider implementing in practice.

Conclusion

We respectfully disagree with the conclusion that “there are currently no existing molecular biomarker tests that have demonstrated analytical validity (AV), clinical validity (CV), and clinical utility (CU) to fulfill the necessary criteria” and that they do not meet Medicare’s criteria for “reasonable and necessary” services. The Cytosponge-TFF3 procedure is a safe and acceptable test to administer. The Cytosponge procedure results in improved detection of Barrett’s Esophagus and offers a more proactive approach for identification and treatment of dysplasia and early cancer.¹⁰

Thank you for the opportunity to submit these comments on the proposed LCD.

References were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA and Noridian:

I am responding to the Proposed LCD: Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia on behalf of the American Foregut Society (AFS). The AFS mission is to help guide both the diagnosis and management of Foregut disease through collaboration between Gastroenterologists and Foregut Surgeons. Our members support the need for a tool to detect Barrett’s esophagus or esophageal adenocarcinoma (EAC).

This belief is rooted in the approximately 20,000 U.S. GERD patients that were projected to be diagnosed with EAC in 2022 this year. The U.S. incidence of EAC has increased 500% over the past four decades. Approximately 16,000 U.S. patients were expected to die from esophageal cancer in 2022. An EAC diagnosis is a near-death sentence in most patients. Over 80% of EAC patients die within five years of diagnosis, placing EAC second only to pancreatic cancer in lethality. This lethality is a direct result of the fact that EAC is nearly always invasive at the time of diagnosis and will have already metastasized to lymph nodes or other organs in nearly 75% of patients.

The clinical utility testing for the early detection of upper gastrointestinal metaplasia, dysplasia, and neoplasia using Esophagogastroduodenoscopy (EGD) is validated in the scientific literature and supported in clinical practice. Recognizing the clinical value of early detection as demonstrated by EGD, the recently released American College of Gastroenterology Guidelines support the use of a nonendoscopic capsule device combined with a biomarker as an acceptable alternative to endoscopy for screening for Barrett’s esophagus, the only know precursor to esophageal cancer. The ability to offer an alternative to EGD for patients to detect the main precursor to cancer allow providers to make personalized treatment decisions for their patients. This would be a significant clinical advance for patients and health care providers. Detecting cancer early aligns with the President’s stated NCI Cancer Moonshot goal to reduce the death rate from cancer by at least 50 percent over the next 25 years.

As such, we support the following changes to the proposed LCD.

1. We recommend that the proposed LCD be updated to reflect the recently released American College of Gastroenterology Guidelines that support the use of a nonendoscopic capsule device combined with a biomarker as an acceptable alternative to endoscopy for screening for Barrett’s esophagus, the only know precursor to esophageal cancer. The updated guidelines reflect the clinical utility of such tests in the daily practice of medicine.
2. We recommend that the language that indicates test in this category would “ideally be highly sensitive (>95%), as false negatives may miss patients with dysplasia, and highly specific (>95%), meaning fewer patients without dysplasia would require follow-up procedures” be removed or modified. We agree that ideal performance criteria should balance such risks and benefits to assure patients are not exposed to unnecessary risks AND are not deprived of the test’s purported benefits, in this case preventing cancer death through early detection of precancer These criteria do not incorporate the following which are necessary to assess risks and benefits of a test

1. The prevalence of the condition being tested in the target population.
2. Current practice regarding detection of the condition.
3. The criteria far exceed the performance of numerous similar tests deemed reasonable and necessary, including most preventative screening tests.
4. By effectively erecting what is likely a biologically insurmountable barrier for a precancer test, the criteria would deprive the target population of the benefits of early precancer detection to prevent EAC.

3. Negative Predictive Value (NPV) and Positive Predictive Value (PPV) incorporate prevalence and are better suited to serve as ideal performance criteria, balancing risks/benefits. As the Carrier Advisor Committee meeting experts noted, since so few at-risk GERD patients currently undergo guideline recommended testing for esophageal precancer, the risk associated with a false negative test compared to current practice is very low. Longstanding guidelines recommend endoscopy to detect non-dysplastic metaplasia, not just dysplasia, in at-risk GERD patients, to allow standard-of-care endoscopic surveillance to detect progression to dysplasia and maximize the opportunity for curative ablation to prevent further progression to EAC. A false positive nonendoscopic biomarker test does not expose the patient to additional risk since, per guidelines, the patient should have had an endoscopy anyway. Triaging patients with a nonendoscopic biomarker test significantly limit a patient’s exposure to negative endoscopies and will increase uptake of screening in patients who may be hesitant to undergo endoscopy as the initial screening modality to test for Barrett’s esophagus. Furthermore, such a non-endoscopic approach provides a significant cost-savings compared to our current approach of screening utilizing sedated endoscopy.

Thank you again for the opportunity to comment on this important topic that focuses on advancing the early detection of EAC in Medicare beneficiaries.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA and Noridian:

We are responding to the Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia proposed local coverage determination (LCD) as participants in the National Institute of Health and National Cancer Institute (NCI) funded Barrett's Esophagus (BE) Translational Research Network (BETRNet) Program. BETRNet aims to speed the translation of important research findings from laboratory and clinical studies into useful medical applications. This includes validated diagnostics, improved patient management, cancer risk stratification and prediction, prevention strategies, and accurate criteria for BE screening and endoscopic surveillance. The conception and development of a methylation biomarker based non-endoscopic technology for detecting Barrett's esophagus happened through support from Division of Cancer Prevention’s Early Detection Research Network (EDRN) and BETRNet programs, and the Division of Cancer Treatment and Diagnosis' Translational Research Program, Specialized Program of Research Excellence (SPORE). The test was highlighted in the NCI 2018 report to the President as a significant advance against cancer.

The need for the early detection is rooted in the approximately 20,000 U.S. GERD patients that were projected to be diagnosed with esophageal adenocarcinoma (EAC) this past year. The U.S. incidence of EAC has increased 500% over the past four decades. Approximately 16,000 U.S. patients were expected to die from esophageal cancer in 2021. An EAC diagnosis is a near-death sentence in most patients. Over 80% of EAC patients die within five years of diagnosis, placing EAC second only to pancreatic cancer in lethality.

This lethality is a direct result of the fact that EAC is nearly always invasive at the time of diagnosis and will have already metastasized to lymph nodes or other organs in nearly 75% of patients.

The clinical utility testing for the early detection of upper gastrointestinal metaplasia, dysplasia, and neoplasia using Esophagogastroduodenoscopy (EGD) is validated in the scientific literature and supported in clinical practice. Recognizing the clinical value of early detection as demonstrated by EGD, the recently released American College of Gastroenterology Guidelines support the use of a nonendoscopic capsule device combined with a biomarker as an acceptable alternative to endoscopy for screening for Barrett’s esophagus, the only know precursor to esophageal cancer. The ability to offer an alternative to EGD for patients to detect the main precursor to cancer allow providers to make personalized treatment decisions for their patients. This would be a significant clinical advance for patients and health care providers. Detecting cancer early aligns with the President’s stated NCI Cancer Moonshot goal to reduce the death rate from cancer by at least 50 percent over the next 25 years.

As such, we support the following changes to the proposed LCD.

1. We recommend that the proposed LCD be updated to reflect the recently released American College of Gastroenterology Guidelines that support, nonendoscopic capsule device combined with a biomarker is an acceptable alternative to endoscopy for screening for Barrett’s esophagus, the only know precursor to esophageal cancer. The updated guidelines reflect the clinical utility of such tests in the daily practice of medicine.
2. It is important to note that BE is the condition in which a specialized columnar epithelium replaces the normal stratified squamous epithelium of the Prolonged acid injury to the squamous epithelium can cause this metaplastic change in about 10% to 15% of individuals suffering from gastroesophageal reflux disease (GERD). This specialized columnar epithelium serves a protective function, as it is more resistant to stomach acid. However, it also increases the risk of esophageal adenocarcinoma. There is a good correlation between the rising incidence of adenocarcinoma of the esophagus with the increasing incidence of GERD. Therefore, timely evaluation and treatment of patients with GERD for Barrett’s esophagus are important. As the title of the LCD indicates, it appears to include the detection of metaplasia which is critical for the early detection of BE. I would suggest that you alter the proposed LCD coverage criteria that states, the test identifies patients with dysplastic disease that are more likely to benefit from invasive procedures from patients without dysplastic disease to reflect the need to assess patients for BE as the ACG guidelines state.
3. We recommend that the language that indicates test in this category would “ideally be highly sensitive (>95%), as false negatives may miss patients with dysplasia, and highly specific (>95%), meaning fewer patients without dysplasia would require follow-up procedures” be removed or modified. We agree that ideal performance criteria should balance such risks and benefits to assure patients are not exposed to unnecessary risks AND are not deprived of the test’s purported benefits, in this case preventing cancer death through early detection of precancer These criteria do not incorporate the following which are necessary to assess risks and benefits of a test.

         a. The prevalence of the condition being tested in the target population.

         b. Current practice regarding detection of the condition

         c. The criteria far exceed the performance of numerous similar tests deemed reasonable and necessary, including most preventative screening tests.

         d. By effectively erecting what is likely a biologically insurmountable barrier for a precancer test, the criteria would deprive the target population of the benefits of early precancer detection to prevent EAC.

4. Negative Predictive Value (NPV) and Positive Predictive Value (PPV) incorporate prevalence and are better suited to serve as ideal performance criteria, balancing risks/benefits. As the Carrier Advisor Committee meeting experts noted, since so few at-risk GERD patients currently undergo guideline recommended testing for esophageal precancer, the risk associated with a false negative test compared to current practice is very low. Longstanding guidelines recommend endoscopy to detect non-dysplastic metaplasia, not just dysplasia, in at-risk GERD patients, to allow standard-of-care endoscopic surveillance to detect progression to dysplasia and maximize the opportunity for curative ablation to prevent further progression to EAC. A false positive nonendoscopic biomarker test does not expose the patient to additional risk since, per guidelines, the patient should have had an endoscopy anyway. Triaging patients with a nonendoscopic biomarker test significantly limit a patient’s exposure to negative endoscopies.

Thank you again for the opportunity to comment on this important topic that focuses on advancing the early detection of EAC in Medicare beneficiaries.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA and Noridian:

I appreciate the opportunity to comment on the Proposed LCD: Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia. I support the use of the Esoguard test in accordance with the recently updated American College of Gastroenterology (ACG) guidelines that state that a swallowable non-endoscopic capsule device combined with a biomarker is an acceptable alternative to endoscopy for Barrett’s esophagus (BE).¹

It is important to note that BE is the condition in which specialized columnar epithelium replaces the normal stratified squamous epithelium of the esophagus. Prolonged acid injury to the squamous epithelium can cause this metaplastic change in 10% to 15% of individuals suffering from gastroesophageal reflux disease (GERD). This specialized columnar epithelium serves a protective function, as it is more resistant to stomach acid. However, it also increases the risk of esophageal adenocarcinoma. There is a good correlation between the rising incidence of adenocarcinoma of the esophagus with the increasing incidence of GERD. Therefore, timely evaluation and treatment of patients with GERD for Barrett’s esophagus are important.²

The title of the LCD includes the detection of metaplasia, which is critical for the early detection of BE. I would suggest that you alter the proposed LCD coverage criteria that states, the test identifies patients with dysplastic disease³ that are more likely to benefit from invasive procedures from patients without dysplastic disease to reflect the need to assess patients for BE as the ACG guidelines state. Remember that approximately 90% of patients with esophageal adenocarcinoma have never had prior diagnosis of Barrett’s esophagus due to lack of screening.

The LCD combines the concepts of screening and surveillance which are both very different. Making this change will allow clinicians the ability to offer an alternative to esophagogastroduodenoscopy (EGD) for patients to detect the main precursor to cancer allow providers to make personalized treatment decisions for their patients. This would be a significant clinical advance for patients and health care providers.

As I hope is apparent, coverage of tests for the early detection of BE are greatly needed, as this technology allows patients to avoid unnecessary invasive procedures.

References were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA and Noridian:

On behalf of Lucid Diagnostics Inc., the developer of the proprietary EsoGuard^® Esophageal DNA Test, I am pleased to submit comments regarding the above-captioned proposed local coverage determination (LCD).

EsoGuard uses next generation sequencing of bisulfate converted DNA to detect the presence of methylation signatures at 31 CPG islands of two genes, Vimentin (mVIM) and CyclinA1 (mCCNA1). EsoGuard is intended to identify at-risk chronic gastroesophageal reflux disease (GERD) patients with nondysplastic metaplasia or dysplasia of the esophagus, known as Barrett’s Esophagus (BE), which are precancerous precursors to highly lethal esophageal adenocarcinoma (EAC). EsoGuard is a single-site assay performed at Lucid’s CLIA-certified, CAP-accredited clinical laboratory in Lake Forest, CA, and, as a result, would be subject to the proposed LCD.

Below, please find our comments on and recommended revisions to the proposed LCD, as well as our rationale for the same.

1. Recommended deletion of language regarding current coverage status of assays subject to the proposed LCD

The “Coverage Indications, Limitations, and/or Medical Necessity” section of the proposed LCD states that “current molecular diagnostic tests that identify individuals with upper gastrointestinal metaplasia, dysplasia, and neoplasia are non-covered by this contractor.” Similarly (though in a non-operative section of the proposed LCD), the “Analysis of Evidence (Rationale for Determination)” section concludes that “although there are several promising molecular biomarker tests designed to further identify at-risk patients, this contractor finds that there are currently no existing tests that have demonstrated AV, CV, and clinical utility (CU) to fulfill the necessary criteria.”

We understand the above-quoted language reflects the current position with respect to coverage of molecular tests for the detection of upper gastrointestinal metaplasia, dysplasia, and neoplasia, and that this proposed foundational LCD is intended to create a pathway to future coverage for tests that meet the criteria set forth in the policy. However, if the above-quoted language is included in the final LCD, it would arguably preclude from adding coverage for new tests—even if such tests successfully complete the Technical Assessment (TA) process—because the LCD identifies these tests as non-covered services. Given the procedural requirements of the 21st Century Cures Act, it could easily take 12 to 18 months to propose and finalize a revised LCD that removes the above-referenced language. To ensure that Medicare beneficiaries have timely access to assays that demonstrate analytic validity, clinical validity, and clinically utility following successful completion of the TA process, we strongly encourage to revise the proposed LCD to (a) clarify that tests meeting the criteria set forth in the LCD will be eligible for coverage, and (b) delete language regarding the current coverage status of tests from the LCD.

Our suggested approach that should consider would be consistent with language recently used by the MolDX Program when outlining coverage and its approach to reviewing new evidence in other foundational LCDs.¹ We also note as an example that the MolDX that deleted language from its recently-finalized “Molecular Syndromic Panels for Infectious Disease Pathogen Identification Testing” coding article,² the draft of which had noted there are currently no covered tests for certain types of tests identified as coverable under the parent LCD.

2. Recommended modification of coverage criteria to clarify intent to cover assays that identify patients with BE

The first two criteria in the “Coverage Indications, Limitations, and/or Medical Necessity” section of the proposed LCD – which would provide coverage for patients actively managed for chronic GERD who have two additional risk factors for BE – indicate that recognizes the clinical utility of, and intends to provide coverage for, tests subject to the proposed LCD that detect upper gastrointestinal, or more specifically, esophageal metaplasia and dysplasia (BE).

Tests that detect BE – which the American College of Gastroenterology (ACG) guidelines recognizes as the “only known precursor to esophageal adenocarcinoma” – are clinically useful, as they allow clinicians to start “endoscopic surveillance…[for] the detection of dysplasia or carcinoma at an early and treatable stage.³ However, the fifth coverage criterion in this section requires that the “test identifies patients with dysplastic disease that are more likely to benefit from invasive procedures from patients without dysplastic disease that are not likely to benefit.”

A nonendoscopic biomarker test designed to detect nondysplastic BE in at-risk patients with a prior diagnosis of chronic GERD, consistent with the ACG updated guideline, would not meet this criterion. We infer from the coverage criteria language and from questions MolDX posed to the invited experts at last year’s Carrier Advisory Committee (CAC) meeting, that MolDX also intends tests designed to detect progression to dysplasia in patients with a prior diagnosis of nondysplastic BE to be subject to this proposed LCD. Nonendoscopic biomarkers tests for nondysplastic BE detection in naïve at-risk chronic GERD patients and dysplasia progression biomarker tests for known nondysplastic BE surveillance are quite distinct in their clinical indication, target population and ideal performance criteria. For example, the ACG updated guideline recommends nonendoscopic biomarker tests to detect BE, one of which is commercially available, but is silent to dysplasia progression biomarker tests which are still in early-stage research.

To appropriately distinguish between these two distinct categories of tests and clarify that intends to cover assays that detect nondysplastic BE, we strongly encourage to revise the draft LCD.

3. Recommended addition of language from March 2022 American College of Gastroenterology (ACG) Updated Guideline

In March 2022, the ACG published an Updated Guideline for the diagnosis and management of BE.⁴ As excerpted below, Recommendation 5 of the Updated Guideline reiterates and strengthens prior guideline recommendations on endoscopic BE screening stating:

We suggest a single screening endoscopy in patients with chronic GERD symptoms and 3 or more additional risk factors for BE, including male sex, age >50 yr, White race, tobacco smoking, obesity, and family history of BE or EAC in a first-degree relative.

Recommendation 6, for the first time, supports nonendoscopic biomarker testing stating:

We suggest that a swallowable, nonendoscopic capsule device combined with a biomarker is an acceptable alternative to endoscopy for screening for BE.

We recommend to update the “Summary of Evidence” section of the LCD to reflect the updated guideline.

4. Recommended deletion of “ideal” test description

The “Analysis of Evidence (Rationale for Determination)” section of the proposed LCD states that a test “would ideally be highly sensitive (>95%), as false negatives may miss patients with dysplasia, and highly specific (>95%), meaning fewer patients without dysplasia would require follow-up procedures.” This language does not appear in an operative section of the proposed LCD – i.e., we do not believe that proposing that an assay must have both 95% sensitivity and 95% specificity to be eligible for coverage. However, to avoid confusion, we strongly recommend the deletion of the above-quoted language.

We strongly agree that Medicare should only cover assays that have established evidence of analytical validity, clinical validity, and clinical utility. The ideal performance criteria embedded in this evidence should balance risks and benefits to assure patients are not exposed to unnecessary risks and are not deprived of the test’s purported benefits, such as preventing esophageal cancer deaths through early detection of esophageal precancer. No benefit accrues to the public by establishing unrealistically and arbitrarily high test performance criteria not rooted in a well-informed risk benefit analysis. We are deeply concerned that this language could establish as a precedent, what may be an insurmountable biologic hurdle for any esophageal precancer test, which would deprive at-risk chronic GERD patients of the potentially life-saving benefits of early precancer detection using modern molecular diagnostic tools. We are not aware of any clinical guidelines, technology assessments, or published clinical literature suggesting that 95% sensitivity and specificity are the minimum performance characteristics for a clinically useful test for the early detection of upper gastrointestinal precancer or cancer in chronic GERD patients, or for any early cancer detection test, for that matter.

We urge to evaluate a test’s performance under this LCD based on the evidence provided, in the context of the test’s real-world risks and benefits, including the prevalence of the condition being tested in the target population and current practice with regard to detection of the condition. More specifically:

• Importance of prevalence in assessing risks and benefits. We note that Negative Predictive Value (NPV) and Positive Predictive Value (PPV), which incorporate disease prevalence and are better suited to an assessment of real-world test performance criteria than intrinsic measures such as sensitivity and specificity, particularly as it relates to balancing potential risks and benefits. For example, based on an estimated 5% to 10% prevalence of BE in the target at-risk chronic GERD population, the published 88% sensitivity and 92% specificity of EsoGuard on nonendoscopic balloon samples corresponds to a >99% NPV, which represents a false negative rate at or below other widespread early cancer detection tests, and an approximately 34% to 54% PPV, which represents a false positive rate also well below other widespread early detection tests.

• The potential risks associated with the test in the context of current practice. The risk of a false positive or false negative test result is also determined by what alternative testing, if any, the patient would actually undergo in current practice. Longstanding guidelines, including the latest ACG update, recommend endoscopy to detect non-dysplastic metaplasia, not just dysplasia, in at-risk GERD patients, to allow standard-of-care endoscopic surveillance to detect progression to dysplasia and maximize the opportunity for curative ablation to prevent further progression to esophageal cancer.⁵ However, studies have shown that less than 10% of patients recommended for BE screening undergo screening endoscopy.⁶ As a result, as last year’s Carrier Advisory Committee meeting experts noted, the risk associated with a false negative nonendoscopic biomarker test is very low because so few patients currently undergo endoscopy.⁷ Similarly, a false positive nonendoscopic biomarker test does not expose the patient to additional risk, since, according to the guidelines, the patient should have undergone an endoscopy if a nonendoscopic biomarker test was not available.

• The potential benefits of nonendoscopic biomarker test. Given longstanding poor adherence to guideline recommendations for BE screening, nonendoscopic biomarker tests covered under the proposed LCD have the potential to increase adherence from <10% to the over 50% level for colonoscopy and other early cancer detection tests. This in turn has the potential to prevent thousands of esophageal cancer deaths, each of which represents a uniquely tragic missed opportunity to detect, surveil and ablate BE. Triaging at-risk chronic GERD patients with a nonendoscopic biomarker test should also significantly reduce patient exposure to negative endoscopies.

References and images were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA and Noridian:

We are writing to express support for Medicare coverage of non-endoscopic, biomarker-based tests for esophageal neoplasia and metaplasia. These tests have the potential to reverse the alarming trends in esophageal adenocarcinoma (EAC) incidence and mortality through improved identification of patients with early-stage EAC and its precursors.

As practicing gastroenterologists, clinical guideline authors, and esophageal disease investigators, we are frontline witnesses to the devastating consequences of esophageal cancer. Despite considerable developments in endoscopic therapies over the past decade, a corresponding improvement in patient outcomes has not yet been realized because the vast majority of EAC cases are detected too late, beyond the stages where curative treatment is possible. Earlier detection of cancer – and detection and treatment of precancerous lesions to prevent cancer – is critical to improving outcomes. This has been demonstrated most successfully by widespread adoption of cervical cancer screening and by development of accurate tests for detection of precancerous cervical cells. Cervical cancer was the leading cause of cancer death for women prior to implementation of regular screening.

EAC arises from Barrett’s Esophagus (BE), a condition in which the normal epithelium of the lower esophagus is replaced by specialized intestinal metaplasia as an adaptive response to acid reflux and chronic gastroesophageal reflux disease (GERD). Endoscopic surveillance of patients diagnosed with non-dysplastic BE (NDBE) is imperative for early detection and treatment of dysplasia, in order to prevent further progression to invasive EAC.

To lessen the burden of esophageal cancer, more patients need to be identified within screening or surveillance settings, rather than upon presentation with symptoms. Currently, over 90% of EAC patients do not have previously recognized BE (due to not participating in preventive endoscopic surveillance programs) and present with advanced-stage malignancy. Despite recommendations by all major gastroenterology societies, only 10-15% of individuals at high risk for BE undergo screening by endoscopic evaluation.¹ Upper endoscopy is an invasive, expensive, and inconvenient procedure which requires trained endoscopists and anesthesia, making it impractical for widespread use. To increase identification of patients with esophageal neoplasia and BE, and to make an impact on esophageal cancer outcomes, we need detection and surveillance methods which are cost effective, safe, accurate, and acceptable to patients. Non-endoscopic, biomarker-based tests meet these criteria. The American College of Gastroenterology’s recently updated guideline for Diagnosis and Management of Barrett’s Esophagus recognizes this and suggests that these tests are now an acceptable alternative to endoscopy for BE screening.²

Over the past decade there have been significant advances in these non-endoscopic, biomarker tests for the detection of BE with or without associated dysplasia/neoplasia.^3-10 Non-endoscopic tests for BE detection would have utility in those with risk factors for BE such as chronic reflux. Biomarkers have been identified which can detect BE, with or without dysplasia, as well as EAC with high sensitivity and specificity (≥ 90%), thereby enabling the identification of patients who need diagnostic endoscopy with biopsy and histopathology to confirm the presence of BE (with or without associated dysplasia). Clinical management then depends on the patient’s diagnosis and may include treatment of dysplasia and/or surveillance for progression at interval that is risk stratified.

Notably, biomarkers have not yet been validated which can distinguish among these disease stages in patients with BE (NDBE vs. LGD vs. HGD vs. EAC), which would enable development of a test for the non-endoscopic surveillance for disease progression of patients diagnosed with NDBE. Consequently, limited evidence currently exists to support the use of non-endoscopic tests in a surveillance population. We are hopeful that the vision of non-endoscopic biomarker tests for surveillance will be achieved in the coming decade, as this could further lessen the endoscopy burden for some patients. Nevertheless, the importance of diagnosing all BE (including NDBE), remains paramount – since all NDBE patients should be in endoscopic surveillance programs where progression to LGD or HGD can be detected early and treated effectively with endoscopic eradication therapy.

For now, we see the promise of biomarker tests for detection of BE (with or without dysplasia) as a potential paradigm shift in the fight against esophageal cancer. The National Health System in the UK also recognizes the value of these tests and has therefore introduced Cytosponge-TFF3 in clinics across Scotland and England. Advances like these are critical to making progress in cancer outcomes, as exemplified in the Biden administration’s Cancer Moonshot goal of reducing the cancer death rate by at least 50 percent over the next 25 years.¹¹ To ensure that patients have access to novel tests like these – and that the cervical cancer success story can be replicated – Medicare coverage is critical for tests which are validated and supported by robust evidence generated in the intended-use population.

References were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA and Noridian:

I am pleased to submit comments regarding the above-captioned proposed local coverage determination (LCD).

My primary professional quest, over more than three decades, has been to save lives through the early detection of precancer and cancer. During this period, I have worked tirelessly as an inventor, entrepreneur, chief executive, board director, corporate and scientific advisor (including as an advisor and director of Lucid Diagnostics Inc.) to advance groundbreaking diagnostic technologies towards widespread clinical use, including technologies which have been credited with preventing hundreds of thousands of cancer deaths through early detection.

The most rigorous guidelines for preventive care, including precancer and cancer screening, are undoubtedly those published by the United States Preventive Services Task Force (USPSTF). Very few in-vitro diagnostic tests for cancer screening have satisfied the stringent criteria that must be fulfilled to be recommended by the USPSTF. I was the driving force behind two of them—the ThinPrep Pap Test for early detection of cervical cancer and the Cologuard test for early detection of colorectal cancer.

ThinPrep became the standard Pap test in the US and much of the world and has been credited with reducing cervical cancer incidence and mortality by over 40% since it was introduced.¹ The original ThinPrep prototypes are part of a collection of historically significant objects at the Smithsonian’s National Museum of American History.

Cologuard combines an assay detecting alterations in stool DNA with an immunochemical test to detect occult blood in the stool. Cologuard is one of the most successful tests in the history of the diagnostics industry.

Lucid Diagnostics’ EsoGuard^® Esophageal DNA Test, which lies within the scope of this LCD, and Cologuard address similar unmet clinical needs and share similar clinical and biologic characteristics. As such, I would first like to describe how Cologuard succeeded in addressing an important unmet clinical need and its performance criteria to meet this need.

If accuracy is the principal goal, colorectal cancer screening is ideally performed by colonoscopy. Because colonoscopy is invasive, requires highly skilled physicians and is fundamentally unpleasant, it is typically performed every ten years. Periodic screening colonoscopy, performed according to clinical practice guideline recommendations, lowers colorectal cancer incidence by detecting precancerous adenomatous polyps and lowers colorectal cancer mortality by shifting the stage at diagnosis from Stages III and IV to Stages I and II.

Yet, despite its clinical effectiveness, screening colonoscopy has not eliminated colorectal cancer deaths. Between 40% and 60% of individuals recommended for screening decline colonoscopy altogether. Some do undergo an index screening colonoscopy but fail to follow up for periodic screening colonoscopies as recommended. Finally, even those who strictly follow recommended periodic screening protocols are at risk of developing interval cancer between screening colonoscopies.

I co-developed and commercialized Cologuard to address this important unmet clinical need and further reduce colorectal cancer mortality.

In its pivotal study,² Cologuard’s sensitivity and specificity for colorectal cancer were 92% and 88%, respectively. Sensitivity for high-grade dysplastic polyps and advanced adenomas were 69% and 42%, respectively. Based on these performance data, Medicare payment and coverage was established, and commercial payors quickly followed suit. Cologuard was subsequently included in USPSTF guidelines as an acceptable alternative to colonoscopy. No other laboratory-based cancer screening test recommended in clinical practice guidelines has performance characteristics (sensitivity and specificity) that even approach Cologuard’s. Although Cologuard’s clinical impact is greatest in those who do not undergo screening colonoscopy, its clinical benefit includes effectively detecting interval adenomas and cancers between screening colonoscopies.

It is important to note that the false negative risk to the patient (missing a colorectal cancer or advanced adenoma) of utilizing Cologuard as alternative to standard of care endoscopy is most appropriately characterized by its negative predictive value (NPV). Cologuard’s sensitivities of 92% and 42% translates into NPVs of >99% for colorectal cancer and 94% for colorectal cancer and advanced adenomas. These NPVs are widely accepted to be sufficient (i.e., indicative of an acceptable false negative risk) for Cologuard to serve as a widespread colorectal cancer screening tool in patients recommended for screening.

Similarly, Cologuard’s false positive risk (unnecessary colonoscopy) is most appropriately characterized by its positive predictive value (PPV). Cologuard’s specificities of 88% and 42% translate into PPVs of 4% for cancer and 24% for either cancer or advanced adenoma, respectively. These PPVs are widely accepted to be sufficient (i.e., indicative of an acceptable false positive risk) for Cologuard to serve as a widespread colorectal cancer screening tool. In fact, one can argue that there is effectively no false positive risk for ColoGuard since the procedure being performed in response to the false positive, a colonoscopy, is one that should have been performed, based on well-established clinical practice guideline recommendations if the patient had not undergone Cologuard testing.

I provide this detailed background on Cologuard’s performance and clinical utility in preventing colorectal cancer deaths to serve as a baseline from which we can evaluate EsoGuard’s in preventing esophageal cancer deaths. I believe the case for EsoGuard is as, if not more compelling, than Cologuard’s.

Esophageal cancer is deadlier than colorectal cancer across all stages. Perhaps most notably, five-year survival for localized (Stage I/II) esophageal cancer is dismal at <50% compared to >90% for colorectal cancer.³ So, in contrast to colorectal cancer, where detecting Stage I/II cancer can lead to a high cure rate and prevent colorectal cancer deaths, the only opportunity to prevent esophageal cancer deaths is to detect, monitor and treat esophageal precancer (nondysplastic metaplasia or dysplasia). Detecting Stage I/II esophageal cancer, on the other hand, is wholly insufficient to meaningfully prevent deaths. Again, to prevent esophageal cancer deaths, one must detect esophageal precancer, to have the opportunity to monitor nondysplastic metaplasia and treat dysplasia before it progresses to adenocarcinoma.

The need to detect esophageal precancer to prevent esophageal cancer deaths has been well recognized for decades. For over a decade, published clinical practice guidelines, including by American College of Gastroenterology (ACG), have recommended GERD patients with well-established risk factors undergo testing to detect esophageal precancer. Until now, these recommendations were to use the only available tool, endoscopy with biopsies. The most recent ACG Updated Guideline reiterates the long-standing recommendation for esophageal precancer testing in at-risk GERD patients and recommends nonendoscopic biomarker tests such as EsoGuard as an acceptable alternative to endoscopy with equivalent strengths of evidence.

One of the most significant differences between Cologuard and EsoGuard is the baseline rate of adherence to guideline recommendations for endoscopy in at-risk patients. At least 50% of patients recommended for testing to detect colorectal cancer or precancer undergo colonoscopy. Cologuard targets the other 50% as an alternative to colonoscopy. In stark contrast, <10% of patients recommended for testing to detect esophageal precancer undergo upper GI endoscopy (EGD). The likely reason for poor adherence is that EGD is performed under intravenous sedation or anesthesia in a specialized facility and that both the patient and someone accompanying them must take a day off; general lack of understanding of the relationship between GERD and esophageal cancer is also a key factor. Ultimately, the reasons are secondary since the recommendations have been in place for over a decade. Whatever the reason, the bottom line is that at-risk patients are not getting tested despite guideline recommendations going back a decade, resulting in approximately 16,000 preventable deaths per year, at least half in Medicare beneficiaries.

A seminal study published in Science Translational Medicine by a highly respected group of investigators led by a long-time colleague, collaborator and renowned cancer geneticist, Dr. Sandy Markowitz, characterized EsoGuard’s performance characteristics.⁴ EsoGuard’s sensitivity and specificity for esophageal precancer and cancer were 95% and 91% respectively (88% and 92% for non-endoscopic balloon capsule samples). The sensitivity was stable across the disease spectrum—94% for esophageal precancer and 96% for esophageal cancer (both 88% for non-endoscopic balloon capsule samples).

As with Cologuard, the false negative risk to the patient (missing an esophageal precancer or cancer) of utilizing EsoGuard as alternative to standard of care EGD is most appropriately characterized by its NPV. Based on an estimated prevalence of 5-10% in the target at-risk population, EsoGuard’s sensitivities of 95% and 88% (nonendoscopic balloon samples) both translate into NPVs of >99%.

Similarly, EsoGuard’s false positive risk (negative EGD) is most appropriately characterized by its positive predictive value (PPV). EsoGuard’s specificities of 91% and 92% (nonendoscopic balloon samples) translate into a PPV range of ~34% to 54%.

This brings us to the language in the proposed LCD that indicates a test in this category would

“…ideally be highly sensitive (>95%), as false negatives may miss patients with dysplasia, and highly specific (>95%), meaning fewer patients without dysplasia would require follow-up procedures.”

I would like to respectfully offer several vigorous objections to this statement.

1. No existing cancer diagnostic test, including well established, widely utilized screening tests that have been shown to reduce mortality (and are covered by Medicare), have performance characteristics that approach these thresholds. In fact, given biologic variability and other factors, no future cancer test, at least one utilizing currently available or contemplated technologies, is likely ever to reach those thresholds. Including this language in the final LCD would establish as a precedent what is likely an insurmountable biologic hurdle, which would deprive patients of the life-saving benefits of early cancer detection. No benefit accrues to the public by establishing unrealistically and arbitrarily high-performance criteria for a test that are not rooted in an appropriate risk benefit analysis.
2. Acceptable performance criteria for any test must consider the prevalence of the condition being tested, alternatives to the test being assessed, and the risks of false positive and negative test results relative the clinical benefits of detecting the condition—in this case preventing esophageal cancer deaths through early detection, monitoring, and treatment of esophageal precancer. The most appropriate metrics are therefore the NPV and PPV of the test.
3. If an NPV of 94% for colorectal cancer or precancer is widely accepted to be sufficient (i.e., indicative of an acceptable false negative risk) for Cologuard to serve as a widespread early colorectal cancer and precancer detection tool to prevent colorectal cancer deaths, then certainly an NPV of 99% should be sufficient for EsoGuard to serve as an esophageal precancer detection tool to prevent esophageal cancer deaths.

In fact, there is a compelling argument that the acceptable NPV should be lower for EsoGuard than Cologuard, based on the marked difference in adherence rates for the endoscopic alternatives to each nonendoscopic biomarker tests, which is the primary driver of false negative risk. Many, if not most patients who receive a false negative Cologuard result would have undergone colonoscopy if Cologuard had not be performed. The colonoscopy likely would have detected the colorectal cancer or precancer and offered a real possibility of a cure.

In contrast, fewer than 10% of patients who receive a false negative EsoGuard result would have undergone EGD, and their precancer or cancer would not have been detected, if EsoGuard were not performed. In fact, the NPV of an esophageal precancer test can be quite a bit lower than 99% and still demonstrate significant clinical utility. This was demonstrated in a very large UK population study⁵ which found that a nonendoscopic immunohistochemical biomarker test, performed on samples collected with a sponge capsule, was able to identify many esophageal precancers and even some cancers, even though this test’s performance criteria (sensitivity 77%, specificity 76%)⁶ falls far short of EsoGuard’s.

4. Similarly, if a PPV of 24% for colorectal cancer and precancer is widely accepted to be sufficient (i.e., indicative of an acceptable false positive risk) for Cologuard to serve as a widespread colorectal cancer and precancer detection tool, then a PPV range of 34% to 54% should be sufficient for EsoGuard to serve as an esophageal precancer detection tool to prevent esophageal cancer deaths.

In fact, as is the case for Cologuard, one can argue that there is effectively no false positive risk for EsoGuard since the procedure being performed in response to the false positive, an EGD, is one that should have been performed if the patient had not undergone EsoGuard testing, based on well-established clinical practice guideline recommendations,

5. Finally, I believe this statement, as well as the fifth criterion in the beginning of the proposed LCD, conflate or combine two very distinct types of tests and the clinical situation to which they are applicable.

Early detection tests, such as EsoGuard, are designed to detect all conditions along the esophageal precancer-cancer spectrum (nondysplastic metaplasia, dysplasia, and adenocarcinoma) in at-risk naïve GERD patients, without a prior diagnosis of conditions along this spectrum. Consistent with flowchart algorithms in guidelines, positive EsoGuard patients are expected to undergo a confirmatory EGD to determine the clinical plan—nondysplastic metaplasia is surveilled with follow-up EGDs, generally every three years; dysplasia is generally ablated endoscopically; and adenocarcinoma is endoscopically or surgically resected. EsoGuard serves as noninvasive test to triage patients to EGD with a substantially higher prior probability than proceeding directly to EGD, which allows most patients to avoid an invasive endoscopy.

Cologuard, as an early detection test, is utilized in precisely the same way. Positive Cologuard patients are expected to undergo a confirmatory colonoscopy to determine the clinical plan—most polyps are endoscopically resected; cancers and some large adenomas are surgically resected. Cologuard serves to triage patients to colonoscopy with a higher prior probability than proceeding directly to colonoscopy, which allows most patients to avoid an invasive endoscopy. The clinical utility of this guideline-based care algorithm is well-established even though its sensitivity for even high-grade dysplasia is only 69%, nowhere close to the 95% bar in the proposed LCD’s “ideal” sensitivity, or EsoGuard’s 94%/88% sensitivity for esophageal precancer.

The same argument holds for the specificity of early detection tests such as EsoGuard. The guidelines recommend testing in patients with a prior diagnosis of GERD to detect nondysplastic metaplasia, not just dysplasia, so nondysplastic metaplasia can be surveilled with serial EGD and progression to dysplasia can be picked up and ablated before it progresses to adenocarcinoma. So, “…meaning fewer patients without dysplasia would require follow-up procedures” is simply not a goal of these tests. In fact, the goal is precisely to identify patients without dysplasia.

Although none are yet commercially available, esophageal progression biomarker tests are being developed and are entirely different than early detection tests, such as EsoGuard, designed to detect nondysplastic metaplasia in GERD patients. Progression biomarker tests are designed to be performed in patients with a prior diagnosis of nondysplastic metaplasia (not GERD) as an alternative to (or between) surveillance EGDs to detect progression to dysplasia. For these tests, having a sufficient PPV “…meaning fewer patients without dysplasia would require follow-up procedures” is in fact a goal.

For the above reasons, I would therefore strongly recommend the following modifications to the proposed LCD.

1. I recommend that the language that indicates a test in this category would “ideally be highly sensitive (>95%), as false negatives may miss patients with dysplasia, and highly specific (>95%), meaning fewer patients without dysplasia would require follow-up procedures” be removed. The appropriate performance criteria for tests subject to this LCD should balance the test’s actual risks (false negative and false positive) and benefits (early precancer detection to prevent cancer deaths) to assure patients are not exposed to unnecessary risks and are not deprived of the test’s purported benefits, in this case preventing cancer death through early detection of precancer. Assessment of the test’s risks and benefits should incorporate prevalence of the condition being tested in the target population (NPV/PPV) and the real-world alternatives to testing (e.g., endoscopy) based on current practice regarding detection of the condition, including the rate of adherence to guideline recommendations for endoscopy in at-risk patients.

2. I recommend that the fifth enumerated criteria be modified to appropriately distinguish between early detection tests in naïve GERD patients and progression tests in patients with established nondysplastic metaplasia.

3. I recommend that proposed LCD be updated to reflect the recently released American College of Gastroenterology Updated Guideline which recommends nonendoscopic biomarker tests, such as EsoGuard, as an acceptable alternative to endoscopy

Thank you for the opportunity to participate in this proposed foundational LCD’s public review process. I am looking forward to following the LCD process to completion. I sincerely hope that the final foundational LCD will provide physicians with the opportunity to use modern molecular diagnostic tools to have the same or greater impact on esophageal cancer deaths in Medicare beneficiaries as ThinPrep Pap testing and Cologuard stool DNA testing have had more broadly on cervical and color cancer deaths, respectively.

References were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA:

Thank you for the opportunity to comment on the Proposed LCD - MolDX: Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia. With an increase of >500% incidence in esophageal adenocarcinoma (EAC) over the past five decades, EAC is the fastest growing cancer in the Western World.¹ Further, even if diagnosed at Stage 1, EAC has a dismal 5-year survival rate of only 47%.² However, if diagnosed while Barrett’s esophagus (BE), numerous studies have demonstrated that 5-year survival increases to >90% in those that receive endoscopic treatment.³ Thus, it is important that the proposed LCD be revised to support the early detection of BE. Supporting the early detection of BE aligns with President Biden’s Moonshot goal to reduce the US cancer death rate by 50%.

Capsulomics is commercializing non-endoscopic, biomarker-based tests for the detection of Barrett’s esophagus, developed by researchers at Johns Hopkins University. Technologies such as ours were invented by leading physicians and researchers determined to cause a paradigm shift, enabling GI physicians and patients the opportunity to receive an early diagnosis and clinical guideline suggested diagnosis, management, and endoscopic treatment, thus saving lives, and improving outcomes. Recognizing the impact of early detection of BE, the American College of Gastroenterology (ACG) has recently updated their guideline for the Diagnosis and Management of BE to state that a swallowable non-endoscopic capsule device combined with a biomarker is an acceptable alternative to endoscopy for Barrett’s esophagus (BE).⁴

Providing Medicare beneficiaries access to tests for the early detection of BE is critical to enable high- risk patients to receive this clinically recommended diagnostic test. Please review and consider the following changes to the Proposed LCD.

Comments:

Current molecular diagnostic tests that identify individuals with upper gastrointestinal metaplasia, dysplasia, and neoplasia are covered when the following criteria are met.

Comment: Please update this language to establish coverage if the test meets the following criteria. This will allow MolDX to cover future tests without having to open the LCD for revision in the future.

We expect these types of tests to meet the following criteria:

1. The beneficiary is being actively managed for chronic gastroesophageal reflex disease (GERD) and/or non-dysplastic Barrett’s esophagus (NDBE); and
2. They also have at least two additional risk factors for Barrett’s esophagus (BE) as described in nationally recognized guidelines.
3. The beneficiary has not been previously diagnosed with dysplasia or esophageal carcinoma; and
4. The beneficiary has not been previously tested by the same test within three years; and
5. The test either
   1. (a) identifies patients with non-dysplastic or dysplastic BE that would benefit from an appropriate procedure, OR
   2. (b) identifies patients with dysplastic disease that are more likely to benefit from invasive procedures from patients without dysplastic disease that are not likely to benefit

Comment: Requiring tests to be able to differentiate patients who have dysplastic vs. non-dysplastic disease does not support the overall goal of early detection. In fact, published data demonstrates that ignoring nondysplastic patients guarantees future EAC deaths.⁵ It is vital that all patients with Barrett’s esophagus need to be endoscopically evaluated to visualize disease and identify segment length as that along with other risk factors may indicate risk of progression and a provider’s decision making around surveillance.⁶ All individuals who meet criteria one and two above should be tested; patients who receive a positive test result should then be referred for EGD and managed by a GI physician.

     6. Test results will be used in determining treatment or management of the

Comment: Patients who receive the test will be triaged; positive test results will be referred for more invasive diagnostic endoscopy

    7. The beneficiary is within the population for which the test was developed and The laboratory providing the test is responsible for clearly indicating to treating clinicians the population and indication for test use.

    8. The test demonstrates analytical validity (AV) including an analytical and clinical validation for any given measured analytes and has demonstrated equivalence or superiority for sensitivity or specificity of detecting dysplasia to other already accepted methods for the same intended use measuring the same or comparable analytes.

    9. Clinical validity (CV) of any analyte measured must be demonstrated in the published peer- reviewed literature, establishing a clear and significant biological/molecular basis for stratifying patients and subsequently selecting (either positively or negatively) their clinical management decision within a clearly defined population.

Comment: Clinical validation should be performed in the population listed in criteria one and two and demonstrate adequate performance such as those listed in the American College of Gastroenterology updated clinical guidelines for Diagnosis and Management of Barrett’s Esophagus.

   10. The test successfully completes a Technical Assessment that will ensure that criteria set in this policy are met to establish the test as Reasonable and Necessary.

In this case, the test would ideally be highly sensitive (>95%), as false negatives may miss patients with dysplasia, and highly specific (>95%), meaning fewer patients without dysplasia would require follow-up procedures.

Comment: We request that you strike the above criteria for sensitivity and specificity of 95%. This is an arbitrary idealistic number that is not based on published guidelines or literature.

Although there are several promising molecular biomarker tests designed to further identify at-risk patients, this contractor finds that there are currently no existing tests that have demonstrated AV, CV, and clinical utility (CU) to fulfill the necessary criteria. This contractor will continue to monitor the evidence and will provide coverage based on the pertinent literature and society recommendations.

Comment: This statement is unnecessary and should be stricken following the same reasoning as the above remarks.

Lastly, as written, the LCD is a non-coverage determination, rather than “coverage if the following criteria are met…,” which is problematic and different from what we’ve seen in categorical LCDs for other types of tests.

References were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA:

I have been utilizing the EsoGuard test in accordance with the recently released American College of Gastroenterology (ACG) guidelines that state that a swallowable non-endoscopic capsule device combined with a biomarker is an acceptable alternative to endoscopy for Barrett’s esophagus (BE).¹ We know that many patients, both at our institution and throughout the US, have not been adequately screened for BE as they do not have access to a convenient, minimally invasive, and low-cost screening test with rapid result turnaround.

I have been using the EsoGuard test for nearly one and a half years, and I have performed nearly 500 tests for our patients. The EsoGuard test has been a vital tool which I utilize to assist making the right decision regarding appropriate candidates for endoscopic evaluation. To ensure that patients have appropriate access to care I would ask that you consider altering the coverage criteria to reflect the need for patients with BE to be tested and screened, and identify intestinal metaplasia early, potentially decreasing the risk of progression to dysplasia or malignancy. The clinical utility testing for the early detection of upper gastrointestinal metaplasia, dysplasia, and neoplasia using esophagogastroduodenoscopy (EGD) has been validated in the scientific literature and supported in clinical practice. Recognizing the clinical value of early detection as demonstrated by EGD, the ability to offer a less-invasive alternative to EGD for patients to detect the main precursor to cancer allows me to tailor more personalized treatment decisions for patients. The use of EsoGuard in clinical practice would indeed be a significant advancement for patients.

Based upon this, my colleagues and I suggest you consider modifying the proposed LCD coverage criteria that states that the test identifies patients with dysplastic disease that are more likely to benefit from invasive procedures from patients without dysplastic disease² to reflect the need to assess patients for BE, as per the updated ACG guidelines. Making such a change will allow clinicians the ability to offer an alternative to EGD for patients to detect the main precursor to cancer allow providers to make personalized treatment decisions for their patients. This would be a significant clinical advance for patients and health care providers.

As I hope is apparent, coverage of tests for the early detection of BE are greatly needed, as this technology, especially the EsoGuard test, broadens our capacity to screen, and help selected patients avoid unnecessary invasive procedures. Importantly, this proposed change will help increase the screening and promote earlier detection of precursors of esophageal adenocarcinoma – a malignancy with dramatic morbidity and mortality in this country.

References were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA:

On behalf of the Coalition for 21st Century Medicine (C21), thank you for the opportunity to submit comments regarding the above-captioned proposed local coverage determination (LCD). C21 comprises many of the world’s most innovative diagnostic technology companies, clinical laboratories, physicians, venture capital companies, and patient advocacy groups. C21’s mission is to improve the quality of health care by encouraging research, development, and commercialization of innovative diagnostic technologies that will personalize patient care, improve patient outcomes, and substantially reduce health care costs.

For more than ten years, C21 has worked with the MolDX program on the development, promulgation, and implementation of policies intended to facilitate appropriate Medicare coverage and reimbursement for high-quality clinical laboratory tests. C21 has a keen interest in the above-captioned proposed LCD, as multiple C21 members are developing tests that would be subject to its terms (if finalized).

In general, C21 supports the content of this foundational LCD, which would create a pathway to Medicare coverage for molecular tests for upper GI metaplasia, dysplasia and neoplasia that meet the criteria set forth in the policy. However, the proposed LCD also includes language that if finalized, could pose a significant obstacle to timely Medicare coverage for these tests. Specifically, the operative “Coverage Indications, Limitations, and/or Medical Necessity” section of the proposed LCD states:

Current molecular diagnostic tests that identify individuals with upper gastrointestinal metaplasia, dysplasia, and neoplasia are non-covered by this contractor.

We understand this language reflects the MolDX program’s current position with respect to coverage of these tests. However, this language is not appropriate for inclusion in a foundational LCD, as it would arguably restrict MolDX’s ability to cover tests in the future – even after such tests successfully complete the Technical Assessment (TA) process. Recent experience suggests that it could take MolDX a substantial period of time – 12 to 18 months, if not more – to issue an updated LCD under the procedures required by 21^st Century Cures. To ensure Medicare beneficiaries have timely access to validated, high-quality assays that have successfully completed the TA process, C21 strongly recommends that MolDX revise the proposed language to (a) remove the language identifying all current tests as non-covered, and (b) revise the remaining language to clarify MolDX’s intent to cover tests in the future if they meet the criteria set forth therein. Potential revised language could look like the following:

Coverage Indications, Limitations, and/or Medical Necessity

diagnostic tests that identify individuals with upper gastrointestinal metaplasia, dysplasia, and neoplasia are covered when ALL of the following criteria are met:

• The beneficiary is being actively managed for chronic gastroesophageal reflex disease (GERD) and/or non-dysplastic Barrett’s esophagus (NDBE); and
• They also have at least two additional risk factors for Barrett’s esophagus (BE) as described in nationally recognized guidelines.
• The beneficiary has not been previously diagnosed with dysplasia or esophageal carcinoma; and

Taking such an approach would be consistent with established MolDX precedent, as MolDX used similar language in its recently-finalized “Molecular Syndromic Panels for Infectious Disease Pathogen Identification Testing” and “Molecular Testing for Solid Organ Allograft Rejection” LCDs, respectively.

Similarly, in the “Analysis of Evidence (Rationale for Determination)” section, we encourage MolDX to delete the language noting that “this contractor finds that there are currently no existing tests that have demonstrated AV, CV, and clinical utility (CU) to fulfill the necessary criteria.” While we understand that this language is not operative for purposes of defining the MolDX coverage criteria, we believe its deletion from the final LCD would be appropriate, as such language will be inaccurate once MolDX approves a test for coverage.

Thank you for considering our comments.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA:

Will the Cytosponge diagnostic test be covered under the proposed LCD, especially in light of the recommendations in the recently published ACG guidelines?

Shaheen et al. Diagnosis and Management of Barrett's Esophagus: An Updated ACG Guideline, The American Journal of Gastroenterology: April 2022 - Volume 117 - Issue 4 - p 559-587, doi: 10.14309/ajg.0000000000001680:

“We suggest that a swallowable, nonendoscopic capsule sponge device combined with a biomarker is an acceptable alternative to endoscopy for screening for BE in those with chronic reflux symptoms and other risk factors.”

The Cytosponge test is a minimally invasive diagnostic test that offers a way to transform care for patients at risk of esophageal cancer by detecting the precancerous condition, Barrett’s Esophagus (BE), as well as dysplasia and neoplasia. The Cytosponge test is listed in the guidelines as having the “greatest experience and largest available evidence”, clearly emphasizing the use of sponge-based sampling devices.

The Cytosponge has been tested in 5 prospective clinical studies over the past 18 years, including the BEST1, BEST2 and BEST3 trials, and has been in widespread real-world use in the UK National Health Service since 2020. The results of the landmark BEST3 trial showed that the Cytosponge-TFF3 test diagnosed 10 times more cases of BE than standard of care. Among the 131 cases of BE diagnosed, 9 had dysplasia (n=4) or stage 1 esophago-gastric cancer (n=5), whereas no participants were diagnosed with dysplastic BE or stage 1 esophago-gastric cancer in the usual care group. These cases all had successful endoscopic therapy, negating systemic chemotherapy required for advanced cancer. The primary endpoint of BEST3 was evaluating the positive predictive value, found to be 59%. Since 2020, the Cytosponge test has been used in over 8000 patients across England and Scotland, facilitated through centralized laboratory testing to ensure quality control. Centralized laboratory testing mitigates issues around biomarker review and standardization of diagnosis demonstrated by real-world data (Tang et al, 2022).

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA:

The Esophageal Cancer Action Network, ECAN, is non-profit patient advocacy group dedicated to saving the lives of those at risk for Esophageal Cancer. ECAN was founded in 2009, when Esophageal Adenocarcinoma was the fastest increasing cancer in the United States – having increased about 700% in recent decades. It remains one of the fastest increasing cancers among American men.

Our board of directors includes many of the top oncologists, gastroenterologists, and surgeons who treat and prevent Esophageal Cancer in the U.S. ECAN is the leading voice for these patients on Capitol Hill. We publicly raise the alarm that the acid reflux that most consider benign can be deadly. We support and educate patients so that they can make the best decisions for their own health. Unfortunately, for those patients who seek to be proactive in preventing Esophageal Cancer, the options are dramatically limited. A sedated upper endoscopy is expensive and is not without risk. For most patients, it is not within reach.

Most cases of Esophageal Cancer are detected at late stages when treatment is rarely life-saving. So it is no surprise that only one out of five of the approximately 20,000 Americans diagnosed with Esophageal Cancer each year will survive five years. Most don’t live for more than 12 months. The vast majority of these patients are diagnosed with Esophageal Adenocarcinoma.

Until recently, the treatment available for Esophageal Cancer patients hadn’t changed in more than a decade. We can hope that approvals for new immunotherapies last year will eventually improve those numbers. But only a small percentage of our patients benefit from these new therapies. And, even if they survive, the treatment and surgery to remove their esophagus that Esophageal Cancer patients usually undergo will impact their health as long as they live. We all know that an ounce of prevention is worth a pound of cure. It is a cliché because it is true.

Our organization believes that the advent of new, minimally invasive detection tests that can find Barrett’s Esophagus and Esophageal Adenocarcinoma without sedation is the most effective way to save the lives of those at risk for Esophageal Cancer. When detected as its only known precursor, Barrett’s Esophagus, effective treatment is widely available that can nearly eliminate the risk of developing this deadly cancer.

But the mere existence of the tests will not save lives. They must be accessible. And that is why your decision regarding coverage of molecular testing for detection of upper gastrointestinal metaplasia, dysplasia, and neoplasia is so critical. Your decision has the power to save the lives of many thousands of Americans. More than 16,000 patients die of Esophageal Cancer every year.

We believe patients at risk for that devastating fate deserve to benefit from the unprecedented scientific advancements in the field of biomarker testing that these new early detection tests represent. The ability to detect this silent killer with convenient, low cost, and quickly responsive tests is the solution we have been seeking for more than a decade. To deprive those at risk of the opportunity to access these tests for lack of coverage would be a travesty.

ECAN is part of a chorus that supports this conclusion. The recently released American College of Gastroenterology Guidelines support the use of a non-endoscopic capsule device combined with a biomarker as an acceptable alternative to endoscopy for screening for Barrett’s Esophagus. Detecting cancer early aligns with the President’s stated NCI Cancer Moonshot goal to reduce the death rate from cancer by at least 50 percent over the next 25 years.

We recognize that your decision will have a broad impact on a variety of molecular testing that is expected to come to market in the coming years. We ask that your standards make it possible for tests that meet reasonable requirements as compared with other such diagnostic tests be provided with coverage.

Today, we specifically ask that coverage be provided for the EsoGuard test based on a report from Case Western Reserve University published in Science Translational Medicine in 2018¹ that demonstrated a sensitivity of 90% and a specificity of 90%.

In the process of diagnosing patients at risk for this deadly disease, these molecular diagnostic tests are an important first step. But they should not be considered the last. Patients with a positive result would undergo an upper endoscopy for confirmation and more detailed biopsies. Only then could an appropriate treatment plan be developed. To expect more from an initial screening test seems unreasonable to us.

We appreciate the opportunity to comment on the Proposed LCD - MolDX: Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia and ask that you make a positive change in the coverage policy.

Thank you for your consideration.

Reference was provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA:

In my current position, I provide care to many patients with Barrett’s esophagus (BE), and I have personally witnessed the life-saving value of early BE detection, upon diagnosis of which I enroll patients in endoscopic surveillance programs, as recommended by national GI medical societies.

I am also a longstanding researcher on the field of Barrett’s esophagus (BE). I have developed epigenetic assays for both early detection of BE and prognostic assays to predict the future risk of neoplastic progression in BE patients. Furthermore, I have served as an author on clinical recommendations for the detection and management of Barrett’s esophagus.

Since >95% of esophageal adenocarcinomas present with previously undiagnosed BE and at advanced, incurable neoplastic stages, it is vitally important that you reconsider the coverage of widespread, inexpensive, convenient, safe, minimally-invasive methods of detecting both dysplastic and non-dysplastic BE. Such methods will enable GI physicians like me to diagnose and treat many additional patients who have not undergone BE screening (and will not likely receive it in the future).

Please consider the following comments regarding Proposed LCD - MolDX: Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia:

1. The beneficiary has been diagnosed with gastroesophageal reflex disease (GERD); and according to all major GI society guidelines, patients with GERD and other risk factors should be screened for BE.¹
2. They also have at least two additional risk factors for Barrett’s esophagus (BE) as described in nationally recognized guidelines.
3. The beneficiary has not been previously diagnosed with dysplasia or esophageal carcinoma; and
4. The beneficiary has not been previously tested by the same test within three years; and
5. The test either

         1. (a) identifies patients with BE that would benefit from an appropriate procedure, OR

         2. (b) identifies patients with dysplastic disease that are more likely to benefit from invasive procedures from patients without dysplastic disease that are not likely to benefit.

I disagree with this statement. Dysplastic patients are not "more likely to benefit" from EGD: rather, all BE patients should be detected BEFORE dysplasia develops, so that they can be entered into surveillance programs.¹ Ignoring all NDBE patients guarantees future EAC deaths.² Similar to Cologuard and FIT testing, all malignant esophageal cancer and premalignant lesions should be detected in the GERD population described above; patients who are tested and found to be positive should then be referred for EGD and management by a GI specialist physician.

      6. Test results will be used in determining treatment or management of the beneficiary.

Yes, test results will indeed be used in determining patient management. A positive initial test will reflex a patient to undergo diagnostic EGD. There is no need for the test to distinguish among histologic stages of BE, since all stages diagnosed by the test will mandate endoscopic diagnosis and subsequent surveillance.

       7. The beneficiary is within the population for which the test was developed and validated. The laboratory providing the test is responsible for clearly indicating to treating clinicians the population and indication for test use.

       8. The test demonstrates analytical validity (AV) including an analytical and clinical validation for any given measured analytes and has demonstrated equivalence or superiority for sensitivity or specificity of detecting dysplasia to other already accepted methods for the same intended use measuring the same or comparable analytes.

Again, I disagree here. The capsule/biomarker test does not have to demonstrate equivalent accuracy to EGD, since EGDs cannot be (and are not currently) performed in the entire at-risk population.

       9. Clinical validity (CV) of any analyte measured must be demonstrated in the published peer-reviewed literature, establishing a clear and significant biological/molecular basis for stratifying patients and subsequently selecting (either positively or negatively) their clinical management decision within a clearly defined population.

      10. The test successfully completes a Technical Assessment that will ensure that criteria set in this policy are met to establish the test as Reasonable and Necessary.

Furthermore, as written, the “Coverage Indications, Limitations, and/or Medical Necessity” section of the proposed LCD states that “current molecular diagnostic tests that identify individuals with upper gastrointestinal metaplasia, dysplasia, and neoplasia are non-covered by this contractor.” Similarly (though in a non-operative section of the proposed LCD), the “Analysis of Evidence (Rationale for Determination)” section concludes that “although there are several promising molecular biomarker tests designed to further identify at-risk patients, this contractor finds that there are currently no existing tests that have demonstrated AV, CV, and clinical utility (CU) to fulfill the necessary criteria.”

I understand that the above-quoted language reflects the MolDX program’s current position with respect to coverage of molecular tests for the detection of upper gastrointestinal metaplasia, dysplasia, and neoplasia, and that this proposed foundational LCD is intended to create a pathway to future coverage for tests that meet the criteria set forth in the policy. However, if the above-quoted language is included in the final LCD, it will arguably preclude MolDX from adding coverage for new tests – even after such tests successfully complete the Technical Assessment (TA) process – because the LCD identifies these tests as non-covered services. Given the procedural requirements of the 21^st Century Cures Act, it could easily take MolDX 12-18 months to propose and finalize a revised LCD that removes the above-referenced language. To ensure that Medicare beneficiaries have timely access to analytically and clinically validated assays following successful completion of the TA process, we strongly encourage MolDX to revise the proposed LCD to (a) clarify that tests meeting the criteria set forth in the LCD will be eligible for coverage, and (b) delete language regarding the current coverage status of tests from the LCD.

References were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.

The following comment was submitted to Palmetto GBA:

Thank you for the opportunity to submit comments on the proposed LCD: MolDX: Molecular Testing for Detection of Upper Gastrointestinal Metaplasia, Dysplasia, and Neoplasia. Exact Sciences Corporation is a molecular diagnostics company working to deliver life-changing innovations in cancer through earlier detection and smarter answers for patients. Headquartered in Madison, WI, the company is the manufacturer of the Cologuard® and Oncotype DX® tests with four CLIA-certified and CAP-accredited labs in three locations across the United States.

We estimate 1 million patients currently live with metastatic, advanced cancer (solid tumors) in the U.S.; more than 600,000 people will die from cancer in the U.S. this year alone. Today, many cancers are detected at an advanced stage when the cancer has become metastatic and therefore more difficult to treat. This is particularly true for cancer types such as ovarian, endometrial, pancreatic, liver, esophageal, and the many other cancers for which population-based screening is not currently recommended. Given that cancer therapies are generally more effective when patients are identified with lower tumor burden (i.e., earlier in the disease’s progression), substantial opportunity exists to positively affect clinical practice through early detection innovation and application.

Survival drops dramatically when cancer is diagnosed late. For esophageal cancer, in the localized stage, 5-year survival is almost 50%, but drops to almost 5% when the disease is diagnosed after it has metastasized.1 If cancer is caught early enough, it is often effectively treated or even cured.

Exact Sciences commends the MolDX® program for the creation of the proposed foundational LCD and supports Medicare coverage of non-endoscopic, biomarker-based tests for esophageal metaplasia, dysplasia, and neoplasia. We recommend revisions be made to the text to improve the clarity of the scope of coverage established by this LCD. We present the following edits to the text of the proposed LCD under the “Coverage Indications, Limitations, and/or Medical Necessity” section:

Current molecular diagnostic tests that identify individuals with upper gastrointestinal metaplasia, dysplasia, and neoplasia are covered when all the following criteria are true:

We understand that proposed LCD language reflects the current position of MolDX with respect to coverage of these tests, however, this language is not appropriate for inclusion in a foundational LCD, as it may restrict MolDX’s ability to cover tests in the future from test manufacturers that successfully complete the Technical Assessment (TA) process. Recent indications from your team are clear that it could take MolDX 12 to 18 months, or more, to issue an updated LCD under the procedures required by 21^st Century Cures. To ensure Medicare beneficiaries have timely access to validated, high-quality assays that have successfully completed the TA process, Exact Sciences recommends that MolDX revise the proposed language to (a) remove the language identifying all current tests as non-covered, and (b) revise the remaining language to clarify MolDX’s intent to cover tests in the future if they meet the criteria set forth therein.

This approach would be consistent with established MolDX precedent, recently set in the “Molecular Syndromic Panels for Infectious Disease Pathogen Identification Testing” and “Molecular Testing for Solid Organ Allograft Rejection” LCDs, respectively.

Exact Sciences also offers the following comments (including proposed modifications, where appropriate) on the text of the proposed LCD under the “Coverage Indications, Limitations, and/or Medical Necessity” specific to the coverage criteria:

The beneficiary is being actively managed for chronic gastroesophageal reflex disease (GERD) and/or non-dysplastic Barrett’s esophagus (NDBE); and

Exact Sciences agrees with this criterion as proposed and supports the use of biomarker testing in these patient populations.

The test either identifies patients who may benefit from surveillance or identifies patients with dysplasia or neoplasia who are more likely to benefit from endoscopic treatment and

We agree with MolDX that “there is a need to improve the ability to detect and prevent EAC at an earlier stage with options of curative therapy.” The criterion as written indicates a test does not meet criterion for coverage unless the test can distinguish between dysplastic and non-dysplastic disease. However, tests that can identify patients with non-dysplastic disease and move them into surveillance with esophagogastroduodenoscopy (EGD) are clinically important. Identification and surveillance of patients diagnosed with NDBE is critical for early detection and treatment of dysplasia to prevent progression to esophageal adenocarcinoma.

Non-dysplastic Barrett’s esophagus should be interpreted as a spectrum. Early detection of NDBE enables physicians to make appropriate decisions for their patients, which may include surveillance. For patients being managed for NDBE, current clinical guidelines from gastroenterology societies^2-4 recommend surveillance intervals be based on patient-specific risk factors, including segment length. NDBE segment length has been associated with a greater risk of progression to dysplasia/EAC, and guidelines^2,6-8 take this into account when defining surveillance intervals. The risk of progression is not negligible, and the diagnosis of NDBE may be important and actionable for some patients. Physicians may make individualized decisions about surveillance based on a number of risk factors. For example, an elderly smoker with long-segment NDBE and a family history of EAC may be managed much differently than a relatively young non-smoker with short-segment NDBE and no family history of EAC.

The test demonstrates analytical validity (AV) including an analytical and clinical validation for any given measured analytes, and has demonstrated equivalence or superiority for sensitivity or specificity of detecting dysplasia to other already accepted methods for the same intended use measuring the same or comparable analytes.

The reference to “clinical validation” within a criterion describing requirements for analytical validity is not clear. Are you referring to analytical validation using clinical (instead of contrived) samples? We believe the criterion as written causes confusion and ask MolDX to consider clarifying the language.

Clinical validity (CV) of any analytes measured must be established through a study published in the peer-reviewed literature for the intended use of the test in the intended population, establishing a clear and significant biological/molecular basis for identifying patients with metaplasia, dysplasia, or neoplasia within a clearly defined population.

We support the need for published clinical validity for a test that can identify patients with metaplasia or dysplasia. The modifications harmonize the language with other recent MolDX LCD’s, including, “MolDX: Minimal Residual Disease Testing for Cancer.” The American College of Gastroenterology’s recently updated guideline for Diagnosis and Management of Barrett’s Esophagus now suggests that minimally invasive tests combined with biomarker testing are an acceptable alternative to endoscopy for the early detection of BE in patients with chronic GERD and additional risk factors.² The clinical guidelines recommend biomarker testing in these at-risk patients at the same strength of recommendation and quality of evidence as EGD, the current standard of care.

Exact Sciences also offers the following comments (including proposed modifications, where appropriate) on the text of the proposed LCD under the “Coverage Indications, Limitations, and/or Medical Necessity” specific to the summary of evidence:

To provide the benefit of detecting precancerous disease amenable to early treatment a test must be able to detect a condition before the patient presents with symptoms, lead to the avoidance of unnecessary and invasive procedures, or to an intervention that yields a superior outcome, and be readily available, feasible and economical. In this case, the test would ideally be highly sensitive, as false negatives may miss patients with dysplasia, or highly specific, meaning fewer patients without dysplasia would require follow-up procedures. Studies must also be carried out in the relevant populations, and caution must be taken in extrapolating results of studies performed in populations already enriched for BE.

Exact Sciences agrees that biomarker tests need to be highly sensitive and highly specific for detection of dysplasia and agree with MolDX that this is addressed in coverage criterion #8 in the MolDX proposed language and with our recommended modifications. Sensitivity and specificity for Medicare-covered diagnostic tests vary and as a leader in the development of molecular diagnostics we want to convey the challenges of developing tests with >95% sensitivity and specificity and believe the idealistic number – which is not based on any clinical guideline or other peer-reviewed publication – should be removed from the rationale. Tests are typically developed to have high sensitivity or high specificity, depending on the intended use population; a test that seeks to optimize both metrics is likely to have reduced performance in both, and is unlikely to meet this very high threshold.^a 

Although there are several promising molecular biomarker tests designed to further identify at-risk patients, this contractor finds that there are currently no existing tests that have demonstrated AV, CV, and clinical utility (CU) to fulfill the necessary criteria. This contractor will continue to monitor the evidence and will provide coverage based on the pertinent literature and society recommendations.

Similar to the modifications requested to the operative text in the “Coverage Indications, Limitations, and/or Medical Necessity” section, we believe this language is not appropriate for inclusion in a foundational LCD, as it may restrict MolDX’s ability to cover tests in the future from other test manufacturers that successfully complete the TA process. While we understand that this language is not operative for purposes of defining the MolDX coverage criteria, we believe its removal from the final LCD would be appropriate, as such language will be inaccurate once MolDX approves a test for coverage.

Thank you for considering our comments on the proposed LCD.

References were provided for review.

Thank you for your thoughtful comments. We have addressed your comments and concerns in Response 1 listed above.