Screening for Colorectal Cancer - Blood-Based Biomarker Tests

November 15, 2020

VIA Electronic Submission to CAGinquiries@cms.hhs.gov

Tamara Syrek Jensen, J.D.
Director, Coverage & Analysis Group
Centers for Medicare & Medicaid Services
Mailstop S3-02-01
7500 Security Blvd.
Baltimore, MD 21244

Re: National Coverage Analysis (NCA) for Screening for Colorectal Cancer - Blood-Based Biomarker Tests [CAG-00454N]

Dear Ms. Jensen:

On behalf of Roche Diagnostics Corporation (Roche), I am pleased to submit comments to the Centers for Medicare and Medicaid Services (CMS) in response to the above-captioned proposed National Coverage Determination. Roche develops innovative products and services that address the prevention, diagnosis and treatment of diseases and medical conditions. As a global leader in the fields of diagnostics and pharmaceuticals, we believe personalized health care will transform lives, improve patient outcomes and reduce costs of care. These beliefs are especially strong as they relate to the early detection of life-threatening cancers at as stage where they may be treated with greater efficacy and at less expense to the Medicare trust funds. Roche appreciates CMS utilizing the National Coverage Analysis (NCA) process to prioritize cancer screening and to develop a proposed decision memo (“the Proposed NCD”) as part of the NCA process that is forward-thinking in its approach to provide coverage to future screening tests.

We offer the following comments on the Proposed NCD in the proposed rule for the agency’s consideration:

Roche supports the proposed NCD and encourages CMS to create clear pathways to national coverage for ctDNA screening for other cancer indications

Roche supports the Proposed NCD, and applauds the Agency's recognition of the medical value of using blood biopsy (circulating tumor DNA or ctDNA) for asymptomatic patients who otherwise meet conditions for coverage. A strong body of evidence exists in support of colorectal cancer (CRC) applications. Given the growing use of ctDNA for indications and oncology applications, we encourage CMS to consider a coverage framework that would enable ctDNA screening in other populations that would also benefit from earlier screening and intervention (e.g., lung cancer, breast cancer).

Consistent with major guidelines, CMS should cover CRC screening beginning at age 45 years

CMS notes in the Proposed NCD that the American Cancer Society guidelines encourage qualified screening starting for CRC at age 45.ⁱ In addition, the U.S. Preventative Services Task Force (USPSTF) currently recommends in a draft statement screening for CRC begin at age 45.ⁱⁱ According to Piscitello et al,ⁱⁱⁱ researchers that lowering the screening age to 45 would add roughly 19 million people to the average-risk population and increase the current number of screening-eligible individuals on immediate implementation by over 60% (from 27 to 44 million).. Piscitello et al note that, since 1994, there is a 51% incidence of colorectal cancer incidence among those less than age 50. In addition, individuals less than 50 are more likely to have advanced-stage colorectal cancer at diagnosis than those over 50.^iv Given the growing incidence and burden of early-onset CRC, any test that fulfills CMS criteria as an appropriate front-line assay for average risk CRC screening should be initiated at 45 years of age.

Roche recommends CMS consider removal of performance benchmarks in the Proposed NCD and add the acceptance of other forms of endorsement from professional societies

Under the Proposed NCD, a test must have “proven performance characteristics for a blood-based screening test with both sensitivity greater than or equal to 74 percent and specificity greater than or equal to 90 percent in the detection of CRC compared to the recognized standard, as minimal threshold levels, based on pivotal studies included in FDA labeling.” Along with these performance criteria, a test would need to be included as a screening option in at least one professional society guideline, professional society consensus statement, or U.S. Preventive Services Task Force (USPSTF) recommendation to meet the requirement for coverage.

We encourage CMS to consider removal of explicit performance criteria on specificity and sensitivity within the Proposed NCD. In lieu of setting pre-defined performance benchmarks in sensitivity and specificity and in recognition that clinical community standards evolve (e.g., current comparators), CMS could rely upon FDA authorization. This reliance considers that due diligence was given to these performance criteria as evidence of comparability and performance satisfactory by the oversight regulatory agency. As evidence that these performance criteria are accepted in the medical community as clinically useful and represent reasonable and necessary services, the Proposed NCD makes a condition of coverage inclusion in a major clinical guideline or a positive USPSTF recommendation.

Relatedly, given the pace of development in this space and the length of time necessary to issue guidelines, CMS should consider alternative pathways for professional societies to provide feedback on whether tests meet the performance criteria aside from inclusion in guidelines. These pathways may include endorsement letters from professional society leadership and/or editorial publication from professional society leadership in peer-reviewed journals.

* * * * *

We appreciate the opportunity to submit our comments on the Proposed NCs. Please contact me at 202-347-0836 or by e-mail at paul.sheives@roche.com if you have any questions about our comments or would like to discuss these further. Thank you for your consideration of our comments.

Sincerely,

Paul Sheives, M.S., J.D.
Director, Federal Health Policy & Reimbursement
Roche Diagnostics Corporation

CC: Joseph Chin, M.D., M.S.
Deputy Director, Coverage and Analysis Group

Melissa A. Evans, Ph.D., M.S.A.E.
Director, Division of Policy & Evidence Review

Lori Ashby, M.A.
Deputy Director, Division of Policy & Evidence Review

Carl Li, M.D., M.P.H.
Medical Officer

Kimberly Long
Lead Analyst

---

i Wolf AMD, Fontham ETH, Church TR, et al. Colorectal cancer screening for average-risk adults: 2018 guideline update from the American Cancer Society. CA: A Cancer Journal for Clinicians. 2018;68(4):250-281. doi:10.3322/caac.21457
ii U.S. Preventive Services Task Force. Draft Recommendation Statement- Colorectal Cancer: Screening. https://www.uspreventiveservicestaskforce.org/uspstf/draft-recommendation/colorectal-cancer-screening3.
iii Piscitello A, Edwards DK. Estimating the screening-eligible population size, aged 45 to 74, at average risk to develop colorectal cancer in the United States. Cancer Prev Res. 2020;13(5):443-448.
iv Virostko J, Capasso A, Yankeelov TE, Goodgame B. Recent trends in the age at diagnosis of colorectal cancer in the US national cancer data base, 2004–2015. Cancer 2019;125:3828–35.

Less

November 15, 2020

VIA Electronic Submission to cms.gov

Tamara Syrek Jensen
Director, Coverage and Analysis Group
Centers for Medicare & Medicaid Services
Mail Stop # S3-02-01
7500 Security Boulevard
Baltimore, MD 21244-1850

RE: Proposed Decision Memo for Screening for Colorectal Cancer – Blood-Based Biomarker Tests (CAG-00454N)

Dear Ms. Syrek Jensen:

On behalf of the Coalition for 21st Century Medicine (C21), we appreciate the opportunity to submit comments regarding the above-captioned proposed decision memo. C21 comprises many of the world’s most innovative diagnostic technology companies, clinical laboratories, physicians, venture capital companies, and patient advocacy groups. Member companies are developing novel tests that are able to detect a large number of different cancers, including colorectal cancer (CRC), using blood-based assays. Given the Coalition’s mission to facilitate development and commercialization of innovative diagnostics to inform important patient management decisions, we have a keen interest in the agency’s National Coverage Determinations (NCDs), including NCDs that address new technologies for cancer screening.

CMS re-opened the CRC screening NCD in response to a request from the manufacturer of a single blood-based CRC screening test.¹ As we noted in our comments on the NCA tracking sheet, other blood-based screening assays are currently under development and may soon be available for the detection of CRC. Therefore, we encouraged CMS to draft NCD language that addresses coverage for blood-based CRC screening assays as a group instead of focusing on any specific company’s test to ensure Medicare beneficiaries have timely access to high-quality tests as they become available.

Consistent with our request, CMS proposes to automatically cover any blood-based CRC screening tests that meets pre-specified criteria (e.g., FDA-approved/cleared, meet minimum performance standards, and supported in professional society guidelines). In general, C21 supports this approach, since it would allow Medicare to cover novel tests as soon as such criteria have been met, and would eliminate the need for CMS to re-start the 9-12 month NCD re-opening process each time the agency adds a new test to the colorectal cancer screening benefit.

With respect to the proposed coverage criteria themselves, C21 offers the following comments for the agency’s consideration:

• Inclusion in professional society guidelines. CMS proposes to condition coverage on “inclusion as a recommended routine colorectal cancer screening test in at least one professional society guideline or consensus statement or [USPSTF] recommendation.” However, colorectal cancer screening tests that successfully obtain FDA approval or clearance and meet CMS’s proposed minimum performance standards (i.e., sensitivity and specificity) will already have substantial evidence supporting their performance and utility. Moreover, nothing in the text of the Social Security Act requires CMS to condition coverage on inclusion in professional society guidelines, which may be updated on an irregular schedule and/or go years between scheduled updates. As a result, C21 encourages CMS to delete the proposed requirement for inclusion in guidelines.

  Alternatively, if CMS believes professional society input is necessary, C21 encourages CMS to be holistic when considering the types of professional society statements that may be sufficient to support coverage, and to consider other evidence of professional society support that can be more quickly obtained– e.g., comment letter, position statement, endorsement letter from society leadership, editorial in a peer-reviewed journal from society leadership – as sufficient to meet this requirement. We further encourage the agency to consider this requirement met if the professional society issues a communication that generally expresses support for a group of tests (e.g., “all tests that meet the CMS criteria set forth in the NCD”), instead of requiring the society to individually recognize each such test by brand name.

• Testing frequency. CMS proposes to cover blood-based biomarker tests “once every three years, or at the interval designated in the [FDA] label if the FDA indicates a specific test interval.” However, FDA labeling does not typically indicate a specific test interval. In the (likely) absence of guidance from FDA, we encourage CMS to cover testing either every three years, or at the interval supported by the test developer’s data (including modeling), a relevant professional society (e.g., in guidelines, position statement, editorial, or comment letter), or in a well-established third-party decision model.

For ease of reference, we have attached a redlined version of the draft NCD language that illustrates our proposed edits.

With respect to the CRC screening NCD more generally (i.e., for screening methodologies including but not limited to blood-based biomarker tests), we encourage CMS to align its screening age requirements with recently published guidelines for CRC screening. The American Cancer Society currently recommends that certain CRC screening modalities be initiated at 45 years of age.² In addition, the recently-released draft USPSTF recommendation statement included a Grade B recommendation for certain CRC screening modalities for patients between 45 and 49 years of age.³ Given these new guidelines and the growing burden of earlyonset CRC, any CRC screening test that fulfills the relevant criteria in the NCD and is supported in a relevant professional society statement (e.g., guideline, consensus statement, position statement, endorsement letter, comment letter, or editorial) for patients age 45-49 should be covered for patients 45-49.⁴

We also encourage CMS to clarify that follow-up colonoscopies performed in response to an abnormal non-invasive screening test will be covered. The follow-up colonoscopy is a critical component of the entire patient screening process and should be treated as part of the screening test. Patient access and completion of the full cycle of screening is at risk when patients are faced with unexpected costs from a follow-up colonoscopy.

C21 thanks CMS for the opportunity to comment on evaluating blood-based CRC screening tests for Medicare coverage, and looks forward to working with the agency as it continues to create coverage policies for cancer screening technologies. As such, C21 encourages CMS to continue prioritizing the evaluation of new screening technologies, including novel blood-based tests that detect a large number of different cancers, and working to develop a pathway to expedite coverage for well-validated tests.

Thank you for considering our comments. Please contact me at hmurphy@c21cm.org should you have any questions or if we can provide you with further information.

Sincerely,

Hannah Murphy
Executive Director
The Coalition for 21st Century Medicine

3. Blood-based Biomarker Tests (effective XX,XX,XXX)

Blood based DNA testing detects molecular markers of altered DNA that are contained in the cells shed into the lumen of the large bowel by colorectal cancer and pre-malignant colorectal epithelial neoplasia.

Effective for dates of service on or after XX,XX,XXX, a blood-based biomarker test is covered as an appropriate colorectal cancer screening test either once every 3 years, or at the interval designated in the Food and Drug Administration (FDA) label if the FDA indicates a specific test interval, or at the interval supported by the test developer’s data (including modeling), a professional society statement (e.g., guideline, consensus statement, position statement, endorsement letter, comment letter, editorial), or in a well-established third-party decision model, for Medicare beneficiaries when performed in a Clinical Laboratory Improvement Act (CLIA)-certified laboratory, when ordered by a treating physician and when all of the following requirements are met:

The patient is:

• age 5045-85 years, and,
• asymptomatic (no signs or symptoms of colorectal disease including but not limited to lower gastrointestinal pain, blood in stool, positive guaiac fecal occult blood test or fecal immunochemical test), and,
• at average risk of developing colorectal cancer (no personal history of adenomatous polyps, colorectal cancer, or inflammatory bowel disease, including Crohn’s Disease and ulcerative colitis; no family history of colorectal cancers or adenomatous polyps, familial adenomatous polyposis, or hereditary nonpolyposis colorectal cancer).

The blood-based biomarker screening test must have all of the following:

• FDA market authorization with an indication for colorectal cancer screening; and
• proven test performance characteristics for a blood based screening test with both sensitivity greater than or equal to 74% and specificity greater than or equal to 90% in the detection of colorectal cancer compared to the recognized standard (accepted as colonoscopy at this time), as minimal threshold levels, based on the pivotal studies included in the FDA labeling; and
• inclusion as a recommended routine colorectal cancer screening test in at least one professional society guideline or consensus statement or United States Preventive Services Task Force (USPSTF) recommendation.

---

1 Epigenomics is not a member of C21, and C21 expresses no opinion with respect to coverage for Epi proColon.
2 See Wolf AMD, Fontham ETH, Church TR, Flowers CR, et al. Colorectal cancer screening for average riskadults: 2018 guideline update from the American Cancer Society. CA Cancer J Clin 2018;68(4):250-281.
3 U.S. Preventive Services Task Force, Draft Recommendation Statement – Colorectal Cancer: Screening (Oct. 2020), https://uspreventiveservicestaskforce.org/uspstf/draft-recommendation/colorectal-cancer-screening3.
4 Blood-based biomarker tests are not currently supported in the ACS or USPSTF (draft) recommendation. As a result, we have proposed to leave the proposed age range as 50-85 in the blood-based biomarker section of the NCD.

Less

I am a gastroenterologist and clinical epidemiologist who has studied CRC screening as well as methods to improve evaluation of screening tests. I was a senior author of Epigenomics' initial study of mSEPT9 (1) and of Exact's study of Cologuard, (2) receiving no payment for either study. I am a member of FDA's Immunology Devices Panel (CDRH).

I write to strongly support and elaborate on the public comments to CMS from Ladabaum et al suggesting that CMS consider assessing new diagnostic tests by "Retaining the traditional pathway (research and development, clinical studies, FDA approval, and inclusion in guidelines culminating in a CMS coverage determination)...." As they note, setting a fixed sensitivity and specificity could result in "incoherent and volatile approvals...and could stifle innovation".

This comment identifies the problems - and considers their magnitude and practical consequences - that could arise from setting a fixed sensitivity and specificity. Also described is how CMS' utilization of the "traditional pathway" might help address those problems.

An unusual feature of the mSEPT9 test - a feature with dramatic consequences largely unintended and unexpected - is its very high false positive rate of roughly 20% for one application of the test.

A test applied yearly with a 20% false positive rate incurs referral to colonoscopy of nearly 50% of persons at 3 yrs and 70% at 5 yrs. (3) Such persons thus avoid colonoscopy for only a small number of years - an important consideration for a test intended for persons who do not want colonoscopy.

Interestingly, the high cumulative rate at which people would start colonoscopy screening is precisely the mechanism by which blood test screening turns out to be so effective in cost-effectiveness analysis. (3,4) I.e., A program of blood test screening almost immediately becomes a program of colonoscopy screening. (4) Such practical consequences would need to be considered in detail by doctors and patients in shared decision-making.

A further and perhaps unexpected consequence of such a high false positive rate is that people are referred for colonoscopy screening almost entirely because of 'chance'. Even if the blood test finds 71% of persons with CRC, the prevalence of CRC at screening age is roughly 0.5-1%, so that only a small number of persons would be detected and referred. The 20% false positive rate in that group means that the vast majority of people referred are referred 'by chance'. (4)

Importantly, the apparent 'discovery' of advanced adenomas - whose removal likely contributes to colonoscopy's efficacy in reducing CRC incidence and mortality - is not because the positive blood test actually finds advanced adenomas but rather is because the positivity rate of the test in persons with advanced adenomas is the same numerically (20%) as for persons with no colonic disease. (1, 5) I.e.,The blood test contributes nothing additional. (4) This impact of chance on the blood test's effectiveness (3) is so striking is that test-developers have asked me whether they should develop a test with high false positive rate and that people believe in and so will enroll in colonoscopy screening. This question - as unsettling as it sounds - is actually not illogical if CMS and other payors or guidelines-makers endorse such a test. Ultimately, however, such endorsement could stifle the development of future tests that provide stronger discrimination.

Improving the specificity from 80% to 90% as CMS proposes, or to 85% as Ladabaum et al suggest, still leads - for a one-year testing interval - to a high cumulative false positive rate. For a test with a 10% false positive rate, the cumulative referral rate is 27% at 3 years and 41% at 5 years; for a test with a 15% false positive rate, the cumulative referral rates are 39% and 55% respectively.

One further unintended consequence of premature approval of a test with weak discrimination and low overall performance is that if such a test is reimbursed and becomes widely used, then it becomes 'standard of care' in a way that may make enrollment difficult or impossible in subsequent prospective studies.

The problems above can be better understood and better addressed not by setting a fixed bar for sensitivity and specificity but by retaining the "traditional pathway" that includes features of the guidelines-making process such as that used by the US Preventive Services Task Force which considers outcomes and practical consequences quantitatively. Such a process can assess, for example, the impact of testing interval and can understand the 'mechanism' of test effectiveness. While CMS might decide to lower the bar in a deliberate manner after considering such quantitative analyses, CMS's rationale - and ultimately its decisions - could more-effectively address a test's practical consequences on the care of patients, on the role of doctors, and on implications for test developers.

REFERENCES

1. Church TR, Wandell M, Lofton-Day C, et al. Prospective evaluation of methylated SEPT9 in plasma for detection of asymptomatic colorectal cancer. Gut. 2014;63(2):317-25.
2. Imperiale TF, Ransohoff DF, Itzkowitz SH, et al. Multitarget stool DNA testing for colorectal-cancer screening. N Engl J Med. 2014;370(14):1287-97.
3. Peterse EFP, Meester RGS, de Jonge L, et. al. Comparing the cost-effectiveness of innovative colorectal cancer screening tests. J Natl Cancer Inst. 2020. Epub 2020/08/08. doi: 10.1093/jnci/djaa103.
4. Ransohoff DF. Evaluating a new cancer screening blood test: Unintended consequences and the need for clarity in policy-making. J Natl Cancer Inst. 2020. Epub 2020/08/08. doi: 10.1093/jnci/djaa104.
5. Potter NT, Hurban P, White MN, et al. Validation of a real-time PCR-based qualitative assay for the detection of methylated SEPT9 DNA in human plasma. Clin Chem. 2014;60(9):1183-91.

Less

15 November 2020

VIA Electronic Submission to cms.gov

Tamara Syrek Jensen, J.D.
Director, Coverage and Analysis Group
Centers for Medicare & Medicaid Services
Mail Stop # S3-02-01
7500 Security Boulevard
Baltimore, MD 21244-1850

RE: Proposed Decision Memo for Screening for Colorectal Cancer – Blood-based Biomarker Tests (CAG-00454N)

Dear Ms. Jensen:

We appreciate the opportunity to submit comments regarding this proposed decision memo (PDM) for screening for Colorectal Cancer - Blood-based Biomarker Tests (CAG-00454N). At Freenome, our mission is to develop clinical tests to help prevent, detect, and treat disease, with an initial focus on colorectal cancer (CRC). Freenome plans to seek FDA approval for the first-ever artificial intelligence/machine learning-enabled multiomics blood test for CRC screening, which we believe will achieve greater clinical sensitivity and specificity than conventional biomarker approaches, particularly for early-stage (I/II) CRC and advanced adenomas. We also believe that our test, once approved, will significantly increase patient acceptance of and adherence to current CRC screening options, which translates directly into decreased patient morbidity and mortality. We therefore have a keen interest in a National Coverage Determination (NCD) for CRC screening tests and sincerely look forward to working with the Agency.

In general, Freenome supports the approach described by CMS in the PDM, which will cover any blood-based CRC screening tests that meet pre-specified criteria and expedites access to safe and effective CRC screening for Medicare beneficiaries. Given that the mortality benefits of CRC screening are widely accepted and reinforced by multiple clinical guidelines (e.g., USPSTF, ACS, USMSTF, and NCCN), we agree that the evidence is sufficient to conclude that there is a mortality benefit for any CRC screening test that has adequate clinical performance. As such, we also believe that CMS’ approach of specifying minimum performance criteria based on clinical sensitivity and specificity is appropriate.

Regarding the proposed coverage criteria, Freenome offers the following recommendations for your consideration:

1. In addition to the performance requirements provided for specificity and CRC sensitivity, CMS should include a performance requirement for advanced adenoma sensitivity ≥ 24% in the intended use population.
2. The method for calculating specificity and CRC sensitivity in the intended use population should be clearly defined and consistently applied to all non-invasive tests, and the proposed specificity requirement should be modified to ≥ 87% to match that of FIT-DNA in the intended use population.
3. The screening interval should not be pre-specified but determined using test characteristics that impact a test's clinical utility, such as clinical performance and patient acceptability/adherence.

Each of these is discussed in greater detail below.

1. In addition to the performance requirements provided for specificity and CRC sensitivity, CMS should include a performance requirement for advanced adenoma sensitivity ≥ 24% in the intended use population.

AAs are clinically important precancerous lesions that are removed during a colonoscopy to prevent CRC. Colonoscopy, sigmoidoscopy, FIT, and FIT-DNA can detect AAs to varying degrees. We therefore recommend that, just as it has done for CRC sensitivity, CMS add a minimum performance requirement for AA sensitivity and clearly define how this criterion must be satisfied. Our rationale is provided below.

The detection and subsequent removal of adenomas, especially AAs, saves lives, with each 1% increase in adenoma detection rate translating into an ~3% decrease in CRC-specific mortality (Corley et al., 2014; Meester et al., 2015). By contrast, each 1% increase in CRC sensitivity translates into an ~0.05% decrease in CRC-specific mortality (Haug et al., 2015). The ~60x greater impact of AA versus CRC sensitivity on CRC-specific mortality highlights the value of including AA sensitivity as a performance requirement.

CMS itself states in the PDM: “in a scenario where a test does not detect large adenomas or early cancer as well as currently covered tests, there may be detriments to patients since the opportunity for cancer prevention may have been missed or cancer treatments may be initiated at a later stage when health outcomes may be reduced.” We therefore recommend that CMS add a new criterion on AA sensitivity (in addition to those already provided for specificity and CRC sensitivity). Since the point estimate for CRC sensitivity proposed by CMS in the PDM is based on the performance of FIT in DeeP-C, we suggest using the same study as the benchmark for AA sensitivity: ≥ 24% (Imperiale et al., 2014). We further recommend that CMS clarify the following:

1. Clearly and consistently define AAs. For the purposes of comparability to FIT and FIT-DNA, we recommend using the definitions used in DeeP-C, in which AAs included:
   1. Adenoma with carcinoma in situ/high grade dysplasia, any size
   2. Adenoma, villous growth pattern (≥ 25%), any size
   3. Adenoma ≥ 1.0 cm in size
   4. Serrated lesion ≥ 1.0 cm in size (including hyperplastic polyps ≥ 1.0 cm in size)
2. Validation of AA performance must be performed in the intended use population (i.e., 50-85, asymptomatic, average risk, and pre-colonoscopy) using colonoscopy with histopathology as the reference method.
3. AA sensitivity is calculated as TP(AA)/[TP(AA) + FN(AA)] using the following 2x2 table:
   

   	Reference Method (Colonoscopy with Histopathology)
   	Advanced adenoma positive	Advanced adenoma negative
   Test-positive	True Positive (TP)	False Positive (FP)
   Test-negative	False Negative (FN)	True Negative (TN)

2. The method for calculating specificity and CRC sensitivity in the intended use population should be clearly defined and consistently applied to all non-invasive tests, and the proposed specificity requirement should be modified to ≥ 87% to match that of FIT-DNA.

In the PDM, CMS proposes minimum performance metrics for a blood-based test for CRC screening that appear to be based on FIT (for CRC sensitivity) and FIT-DNA (for specificity). We would like to make the following recommendations regarding this proposal:

1. CMS should clarify that these performance criteria must be established in the intended use population (i.e., 50-85, asymptomatic, average risk, and pre-colonoscopy) using colonoscopy with histopathology as the reference method.
2. CMS should clarify how it intends to define specificity (see recommendation below).
3. CMS should modify the specificity requirement to 87%, not 90%, to match the clinical specificity of FIT-DNA in the intended use population.

Our rationale is provided below.

• Colonoscopy with histopathology remains the standard for establishing the final diagnosis in patients whose lesions were found by CRC screening and so should serve as the reference method to establish the performance characteristics for any non-invasive CRC screening test (as it was in DeeP-C).
• CMS must define exactly how it plans to calculate specificity and to do so consistently because the clinical implications (i.e., a diagnostic colonoscopy) of a positive result are identical for any non-invasive screening test, whether stool, blood, or otherwise. Based on the data used to support its approval (Food and Drug Administration - Center for Devices and Radiological Health, 2014; Imperiale et al., 2014) , the specificity and CRC sensitivity of FIT-DNA are 87% (i.e., not 90%) and 92%, respectively. Comparable figures from the same study for FIT are 95% and 74%, respectively.

Shown below is the histopathological classification scheme used for FIT and FIT-DNA in DeeP-C [ (Food and Drug Administration - Center for Devices and Radiological Health, 2014), Table 6].

• Category 1: CRC, all stages (I-IV)
• Category 2: AAs, including the following subcategories
  • Adenoma with carcinoma in situ/high grade dysplasia, any size
  • Adenoma, villous growth pattern (≥ 25%), any size
  • Adenoma ≥ 1.0 cm in size
  • Serrated lesion ≥ 1.0 cm in size (includes hyperplastic polyps ≥ 1.0 cm)
• Category 3: 1 or 2 non-advanced adenomas (NAAs) > 0.5 cm in size, or < 1.0 cm size
• Category 4: ≥ 3 NAAs, < 1.0 cm
• Category 5: 1 or 2 NAAs, ≤ 0.5 cm in size
• Category 6: Negative (NEG) - No neoplastic findings
  • 6.1: Negative upon histopathological review (includes hyperplastic polyps < 1.0 cm)
  • 6.2: No findings on colonoscopy, no histopathological review

Also shown below is the specificity and sensitivity for FIT-DNA based on different definitions of “test-positive” and “test-negative” (Food and Drug Administration - Center for Devices and Radiological Health, 2014; Imperiale et al., 2014). The gray highlighted column reflects how FIT-DNA is reported in real-world clinical practice.

	Definition of “Test-positive” and “Test-negative”
	Positive = CRC Negative = AA+NAA+NEG	Positive = CRC+AA Negative = NAA+NEG	Positive = CRC+AA+NAA Negative = NEG
Sensitivity	92%	46%	24%
Specificity	84%	87%	88%

As shown above, FIT-DNA does not achieve 90% specificity in ANY of these scenarios (so would itself not meet the performance criteria proposed in this PDM). Only if one considers category 6.2 (defined as “no findings on colonoscopy, no histopathological review”; in other words, a “completely clean” colonoscopy with no adenomatous or non-adenomatous polyps) as “true negatives” does FIT-DNA achieve a specificity of 90% (Food and Drug Administration - Center for Devices and Radiological Health, 2014; Imperiale et al., 2014) . However, this is NOT how FIT-DNA test results are reported to providers and patients in the real world. We therefore recommend that specificity be calculated in the following manner, consistent with how it was calculated for the co-primary specificity endpoint in DeeP-C and is currently reported in clinical practice for FIT-DNA:

Specificity to advanced colorectal neoplasia (i.e., CRC + AA) =

TN (NAA+NEG)/[TN (NAA+NEG) + FP (NAA+NEG)] using the following 2x2 table

	Reference Method (Colonoscopy with Histopathology)
	Advanced Colorectal Neoplasia positive	Advanced Colorectal Neoplasia negative
Test-positive	True Positive (TP)	False Positive (FP)
Test-negative	False Negative (FN)	True Negative (TN)

If CMS wishes to use a 90% specificity threshold for blood but not stool tests, then we believe that the Agency should apply a different methodology for calculating specificity: Only patients with “completely clean” colonoscopies (i.e., category 6.2 above) should be considered “true negatives”. In such a case, we request that the coverage language be modified accordingly.

3. The screening interval should not be pre-specified but determined using test characteristics that impact a test's clinical utility, such as clinical performance and patient acceptability/adherence.

In the PDM, CMS proposes to cover a blood-based test for CRC screening “once every 3 years, or at the interval designated in the Food and Drug Administration (FDA) label if the FDA indicates a specific test interval.” While we understand and agree with CMS’ intent, we strongly urge CMS not to pre-specify the test interval for the following reasons:

1. The appropriate screening interval is ultimately determined by the interaction of multiple factors, most especially clinical performance (e.g., sensitivity and specificity), patient acceptability/adherence, and cost.
2. Based on the precedents set by FIT-DNA and m SEPT9 , the FDA label typically does NOT specify a test interval.

We recommend that CMS modify this language to say: “at the interval supported by the test developer’s data (including modeling), or by a professional society guideline or consensus statement or USPSTF. If these approaches support different intervals, the most frequent interval will be covered.”

Regarding the first point above, the clinical utility of any CRC screening test will ultimately depend on the interaction of natural history, clinical performance, screening frequency (i.e., test interval), patient acceptability and adherence, and cost. Especially for CRC, where the benefit of detecting early-stage (I/II) CRC and AAs is well accepted, screening frequency is a critical consideration (Colditz et al., 1997). However, the public health consequences of screening frequency are meaningfully influenced by other factors, especially test performance and patient acceptability/adherence, as illustrated by the range of test intervals recommended by clinical guidelines from USPSTF, ACS, USMSTF, and NCCN, among others, ranging from FIT every year to colonoscopy every 10 years (Bibbins-Domingo et al., 2016). The impact of test interval and programmatic screening is also illustrated by longitudinal studies of FIT showing that the “cumulative detection rate” (a combination of adherence, screening frequency, and test performance) of annual FIT can be comparable to that of once-a-decade colonoscopy (Crotta et al., 2012; Zorzi et al., 2018). Therefore, the appropriate screening interval cannot and should not be pre-specified a priori in the absence of understanding these other variables.

With respect to the second point above, while the FDA is well-positioned to evaluate the appropriate interval for a CRC screening test, the Agency did not specify this in the labels of either FIT-DNA (Food and Drug Administration - Center for Devices and Radiological Health, 2014) or m SEPT9 (Food and Drug Administration - Center for Devices and Radiological Health, 2016). Indeed, evaluation of appropriate test interval and programmatic performance was a post-approval requirement for both tests. As we believe CMS already understands based on the work it did in deciding on coverage for FIT-DNA, clinical trials to determine the optimal intervals for screening tests are prohibitively costly and complex, so modeling is the best practical alternative. In fact, to date, the appropriate screening interval for FIT-DNA and m SEPT9 have both only been established by population modeling, not clinical trials. Furthermore, the recommended interval for CRC screening tests has typically been established by clinical guidelines from USPSTF, ACS, USMSTF, and NCCN, among others. We are concerned that the current proposal for establishing the test interval is too restrictive and may inadvertently undermine the clinical utility of future blood tests. We therefore recommend that CMS use critical factors, such as clinical performance and patient acceptability/adherence, from multiple bodies of evidence (e.g., population modeling and/or clinical studies) to determine the appropriate interval. We also recommend that CMS defer to the recommendations of subject matter experts, including but not limited to the FDA, USPSTF, ACS, USMSTF, and NCCN.

Conclusions

In summary, Freenome’s recommendations are as follows:

1. In addition to the performance requirements provided for specificity and CRC sensitivity, CMS should include a performance requirement for advanced adenoma sensitivity ≥ 24% in the intended use population.
2. The method for calculating specificity and CRC sensitivity in the intended use population should be clearly defined and consistently applied to all non-invasive tests, and the proposed specificity requirement should be modified to ≥ 87% to match that of FIT-DNA.
3. The screening interval should not be pre-specified but determined using test characteristics that impact a test's clinical utility, such as clinical performance and patient acceptability/adherence.

We sincerely commend CMS for creating a clear and consistent path for coverage of blood-based CRC screening tests and look forward to working with the Agency now and in the future. Freenome believes providing access to more safe and effective CRC screening options will save and improve the lives of millions of Medicare beneficiaries, reduce burdens across the entire spectrum of stakeholders, and facilitate innovation. Thank you for considering our comments. Please contact me at girish.putcha@freenome.com should you have any questions or seek additional information.

Sincerely yours,
Girish Putcha, M.D., Ph.D.
Chief Medical Officer
Clinical Laboratory Director
Freenome

Bibbins-Domingo, K., Grossman, D. C., Curry, S. J., Davidson, K. W., Epling, J. W., García, F. A. R., Gillman, M. W., Harper, D. M., Kemper, A. R., Krist, A. H., Kurth, A. E., Landefeld, C. S., Mangione, C. M., Owens, D. K., Phillips, W. R., Phipps, M. G., Pignone, M. P., & Siu, A. L. (2016). Screening for Colorectal Cancer: US Preventive Services Task Force Recommendation Statement. JAMA, 315(23), 2564–2575. https://doi.org/10.1001/jama.2016.5989

Colditz, G. A., Hoaglin, D. C., & Berkey, C. S. (1997). Cancer incidence and mortality: The priority of screening frequency and population coverage. The Milbank Quarterly, 75(2), 147–173. https://doi.org/10.1111/1468-0009.00050

Corley, D. A., Jensen, C. D., Marks, A. R., Zhao, W. K., Lee, J. K., Doubeni, C. A., Zauber, A. G., de Boer, J., Fireman, B. H., Schottinger, J. E., Quinn, V. P., Ghai, N. R., Levin, T. R., & Quesenberry, C. P. (2014). Adenoma Detection Rate and Risk of Colorectal Cancer and Death. New England Journal of Medicine, 370(14), 1298–1306. https://doi.org/10.1056/NEJMoa1309086

Crotta, S., Segnan, N., Paganin, S., Dagnes, B., Rosset, R., & Senore, C. (2012). High Rate of Advanced Adenoma Detection in 4 Rounds of Colorectal Cancer Screening With the Fecal Immunochemical Test. Clinical Gastroenterology and Hepatology, 10(6), 633–638. https://doi.org/10.1016/j.cgh.2012.02.030

Food and Drug Administration - Center for Devices and Radiological Health,. (2014). Summary of Safety and Effectiveness Data (SSED): Stool DNA-Based Colorectal Cancer Screening Test (PMA Number: P130017). Food and Drug Administration. https://www.accessdata.fda.gov/cdrh_docs/pdf13/P130017b.pdf

Food and Drug Administration - Center for Devices and Ragiological Health,. (2016). Summary of Safety and Effectiveness Data (SSED): Blood-based Qualitative Colorectal Cancer Screening Test (PMA Number P130001B). Food and Drug Administration. https://www.accessdata.fda.gov/cdrh_docs/pdf13/P130001B.pdf

Haug, U., Knudsen, A. B., Lansdorp-Vogelaar, I., & Kuntz, K. M. (2015). Development of new non-invasive tests for colorectal cancer screening: The relevance of information on adenoma detection: New tests for colorectal cancer screening. International Journal of Cancer, 136(12), 2864–2874. https://doi.org/10.1002/ijc.29343

Imperiale, T. F., Ransohoff, D. F., Itzkowitz, S. H., Levin, T. R., Lavin, P., Lidgard, G. P., Ahlquist, D. A., & Berger, B. M. (2014). Multitarget Stool DNA Testing for Colorectal-Cancer Screening. New England Journal of Medicine, 370(14), 1287–1297. https://doi.org/10.1056/NEJMoa1311194

Meester, R. G. S., Doubeni, C. A., Zauber, A. G., Goede, S. L., Levin, T. R., Corley, D. A., Jemal, A., & Lansdorp-Vogelaar, I. (2015). Public health impact of achieving 80% colorectal cancer screening rates in the United States by 2018. Cancer, 121(13), 2281–2285. https://doi.org/10.1002/cncr.29336

Zorzi, M., Hassan, C., Capodaglio, G., Fedato, C., Montaguti, A., Turrin, A., Rosano, A., Monetti, D., Stocco, C., Baracco, S., Russo, F., Repici, A., & Rugge, M. (2018). Long-term performance of colorectal cancer screening programmes based on the faecal immunochemical test. Gut, 67(12), 2124–2130. https://doi.org/10.1136/gutjnl-2017-314753

Less

Dear Ms. Jensen,

The American Cancer Society (the Society) Society and American Cancer Society Cancer Action Network (ACS CAN) welcome the opportunity to respond to the Proposed Decision Memo (PDM) for Screening for Colorectal Cancer - Blood-Based Biomarker Tests (CAG-00454N). The Society is the nation’s largest voluntary health organization, dedicated to eliminating cancer as a major health outcome through research, education, and service. ACS CAN is the non-profit, non-partisan advocacy affiliate of the Society. ACS CAN advocates for legislative, regulatory, and policy solutions that will make cancer a national priority.

We appreciate CMS’ interest in expanding colorectal cancer screening options. Colorectal cancer is one of the few cancers that can be completely prevented through screening. Polyps, or abnormal precancerous growths, can be detected during the screening process and entirely removed, thereby stopping any cancer formation. Regular screening is the most effective way of detecting precancerous growths and early colorectal cancer. Cancers that are found at an early stage can be treated more easily, and lead to greater survival.¹ Yet colorectal cancer remains the second leading cause of cancer death in men and women combined in the U.S.²

We appreciate CMS’ interest in pursuing coverage options for blood-based biomarker screenings tests. We offer the following comments on the process and standards CMS proposes with respect to these tests. ACS does not currently recommend for or against any blood-based biomarker test for colorectal cancer screening. Therefore, we are not commenting on the coverage determination with respect to the Epi proColon® test. Our comments are general in nature and address observations we believe CMS should consider before issuing a final Decision Memo. We also critique the methodology that CMS used in developing the PDM.

Revised screening age: With respect to colorectal cancer screening, we note that the American Cancer Society lowered the age to begin screening in 2018 to age 45,³ and the United States Preventive Services Task Force (USPSTF) has proposed a change in their recommendations to lower the age to begin screening from 50 to age 45. As such, we strongly urge CMS to comport its coverage determinations to be in line with Task Force and ACS guidelines with respect to this, and future, PDMs regarding colorectal cancer screening tests.
FDA approval: We agree that a blood-based biomarker screening test should have FDA-authorization. We also think it is sensible to require a recommendation from at least one professional society or the USPSTF, although we think it is worth considering limiting that criteria to an organization that adheres to high standards in guideline development as measured by AGREE II standards.⁴

Use of positive predictive value: It is unclear why CMS would include the positive predictive value (PPV) as a test performance characteristic, and we note that threshold measures of predictive value are not mentioned after this statement. First, it is not mentioned in the performance criteria described on page 2. Second, a specific PPV measure is not mentioned i.e., whether the PPV is based on the initial screening exam results (PPV1), or referral to biopsy (PPV2), or biopsy results (PPV3). Third, whichever PPV is applied, it is important to recognize that PPV is strongly associated by the underlying prevalence of disease, which varies considerably within the target population. This is commonly misunderstood, as if the PPV were an alternative way to measure test sensitivity. If a test had good sensitivity and specificity, the PPV can give some insights into the cost-effectiveness of detection in a target group based on the prevalence of disease. This is useful mostly at the time when the efficacy of screening is being determined. However, when measuring the performance of a new test against existing, proven tests, and when screening is established, it doesn’t seem that any measure of the PPV is likely to be very informative compared to the basic screening outcomes (sensitivity and specificity). For the population covered by Medicare, prevalence of any cancer will be higher than the broader target population usually set by screening guidelines. For PPV1 , the value is very low for all screening tests given the fact that most initial tests with positive findings proceed to a 2nd test to reconcile these findings, with very few patients subsequently being referred to biopsy and being determined to have cancer. In this instance, if a blood-based biomarker test has somewhat similar test or program sensitivity and specificity, which would be the first consideration, it is unclear what PPV adds to the determination of acceptability.

Although we agree that test adherence is not a validation measure, it nonetheless is an important and relevant measure, and one that is increasingly appreciated in national programs. Few citations in the screening literature are as commonly cited as Wilson and Jungner’s Principles and Practice of Screening for Disease, which is highlighted in the PDM. In the list of assessment categories is “acceptability,” which could be interpreted as willingness to undergo screening with a particular test. The success of a screening program at its core is attributable to the sensitivity of the test to detect the cancers that are potentially lethal, and the attendance rate (i.e., “adherence,” or “acceptability”). Tests for colorectal cancer screening historically have been evaluated based on test sensitivity and specificity, and there has been a wide range of performance between the older generation of stool tests and endoscopy. More recently, when modeling has been added to the evidence, long-term comparisons of the recommended screening tests have shown similar performance in terms of deaths avoided and life-years gained. While these ideal scenarios are useful for head-to-head comparisons, they also have been criticized for not being reflective of reality, i.e., experience to date, both in terms of the attendance rate recommended for the different tests (1, 3, 5, and 10 years), and the acceptability of the tests. In some European countries, new trials are focusing on choice of tests, and looking not only at test sensitivity and specificity, but also population-based adenoma and cancer detection rate based on the uptake rate, i.e., the adherence rate. Here the thinking is clear, if only one test can be offered, if a test with lower performance has a higher uptake rate than a higher performance test, then on a population basis the lower performance test may be the better option. Moreover, if offering a choice between a lower performing test and a higher performing test would measurably increase overall adherence and detection rate of priority lesions, then this could be more productive in terms of avoiding adverse outcomes than continuing to hope that uptake of a less popular test will increase.

Given that a sizable fraction of the population, including the Medicare population, does not follow through on recommendations for cancer screening, and vary considerably in which test they prefer, CMS should consider test adherence in their approach to coverage. If a 10% or more increase in screening could be achieved with a blood-based biomarker test in a cohort of older, unscreened individuals, then even with lower performance compared with another, recommended screening test, “the best test may be the one that gets done.”

Level of specificity: We have concerns with CMS’ characterization of the level of specificity required for screening test coverage. The PDM states, “[t]he clinical implications of low specificity [emphasis added] for a CRC screening test is missing individuals with colorectal cancer. Low specificity increases the false negative rate leading to a negative screening test result for an individual with colorectal cancer. The individual who has a follow up diagnostic colonoscopy could be positive for colorectal cancer, despite having a negative CRC screening test. Thus, a person with colorectal cancer will be missed since the screening test was incorrectly negative.” It appears that the PDM has confused specificity with sensitivity in this instance. A test with low specificity actually could result in higher sensitivity, but the emphasis here is that low sensitivity is associated with false negatives, not low specificity. Further, it is not clear why a person would have a diagnostic colonoscopy after a negative screening test. Even further, we should not refer to a follow-up colonoscopy after a positive stool or blood-based biomarker test as “diagnostic” since it simply is part of the continuum of screening, i.e., additional “information gathering” related to whether or not the patient has a suspicious lesion that should be examined by a pathologist. If the patient only has a screening colonoscopy and advanced neoplasia is excised, the test, by definition under the Patient Protection and Affordable Care Act, does not become a diagnostic test. In summary, when colonoscopy is used as a screening test, or as a follow-up test to a positive non-colonoscopy test, no diagnosis takes place during the colonoscopy.

As noted earlier, in this instance, the blood-based biomarker test is expected to perform better than FIT to potentially qualify for inclusion as a screening option. Here it is stated that the biomarker test does not have better sensitivity compared with FIT, and it has lower specificity. The next statement then dismisses the biomarker test as having “no indirect evidence that using the blood-based Epi proColon® test is appropriate for prevention or early detection of colorectal cancer among Medicare beneficiaries.” It is not credible to state that indirect evidence only exists if a new test has the same or greater sensitivity and specificity of the comparison test. CMS should consider the following questions:

1) If the biomarker test had superior sensitivity and specificity to the lowest performing covered alternative test, but low acceptance by the target population, would it be acceptable for inclusion among the options for screening?

2) If CMS were to consider:

1. the convenience and greater acceptability of a blood-based test in an unscreened population as a desirable feature, and
2. if the biomarker test, with similar sensitivity to an alternative approved test, would result in an increase in the detection of occult cancer in the population that otherwise would not occur, and
3. if lower specificity would result in a higher rate of follow-up testing with colonoscopy compared with the alternative test (which would not take place based on the patient’s prior history of no screening), which, if normal, would likely be the last colorectal screening test a patient would ever receive given their age, and thus,
4. isn’t it possible that this is desirable outcome?

CMS should take into consideration also that measures of sensitivity and specificity are highly variable for the same test across studies and study populations. Moreover, as noted above, a screening test may vary in sensitivity and specificity within a target group, and the consequences of a false negative and false positive also can vary, with implications for the thresholds for each. The point we would emphasize further is that colorectal cancer screening is unique. When follow-up colonoscopy determines that the initial non-colonoscopy screening test was a false positive (i.e., the colonoscopy is a true negative), the patient has completed screening and rather than returning to screening in 1-3 years (based on the interval set for the non-colonoscopy test), they are not due for repeat screening with any colorectal cancer screening test for a period of 10 years. For a Medicare beneficiary, the most recent test could be their last colorectal cancer screening test. If colorectal cancer screening rates could be increased due to greater acceptability of a lower performing test, then the overall net benefit at the population level is highly effective. Although concerns about lower sensitivity and missed cancers are apt, in an unscreened population, all cancers are missed until they become symptomatic. Thus, in our judgement, it is a mistake to neglect the very sound advice from Wilson and Jungner⁸ that acceptability is a key factor in judging whether a screening test should be offered to a target population.

Appropriateness of blood-based biomarker screening test for the Medicare population: The PDM seeks comment on whether the evidence is sufficient to determine that colorectal cancer screening using a blood-based biomarker screening test is appropriate for Medicare beneficiaries. We are concerned that this question is worded in a way that could apply to any blood-based biomarker, rather than the biomarker that is being reviewed in this PDM. As such, the question seems to overreach in anticipation of future blood-based biomarker tests that will need to be evaluated by CMS. Further, it states indirectly that if a test does not meet thresholds for both sensitivity and specificity, then it would not be judged as having any potential benefit at all. As we stated earlier, to date no cancer screening test has been evaluated for inclusion or exclusion when compared with another based solely on these criteria.

There are no standard thresholds for sensitivity and specificity across cancer screening tests. This is likely due to the fact that characteristics of the disease, variation in the consequences of false negatives and false positives, costs measured in various ways, all are weighed collectively to reach a decision to include or exclude a cancer screening test in recommendations. A new test with significantly lower sensitivity compared with existing tests will always be impermissible. However, if the test has slightly lower sensitivity, and significantly better specificity, it might be considered, especially if the consequences of a false negative are not dire, and false positives carried significant undesirable outcomes. Although a test with lower sensitivity and lower specificity, and the same requirements/process for test completion (say, another stool test), would not be considered, tests with lower sensitivity and specificity could be considered if they would be more acceptable than alternative tests to a sizable percentage of the persistently unscreened cohort simply because they could measurably increase the screening rate and the rate of detection of important occult lesions. We would advise CMS to not strictly evaluate a blood-based biomarker test based on fixed thresholds of sensitivity and specificity, especially when there are a range of conventional tests in the market using the same technology that vary in their performance and have not been validated in population-based studies, i.e., the majority of FDA-authorized stool tests.

A fixed threshold for sensitivity and specificity could also have adverse consequences for discovery. At this time, FDA clears stool test for the detection of occult blood based on the lowest level of performance of a cleared test (a predicate), and thus, although there may be incentive for a manufacturer to produce a higher performing test, there also is an incentive to produce a test that is equal to or only slightly better than the lowest performing test, especially since patients, clinicians, and purchasers are not aware of the performance characteristics or how they were determined, only that the test has been cleared by the FDA. In a sense, with this proposal, CMS would be adopting a similar strategy, albeit one in which a blood-based biomarker test has had to achieve a higher benchmark than other, conventional reimbursed tests.

Screening rates: Finally, we note the screening rate of 79% screened annually is mistaken. In the BRFSS, 79% of Medicare aged adults (65-75) report being up to date on colorectal cancer screening, which could include colonoscopy within the past 10 years, sigmoidoscopy or CT colonography within the past 5 years, and stool testing (depending on the test) within the past 1-3 years. This is different from annual screening. In addition, The PDM should acknowledge that screening rates derived from sample surveys are overestimated. For example, HEDIS screening rate data for Medicare HMOs and PPOs in 2018 was 72%, and these rates are direct measures of screening from chart or billing data.

Conclusion
The ACS and ACS CAN appreciate the opportunity to provide comments on the Proposed Decision Memo for Screening for Colorectal Cancer - Blood-Based Biomarker Tests (CAG-00454N). Our comments should not be interpreted as agreement or disagreement with the proposed decision related to coverage of Epi proColon®. Rather, our comments are 1) editorial with respect to the information included in the PDM, and 2) specific with respect to the proposal for a pathway for evaluation for future candidate blood-based biomarker screening tests for colorectal cancer screening, and by extension, screening for other cancers.

We disagree with the CMS’s proposal to set arbitrary thresholds for sensitivity and specificity as a basis for blood-based biomarker screening tests. We also disagree with the decision to not consider acceptability, i.e., adherence data, as a consideration in the evaluation of the screening test under evaluation.

Sincerely,
/filed electronically/
Robert A Smith, PhD
Senior Vice President, Cancer Screening
American Cancer Society

References
1. American Cancer Society. Cancer Prevention & Early Detection Facts & Figures 2019-2020. Atlanta: American Cancer Society; 2019.
2. American Cancer Society. Cancer Facts and Figures 2019. Atlanta: American Cancer Society; 2019.
3. Wolf AMD, Fontham ETH, Church TR, et al. Colorectal cancer screening for average-risk adults: 2018 guideline update from the American Cancer Society. CA Cancer J Clin. 2018;68: 250-281.
4. Brouwers MC, Kerkvliet K, Spithoff K, Consortium ANS. The AGREE Reporting Checklist: a tool to improve reporting of clinical practice guidelines. Bmj. 2016;352: i1152.
5. Imperiale TF, Ransohoff DF, Itzkowitz SH, Turnbull BA, Ross ME. Fecal DNA versus fecal occult blood for colorectal-cancer screening in an average-risk population. N Engl J Med. 2004;351: 2704-2714.
6. Collins JF, Lieberman DA, Durbin TE, Weiss DG. Accuracy of screening for fecal occult blood on a single stool sample obtained by digital rectal examination: a comparison with recommended sampling practice. Ann Intern Med. 2005;142: 81-85.
7. Levin B, Lieberman DA, McFarland B, et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. CA Cancer J Clin. 2008;58: 130-160.
8. Wilson JMG, Jungner G. Principles and practice of screening for disease. Geneva, Switzerland: World Health Organization, 1968.

Less

November 14, 2020

VIA Electronic Submission to cms.gov

Tamara Syrek Jensen, J.D.
Director, Coverage and Analysis Group
Centers for Medicare & Medicaid Services
Mail Stop # S3-02-01
7500 Security Boulevard
Baltimore, MD 21244-1850

RE: Proposed Decision Memo for Screening for Colorectal Cancer – Blood-Based Biomarker Tests (CAG-00454N)

Dear Ms. Syrek Jensen:

GRAIL, Inc. (GRAIL) appreciates the opportunity to comment on the Proposed Decision Memo for Screening for Colorectal Cancer – Blood-Based Biomarker Tests (CAG-00454N). GRAIL is a healthcare company whose mission is to detect cancer early, when it can be cured. Through our population-scale clinical studies and state-of-the-art machine learning and data science, we have developed the Galleri™ multi-cancer early detection (MCED) blood test. Our breakthrough-designated blood test has demonstrated in a clinical study the ability to detect more than 50 types of cancer including colorectal cancer, across all stages, and to localize the cancer signal with a high degree of accuracy [1], [2].

GRAIL applauds the agency for prioritizing Medicare beneficiary access to new cancer screening technology upon approval from the U.S. Food and Drug Administration (FDA) and for recognizing that blood-based tests can improve early detection. By establishing a coverage pathway for FDA-cleared or approved blood-based colorectal cancer (CRC) screening tests that meet certain criteria, CMS can reduce unnecessary delays in beneficiary access to innovative screening technology. As the agency works to improve beneficiary access to colorectal cancer screening, the lack of available screenings for most cancer types highlights a significant unmet need for new screening technology in the Medicare program. Novel MCED technology promises to help significantly bend the curve on cancer mortality and dramatically improve the population cancer detection rate.

I. Support CMS Prioritizing Coverage for Innovative FDA-Authorized Cancer Screening Tests

CMS proposes to automatically cover any blood-based CRC screening tests that meet certain pre-specified criteria. GRAIL supports CMS’ goal to establish a prospective coverage framework for FDA-cleared or approved blood-based biomarker tests for colorectal cancer screening. We believe that the use of prospective coverage criteria for a category of devices is a beneficial approach that provides transparency and predictability for innovative test makers, providers, and beneficiaries with respect to coverage of tests. Such an approach is important to avoid coverage gaps for tests that the FDA has determined to be safe and effective. We recommend that CMS take a holistic approach to establishing prospective coverage criteria that drives towards an increased cancer detection rate, increases in screening adherence, and the potential of new tests to reduce late-stage cancer diagnoses, particularly in communities with significant cancer disparities.

Along with streamlining beneficiary access, a prospective coverage criteria framework avoids duplicative coverage reviews and helps conserve CMS’ resources. We note that the agency has used the prospective coverage criteria approach successfully in recent years to establish national coverage of FDA-cleared or approved companion diagnostics using an NGS platform for beneficiaries with qualifying types and/or stages of cancer [3]. We commend the agency for continuing to find ways to streamline coverage of new cancer screening technology for Medicare beneficiaries to avoid unnecessary access delays.

Finally, we also encourage CMS to clarify that colonoscopies performed as a follow-up to other non-invasive colorectal screening tests will be covered and patients will not face cost sharing. A follow-up colonoscopy is a critical component in the colorectal cancer screening process and should be treated as part of the screening encounter. To increase cancer detection, patients should not face unexpected costs from a follow-up colonoscopy, as such costs can be a barrier to patient access.

II. The Critical Need for Multi-Cancer Early Detection in Medicare

Early detection is among the most critical advances in the War on Cancer today [4]. Detecting cancer early, before it has spread through the body, saves lives. According to practice guidelines and literature, nearly all early cancers have effective treatments, even if monitoring is recommended for some. Given that treatment is more effective when cancer is detected early, existing early detection tests such as colonoscopies have led to a substantial reduction in mortality.

Unfortunately, the important benefits of early cancer detection have reached too few Americans. The early cancer detection tests currently available and covered by Medicare detect only five (5) cancers, which means there is no early detection for most cancer types. Over 70% of the 600,000 cancer deaths each year are from cancers without recommended early detection tests [5]. Thus, there is a clear need for new early cancer detection among Medicare beneficiaries. In fact, age is the leading risk factor for cancer, which places Medicare beneficiaries at an elevated risk. Diagnosing and treating cancer earlier can mean better health outcomes for patients and less expensive treatments for the Medicare program [6].

While advances in CRC screening can expand the benefits of early detection of CRC to more individuals, patients do not know which cancer they may be harboring when they get screened. Additionally, screening adherence for the single cancer screening tests covered by Medicare remains a chronic challenge. By transitioning away from screening for individual cancers and instead screening individuals for all possible cancers we could dramatically improve the cancer detection rate (CDR) in the population. The CDR is a measure of population yield of a screening program – the number of cancers detected out of those expected. While we have made incredible strides in existing screenings, the CDR is still only about 16 percent nationwide [7], and this alone will not address the larger public health crises caused by cancer. Advances in genomic medicine provide a unique opportunity to increase the CDR. Modeling estimates that combining MCED tests with current screening tests could detect 50 percent of all cancers, including lethal cancers like pancreatic, stomach, and lung with five-year survival rates under 50 percent. From a public health perspective, recognizing that it is not feasible to develop new screening approaches for each individual cancer, a 50 percent CDR could lead to the early detection of many cancers at a point at which potentially curative treatment is available.

GRAIL’s novel genomic approach has shown the ability to detect more than 50 cancers by looking at DNA methylation in the blood, recognizing the patterns associated with cancer (cancer signals), and localizing the cancer signals to specific tissues of origin. The Galleri test is designed to complement, not replace, existing guideline-recommended early detection tests like colonoscopies, and can detect cancers without existing early detection tests today. An innovation like the Galleri MCED test can make testing more accessible and help fill the gaps that remain in the use of existing early detection tests. Our data suggest that the impacts of this technology have the potential to be substantial – models indicate that if all cancers diagnosed after metastasis could instead be diagnosed at earlier stages, the U.S. could see up to a 24% reduction in five-year cancer mortality [8]. Thus, innovative blood-based MCED screening tests, like Galleri, are a necessary component of the strategy to bend the cancer mortality curve.

III. Conclusion

GRAIL appreciates CMS’ efforts to prioritize cancer early detection and support the uptake of new technology that can expand access to more individuals and increase screening adherence. We also support CMS’ goal of reducing unnecessary coverage delays following FDA approval to increase access for millions of beneficiaries. With age as the leading risk factor for cancer, a new multi-cancer early detection approach has the potential to improve health outcomes for Medicare beneficiaries and take a more public health-focused approach to early detection.

Thank you for considering our comments. If you have any questions or would like additional information, please contact me at jofman@grailbio.com.

Sincerely,

Joshua J. Ofman MD, MSHS
Chief Medical Officer and External Affairs
GRAIL

---

[1] “GRAIL Announces Significant Progress with Multi-Cancer Early Detection Test Including FDA Breakthrough Device Designation,” May 13, 2019. Available at https://grail.com/press-releases/grail-announces-significant-progress-with-multi-cancer-early-detection-test-including-fda-breakthrough-device-designation/
[2]For more on GRAIL’s clinical studies and publications, please see https://grail.com/science/publications/
[3] Available at https://www.cms.gov/medicare-coverage-database/details/nca-details.aspx?NCAId=296&type=Closed&bc=AAgAAAAACAAA&.
[4] Seth A. Seabury et al., Quantifying Gains in the War on Cancer Due to Improved Treatment and Earlier Detection, 19 Health Econ. Policy 141-156 (2016), available at https://www.ncbi.nlm.nih.gov/pubmed/31419891.
[5] Data on file from SEER 18 Regs Research Data, Nov 2017 Sub. Includes persons aged 50-79. Estimated deaths per year in 2020 from ACS Cancer Facts and Figures Report 2020. Recommended early detection tests include breast, cervical, colorectal, and high-risk lung cancer.
[6] Mariotto AB, Enewold L, Zhao J, Zeruto CA, Yabroff KR. Medical care costs associated with cancer survivorship in the United States. Cancer Epidemiol Biomarkers, Prev 2020;29:1304–12.
[7] Ofman, Josh MD “The cancer detection rate—a public health approach to early detection” The Cancer Letter, October 2, 2020. Available at https://cancerletter.com/articles/20201002_7/
[8] Clarke, C. A., et al., Projected Reductions in Absolute Cancer–Related Deaths from Diagnosing Cancers Before Metastasis, 2006–2015. Cancer Epidemiology Biomarkers & Prevention. May 2020. Available at: https://cebp.aacrjournals.org/content/29/5/895?iss=5#sec-13

Less

November 15, 2020

Tamara Syrek Jensen, JD
Director, Coverage and Analysis Group
Centers for Medicare & Medicaid Services
7500 Security Boulevard
Baltimore, MD 21244

RE: Proposed Decision Memo for Screening for Colorectal Cancer Blood-Based Biomarker Tests (CAG-00454N)

Dear Ms. Syrek Jensen:

The Association for Molecular Pathology (AMP) appreciates the opportunity to provide comments on the Centers for Medicare & Medicaid Services’ (CMS) proposed decision memo for screening for colorectal cancer blood-based biomarker tests. AMP is an international medical and professional association representing approximately 2,500 physicians, doctoral scientists, and medical technologists who perform or are involved with laboratory testing based on knowledge derived from molecular biology, genetics, and genomics. Membership includes professionals from academic medicine, hospital-based and private clinical laboratories, the government and the in vitro diagnostics industry.

AMP commends CMS for providing a proactive coverage standard that recognizes the value of blood-based biomarker tests as another non-invasive option for the early detection of colorectal cancer. This is an evolving area of medicine and having an established pathway to national coverage will expand patient access to medically-appropriate screening tests as they are introduced clinically. Additionally, many Medicare beneficiaries fail to comply with recommendations for colorectal cancer screening and providing more methods for individuals to be screened will likely help increase access, adherence, and earlier detection of colorectal cancer.

While we are appreciative of this coverage standard, we do have some concerns regarding the proposed coverage policy and its implications for future coverage policies. As such, AMP would like to provide comments on the following proposed coverage criteria:

• FDA market authorization with an indication of colorectal cancer screening;
• Proven test performance characteristics for a blood based screening test with both sensitivity greater than or equal to 74% and specificity greater than or equal to 90% in the detection of colorectal cancer compared to the recognized standard (accepted as colonoscopy at this time), based on the pivotal studies included in the FDA labeling; and
• Inclusion as a recommended routine colorectal screening test in at least one professional society guideline or consensus statement or USPSTF recommendation.

1. FDA market authorization with an indication of colorectal cancer screening

Requiring FDA market authorization unreasonably narrows patient access to clinically useful blood-based colorectal cancer screening tests, and as drafted, this policy eliminates any possible path for tests without FDA authorization; it states that “all other indications for colorectal cancer screening not otherwise specified in the ACT and regulations, or otherwise specified above remain nationally non-covered.” Blood-based biomarker screening tests are a growing and dynamic type of testing. To ensure that clinically and analytically-validated colorectal screening tests are recognized by Medicare as a viable, covered service for beneficiaries, CMS should provide coverage pathways within the NCD for other tests, such as those performed in a Clinical Laboratory Improvement Amendments (CLIA)-certified laboratory, in addition to those tests that voluntarily seek authorization by the Food and Drug Administration (FDA).

Both the CMS Coverage and Analysis Group and the Department of Health and Human Services (HHS) have existing policies that support coverage pathways for laboratory developed testing procedures and clarification that such tests do not require FDA authorization. First, the National Coverage Determination (NCD) for Next Generation Sequencing (NGS) (90.2) provides coverage pathways for both FDA-approved or –cleared tests as well as laboratory developed testing procedures, the latter via promulgation by Medicare Administrative Contractors (MACs)¹. Secondly, after review by the Office of the General Counsel this year, HHS “determined that the Food and Drug Administration (“FDA”) will not require premarket review of laboratory developed tests (“LDT”) absent notice-and-comment rulemaking, as opposed to through guidance documents, compliance manuals, website statements, or other informal issuances.”² This rescission of guidance documents and other informal policy specifically states that laboratory developed testing procedures will remain subject to regulation by CMS CLIA, which assures laboratory performance standards and laboratory developed testing procedures’ accuracy, validity, and reliability. Therefore, mandating FDA authorization for laboratory developed testing procedures as a prerequisite to receiving national coverage under Medicare is incongruent with and unnecessary under current HHS policy. We ask that the NCD be modified to be consistent with current regulatory policy within HHS.

2. Proven test performance characteristics for a blood based screening test with both sensitivity greater than or equal to 74% and specificity greater than or equal to 90% in the detection of colorectal cancer compared to the recognized standard (accepted as colonoscopy at this time), based on the pivotal studies included in the FDA labeling

AMP has significant concerns with CMS dictating sensitivity and specificity requirements in any coverage policy and believes it is not the role of the Coverage and Analysis Group at CMS to evaluate tests’ sensitivity and specificity requirements in order to determine a test’s medical usefulness. Diagnostic performance, such as test sensitivity and specificity, should be determined by FDA in the event the test is an in vitro diagnostic test kit, or under the CLIA program if the test is a laboratory developed testing procedure. The Coverage and Analysis Group dictating performance standards is outside its mandate to evaluate the clinical utility of a service and even more concerning, staff lack the scientific expertise to establish performance characteristics. This should be left to agencies and divisions with those oversight authorities, i.e. FDA and the CLIA Program. Moreover, we believe this criterion sets a dangerous precedent for future coverage policies. For these reasons, we strongly recommend deletion of this criterion. AMP welcomes the opportunity to discuss this issue and our concerns further with CMS.

3. Inclusion as recommended routine colorectal screening test in at least one professional society guideline or consensus statement or USPSTF recommendation

We are pleased to see these guidelines recognized as a criterion for coverage in this policy. AMP strongly supports this coverage criterion and specifically, the inclusion of a recommended routine colorectal screening test in at least one professional society guideline or consensus statement. AMP continues to believe that coverage policies should align with evidence-based guidelines written by professional medical societies. When guidelines support the utility of a test, they also support coverage for that test. For example, specific to this coverage policy, the current National Comprehensive Cancer Network guidelines for Colorectal Cancer Screening acknowledge that blood-based screening tests for colorectal cancer are an emerging testing option.³

AMP also invests in a robust program to leverage the field’s expertise to develop Clinical Practice Guidelines and Reports, which assist laboratory and other health care professionals by providing guidance and recommendations for particular areas of practice. The AMP Clinical Practice Guidelines Program is comprised of multiple AMP-led working groups that plan, organize and coordinate efforts such as practice guidelines, sample exchanges, reporting surveys, validation and quality control measures. The majority of these projects include representation from other professional organizations and groups. AMP’s Clinical Practice Guidelines Program as well as other efforts from additional organizations such as the National Comprehensive Cancer Network (NCCN), the American Society of Clinical Oncology (ASCO), American College of Genetics and Genomics (ACMG), and the College of American Pathologists (CAP) are key to establishing parameters for clinical practice.

Thank you again for the opportunity to review and provide comments on this proposed decision memo. AMP appreciates CMS’ efforts to ensure access to medically necessary testing and provide coverage for testing more broadly. We welcome the opportunity to work with you and should you have any questions or require additional information, please direct your correspondence to Tara Burke, Senior Director of Public Policy and Advocacy, at tburke@amp.org.

Sincerely,

Karen E. Weck, MD FCAP
President, Association for Molecular Pathology

---

1 https://www.cms.gov/medicare-coverage-database/details/ncd-details.aspx?NCDId=372
2 https://www.hhs.gov/coronavirus/testing/recission-guidances-informal-issuances-premarket-review-lab-tests/index.html
3 https://www.nccn.org/professionals/physician_gls/pdf/colorectal_screening.pdf

Less

CellMax applauds CMS for proposal of rolling coverage of an NCD for colorectal cancer screening with liquid biopsy and appreciates the opportunity to comment on the proposal. Colorectal cancer detection is a major clinically unmet need. There will be an estimated 147,950 new CRC cases and 53,2000 deaths from colorectal cancer (CRC) in 2020 (https://seer.cancer.gov/statfacts/html/colorect.html). Colorectal cancer is the second most deadly cancer but sadly among the most preventable cancers. In addition, there are examples of both testing under-surveillance and over-surveillance that would be informed with an accurate risk-based actionable non-invasive test. The primary reasons for the number of incident cases and deaths are that approximately 30% of individuals do not follow recommended guidelines for screening (1) and only 50% abide by multi-society guidelines for follow up colonoscopy after polypectomy (2,3). The reasons for not having colonoscopy or undergoing stool testing are well studied and known but efforts to increase testing have only accomplished modest increases. The COVID-19 pandemic has decreased the average weekly colorectal cancer (CRC) colonoscopy screening volumes by an estimated 90% and whether and when screening will return is unclear further highlighting the unmet need for a non-invasive test (4). CellMax also supports the requirement for FDA authorization and inclusion in professional society guidelines or consensus statements or United States Preventative Services Task Force (USPSTF) recommendation though we note the often protracted times of updated guidelines.

In order to impact outcomes, CellMax does, however, recommend CMS edit the NCD for colorectal cancer screening to

i) include detection of advanced adenomas to reflex those people with adenomas to actionable colonoscopy,
ii) define specificity relative to the absence of any adenoma or colorectal cancer detected by colonoscopy to avoid penalizing a test for detection of early stage disease and
iii) increase the required sensitivity for colorectal cancer above 74% to minimize cases of cancer that go undetected and the specificity above 90% to minimize unnecessary reflex colonoscopies.

A blood-based test with the NCD proposed sensitivity and specificity would not be in equipoise for those tested with FIT or colonoscopy.

We suggest these changes for multiple reasons:

First, guidelines from multiple organizations outline dual objectives for screening. The dual objectives include adenoma detection to avoid future cancer development as well as cancer detection in as an early a stage as possible to increase overall survival and decrease severity and adverse effects of therapy. Since there would be no performance for adenomas required, even advanced adenomas, adenomas would go unreported and these patients would not be in equipoise with the widely endorsed colonoscopy testing, the alternative test/procedure. The National Polyp Study reported an approximately 50% reduction in mortality that accrued from removal of all adenomas (5). In addition, organizational guidelines recommend that screening be informed by the increasing understanding of natural history of the adenoma-carcinoma pathway. Our new understanding of the adenoma-carcinoma pathway includes i) early ‘big bang’ evolution rather than linear driver sweep evolution (6-8), ii) early dissemination of cells that become metastatic later in course of illness (7, 9-13), iii) dormancy of disseminated cells that makes their detection challenging (14-15) and iv) ongoing communication between disseminated cells and primary adenoma as well as playing a role in conditioning of pre-metastatic sites (16-17).

A non-invasive test that only detects frank cancer would not in equipoise for screened individuals with FIT or colonoscopy for the tested population. CMS should require a minimal performance for advanced adenoma detection. CMS and the FDA are to be recognized for including advanced adenoma on the intended use of the Cologuard test (albeit at only 42% sensitivity (18)) so it is unclear why detection of advanced adenomas would go unreported in the proposed NCD. The annual progression rate from advanced adenoma to cancer ranges from 3-6%. (19) with a strong age gradient. The cumulative progression rate from advanced adenoma would depend on the interval of time until the next test; for example, a 3 year interval as proposed by NCD could result in a progression rate from advanced adenoma of 9-18%. In brief, we now know that adenomas are not benign as their name implies.

The adenoma-carcinoma pathway spans approximately 5-10 years but varies with the individual and rather than consisting of clearly delineated steps or extended dwell times, consists of co-evolution of neoplastic cells, modifications of extracellular matrix, tumor vasculature re-modeling and increasing role of resident and recruitment of fibroblasts and immune cells to the microenvironment (20). Along with the genetic and epigenetic changes in neoplastic cells and cross talk in the proliferating lesion, there is an emergence of increasingly malignant features. Removal of adenomas is therefore best thought of as treatment of early disease rather than before the disease process. In addition, those tested with a binary test of cancer or no cancer would not be in equipoise with colonoscopy since colonoscopy provides prognostic information for interval testing (when the next test should be performed) whereas the proposed blood-based test would only provide binary diagnostic (cancer or no cancer) rather than prognostic information. Unless a blood test can detect adenomas, it should not be allowed to replace colonoscopy.

Second, setting the sensitivity and specificity at 74% and 90%, respectively needs to be translated into the patient applicable Positive and Negative Predictive Values which are the clinically relevant diagnostic metrics. At a prevalence of 0.5% and an assay with 74% sensitivity and 90% specificity, the Positive Predictive Value (PPV) would only be 3.6% and Negative Predictive Value (NPV) would be 99.8%. Said another way, in every 1000 people tested even though 4 would be identified as true positive and 896 would be true negatives for CRC, there would be 1 case of CRC that would go undetected (false negative) and 100 cases that would not be CRC (false positives). While the false positives may be due to advanced adenomas, there is no assurance of that and in the absence of reporting adenomas, critical screening test performance information would be absent.

Test performance across the stages of CRC is critical to consider as well. The predictive values are further reduced for stage I CRC for FIT and Cologuard because these assays have considerably lower sensitivity for stage I CRC, in particular T1-stage I relative to stages II-IV. For example, a meta-analysis of quality studies suggested that the sensitivity of FIT was approximately 8% lower in CRC stage I compared with stages II-IV. T1 stage I may be as low as 40% for FIT (21). In comparison, the prevalence of advanced adenomas is approximately 6% leading to marked increases in Positive Predictive Values.

Third, reimbursement of a test for cancer without performance for adenoma risks may undergo indication drift-in both directions (22). Patients with unrelated severe comorbidities whose overall survival may not be impacted by CRC screening may wish to undergo screening if a blood-based test were available and patients who are candidates of more interventional procedures such as colonoscopy may wish to be tested in lieu of colonoscopy leading to a double expense or avoidance of a procedure that would save their lives.

In summary, CellMax recommends CMS edit the NCD proposal for screening to include
i) a minimal detection rate for advanced adenoma of more than 60%,
ii) define specificity relative to the absence of any adenoma or colorectal cancer detected by colonoscopy to avoid penalizing a test for detection of early stage disease and
iii) require a minimal overall detection rate of 80% with sensitivity of stage I CRC of at least 70%. Further, since only 50% of individuals are compliant to recommended timing for follow up colonoscopy after polypectomy, CMS should also propose an NCD for non-invasive surveillance testing recognizing that approximately 25% of colonoscopies are follow up colonoscopies after polypectomy (ref). Lastly, CellMax urges changes in overall reimbursement for CRC screening and surveillance that would lower co-pays for recommended colonoscopies after a positive non-invasive test such as FIT or future blood-based tests. Sadly, CRC is a preventable cancer that disproportionately impacts the vulnerable population of uninsured and under insured.

References
1. Joseph et al. Vital Signs: Colorectal Cancer Screening Test Use — United States, 2018 MMWR 69, 253 (2018).
2. Lieberman and Gupta Does Colon Polyp Surveillance Improve Patient Outcomes? Gastroenterology 158, 436 (2020).
3. Gupta et al. Recommendations for Follow-Up After Colonoscopy and Polypectomy: A Consensus Update by the US Multi-Society Task Force on Colorectal Cancer Gastroenterology 158, 1131 (2020).
4. Gupta and Lieberman Screening and Surveillance Colonoscopy and COVID-19: Avoiding More Casualties Gastroenterology 159, 1205 (2020).
5. Zauber et al. Colonoscopic Polypectomy and Long-Term Prevention of Colorectal-Cancer Deaths NEJM 366, 687 (2012).
6. Sottoriva et al. A Big Bang model of human colorectal tumor growth Nat Genet 47, 209 (2015).
7. Hu et al. Quantitative evidence for early metastatic seeding in colorectal cancer Nat Genet 51, 1113 (2019).
8. Ryser et al. Spatial mutation patterns as markers of early colorectal tumor cell mobility 115, 5774 Proc Natl Acad Sci USA (2018).
9. Hannahan and Folkman Patterns and Emerging Mechanisms Review of the Angiogenic Switch during Tumorigenesis Cell 86, 353 (1996).
10. Pantel et al. Circulating Epithelial Cells in Patients with Benign Colon Diseases Clin Chem 58, 936 (2011).
11. Yang et al. HIFs, Angiogenesis, and Cancer J Cell Biochem 114, 967 (2013).
12. Paduch The role of lymphangiogenesis and angiogenesis in tumor metastasis Cell Oncol 39, 397 (2016).
13. Hu and Curtis Looking backward in time to define the chronology of metastasis Nature Comm 11, 3213 (2020).
14. Yadav et al. The Biology and Therapeutic Implications of Tumor Dormancy and Reactivation Frontiers in Oncol 8, 1 (2018).
15. Yeh and Ramaswamy Mechanisms of Cancer Cell Dormancy—Another Hallmark of Cancer? Can Res 75, 5014 (2015).
16. Maia et al. Exosome-Based Cell-Cell Communication in the Tumor Microenvironment Frontiers in Cell and Development Biol 6, 1 (2018).
17. Mannavola et al. Revisiting the Role of Exosomes in Colorectal Cancer: Where Are We Now? Front Oncol 9, 1 (2019).
18. Imperiale et al. Multitarget Stool DNA Testing for Colorectal-Cancer Screening NEJM 370, 1287 (2014).
19. Brenner et al. Risk of progression of advanced adenomas to colorectal cancer by age and sex: estimates based on 840 149 screening colonoscopies Gut 56, 1585 (2007).
20. Tape The Heterocellular Emergence of Colorectal Cancer Trends in Cancer 3, 79 (2017).
21. Niedermaier et al. Stage-Specific Sensitivity of Fecal Immunochemical Tests for Detecting Colorectal Cancer: Systematic Review and Meta-Analysis Amer J Gastroenterology 115, 56 (2020).
22. Parikh and Prasad Blood-Based Screening for Colon Cancer A Disruptive Innovation or Simply a Disruption? JAMA 315, 2519 (2016).

Less

Tamara Syrek Jensen
Director, Coverage and Analysis Group
Centers for Medicare & Medicaid Services
Mail Stop # S3-02-01
7500 Security Boulevard
Baltimore, MD 21244-1850

RE: National Coverage Analysis (NCA) Tracking Sheet for Screening for Colorectal Cancer – Blood-Based Biomarker Tests (CAG-00454N)

Dear Ms. Jensen:

On behalf of Guardant Health, I would like to submit our comments related to the National Tracking Sheet for Colorectal Cancer (CRC) screening – Blood-Based Biomarker Tests. Guardant Health is a company that is dedicated to blood based liquid biopsy testing focused across the entire lifecycle of solid tumor oncology diagnostics. Our current clinically available testing focuses on advanced cancer to assist physicians to determine the optimal treatment regimens for patients with advanced cancer. We also have products under development to detect Minimum Residual Disease (MRD) in early stage cancer patients as well as for cancer screening. Our CRC screening program started enrollment on an approximately 10,000-patient clinical validation study in October of 2019 with an expected completion date in 2021.

Guardant would like to express our support for CMS creating a national coverage policy that can allow patient access to new technologies in CRC screening without redetermining NCD 210.3. We believe that this will enable faster access to care as new technologies emerge which demonstrate suitable clinical performance along with FDA PMA approval. Guardant Health is supportive of finalizing the NCD with a few minor clarifications/modifications.

The first area of clarification is around the statutory requirement that enables CMS to cover colorectal screening tests “in consultation with appropriate organizations”(1). Guardant would propose that the evidence review conducted by the FDA is far and away the most rigorous review of a technology. Confidential information is submitted to enable a detailed technical assessment that is far more nuanced than a professional society review of a technology. As a result, FDA PMA approval alone should be viewed as meeting the statutory requirement that CMS consults with an appropriate organization. This proposed process would also be aligned with CMS’s Patient’s over Paperwork initiative, enabling the fastest access to new technologies that meet the NCD technical performance criteria of 74% sensitivity and 90% specificity. We believe this approval paradigm will offer the fastest access to patients who may otherwise not be screened. The latest data(2) from the American Cancer Society still demonstrates that 27% of Medicare patients are not up to date with screening. Rapid access to multiple non-invasive screening modalities should heighten screening compliance.

If consulting with a non-governmental organization is deemed to be mandatory, we would propose to broaden the way that a professional society could endorse a new technology. Guardant believes that blood-based CRC testing will be included in several guidelines in the future; however, the frequency of guideline updates often is mismatched with the speed of development of new technologies. For instance, the USPSTF guidelines are updated roughly every 5 years. If a technology were to miss that approval window it may take 4+ years for inclusion and coverage and the pathway for early review is unclear. To ensure access for Medicare beneficiaries, we feel guideline inclusion should not be the only pathway for a professional society to provide support for an FDA PMA approved test. The additional ways that CMS should accept are:

• Endorsement letter from a professional society guideline author
• Endorsement letter from the leader or co-leader of a professional society
• Positive editorial published in a peer reviewed journal related to the use of the test by a guideline committee author or leader
• Endorsement from the GI Oncology department or chief quality officer at a National Cancer Institute (NCI) cancer center

The majority of guidelines do not reference the brand names of specific technologies; most commonly they refer to the technology class. Once the technology class, such as blood-based CRC screening is recommended within certain parameters, all tests that meet the criteria should be viewed as endorsed for CMS coverage.

Additionally, the organizations that should qualify to make these endorsements or guideline recommendations should be publicly available. CMS should initially start with a broad list of existing professional societies. Once the NCD is finalized there should be a mechanism for a society to be nominated to be included in the list either by themselves or by other organizations. If CMS were to decide to remove a society from the list, there should also be a public comment period process where the public could weigh in to ensure the organization was not removed from the list without appropriate diligence.

At a minimum, the initial list of professional societies that we would recommend for endorsement of a test, in no particular order, are the following:

• United States Preventative Services Task Force (USPSTF)
• United States Multi-Society Task Force on Colorectal Cancer (UMSTF)
• American Society of Gastrointestinal Endoscopy (ASGE)
• American College of Gastroenterology (ACG)
• American Gastroenterological Association (AGA)
• American Medical Association (AMA)
• American Academy of Family Physicians (AAFP)
• The American College of Obstetricians and Gynecologists (ACOG)
• American College of Physicians (ACP)
• American Society of Colon and Rectal Surgeons (ASCRS)
• American Cancer Society (ACS)
• National Comprehensive Cancer Network (NCCN)
• American College of Radiology (ACR)
• National Cancer Institute (NCI)
• Cancer centers affiliated with the NCI

Attempting to construct a process to vet professional societies to work with will create additional regulatory hurdles, which, as mentioned previously, is counter to the Patients over Paperwork initiative. Guardant strongly urges CMS to view FDA PMA approval alone as meeting the statutory requirement for consultation.

The second topic that we feel needs further clarification is related to the frequency of testing. In several instances the FDA label does not include the frequency of use for a test. Additionally, after a product has been on the market for several years, further research may demonstrate the optimal frequency for which to run the test. While the FDA label can offer one opportunity to demonstrate the appropriate frequency, professional societies can also weigh in on the appropriate frequency to perform testing for Medicare beneficiaries. We request that, if a professional society recommends that the frequency of testing is more often than either the proposed three-year interval or the FDA label, CMS coverage is extended to the most frequent of the recommendations.

The third topic relates to patient inclusion criteria and the definition of screening. The recently updated draft USPSTF guidelines assigned a “B” grade for screening of patients between the ages of 45-49. We feel that it is appropriate to extend the age range for all Medicare covered CRC screening technologies to 45-85. Additionally, with the increase of technologies whose abnormal findings require confirmatory colonoscopy we think CMS should provide clarity that the follow-up colonoscopy from an abnormal finding of a blood or stool-based test should still be considered part of the screening regimen. Access to care may be put at risk if there is an assigned patient responsibility related to the follow-up colonoscopy.

Guardant Health is strongly committed to conquering cancer with data and our pipeline of products is dedicated to that mission. This NCD will go a long way towards enabling early detection of cancers when they can hopefully be overcome, significantly extending the lives of patients. We look forward to continuing our strong working relationship with CMS as we develop FDA approved screening and diagnostic applications. Thank you for considering our comments in your final determination.

Sincerely,

Helmy Eltoukhy
CEO
Guardant Health

References:
1 https://www.ssa.gov/OP_Home/ssact/title18/1861.htm
2 https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/colorectal-cancer-facts-and-figures/colorectal-cancer-facts-and-figures-2020-2022.pdf

Less

November 13, 2020

Tamara Syrek Jensen, J.D.
Director, Coverage & Analysis Group
Centers for Medicare & Medicaid Services
Mailstop S3-02-01
7500 Security Blvd.
Baltimore, MD 21244

Sent electronically

Re: Proposed Decision Memo for Screening for Colorectal Cancer - Blood-Based Biomarker Tests (CAG-004454N)

Dear Ms. Syrek Jensen:

The Personalized Medicine Coalition (PMC), a multi-stakeholder group comprising more than 200 institutions across the health care spectrum, thanks the Centers for Medicare and Medicaid Services (CMS) for the opportunity to comment on the recently released proposed decision memo for screening of colorectal cancer (CRC) among Medicare beneficiaries using blood-based biomarker tests. PMC appreciates CMS utilizing the National Coverage Analysis (NCA) process to prioritize cancer screening and to develop a proposed decision memo as part of the NCA process that is forward-thinking in establishing coverage for some new screening tests that come to market. The proposed decision memo recognizes that blood-based tests can improve early detection and personalized treatment of multiple cancers, including CRC, and it establishes a coverage approach that could reduce delays in beneficiaries’ access to innovative screening tests that receive approval from the U.S. Food and Drug Administration (FDA).

Personalized medicine is an evolving field that uses diagnostic tools to identify specific biological markers, often genetic, to help determine which medical treatments and procedures will be best for each patient. By combining this information with an individual’s medical history, circumstances, and values, personalized medicine allows doctors and patients to develop targeted prevention and treatment plans.

Personalized medicine is helping to shift the patient and provider experiences away from trial-and-error treatments of late-stage diseases in favor of more streamlined approaches to disease prevention and treatment, which will lead to improved patient outcomes, a reduction in unnecessary treatment costs, and better patient and provider satisfaction. PMC’s members are leading the way in personalized medicine and recommend that patients who may benefit from this approach undergo appropriate testing and tailored treatment as soon as possible during their clinical experiences.

Statement of Neutrality

Many of PMC’s members will present their own responses to CMS and will actively advocate for those positions. PMC’s comments are designed to provide feedback so that the general concept of personalized medicine can advance, and are not intended to impact adversely the ability of individual PMC members, alone or in combination, to pursue separate comments with respect to the proposed decision memo for screening for CRC using blood-based biomarker tests.

Burden of Colorectal Cancer and Value of Earlier Detection

The Colorectal Cancer Alliance estimates that there are more than one million CRC survivors in the United States today. The survival rate for CRC has been increasing in part due to increased awareness, screening and improved treatment options. Earlier detection and treatment are particularly important for the older adult population. For colon cancer, the median age of diagnosis is 68 years of age in men and 72 years of age in women. For rectal cancer, the median age at diagnosis is 63 in both men and women.

Despite strides made to increase survival rates, CRC remains the second leading cause of cancer death among men and women combined in the United States. Nearly half of CRC cases and deaths would be preventable with improved screening. Molecular screening tests, including blood-based cancer tests, are emerging as an additional way to screen for cancer. This broad category of tests will continue to evolve. Many tests in development have the potential to be options for patients who otherwise would not be screened with available screening tools such as colonoscopy, sigmoidoscopy, and fecal occult blood testing. PMC appreciates CMS recognizing the value of earlier detection and seeking to address a major factor contributing to gaps in screening , namely patient concerns over coverage for a test, with the policy outlined in the proposed decision memo.

Considerations for Final Coverage Determination

PMC applauds CMS for responding to stakeholder feedback provided upon opening the NCA, which emphasized opportunities that blood-based tests present for improving early detection and treatment of CRC. When issuing a final coverage determination, PMC urges CMS to consider the following issues in order to allow more streamlined coverage of new blood-based tests as they are approved.

PMC understands that CMS is proposing to cover “blood-based biomarker test as an appropriate colorectal cancer screening test once every 3 years, or at the interval designated in FDA label if the FDA indicates a specific test interval,” for patients age 50-85 years “when performed in a Clinical Laboratory Improvement Act (CLIA)-certified laboratory and when ordered by a treating physician.” A 3-year screening interval reflects current evidence for frequency of testing using an FDA-approved stool-based DNA test but may not be appropriate for blood-based testing where testing intervals have not yet been established. CMS indicates that it will cover blood-based CRC screening tests at another screening interval if it is designated in the FDA label for a new blood-based CRC screening test. However, the frequency of a test is not always dictated by an FDA label. For this reason, CMS’ final coverage decision should allow for test frequency to be adjusted based on professional society recommendations or test-specific data (e.g., modeling studies). Additionally, PMC encourages CMS to align its screening age requirements with current recommendations for CRC screening. The American Cancer Society now recommends certain CRC screening modalities for individuals at age 45 and the U.S. Preventive Services Task Force (USPSTF) proposed a Grade B recommendation for certain CRC screening modalities for patients between ages 45 and 49. We believe any CRC screening tests, including blood-based biomarker tests, that are supported by relevant professional societies and fulfill criteria established by the final coverage determination should be covered for patients as young as age 45.

In order to increase beneficiary access without having to reconsider the final coverage determination, CMS is also proposing a pathway for future coverage if specific threshold test performance criteria are met. A test must have “proven performance characteristics for a blood-based screening test with both sensitivity greater than or equal to 74 percent and specificity greater than or equal to 90 percent in the detection of CRC compared to the recognized standard, as minimal threshold levels, based on pivotal studies included in FDA labeling.” Recognizing the unique balance of benefits and harms for cancer screening tests and the opportunity to increase screening adherence with new technology, PMC has concerns with the performance criteria in this coverage policy. PMC urges CMS to discuss these concerns further with stakeholders and then consider removal of this criteria in the final NCD issued by the agency.

Along with the proposed performance criteria, a test would need to be included as a screening option in at least one professional society guideline, professional society consensus statement, or USPSTF recommendation to meet the requirement for coverage. Given the pace of development in this space and the length of time necessary to issue guidelines, CMS should consider alternative pathways for professional societies to provide feedback on whether tests meet the performance criteria aside from inclusion in guidelines. These pathways may include endorsement letters from professional society leadership and/or editorial publication from professional society leadership in peer-reviewed journals.

Conclusion

Thank you for releasing the proposed decision memo on blood-based biomarker tests for CRC screening and for considering our comments. PMC welcomes the opportunity to serve as a resource for you in continuing to shape coverage policies that impact beneficiary access to personalized medicine tests and treatments so that they achieve the goal we share with CMS of delivering appropriate, efficient, and accessible health care to patients. If you have any questions about the content of this letter, please contact me at 202-499-0986 or cbens@personalizedmedicinecoalition.org.

Sincerely yours,

Cynthia A. Bens
Senior Vice President, Public Policy

References:
Centers for Medicare & Medicaid Services. Proposed Decision Memo for Screening for Colorectal Cancer - Blood-Based Biomarker Tests. October 16, 2020. https://www.cms.gov/medicare-coverage-database/details/nca-proposed-decision-memo.aspx?NCAId=299.
Colorectal Cancer Alliance. Statistics and Risk Factors. https://www.ccalliance.org/colorectal-cancer-information/statistics-risk-factors. Accessed November 13, 2020.
Colorectal Cancer Alliance. Statistics and Risk Factors. https://www.ccalliance.org/colorectal-cancer-information/statistics-risk-factors. Accessed November 13, 2020.
Centers for Disease Control and Prevention. Vital Signs Colorectal Cancer. July 2011. https://www.cdc.gov/vitalsigns/cancerscreening/colorectalcancer/index.html.
Colorectal Cancer Alliance. Screening FAQs. https://www.ccalliance.org/screening-prevention/screening-faqs. Accessed November 13, 2020.
American Cancer Society. Guideline for Colorectal Cancer Screening. https://www.cancer.org/cancer/colon-rectal-cancer/detection-diagnosis-staging/acs-recommendations.html. Accessed November 13, 2020.
U.S. Preventive Services Task Force. Draft Recommendation Statement Colorectal Cancer Screening. October 27, 2020. https://uspreventiveservicestaskforce.org/uspstf/draft-recommendation/colorectal-cancer-screening3#:~:text=In%20the%20current%20draft%20recommendation,years%20(B%20grade%20recommendation).

Less

November 13, 2020

Tamara Syrek Jensen, JD
Director, Coverage & Analysis Group
Center for Clinical Standards and Quality
Centers for Medicare & Medicaid Services
Mailstop S3-02-01
7500 Security Blvd
Baltimore MD 21244

RE: Proposed Decision Memo for Screening for Colorectal Cancer - Blood-Based Biomarker Tests (CAG-00454N)

Dear Ms. Jensen:

The Advanced Medical Technology Association (AdvaMed) is pleased to offer the following comments on the Centers for Medicare & Medicaid Services’ (CMS) proposed decision memorandum for Screening for Colorectal Cancer-Blood-Based Biomarker Tests (CAG-00454N).¹

AdvaMed member companies produce the medical devices, diagnostic products, and health information systems that are transforming health care through earlier disease detection, less invasive procedures, and more effective treatments. AdvaMedDx functions as an association within AdvaMed and its member companies produce advanced in vitro diagnostic tests that facilitate evidence-based medicine, improve quality of patient care, enable early detection of disease and often reduce overall health care costs. Our membership includes manufacturers engaged in the development of innovative diagnostics that support the advancement of public health, including next generation sequencing-based technologies. Throughout this letter, AdvaMed refers to both AdvaMed and AdvaMedDx.

We appreciate this opportunity to provide comments on the proposed NCD. We agree that CMS should expand the scope of colorectal cancer (CRC) screening to include coverage of blood-based screening tests. As CMS points out in its proposed decision memo, consistent CRC screening over the past two decades has significantly decreased mortality due to colorectal cancer.

Medicare covers a wide range of CRC screening tests meeting certain frequency and eligibility criteria, and this proposal would expand that list to include FDA-market authorized blood-based biomarker screening tests that meet specified criteria, including specific test performance characteristics (sensitivity greater than or equal to 74% and specificity greater than or equal to 90% in the detection of colon cancer as opposed to the recognized standard. Further, CMS proposes that the blood-based screening test must be included as a “recommended routine colorectal cancer screening test in at least one professional society guideline or consensus statement or United States Preventive Services Task Force (USPSTF) recommendation.”

AdvaMed is supportive of this movement toward coverage of FDA-approved liquid biopsy CRC screening tests. If finalized, the NCD would result in streamlined coverage for blood-biomarker CRC screening tests that are approved by the FDA, meet the sensitivity and specificity criteria outlined in the NCD, and that are recommended by the USPSTF or other professional society guideline or consensus statement. It would create a new precedent under which newly approved tests meeting the sensitivity and specificity criteria outlined in the NCD will have some certainty regarding coverage.

AdvaMed has long supported development of coverage policies that expand patient access to new tests and technologies based on advancements in technology and the medical landscape, as supported by the public literature, clinical evidence, and input from the relevant physician and specialty societies. We therefore support the direction of the proposed NCD, which would potentially provide speedy coverage of beneficial cancer screening tests.

AdvaMed greatly appreciates the opportunity to provide comments on this draft coverage determination. If you have questions regarding these comments or if you require additional information, please contact me at (202) 434-7219 or cbranham@AdvaMed.org.

Sincerely,

Chandra N. Branham, JD
Vice President, Payment & Health Care Delivery Policy
AdvaMed

---

1 See https://www.cms.gov/medicare-coverage-database/details/nca-proposed-decision-memo.aspx?NCAId=299, Centers for Medicare & Medicaid Services Proposed Decision Memo for Screening for Colorectal Cancer - Blood- Based Biomarker Tests (CAG-00454N), accessed October 16, 2020.

Less

With respect, the criteria for an acceptable blood-based biomarker for CRC screening outlined in paragraph B.3 appear to be arrived at arbitrarily. Mathematically, the performance you prescribe for that assay, when used at 3-year intervals, would be substantially inferior to the performance of the existing Epi proColon (mSeptin9) assay when used annually, in terms of cancers identified and CRC deaths avoided.

Likewise, the rationale for the 90% specificity requirement is not clear. Assuming it is intended to limit unnecessary follow-on colonoscopies, in absolute terms, the marginal impact on the number of follow-on colonoscopies performed for a blood test with 90% specificity compared to that of Epi proColon at approximately 80% is negligible.

The recognized “gold standard” for CRC screening in the average risk population is colonoscopy - which every average risk, age-eligible person would undergo ideally. Unfortunately, at least 50% of that population refuse it. Adding the adherence rates of all other screening modalities combined raises that number to less than 70%. Clearly, the blood test has the potential to reach a huge unscreened population and dramatically improve screening rates.

Epi proColon is intended for average-risk patients, aged 50 – 75, who cannot or will not undergo screening by colonoscopy or FIT. It must be prescribed by a physician. The follow-on recommendation for a patient with an Epi proColon positive test result is colonoscopy – as it is with positive results from all other screening modalities. Thus, screening with Epi proColon will lead to more colonoscopies, not less.

CMS coverage for this test will save lives! I urge you to reconsider your recommendation and pay for this test.

Respectfully,

Less

Via email (and submitted online)
November 13, 2020

Ms. Tamara Syrek-Jensen, JD
Director, Coverage & Analysis Group
Centers for Medicare & Medicaid Services
Mail Stop S3-02-01
7500 Security Boulevard
Baltimore, MD 21244–1850

Re: CAG-00454N: Screening for Colorectal Cancer – Blood-Based Biomarker Tests

Dear Ms. Syrek-Jensen:

Thank you for the opportunity to comment on CAG’s Proposed Decision Memo for Screening for Colorectal Cancer – Blood-Based Biomarker Tests (CAG-00454N) (the Memo).

As noted in our previous comment on the National Coverage Analysis, we believe that blood-based biomarker tests offer the potential to significantly increase access to and participation in CRC screening, and to further reduce CRC-associated mortality. We support the Memo’s test-agnostic approach to providing coverage for blood-based biomarker tests that have received Premarket Approval (PMA) from FDA. However, we respectfully suggest that two other criteria enumerated in the Memo as prerequisites for coverage, namely the specified performance criteria and inclusion in one or more clinical practice guidelines, warrant further clarification or consideration. We also believe that draft language suggesting that coverage be limited to “once every 3 years, or at the interval designated in the Food and Drug Administration (FDA) label” should be modified. We have provided detailed comments with respect to these three topics below.

Perfromance Criteria:

The Memo proposes that blood-based biomarker screening test must have:

“proven test performance characteristics for a blood based screening test with both sensitivity greater than or equal to 74% and specificity greater than or equal to 90% in the detection of colorectal cancer compared to the recognized standard (accepted as colonoscopy at this time), based on the pivotal studies included in the FDA labeling:”

Based on the detailed analysis presented in the body of the Memo, it is clear that this sensitivity figure of 74% refers to sensitivity for the detection of colorectal cancer. However, we believe that the current wording related to specificity is less clear.

The detailed information within the Memo indicates that the 90% figure relates to a test’s specificity as measured in patients with a “negative colonoscopy” (i.e. study subjects without any neoplastic or non-neoplastic colorectal pathology detected during their colonoscopy) – see Table 4 and the two paragraphs immediately following Table 4, which indicate that the 90% figure relates to “negative colonoscopy.”

This is different than specificity for subjects without CRC. As detailed in the Cologuard Summary of Safety and Effectiveness, the specificity for subjects without CRC in the Cologuard FDA pivotal DEEP-C study was 84.4%. Conversely, the Cologuard specificity among patients with a “negative colonoscopy” (defined in the DEEP-C study as subjects with “No findings on colonoscopy, no histopathological review”) was 89.8%.

Accordingly, to clarify the intended performance criteria we suggest that the following summary language be substituted for the current wording:

“proven test performance characteristics for a blood based screening test with both sensitivity for colorectal cancer greater than or equal to 74% and specificity among patients with a negative colonoscopy greater than or equal to 90% compared to the recognized standard (accepted as colonoscopy at this time), based on the pivotal studies included in the FDA labeling;”

We believe that publication of pivotal study performance data in a peer-reviewed journal is an appropriate way to make the study results available to CMS and to the public.

Guidelines Inclusion:

The Memo proposes that a blood-based biomarker screening test must be included “as a recommended routine colorectal cancer screening test in at least one professional society guideline or consensus statement or United States Preventative Services Task Force (USPSTF) recommendation.” We believe that the proposed guideline or consensus statement inclusion requirement is unnecessary, without precedent for the coverage assessment of innovative CRC screening technologies, and potentially problematic. We therefore recommend that this specific requirement be dropped from the criteria required for coverage.

Two key questions lie at the heart of a National Coverage Determination for a new CRC screening technology:

1. Is the evidence sufficient to determine that the test is a suitable colorectal cancer screening test for prevention or early detection in Medicare beneficiaries?
2. Is the evidence sufficient to determine that colorectal cancer screening using the test is appropriate for Medicare beneficiaries?

These evidence-based questions have been carefully and thoroughly reviewed in the proposed Memo. A blood-based biomarker screening assay that meets the specified performance criteria will perform in a manner consistent with annual FIT testing. The Memo effectively reflects the large body of evidence that demonstrates the life-saving impact of FIT testing on the field of CRC screening. The Coverage and Analysis Group has developed sufficient criteria to support coverage without reliance on the assembly or review of evidence by third-party organizations that develop guidelines or consensus statements. Further, unlike other types of screening tests, the statutory language providing CMS with the authority to review new CRC screening test technologies via NCD does not require support in USPSTF or other professional society guidelines.

Importantly, prior NCDs developed for innovative CRC screening tests have not set as a precondition for coverage inclusion of the test in a guideline or consensus statement. See CAG-00180N, Screening Immunoassay Fecal-Occult Blood Test (FIT), and CAG-00440N, Screening for Colorectal Cancer – Stool DNA Test (Cologuard), for examples where new technologies received a positive coverage determination without a requirement of prior inclusion in a guideline or consensus statement. On the contrary, at the time of review and approval FIT had not been included in any guideline or consensus statement and the Cologuard assay under consideration was not explicitly included in guidelines (although a predecessor assay based on similar principles had been included in several guidelines). In both cases, inclusion within USPSTF guidelines only came about years after a positive NCD and following widespread use of the technologies by Medicare beneficiaries. CMS’s willingness to move forward with coverage in the absence of guidelines or consensus statement inclusion was key to the widespread utilization of FIT testing, which was subsequently acknowledged as leading to improved screening rates for the Medicare population and other Americans of screening age.

Additionally, this criterion is problematic because guidelines are updated relatively infrequently. For example, the USPSTF typically updates CRC screening guidelines only once every five to seven years. Other organizations have similar intervals in their review and update processes. Requiring prior inclusion in one or more of the guidelines will likely prevent access by Medicare beneficiaries to innovative, effective blood-based biomarker tests for several years following FDA premarket approval. We note that such a limit on access would conflict with the stated intent of the proposed Memo: to provide a clear and streamlined path to coverage for new tests that meet the proposed performance criteria.

Lastly, certain of the guidelines organizations are facing conflict of interest concerns that raise potential questions regarding equality of access to or inclusion in their CRC screening guidelines. While we would not venture to suggest that any of the guidelines organizations would allow these potential conflicts to interfere with a fair review of new blood-based screening tests, the fact that substantial potential conflicts exist should lead CMS to avoid a requirement that guidelines inclusion be a prerequisite for coverage under this NCD.

Test Interval:

We recommend that the Proposed Decision language regarding test interval should be changed from the current “once every 3 years, or at the interval designated in the Food and Drug Administration (FDA) label if the FDA indicates a specific test interval,” to “at the interval designated in the Food and Drug Administration (FDA) label, if the FDA indicates a specific test interval, or such interval as CMS determines appropriate,”. We suspect the inclusion of a default three year coverage interval is anchored in the prior coverage decision for Cologuard, and so may not be appropriate for future blood-based biomarker screening tests – depending upon the specific performance characteristics and payment rates of the tests in question.

The three year screening interval established for Cologuard was based upon the manufacturer’s recommendation and reflected the acknowledged tradeoff posed by Cologuard between very high sensitivity (92% for CRC) and relatively high cost test (~$500). As noted in CAG-00440N:

“The post approval study required by the FDA [for Cologuard®] is designed to evaluate the validity of screening every three years to ensure that clinically important findings are not missed. With the natural history of most colorectal neoplasia and improved test performance, screening every three years appears to be a reasonable initial frequency [emphasis added] to detect clinically meaningful changes. CMS will re-evaluate the screening interval after the completion of the post approval study and modify coverage if appropriate.”

Cost effectiveness modeling of Cologuard would not have supported a one year interval (See Peterse EFP et al., Comparing the cost-effectiveness of innovative colorectal cancer screening tests, JNCI 2020).

A blood-based biomarker screening test, on the other hand, may have performance characteristics similar to FIT, as suggested in the proposed Memo, i.e. sensitivity for CRC at or above 74% and specificity among patients with a negative colonoscopy of 90%. Such a test, assuming it can be delivered at a reasonable cost (i.e. $200 or less), would warrant coverage at an interval consistent with FIT testing, i.e. on an annual basis. Alternatively, a different blood-based biomarker test may offer sensitivity for CRC at a level close to Cologuard®, i.e. 90%+, with acceptable specificity, but at a significantly higher cost that would support testing only on an interval similar to Cologuard®, i.e. every three years.

While we strongly support the general concept of a test-agnostic NCD based upon performance levels demonstrated in a pivotal FDA premarket approval study, we believe that CMS should evaluate coverage frequency on a test by test basis. This can be done relatively easily via assessment of the test’s performance in a validated CRC screening model. Examples of these type of models include the three models that are evaluated as part of The Cancer Intervention and Surveillance Modeling Network (CISNET) consortium – CRC-SPIN, SimCRC, and MISCAN – as well as models developed and published in peer-reviewed journals by academic research groups (see, for example, Ladabaum U et al, Contrasting Effectiveness and Cost-Effectiveness of Colorectal Cancer Screening Under Commercial Insurance vs. Medicare, Am J Gastro 2018, and D’Andrea E et al., Quantifying the impact of adherence to screening strategies on colorectal cancer incidence and mortality, Cancer Med 2019).

These models provide clear and helpful data regarding the appropriate interval for test utilization. For example, in the current draft update to the USPSTF CRC Guidelines released for public comment on October 27th, the Task Force notes that while sDNA-FIT testing (Cologuard) can be used at either a one year or three year interval, “[m]odeling suggests that screening every 3 years does not provide a favorable (i.e., efficient) balance of benefits and harms, compared with other stool-based options (i.e., annual FIT or annual sDNA-FIT).” We further note that 42 CFR 410.37 (a)(1)(v) directs CMS to assess the cost-effectiveness of any new CRC screening option that is considered for coverage (“. . . with such frequency and payment limits as CMS determines appropriate, . . . “), and this exercise involves evaluation of the appropriate interval and payment for a given test based on that test’s performance and cost characteristics.

Summary:

In summary, we respectfully recommend revision of the proposed Memo as outlined above to: 1) clarify that test specificity is to be measured in patients with a negative colonoscopy, 2) delete the requirement that the test be included in one or more guidelines or consensus statements, and 3) provide CMS the flexibility to determine the appropriate coverage interval for a test, based on test performance and payment parameters.

We appreciate the opportunity to provide comments on this Proposed Decision Memo. If you have any questions or would like to discuss this issue further, please contact me at (908) 300-8673.

Sincerely,

Robert C. Bruce
Vice President, Reimbursement

Less

Thank you for the opportunity to comment on CAG’s Proposed Decision Memo for Screening for Colorectal Cancer – Blood-Based Biomarker Tests (CAG-00454N).

Quest Diagnostics remains committed to ensuring patients in the U.S. have broad access to innovative colorectal cancer diagnostics with existing test options like Fecal Immunochemical Testing, and in collaboration with Clinical Genomics to develop a blood-based biomarker screening assay.

We continue to believe that blood-based biomarker tests offered as part of an annual wellness program or an annual visit to a primary care physician have the potential to significantly increase access to and participation in CRC screening, and to further reduce CRC-associated mortality. As stated in our previous comment we agree with the test-agnostic approach to providing coverage for blood-based test with Premarket Approval from FDA. We do ask for further clarification and respectfully recommend revision of the proposed Memo to clarify test specificity is to be measured in patients with a negative colonoscopy, removal of the requirement that the test be included in one or more guidelines or consensus statements, and provide CMS the flexibility to determine the appropriate coverage interval for a test, based on test performance and payment parameters.

Thank you again for the opportunity to comment. Please feel free to reach out directly at Yuri.A.Fesko@QuestDiagnostics or call 919.812.7803 if you would like to discuss.

Less

Durado D Brooks, MD, MPH
Deputy Chief Medical Officer
Exact Sciences
Date:11/13/20

Comment: RE: Proposed National Coverage Determination for Screening for Colorectal Cancer-Blood-Based Biomarker Tests (CAG-00454N)

Summary:

Exact Sciences Corporation appreciates the opportunity to comment on the proposed national coverage determination (“NCD”) for blood-based colorectal cancer (CRC) screening tests. As a company committed to increasing CRC screening rates and developing our own blood-based CRC screening test, we support the proposed NCD and recommend that CMS establish more specific coverage criteria that balance the benefit of blood-based screening against the potential harm of using a less-accurate test.

Elaborating on the CMS recommendations, we propose the following additional criteria for evaluating Medicare coverage of a blood-based CRC screening assay:

1) CMS should set minimum performance requirements that demonstrate head-to-head non-inferiority to FIT for early-stage CRC and advanced adenoma detection
2) CMS should use recommended tests as benchmarks to inform the covered screening interval, with guidelines as an alternative coverage pathway
3) CMS should require FDA-approval of the assay for front-line, average-risk screening
4) CMS should cover CRC screening beginning at age 45 years or as recommended by major guidelines.
5) CMS should clarify that follow-up colonoscopy to an initial non-invasive test is part of the screening process and a covered preventive service

Detailed Comments:

1) CMS should set minimum performance requirements that demonstrate head-to-head non-inferiority to FIT for early-stage CRC and advanced adenoma detection

We agree with CMS that FIT is the appropriate minimum performance comparator for a blood-based CRC screening test. However, FIT should be the benchmark not only for all-stage CRC sensitivity, but also for early-stage CRC and advanced adenoma (AA) sensitivity, which is the primary benefit of screening.

Given the evidence supporting the clinical utility of FIT (1), any test with equivalent performance metrics would be expected to provide similar clinical benefits. Non-inferiority to FIT would also ensure that any patients switching from a guideline-supported strategy to a new blood-based screening strategy would not be subject to additional harms by way of missed CRC or advanced adenoma.

The effectiveness and clinical utility of any new non-invasive screening test, including blood-based, will be dependent on its ability to detect early stage cancers and precancerous lesions (2,3). Advanced adenoma detection significantly influences the predicted benefits derived from front-line CRC screening modalities (4). Therefore, a blood-based test should demonstrate the ability to detect early stage CRC and AA, at sensitivities similarly non-inferior to FIT. However, the reported early-stage and advanced adenoma detection rates of several blood-based CRC screening tests in development are well below those of established CRC screening tests like FIT (2).

Evidence of non-inferiority should be established with a prospectively designed study that compares CRC blood test performance head-to-head with FIT performance, using colonoscopy as a reference standard, in the same patient population. Such a study should be performed head-to-head due to the variability of FIT performance from study to study internationally (5) and in the US (6,7).

Therefore, CMS should include 3 additional minimum sensitivity requirements for coverage of a blood-based CRC screening test:

• Head-to-head performance comparison to FIT (due to FIT variability)
• Advanced adenoma sensitivity non-inferior to FIT
• Early stage cancer sensitivity non-inferior to FIT

2) CMS should use recommended tests as benchmarks to inform the covered screening interval, with guidelines as an alternative coverage pathway

Established screening intervals of guideline-recommended tests should serve as the benchmarks for interval coverage determinations. A CRC blood test with sensitivity non-inferior to FIT would therefore be anticipated to have a 1 year screening interval, one with sensitivity non-inferior to mt-sDNA should have a 3 year interval, and one with sensitivity non-inferior to colonoscopy should have a 10 year interval (8).

Although sensitivity is an important safety requirement, specificity is also critical, especially when considering interval. An annual blood-based screening test with low specificity, and hence a high false-positive rate, would ultimately send the majority of patients to colonoscopy after only a few rounds of testing, as modeled in the Peterse et al. study (9). Adoption or approval of a CRC screening test that is predicted to derive its benefit from high rates of colonoscopy referral could derail efforts to develop assays that achieve a high degree of discrimination (10). Such a modeling strategy is unlikely to work in the real-world and is little differentiated from a colonoscopy screening strategy.

To control for this, we suggest tying both specificity and sensitivity to interval. The appropriate interval would be defined by head-to-head non-inferiority with an existing recommended test for early stage cancer and advanced adenoma sensitivity and specificity as follows:

• 1 year interval: Sensitivity and specificity non-inferior to FIT
• 3 year interval: Sensitivity and specificity non-inferior to mt-sDNA
• 5 year interval: Sensitivity and specificity non-inferior to flexible sigmoidoscopy
• 10 year interval: Sensitivity equivalent to colonoscopy

Comparison with existing screening tests may not work in all cases (e.g. a 2-year interval). Therefore, major national guidelines are an appropriate alternative pathway to coverage and interval (although the minimum test sensitivity should still be met). Major guidelines are more appropriate for determining interval because the FDA does not have an interval recommendation for colonoscopy, FIT or mt-sDNA. In the absence of detailed sensitivity/specificity/interval criteria in the NCD, a major guideline recommendation should be a requirement for coverage.

3) CMS should require FDA-approval of the assay for front-line, average-risk screening

We agree with CMS that any blood-based CRC screening strategy should be deemed clinically appropriate by the FDA as a front-line option for population-level screening of all average-risk individuals. Tests that are limited by their label to a subset of the screening population (e.g. patients that have declined all other guideline-recommended screening tests) would be difficult to enforce in clinical practice, creating substantial risk for inappropriate use and attendant harms. It would also be administratively burdensome and costly to collect reliable documentation that a patient has declined all other guideline-recommended screening options prior to opting for a blood-based screening test. These challenges could ultimately hamper screening uptake.

4) CMS should cover CRC screening beginning at age 45 years, as recommended by major guidelines.

Given the growing incidence and burden of early-onset CRC, any test that fulfills CMS criteria as an appropriate front-line assay for average risk CRC screening should be initiated at 45 years of age, aligning with the recent ACS guidelines (11) and draft USPSTF recommendation statement (12).

5) CMS should clarify that follow-up colonoscopy to an initial non-invasive test is part of the screening process and a covered preventive service

Patient failure to complete the screening process poses serious public health risks. According to one study, patients who failed to follow-up on an abnormal CRC screening result had a seven-fold higher risk of dying of CRC than those who completed the follow-up process (13). And yet, approximately 3 in 10 patients with an abnormal FIT result have no documented follow-up colonoscopy within the recommended timeframe (14,15). As it stands, follow-up colonoscopies can carry significant coinsurance costs for many Medicare beneficiaries, depending on whether they are considered diagnostic. This cost obligation can discourage completion of the CRC screening process.

There is growing appreciation in the public health community about the need to address this problem. According to the American Cancer Society’s 2018 Colorectal Cancer Screening Guidelines (11), “[t]he follow-up colonoscopy should not be considered a ‘diagnostic’ colonoscopy but, rather, an integral part of the screening process, which is not complete until the colonoscopy is performed”.

A novel blood-based screening tests that requires a colonoscopy to confirm positive results could be subject to similar failures to follow-up. We therefore urge CMS to use the opportunity presented by this NCD to clarify that coverage for CRC screening using a non-invasive strategy includes both the initial screening test and any necessary follow-up colonoscopy to assess a positive result.

In closing, Exact Sciences is committed to reducing CRC burden, and believes a blood-based CRC screening strategy that meets the performance and clinical use characteristics outlined above could complement the established portfolio of effective tests in facilitating this goal. As a reflection of this commitment, Exact Sciences continues to make improvements to Cologuard (16) and advance development of a blood-based CRC screening test. Progress is being made in reducing CRC mortality (17) and increasing CRC screening rates (18). Cologuard alone has screened over 4.5 million people since 2014.

Because of the biology of colorectal cancer, it will be challenging for blood-based tests to detect early stage disease at similar rates as stool-based tests. However, choice is important and offering a blood-based test has the potential to increase screening rates, although real-world adherence to blood draw is also imperfect (19). In weighing the strength and balance of the evidence, CMS should carefully review the trade-offs between test-performance characteristics, benefits, and harms of a blood-based CRC screening test that is less accurate than stool-based screening tests before reaching a final coverage determination.

Exact Sciences thanks CMS for the opportunity to provide public comment and looks forward to working closely with the agency in the future.

Sincerely,

Durado D Brooks, MD, MPH
Deputy Chief Medical Officer

Exact Sciences References:
1. Robertson DJ, Lee JK, Boland CR, et al. Recommendations on fecal immunochemical testing to screen for colorectal neoplasia: a consensus statement by the US Multi-Society Task Force on colorectal cancer. Gastrointestinal Endoscopy. 2017;85(1):2-21.e3.
doi:10.1016/j.gie.2016.09.025 2. Ebner DW, Kisiel JB. Stool-Based Tests for Colorectal Cancer Screening: Performance Benchmarks Lead to High Expected Efficacy. Current Gastroenterology Reports. 2020;22(7). doi:10.1007/s11894-020-00770-6
3. Haug U, Knudsen AB, Lansdorp-Vogelaar I, Kuntz KM. Development of new non-invasive tests for colorectal cancer screening: The relevance of information on adenoma detection. International Journal of Cancer. 2015;136(12):2864-2874. doi:10.1002/ijc.29343
4. Meester RGS, Doubeni CA, Lansdorp-Vogelaar I, et al. Variation in adenoma detection rate and the lifetime benefits and cost of colorectal cancer screening: A microsimulation model. JAMA - Journal of the American Medical Association. 2015;313(23):2349-2358. doi:10.1001/jama.2015.6251
5. Imperiale TF, Gruber RN, Stump TE, Emmett TW, Monahan PO. Performance characteristics of fecal immunochemical tests for colorectal cancer and advanced adenomatous polyps: A systematic review and meta-analysis. Annals of Internal Medicine. 2019;170(5):319-329. doi:10.7326/M18-2390
6. Shapiro JA, Bobo JK, Church TR, et al. A Comparison of Fecal Immunochemical and High-Sensitivity Guaiac Tests for Colorectal Cancer Screening. American Journal of Gastroenterology. 2017;112(11):1728-1735. doi:10.1038/ajg.2017.285
7. Imperiale TF, Ransohoff DF, Itzkowitz SH, et al. Multitarget Stool DNA Testing for Colorectal-Cancer Screening. New England Journal of Medicine. 2014;370(14):1287-1297. doi:10.1056/nejmoa1311194
8. Bibbins-Domingo K, Grossman DC, Curry SJ, et al. Screening for colorectal cancer: US preventive services task force recommendation statement. JAMA - Journal of the American Medical Association. 2016;315(23):2564-2575. doi:10.1001/jama.2016.5989
9. Peterse EFP, Meester RGS, de Jonge L, et al. Comparing the cost-effectiveness of innovative colorectal cancer screening tests. JNCI: Journal of the National Cancer Institute. Published online August 6, 2020. doi:10.1093/jnci/djaa103
10. Ransohoff DF. Evaluating a new cancer screening blood test: Unintended consequences and the need for clarity in policy-making. JNCI: Journal of the National Cancer Institute. Published online August 6, 2020. doi:10.1093/jnci/djaa104
11. Wolf AMD, Fontham ETH, Church TR, et al. Colorectal cancer screening for average-risk adults: 2018 guideline update from the American Cancer Society. CA: A Cancer Journal for Clinicians. 2018;68(4):250-281. doi:10.3322/caac.21457
12. U.S. Preventive Services Task Force. Draft Recommendation Statement- Colorectal Cancer: Screening. https://www.uspreventiveservicestaskforce.org/uspstf/draft-recommendation/colorectal-cancer-screening3.
13. Doubeni CA, Fedewa SA, Levin TR, et al. Modifiable Failures in the Colorectal Cancer Screening Process and Their Association With Risk of Death. Gastroenterology. 2019;156(1):63-74.e6. doi:10.1053/j.gastro.2018.09.040
14. Chubak J, Garcia MP, Burnett-Hartman AN, et al. Time to colonoscopy after positive fecal blood test in four U.S. health care systems. Cancer Epidemiology Biomarkers and Prevention. 2016;25(2):344-350. doi:10.1158/1055-9965.EPI-15-0470
15. McCarthy AM, Kim JJ, Beaber EF, et al. Follow-Up of Abnormal Breast and Colorectal Cancer Screening by Race/Ethnicity. American Journal of Preventive Medicine. 2016;51(4):507-512. doi:10.1016/j.amepre.2016.03.017
16. Domanico M, Kisiel J, Gagrat Z, et al. ACG 2019 Annual Scientific Meeting2019 | eventScribe Itinerary Planner. Accessed November 9, 2020. https://www.eventscribe.com/2019/ACG-Abstracts/Presinfo/PresInfo.asp?pres=594316
17. National Cancer Institute, Surveillance E and ERP. Cancer Stat Facts: Colorectal Cancer. Published 2020. Accessed November 9, 2020. https://seer.cancer.gov/statfacts/html/colorect.html
18. Centers for Disease Control and Prevention. Use of Colorectal Cancer Screening Tests (2018 Behavioral Risk Factor Surveillance System) | CDC. Published October 22, 2019. Accessed November 9, 2020. https://www.cdc.gov/cancer/colorectal/statistics/use-screening-tests-BRFSS.htm
19. Moffet HH, Parker MM, Sarkar U, et al. Adherence to laboratory test requests by patients with diabetes: The diabetes study of Northern California (DISTANCE). American Journal of Managed Care. 2011;17(5):339-344. Accessed November 9, 2020. /pmc/articles/PMC3189790/?report=abstract

Less

To Whom it may concern,

A Key FDA indication for Epi proColon that you first mention on page 50 of the decision summary, is that it is indicated for patients who have been offered and declined available CRC screening tests that are recommended by appropriate clinical guidelines, such as colonoscopy and FIT. On page 7 of the summary you state that in 2018 79.2% of individuals age 65-75 yrs of age were up to date with colorectal cancer screening. This strongly implies that no more than 21% of this important age group would be eligible to be offered Epi proColon. I would like to mention two concerns I have with this assumption.

First, the source of your statistics is a March 2018 CDC Vital Signs publication. A venerable source, but they clearly state that their data was collected subjectively. When they say that 79.2% of individuals were up to date with CRC screening, they got that information from phone surveys. For a sensitive issue such as CRC, what percentage of that 79.2% was less than truthful? A lot, I think. The second issue I have with this data is that it was collected pre COVID-19 pandemic. So many of our senior citizens have significantly limited their activities this year, and will continue to do so until the pandemic ends. I hope those reading this can have some empathy for people pushing back on going into an endoscopy suite, the travel there, encountering all of those heath care workers, and then being under anesthesia during a pandemic, for a procedure they may deem optional. Thus, I present the logical theory that, as of today and for the near term, the percentage of individuals age 65-75 yrs of age up to date with crc screening, is significantly below any previous estimates, especially the 79.2% stated in this summary.

Given these troubling times in our country’s history, and with the crc unscreened population larger than ever before, I urge CMS to agree to pay for Epi proColon for our Medicare population. If not permanently, at least until the end of the COVID 19 pandemic.

Thank you for considering my comment.

Less

As a seasoned professional in IVD diagnostics, I was both surprised and perplexed by the CMS response for national coverage for this FDA-approved blood test. The FDA is the branch that reviews and determines safety and efficacy for new IVDs based on submitted evidence for the intended use populations, not CMS.

Not new news - the problem with colorectal cancer screening right now, today – is that it simply doesn’t and hasn’t happened to a third of the eligible population – the staunch refusers. Despite impassioned appeals, speeches, parades and guidelines, neither colonoscopy nor stool tests provide the needed impetus to participate in screening. As a result, the public health impact is large and widely documented. As referenced throughout these public comments, the statistics assign significant morbidity (43% new cases; Doubeni et al, 2019), mortality (76% deaths; Doubeni et al, 2019) and CRC-associated healthcare costs to the unscreened.

Along comes the first FDA-approved (2016) CRC screening blood test developed for the unscreened, Epi proColon. The test demonstrates increased adherence (Liles et al, 2017) in this very specific non-adherent population. As a patient-accepted method of testing, a blood test method can go where no imaging nor stool screening test has gone before - to a blood collection center. It’s not only intuitive that a blood test has the ability to increase population screening and thereby population health for CRC, but blood testing IS the method of choice that is sparking new IVD “liquid biopsy” research and development – the sample is easy to give and easy to get.

A recent National Cancer Institute (NCI) microsimulation study (Peterse, 2020) describes this blood test as suitable for increasing screening participation and attracting the non-adherent to screening. The author concluded, “for people who are unwilling to be screened with FIT or colonoscopy, annual screening with the mSEPT9 is the test of choice given its cost-effectiveness profile compared to CTC, PillCam and mtSDNA. The number of CRC cases and deaths averted, and the number QALYG from annual mSEPT9 screening are even higher than from annual FIT screening.” Another cited study found a 25% uptake of the blood test in those who refused FIT testing (Symonds E, 2017)

Have you considered what the sensitivity of a test or its clinical effectiveness is for finding cancer if no one participates? ZERO% Adherence is the key to the clinical and population effectiveness for any test.

When thinking about the impact on Public Health, the recent NCI microsimulation study (Petarse et al, 2020) found that use of the Epi proColon blood test results in 80% less CRC deaths and 60% less CRC cases compared to no screening. That’s a big impact.

Epi proColon is the first blood test for CRC screening but there will be more. It’s the first testing method that has the ability to grab and retain the attention of those who refuse to do a colonoscopy or a stool test. The best test still is the test that gets done.

Please don’t deny an unscreened person an opportunity to be screened for CRC. We know many are unable to pay for healthcare and coverage is important. I urge you to provide a national coverage decision for Epi proColon.

Less

Response to comments by Dr. Uri Ladabaum and the Early Detection Research Network (EDRN) Investigators:

We appreciate the analysis and comments provided by the members of the EDRN group.
We concur with many of the points outlined in their comments and provide the following for your additional consideration:

• The comments recognize that there is a significant population who are not reached by current endoscopic and stool-based methods. Given FDA’s indication for use of Epi proColon in this population, this clearly identifies a population that would benefit from Epi proColon (those with a history of not completing CRC screening), addressing the CMS language suggesting that there is no identifiable beneficiary population.
• In their comments regarding performance criteria, the EDRN commenters mention that “...a test with sensitivity and specificity reflecting those of Epi proColon® (sensitivity for colorectal cancer 68%, specificity 79%, positivity rate in persons with adenoma = 1-specificity) can achieve a programmatic benefit that is comparable to that of FIT” and we agree with this position.
• However, they also mention that “If CMS determines that the sensitivity and specificity of Epi proColon® are not acceptable, then we urge CMS to consider covering blood tests whose performance characteristics fall between those of Epi proColon® and the available stool tests, because of the benefit that such blood tests can deliver in persons who will not take up stool testing….such as approximately 70% sensitivity and 85% specificity.”
• As an alternative to identifying specific arbitrary sensitivity and specificity point estimates, that override FDA approval, we recommend utilizing CISNET (a consortium also funded by the NCI) microsimulation models to measures clinical outcomes (harms and benefits and optimal interval).
• This approach aligns with the commentors statement “Setting minimum parameters, especially in the early days of a nascent field, could lead to incoherent and volatile approvals across time and populations, and could stifle innovation” by utilizing validated peer-reviewed methods to measures effectiveness (clinical utility) against the desired goal of screening i.e. reduction of incidence and mortality of disease.
• The commenters suggest that if there is a specificity requirement it could be reduced to 85%. Using the MISCAN analysis and estimating the impact on colonoscopy rates, suggests that this incremental 5% in specificity would result in a miniscule difference in the number of colonoscopies required over a lifetime of screening compared with Epi proColon, and an insignificant difference in the number of actual adverse events, because of the low adverse event rate of colonoscopy. The average number of colonoscopies over an average individual’s lifetime of CRC screening per the CISNET decision modeling data is 3.8 for Epi proColon, 3.3 for a 90% specific test run annually, and approximately 3.6 – 3.7 for an 85% specific test. The average number of colonoscopies using the gold-standard of a colonoscopy every ten years is 4.7 as a reference based on the same modeling data published in August 2020 in the Journal of the National Cancer Institute.
• The FDA approach to establishment of acceptable performance levels and benefit:harms ratios for a test are rigorous, such that a test receiving approval with any CRC screening indication should be considered for coverage. This aligns with the broader effort of providing CMS coverage for FDA approved breakthrough devices.
• Absent clinical trial data, clinical utility should be assessed based on longstanding and extensively validated CISNET decision modeling using the FDA established performance characteristics. Given decision modeling is a key component of assessing clinical utility, any FDA-approved test that scores well in the model and meets the other criteria should be covered.
• This combination will provide CMS with an accurate assessment of benefit and harms for the covered population.
• We commend the recommendation that CMS avoid setting thresholds that are too stringent, thereby depriving Medicare beneficiaries of this life-saving screening modality. As referenced by the commentors “Epi proColon can achieve a programmatic benefit that is comparable to that of FIT” and thus we believe Epi proColon should be covered in the final decision rather than excluding it for barely missing arbitrary cutoffs that have been demonstrated to have a very small impact on the overall harms and benefits of the test when compared to other currently covered screening methods.
• Epi proColon is an FDA approved product that is available today. It is disheartening to us to imagine that this test wouldn’t be covered based upon arbitrary point estimate standards in hopes that future tests may meet. How many people will unnecessarily die between now and some unknown event in the future? If covered, Epi proColon will save lives imminently while other tests are a long away from FDA approval and today can only boast about aspirational goals.

Jorge Garces, PhD
CSO & President,
Epigenomics AG.

Theo deVos, PhD
VP Clinical & Scientific Affairs
Epigenomics Inc.

Less

Dear CMS,

As president of the Colon Cancer Coalition, I urge you to reconsider your recent decision not to cover Epi proColon for Medicare patients who refuse other traditional colorectal cancer screening methods. Our organizational focus is to increase screening for colorectal cancer and save lives through early detection and treatment of this devastating disease. We provide resources and education to increase screening, as well as education about the signs and symptoms of this very treatable disease when it is caught in early stages. Our hope is for all people will understand their personal risk factors and get the right screening at the right time. As a patient advocacy organization, we believe it is important that ALL possible screening options are available to patients.

Unfortunately, 1 in 3 people refuse to complete a colonoscopy or take-home stool-based test. When screening is delayed and unscreened individuals finally acknowledge colorectal symptoms they are experiencing, the cancer has progressed to a later stage resulting in worse outcomes. If we can provide a screening tool that patients who refuse colonoscopies or stool-based tests will complete, we can then increase the colorectal cancer screening rates and make a notable impact to the lives of colorectal cancer patients.

The Epi proColon blood test is intended for patients who refuse to be screened by traditional methods. A recent publication by the National Cancer Institute concluded that Epi proColon is the ‘Test of Choice’ for patients who refuse colonoscopy and FIT. In the same publication, Epi proColon performed better than other covered tests such as Cologuard and FIT – and much better than No Screening (which is the intended use for Epi proColon). Epi proColon helps solve a healthcare challenge and performs better than currently covered tests.

CMS has an initiative to cover FDA-approved ‘breakthrough’ technologies. We believe a blood test like Epi proColon that targets unresolved unscreened challenges, that has proven benefits over currently covered products, and is something hard-to-reach patients are willing to complete, should be considered a ‘breakthrough.’

Lastly, the CMS proposal mentions ‘inclusion’ in guidelines (USPSTF, ACS, MSTF). As comparison, Cologuard was not included in ANY guidelines (only FDA approval) when CMS decided a positive coverage determination. Waiting for all guidelines will be years and, in the meantime, unscreened patients will exhibit with cancer and many will die needlessly. NOTE: NCCN has included Epi proColon – not for routine testing but for patients who refuse testing which is aligned with the FDA intended use.

We believe Medicare should cover Epi proColon, and urge the CMS to reconsider their position.

Sincerely,
Chris Evans
President & CFO
Colon Cancer Coalition

Less

November 12, 2020

Re: National Coverage Analysis (NCA) for Screening for Colorectal Cancer - Blood-Based Biomarker Tests (CAG-00454N)

Fight Colorectal Cancer (Fight CRC) is the leading patient-empowerment colorectal cancer advocacy organization based in the United States. We fight to cure colorectal cancer and serve as relentless champions of hope for all those affected by this disease through informed patient support, impactful policy change, and breakthrough research endeavors.

We are responding to the National Coverage Decision (NCD) regarding Screening for Colorectal Cancer - Blood-Based Biomarker Tests. Thank you for the opportunity to provide comment. We appreciate your work to develop general standards and guidelines for blood-based screening for future considerations. Fight CRC believes that additional effective screening modalities are critical for the colorectal cancer community. However, we respectfully request the following from CMS:

1) Additional detail regarding how these specific criteria were established.

2) Explanation of the inclusion and exclusion criteria for the data utilized, as it appears not all current data sources were included in this analysis. Additional detail would be helpful for this and future NCD opportunities.

In the wake of COVID-19, now more than ever, we are committed to advocating for the advancement of science, and increased and improved screening modalities. Further understanding of the CMS process and decision path is greatly appreciated.

Fight CRC believes in fully disclosing all potential conflicts of interest. As an organization, we have received sponsorships and/or educational grants from companies who have an interest in novel screening methods for colorectal cancer, including: Exact Sciences, Quest Diagnostics, Boston Scientific, Clinical Genomics, Epigenomics, Fuji Endoscopy and Medtronic. Neither these companies nor any of our corporate supporters have influenced our comments on this issue.

We appreciate your consideration of our comments as you further consider the NCD for methylated SEPT9 DNA.

Sincerely,
Anjee Davis, President
Fight Colorectal Cancer

Less

November 12, 2020

Re: Public Comment for National Coverage Analysis (NCA) Tracking Sheet for Screening for Colorectal Cancer - Blood-Based Biomarker Tests (CAG-00454N)

Dear Sir/Madame,

I am a Professor Medicine at University of California San Diego, and conduct research on colorectal cancer screening and prevention, including on strategies to optimize participation in screening.

My primary recommendation on reviewing the coverage analysis is to urge CMS to cover all FDA approved screening tests based on their approved FDA label. For the EpiProcolon test, this would allow for Medicare beneficiaries to be offered the test if they have refused other screening tests such as colonoscopy or a fecal immunochemical test. Having highly acceptable options available to patients who are unwilling to do currently available tests could optimize participation in screening, and ultimately, early detection of colorectal cancer.

I have a few additional suggestions for improving the decision memo:

1. The third paragraph of page 42 (beginning with “The clinical implications of low specificity”) appears to describe the clinical implications of low sensitivity. Please review carefully to make sure the correct parameter is being discussed.
2. On page 48, a potential specificity benchmark for specificity of a blood based biomarker is recommended. I recommend clarifying the definition of specificity in the context of a biomarker for colorectal cancer screening. Specificity could be defined by: a) the proportion of people with a normal colonoscopy, defined by absence of any adenoma or colorectal cancer, who have a normal test result; b) the proportion of people with a normal colonoscopy, defined by absence of any advanced adenoma or colorectal cancer, who have a normal test result; or c) the proportion of people with a normal colonoscopy, defined by absence of a colorectal cancer, who have a normal test result. I favor choosing a or b, and against choosing option c, as a blood based biomarker that picks up adenomas should not be “penalized” for picking up these lesions.

Sincerely,

Samir Gupta, MD, MSCS, AGAF
Professor, Division of Gastroenterology, Department of Internal Medicine
University of California San Diego
Email: s1gupta@health.ucsd.edu

Less

Dear Medicare:

Epi proColon is a win-win for Medicare and its recipients. Let’s face it, many people are apprehensive about the procedures to screen for colorectal cancer, but what if colorectal cancer screening was part of routine blood work. You would definitely have a higher rate of screening and a lower rate of late diagnosis and death. The average age of colon cancer diagnosis is 68 for men and 72 for women, an age group that is also at a higher risk of treatment complications. Screening for colorectal cancer as part of routine blood work would save lives and would also decrease Medicare care costs of screening procedures and their complications, in addition to the cost associated with a late diagnosis. Medicare needs to pay for Epi proColon to save lives.

Regards,

Charles L Wimbley

Less

Executive Summary

As investigators devoted to the early detection and prevention of colorectal cancer, we submit this comment on behalf of the Early Detection Research Network (EDRN) funded by the National Cancer Institute (NCI) to urge that CMS reconsider the proposed thresholds for colorectal cancer blood-based biomarkers of “sensitivity greater than or equal to 74% and specificity greater than or equal to 90% in the detection of colorectal cancer compared to the recognized standard” (CAG-00454N, Proposed Decision Memo for Screening for Colorectal Cancer - Blood-Based Biomarker Tests). This brief comment presents a high level view of our concerns. We are ready to draft a more detailed document if CMS would find it useful, and we expect to expand on this comment in a submission to an academic journal.

We submit to CMS that (see “Comment in Detail” below):

Setting minimum parameters, especially in the early days of a nascent field, could lead to incoherent and volatile approvals across time and populations, and could stifle innovation.

Retaining the traditional pathway (research and development, clinical studies, FDA approval, and inclusion in guidelines culminating in a CMS coverage determination) should be considered.

If CMS decides to set minimum thresholds despite the multiple concerns raised by us and others, then we urge it to:

1. appreciate the potential public health impact of a blood test that brings unscreened persons into screening even if its test performance characteristics do not match those of current stool-based tests,
2. define specificity with respect to detection of colorectal cancer and pre-invasive precursors, most importantly advanced adenomas,
3. account for the economic consequences to CMS, and the balance of benefits vs. risks to patients,
4. and consider thresholds that are more liberal than the current performance of stool-based tests (e.g. sensitivity for colorectal cancer of approximately 70% and specificity for colorectal cancer and pre-invasive precursors, most importantly advanced adenomas, of approximately 85%).

Comment in Detail

Public Health Impact

Despite randomized, controlled trials demonstrating that screening reduces colorectal cancer incidence and mortality, 28% of 65-75 year-olds (i.e. Medicare beneficiaries of screening-eligible age) and 46% of 50-64 year-olds in the U.S. are not up to date with colorectal cancer screening recommendations,¹ and participation rates are worse in disadvantaged populations.1 Blood-based colorectal cancer screening tests have the potential to improve participation with screening, and thereby yield substantial public health benefit.^{2, 3}

In approving Epi proColon®, the FDA recognized that a blood-based screening test with performance characteristics that do not meet those of stool tests could be beneficial if used by persons who otherwise would not participate in colorectal cancer screening (Epi proColon® is FDA-approved “to screen adults of either sex, 50 years or older, defined as average risk for CRC, who have been offered and have a history of not completing CRC screening”).

While it remains unknown what fraction of the Medicare population would accept a blood-based screening test but not FIT, the FIT + stool DNA (FIT-DNA, or Cologuard®) test, or colonoscopy, and then accept colonoscopy if the blood-based screening test were positive, screening in such persons, with proper shared decision making and consent, is preferable to no screening.

Specificity for Colorectal Cancer and Adenomas vs. Only Colorectal Cancer

In setting a threshold “specificity for colorectal cancer,” the CMS proposal implies that a blood test should deliver a negative result in persons with adenomas. The FIT-DNA test on which CMS based its proposed specificity threshold of =90% achieved a specificity of 89.8% only with the most expansive definition for specificity, which counted all detected precancerous polyps as true positives.⁴ The test’s specificity for colorectal cancer alone was not directly reported in the pivotal trial for FIT-DNA,⁴ but based on the data in the paper, we calculate it as 84.4%, which does not meet the proposed CMS threshold.

We urge CMS to consider the specificity of blood tests for colorectal cancer and pre-invasive precursors, most importantly advanced adenomas, given that adenoma detection and removal can prevent colorectal cancer. Focusing on specificity for colorectal cancer would inadvertently penalize a test that detects more advanced adenomas than a comparator that is positive when there is an advanced adenoma only at the same rate that it is positive when the colon is normal. For instance, in a hypothetical 10,000-person cohort in which 50 have colorectal cancer, 750 have advanced adenoma, 2,750 have non-advanced adenoma, and 6,450 have a normal colon, a hypothetical test A with positivity rate of 10% in normal colon, 10% in non-advanced adenoma, and 10% in advanced adenoma would have 90% specificity for colorectal cancer; in contrast, a hypothetical test B with positivity rate of 10% in normal colon, 10% in non-advanced adenoma, and 30% in advanced adenoma would have 88% specificity for colorectal cancer.

Economic Consequences to CMS

Although CMS might be concerned about expenditures incurred as blood-based screening tests enter the market, we urge CMS to consider the potential impact on total expenditures related to colorectal cancer treatment as well as screening, and the balance of benefit vs. risk to patients. It has been estimated that initiating stool-based or colonoscopy-based colorectal cancer screening in previously unscreened Medicare enrollees, as well as continuing low-cost screening in Medicare enrollees who were screened before entering Medicare, are both cost-saving due to averted colorectal cancer treatment costs.⁵

Revised specific thresholds, if CMS decides that setting thresholds now is desirable

The CMS proposed thresholds for blood-based biomarkers are based on comparisons with currently available and approved non-invasive stool tests, FIT and FIT-DNA,^{4, 6} and do not consider the effect of programmatic screening. This is especially important when considering tests with two different recommended screening intervals.

As shown in a recent decision analytic modeling study, a test with sensitivity and specificity reflecting those of Epi proColon® (sensitivity for colorectal cancer 68%, specificity 79%, positivity rate in persons with adenoma = 1-specificity) can achieve a programmatic benefit that is comparable to that of FIT.³ While the benefit of such a blood test is derived in part by a relatively high programmatic referral rate to colonoscopy due to its level of specificity,⁷ the consequences of this must be put in perspective: colonoscopy carries a low risk of complications and is already the most common colorectal cancer screening test in the U.S.; when colonoscopy is normal, it leads to a prolonged period of at least 10 years during which screening does not need to be repeated; when colonoscopy detects adenomas, adenoma removal followed by surveillance can lead to reductions in colorectal cancer incidence and mortality.

If CMS determines that the sensitivity and specificity of Epi proColon® are not acceptable, then we urge CMS to consider covering blood tests whose performance characteristics fall between those of Epi proColon® and the available stool tests, because of the benefit that such blood tests can deliver in persons who will not take up stool testing. We thus propose that CMS consider thresholds for coverage of sensitivity for colorectal cancer of approximately 70% (instead of 74%), and specificity for colorectal cancer and pre-invasive precursors, most importantly advanced adenomas, of approximately 85% (instead of sensitivity for colorectal cancer of 90%).

Conclusion

In conclusion, we urge CMS to avoid setting thresholds that are too stringent, which could stifle innovation in a field that is not yet mature, and which might deprive Medicare beneficiaries from the potential benefits of blood-based colorectal cancer screening.

Uri Ladabaum, M.D., M.S.
Professor of Medicine
Stanford University School of Medicine

Robert E. Schoen, M.D., M.P.H.
Professor of Medicine
University of Pittsburgh

John A. Baron M.D.
Professor Emeritus of Medicine
University of North Carolina at Chapel Hill

Dean E Brenner, M.D.
Talpaz Professor of Translational Oncology
Professor of Internal Medicine and Professor of Pharmacology
University of Michigan

Robert S. Bresalier, M.D.
Professor of Medicine
Distinguished Professor in Gastrointestinal Oncology
The University of Texas MD Anderson Cancer Center

Timothy R. Church, Ph.D.
Professor, Environmental and Occupational Health
University of Minnesota School of Public Health
University of Minnesota Masonic Cancer Center

Seth Crockett, M.D., M.P.H.
Associate Professor of Medicine
University of North Carolina at Chapel Hill

Ziding Feng, Ph.D.
Professor, Public Health Sciences Division
Co-Program Head, Biostatistics Program
Fred Hutchinson Cancer Research Center

David Ransohoff, M.D.
Professor of Medicine and Epidemiology
University of North Carolina at Chapel Hill

Mack T. Ruffin IV, M.D., M.P.H.
Professor and Chair, The Hershey Company Professor
Department of Family and Community Medicine
Penn State Health, Hershey Medical Center

Aasma Shaukat, M.D., M.P.H.
Professor of Medicine and Public Health,
University of Minnesota

Sapna Syngal, M.D., M.P.H
Professor of Medicine
Dana-Farber Cancer Institute
Brigham and Women’s Hospital and Harvard Medical School

Disclosures:
DB: consultant (Clinical Genomics)
DB: consultant (Clinical Genomics); research support (Clinical Genomics, Volition)
TC: advisor (GRAIL); consultant (Clinical Genomics, Guardant Health); data monitoring board (Medtronic, Spectranetics, Janssen)
SC: participation in clinical research without direct funding (Guardant, Freenome, EXACT Sciences)
ZF: research contract (EXACT Sciences)
UL: advisor (UniversalDx); consultant (Clinical Genomics, Guardant Health, Freenome, Medtronic, CheckCap, Leerink)
RS: research support (Freenome, Medtronic, Immunovia)
AS: consultant (Freenome); research support (Freenome)

References
1. de Moor JS, Cohen RA, Shapiro JA, et al. Colorectal cancer screening in the United States: Trends from 2008 to 2015 and variation by health insurance coverage. Prev Med 2018;112:199-206.
2. Ladabaum U, Allen J, Wandell M, et al. Colorectal Cancer Screening with Blood-Based Biomarkers: Cost-Effectiveness of Methylated Septin 9 DNA vs. Current Strategies. Cancer Epidemiol Biomarkers Prev 2013;22:1567-76.
3. Peterse EFP, Meester RGS, de Jonge L, et al. Comparing the cost-effectiveness of innovative colorectal cancer screening tests. J Natl Cancer Inst 2020. In press.
4. Imperiale TF, Ransohoff DF, Itzkowitz SH, et al. Multitarget stool DNA testing for colorectal-cancer screening. N Engl J Med 2014;370:1287-97.
5. Ladabaum U, Mannalithara A, Brill JV, et al. Contrasting Effectiveness and Cost-Effectiveness of Colorectal Cancer Screening Under Commercial Insurance vs. Medicare. Am J Gastroenterol 2018;113:1836-1847.
6. Lee JK, Liles EG, Bent S, et al. Accuracy of fecal immunochemical tests for colorectal cancer: systematic review and meta-analysis. Ann Intern Med 2014;160:171.
7. Ransohoff DF. Evaluating a new cancer screening blood test: Unintended consequences and the need for clarity in policy-making. J Natl Cancer Inst 2020. In press.

Less

One of the statements supporting your recommendation was that Epi proColon (mSeptin9) has not been included in colorectal cancer screening guidelines. Notwithstanding the fact that mSeptin9 was recommended in recent NCCRT guidelines, CMS has never before required inclusion in guidelines as a criterion for coverage; the most recent example being the recommendation that the stool DNA test be covered despite its not being included in guidelines at the time of the recommendation.

The recent JNCI study (Peterse, et al.), which followed CISNET modeling to compare various colorectal cancer screening modalities (FIT, mtSDNA, CTC, PillCam), clearly demonstrated that screening with mSeptin9 at one-year intervals was “the test of choice” for patients unwilling to participate in screening by colonoscopy or FIT.

A microsimulation model that compared all screening modalities, taking screening adherence into account (Cancer Medicine D’Andrea, et al., October 2019), demonstrated that mSeptin9 found more cancers, averted more CRC cases and deaths, and resulted in more Life Years Gained than any other modality when used at one-year intervals, and was superior to all but colonoscopy when used at 2-year and 3-year intervals.

At least 30% of the screening eligible population in the U.S. refuses screening. Having a covered blood test available should close that gap dramatically.

Your recommendation that it not be covered is bewildering. I respectfully urge to reconsider your recommendation and cover this test. This test will save lives!

Wendy L. Williams

Less

Tamara Syrek Jensen, JD, Director
Coverage and Analysis Group
Centers for Medicare & Medicaid Services
7500SecurityBoulevard
Baltimore, Maryland21244

Dear Ms. Jensen,

Thank you for the opportunity to comment on the proposed National Coverage Analysis for Screening for Colorectal Cancer - Blood-Based Biomarker Tests (CAG-00454N). The College of American Pathologists (CAP) appreciates CMS’ efforts to expand access to colorectal cancer screening among Medicare beneficiaries.

As the third most common cancer in both men and women, colorectal cancer (CRC) is also one of the most preventable cancers with proper screening and early detection of adenomatous polyps. However, according to the Centers of Disease Control and Prevention (CDC), about one in three adults between the ages of 50 and 75 years old are not getting screened for colorectal cancer as recommended.

As CMS moves to develop a national policy to broaden and standardize coverage for blood-based biomarker tests for CRC, we ask that CMS remove any coverage requirement language that would preclude FDA approved technologies that demonstrate clinical utility from receiving coverage under the NCD.

Also, we recommend that the final NCD include language allowing Medicare Administrative Contractors (MACs) discretion to develop local coverage determinations for CRC blood-based biomarker tests using their own technical assessment process, and independent of the tests’ FDA-approval status.

Thank you for your willingness to consider our comments on the proposed coverage analysis memo. Please contact Nonda Wilson at nwilson@cap.org with any questions.

Sincerely,

College of American Pathologists

Less

Guideline requirement concerns. The NCD for a previous test (Cologuard) was not predicate on guideline inclusion at the time of coverage determination. This requirement does not align with CMS’s stated goals for breakthrough technologies as these novel tests would not yet be included in guidelines at the time of FDA market approval.I believe that this requirement will result in a significant delay for a breakthrough technology to be reimbursed, given that the schedule for guideline update is typically a multiyear process. Currently there is no alternative for those individuals seeking colon cancer screening but not willing to use the traditional testing mechanisms. Thank you for your consideration.

Less

Obviously the people making the decision to exclude this from Medicare have NOT had endure the prep for a colonoscopy. It’s very hard for older adults. [PHI Redacted] This is insanity. I highly object. Not fair at all! A simple blood test would prevent the rigors of colonoscopy prep as well as the cost involved for having one done. Seems to me payment for that test would be much more expensive than a blood test! Drs recommend colonoscopy every ten years. Sounds like a conspiracy to me.

Less

As a healthcare professional, I believe that many options should be offered to the patient for diagnostic screening. A colonoscopy requires careful preparation and some patients are unable to comply with the time centered preparation. The cologard is good, as is the fecal occult blood test. The latter two are not as good at finding the malignancies as the colonoscopy nor do they provide a means of removal of polyps which can progress to malignancy.. With the progress in testing different substances present in the blood, we have new biomarkers that can possibly find the presence of malignancies when they are still small and contained and treatable.

Less

CMS should consider the evidence for coverage off FDA-approved septin9 testing under the conditions of current FDA labeling, which are for a small population untested by any other methods. This decision must be driven by science, but the weighting of risk and value for this particular test proposition also requires judgment as well. There is a large population not currently testing, and CMS offers no redress for this problem other than the potential to cover future tests at an unspecified time. More sophisticated modeling studies have been published and submitted for CMS's review by others.

In addition, all tests (other than colonoscopy) generate considerable colonoscopy candidates. Typically, there is about a 1% colon cancer rate or 10 patients per 1000. If we set sensitivity at 74% and specificity at 90%, there is essentially a 10% false positive rate. In our simple model, in those 1000 patients, there will be 7 colon cancers detected and 100 non-cancer patients flagged, or roughly 7 cancers per 107 patients flagged. FIT, Cologuard, etc, remain screening tests just like mammography or PSA or LDCT or pap smears. Sept9 is also a screening test.

While defining the positive rate only against colon cancer, the potential of flagging adenoma patients for intervention is missed. CMS should consider the important and recognized role of flagging adenoma and advanced adenoma patients for intervention, and CMS should encourage development and success in this area, which the NCD as drafted does not do.

Less

I represent the Colon Cancer Foundation, a 501c3 organization. Our mission is to support research into the causes and cures for colorectal cancer; increasing public awareness; educating the public about the importance of screening and early detection in the fight against colorectal cancer. Colorectal cancer screening saves lives and prevents cancer. The fact remains – 1 in 3 Americans STILL remain unscreened. Even worse, a majority of the cases in the unscreened population are late stage diagnosis, and deaths. A blood based test would address the patient barriers associated with traditional screening methods. Thus, Epi proColon should be a consideration to get non-compliant patients screened. Comparably, an annual Epi proColon FIT – performed better than currently covered screening methods - Cologuard, FIT, and CTC. Per the National Cancer Institute CISNET-MISCAN publication (Peterse et al – 2020) which studies reduction of incidence, reduction of mortality and life-years gained, annual Epi proColon, at the current performance, is the ‘Test of Choice’ for patients refusing colonoscopy and FIT. Most importantly, Epi proColon performed significantly better than ‘No Screening’. One goal of Colon Cancer Foundation is to get everyone access to screening. Medicare should be considering and covering all methods to reach all beneficiaries and save lives. Any delay would result in needless deaths due to CRC. CMS needs to take the NCI CISNET model into consideration and cover the ‘Test of Choice’ Epi proColon for the non-compliant patients.

Less

I am the CEO of a small IVD company that is developing a stool-based, preventive screening test for colorectal cancer. First and foremost, we believe this proposal is a great step forward in terms of setting parameters under which future tests will be covered, and will ensure that developers of tests have a clear pathway and clear endpoints to receive CMS coverage. This type of rule will ensure that patients have access to new alternatives for colorectal cancer screening and prevention. However, we believe that the final proposed rule should address the following three points: cover both liquid biopsy and stool-based assays on a rolling basis to prevent NCDs for advanced stool-based assays from becoming outdated, consider the critical impact of annual measurements for some tests and every-third-year measurements for other tests, and require the assessment of adenomas, which should also be a trigger for colonoscopy and intervention.

1) An "umbrella" policy is an important advance because it reduces the administrative burden for both CMS and developers, and speeds access to testing. However, it is unclear why the policy would only apply to blood-based tests. The patient requirements are identical to those covered under the existing NCD for all Colorectal Cancer Screening Tests and the sensitivity / specificity requirements are derived directly from the existing stool-based assays. We feel that this rule should apply to all novel tests, regardless of sample type. For example, the current NCD for a FIT-DNA test is specific to Cologuard version 1, and a successor version or competitor with equal performance would not be covered.

2) We believe that the accuracy benchmarks used for new tests are inconsistent across the different tests because they are used at different intervals (one year vs three years). Specifically, a blood-based test that has a 74% sensitivity for cancer and a three-year screening interval cannot be directly compared to a FIT test that has a 74% sensitivity for cancer and a one-year screening interval. In this example, the programmatic or multi-year cancer sensitivity of a FIT, even factoring in imperfect compliance with annual screening, is well in excess of 74%. Further, there are now a number of fecal immunochemical tests that offer sensitivity much higher than 74%, including several tests that report >85% sensitivity for cancer while maintaining a specificity above the specificity hurdle outlined by the CMS in the proposed rule. At a minimum, we believe that the accuracy benchmarks for sensitivity should be generated using consistent specificity metrics. In P130017, the Cologuard FDA approval, the FIT sensitivity of 74% was compared to the FIT-DNA test sensitivity of 92%. However, this comparison was made at different specificity metrics. When evaluating the ROC curve the cancer sensitivity of the FIT at the same specificity of the Cologuard test would rise to about 80%. We would encourage CMS to raise the sensitivity bar to ensure that new tests add meaningful value to clinical practice.

3) One of the most critical attributes of colorectal cancer screening tests is the sensitivity for adenomas. In the words of the US Multi-Society Task Force on Colorectal Cancer, "colon cancer prevention should be the primary goal of screening", making sensitivity for adenomas critical. Yet adenomas are not part of the coverage criteria at all. While the proposed rule mentions the importance of adenoma detection, it is concerning that no standard for advanced adenoma sensitivity in outlined under the NCD proposal. In P130017, Cologuard demonstrated a 42% sensitivity for advanced adenomas and FIT demonstrated a 24% sensitivity for advanced adenomas. If you adjust the FIT advanced adenomas sensitivity for screening interval and specificity (see section 2), the programmatic sensitivity for the FIT is close to Cologuard's sensitivity. We believe that at the very minimum, new tests should be able to detect advanced adenomas with at least the same sensitivity as Cologuard and FIT. We strongly encourage that the final rule include an appropriate benchmark for advanced adenoma sensitivity that novel tests must meet for coverage. This standard could be 40% on a programmatic basis.

We believe that this rule could have a tremendous impact on patient access to colorectal cancer screening alternatives, but we strongly recommend that the final rule also encompass other sample types, raise the bar for colorectal cancer sensitivity, and address adenoma detection.

Less

Within the CMS NCD proposal, there is a requirement for inclusion in clinical guidelines. This requirement not readily available for ‘breakthrough’ FDA approved technologies, inconsistent with the previous Cologuard NCD, and/or should align with the FDA intended use. CMS Breakthrough initiative recognizes FDA approved technologies to fast track into practical use. At time of FDA approval, no guidelines will have included a new technology as a standard of care or treatment. Thus, guideline inclusion should not be a requirement. Consistent with this observation, Cologuard did not have USPSTF or other society guideline inclusion at time of FDA-approval and subsequent positive CMS NCD. At the present, National Comprehensive Cancer Network (NCCN) does include Epi proColon parallel to the FDA intended use statement. Epi proColon is not for routine screening but for patients who refuse colonoscopy and stool testing.

Guidelines requirement should not a requirement for NCD. Or – CMS should recognize NCCN inclusion to address the unscreened population challenge.

Less

CMS,

I am writing to you to share my experience with Epi proColon and the patient population that I work with. I practice as a clinical consultant pharmacist with over 25 years of experience with developmentally delayed individuals. This population of patients have inherent difficulties with some of the most mundane of tasks; taking medications, getting dressed, communicating discomforts, dislikes and needs are just the tip of the iceberg. My experience with my patients, most of whom are in the fifth decade of life or greater but have the mental capacity of preschooler. If you could only imagine, the difficulty that is incurred when attempting to screen for colorectal cancer via tradition methods. The oral prep for a colonoscopy in this patient demographic is often difficult to impossible for the patients to tolerate. This often results in unsuccessful colonoscopies with the added risk that are incurred with unnecessary anesthesia. This blood based test, clearly has a role for patients that are non-compliant for traditional testing methods. Not withstanding, the vast majority of this patient population are covered by Medicare. It is my hope that you would reconsider CMS coverage for Epi pro-Colon.

Respectfully Submitted
Dr. Kelly Morris-Stephens
Clinical Consultant Pharmacist
Pharma-Care Inc.
136 Central Ave
Clark, NJ 07066
609-977-3506

Less

As a follow up to the recent decision by CMS to not fund the molecular blood test, Epi proColon®, manufactured by Epigenomics, I am providing CMS with my comments. As full disclosure, I am co-owner of Polymedco, a diagnostic firm based in New York since 1980. Polymedco has been involved in colorectal cancer screening (CRC) since 2006, marketing the product OC-Auto® FIT. OC-Auto FIT is the leading fecal immunochemical test in the world and in the U.S. market producing many millions of results each year. OC-Auto FIT has literally hundreds of studies worldwide published and includes many significant studies published in the United States.

I say all this because over the last 14 years, it has been clear to us that having a sensitive and specific screen for human hemoglobin in stool has led to a reduction of the incidence of CRC by 26%, and mortality, by 53%, which results were printed in the Journal Gastroenterology in January 2018 in a longevity study published by Kaiser Permanente of California using millions of patients over a 16 year period. So, what does that have to do with Epi proColon?

I am here clearly to say that the more our fellow citizens, from the age of 45 through 75 or older, are screened annually with the FIT test, the medical system would result in a significant reduction of the incidence of CRC and its mortality. That being said, the CDC has said that we have improved our screening rates through all methods to approximately 69% of our age appropriate population. Yet, 30% remains continually unscreened. What better way of capturing greater screening rates than by simply drawing blood and identifying with a fairly high proficiency of those who should go to the next step of colonoscopy. CMS approved over three years ago reimbursement for a stool molecular test called Cologuard® by Exact Sciences. Its reimbursement is well over $500 per test and has truly no better performance characteristics of sensitivity or specificity than Epi proColon.

Yes, it is true that the early studies for Septin-9 under the tradename Epi proColon exhibited a rather high false positive rate but the generation of Epi proColon that was approved by the FDA in 2016 and given a CPT code of 81327 by CMS, has every bit the level of sensitivity and specificity than the significantly more expensive Cologuard test. Not to mention, the huge difference in the ease of testing of a simple blood draw versus a patient who is collecting a rather large stool sample with 37 steps involved and is ran at a single lab by Exact Sciences in Wisconsin. Anyone surveying gastroenterologists who are screening for colorectal cancer will clearly hear that 20% of all the positives, or more, from Cologuard tests are simply false positives.

It was said back in 2009 by a leading gastroenterologist, one of the real important factors to any screen test, “is based on the test that gets performed” and certainly Epigenomics’ screen for CRC, called Epi proColon, is far easier for the patient and can be run in hundreds of laboratories throughout the United States. It also is at least $300 less in per result cost.

I thank you for taking the time to read these points and consider these arguments.

Respectfully,
Pete Welsh
Executive Vice President
Polymedco Cancer Diagnostic Products, LLC

Less

dear mcd , the money that you are trying to keep from the people is tax money that we and they have payed in , this money is or was to stay in an acount for the people in need and you keep taking out , stop being gready , and givethe people what you had promes us when you first begen this couverige , the money belongs to the people , the lord says if you gane the hole world and loes your sole , what have you gand , not a thing , stop robing people , if it was not for us the people you would not have a job , you onle intertan you frinds with the money you keep enny was , so stop robing the people , there a nough to go around

Less

It is so important to have as many options as possible for patients to be screened for colon cancer. Of course nothing replaces colonoscopy as the gold standard, but Epi ProColon is another tool in our toolbox to fighting advanced stage cancers. Not everybody wants to do ColoGuard and the availability of Epi ProColon is even more crucial for those in frail health or just plain stubborn and won't get colonoscopy. We know FOBT is not adequate and also poor as screening tools go. I take care of so many without insurance, access, minority and LEP. There is no valid reason NOT to cover this test.

Less

If the goal is to examine 80% of the population group relevant for screening and the current tests are just as expensive or the gold standard colonoscopy is extremely expensive, then I don't understand the calculation.
The necessary microsimulations and further studies were carried out to refute the cost / benefit by offering Epi ProColon annually, it beats the inventory tests and is cost effective.
The goal is to screen more people who reject the current screening tests and are looking for an alternative.
If these people did not refuse the screening, the costs for the current tests or the colonoscopy would be incurred directly.
At the end of the day, the motto that counts is: "The best test is the one that is taken".
The goal should be to save life and thus preserve sustainable value in society and make it accessible to everyone.
Of course, under the premise of taking the cost / benefit ratio into account.
If you disregard the relevant studies in the decision, and uses performance values ??that seem arbitrary, as a normal mortal one is disappointed with a health institution like the CMS.
I appeal to you to reimburse the Epi ProColon test according to the FDA-approved criteria - the test is for objectors.
It would also be conceivable that the reimbursement should run for 3 years until Epi ProColon fulfills the arbitrary performance values ??of you and then not only the refusers can be used but for everyone.

Best regards

Less

Since Epi proColon is the "Test of Choice" for those refusing colonoscopy and stool testing, then you should authorize this test in an effort that early detection will reduce the deaths of 33% diagnosed with colorectal cancer. This is similar to the controversy over approving payment for the CA-125 blood test for early detection of ovarian cancer which would give women the same chance of survival as men who given the PSA blood test to detect prostrate cancer.

The number one issue of Medicare should be saving lives through early detection and preserving the best quality of life possible. In summary, early detection of colorectal cancer not only saves lives, it will ultimately be a great cost savings measure over the long term period of treatment.

Less

To Whom it May Concern:
I would like CMS to cover the epi-procolon blood test. [PHI Redacted] I am so thankful that a "blood test" is available now. So much easier, convenient and I know this will be successful.. A blood based test would address these non-compliant patients by including Epi proColon in routine bloodwork. Data proves that 99.5% of patients would do a blood test for cancer screening, a simple and easy process. CMS must pay for this cost effective and critically needed blood test. [PHI Redacted] Thank you for your consideration. I look forward to your positive response for coverage.
Thanks,
Carol M. Zeidlhack

Less

As a citizen, I would like to commend CMS for their efforts to provide us with quality services and products, and protect us from those in healthcare who either use overt or passive methods to line their pockets with ill-gotten gains. And having spent my entire career working in healthcare, I would like to point out some concerns I have with the Proposed Decision Memo in question.

A few years ago, I came across an FDA-approved blood test (Epi proColon) with clinical studies compared to Colonoscopy and fecal immunochemical test (FIT). Per the Johnson (2014) study, Epi proColon and FIT had similar detection rates for cancer and lesions with a higher specificity (NOTE: I believe this was stated incorrectly within the proposed CMS coverage determination).

Recently, I also discovered a National Cancer Institute (NCI) comparison model in which Epi proColon performed better than mtSDNA, PillCam, CTC, and FIT. I believe the article endorsed Epi proColon as the ‘test of choice’ for people who didn’t want to get screened. (NOTE: This study was in the references but not cited)

This really concerns me – Epi proColon is excluded from the proposed Medicare NCD proposal. How can the ‘test of choice’ that performs as good or even outperforms other Medicare covered tests be excluded? This does not seem logical.

Medicare needs to pay for Epi proColon. It seems clinically effective and cost effective, according to the studies/models. If screening saves lives and healthcare dollars AND there are still people who are unscreened, then Epi proColon would be a simple option to get everyone screened and refer the higher risk Medicare recipients to colonoscopy.

Please, please, please – cover Epi proColon (especially during a pandemic)!!!

Yours Truly,

Rosalinda Tamayo
Senior Medical Technologist (Retired)

Less

According to the National Cancer Institute publication, Epi proColon is the ‘Test of Choice’ for patients who refuse colonoscopy or FIT. [PHI Redacted] Comparative performance of Epi proColon was better than No Screening and is cost effective. Epi proColon is a tool which healthcare professionals NEED to get their non-compliant patients screened. [PHI Redacted] Medicare would benefit (clinically and economically) from getting the unscreened patients screened rather than treating late stage cancer with suboptimal outcomes. Medicare should pay for Epi proColon and reconsider their latest decision.

Less

The USPSTF established recommendation include the use of blood based testing for colorectal cancer. CMS' decision to deny coverage seems arbitrary. In particular, this is a disservice to Primary Care Physicians who would use this test to close care GAPs. CMS appears to place more weight on opinions from societies entrenched in preserving endoscopy rather then helping primary care. Strategies that prevent common disease or identify early disease, especially when there is evidence of effective treatment should be given special consideration. CMS needs to support the assertion that for CRC screening, "the best test is the one that gets done." One third of the population refuses colonoscopy or stool based testing. Patients find blood based testing an acceptable alternative. Alex Fishberg, MD President Elect, AAPCE

Less

The proposal to not cover the Epi proColon blood test defies common sense for increasing CRC screening rates in average risk patients who refuse a FIT (including Cologuard) and Colonoscopy. How does this decision support increasing CRC screening rates in the unscreened Medicare population and saving on downstream medical costs when late stage disease is present? Peterse E et al. 2020. (Comparing the cost-effectiveness of innovative colorectal cancer screening tests. JNCI: Journal of the National Cancer Institute, djaa103. doi.org/10.1093/jnci/djaa103) clearly demonstrated the beneficial performance of Epi proColon compared with other CMS covered non-invasive screening methods (Cologuard, FIT) in terms of reduction in both cancer incidence (cases) and mortality (deaths). The authors concluded that Epi proColon is the test of choice for the unscreened. All the screening tests currently available for CRC are flawed in one aspect or another, yet more options for patients to be screened for CRC is the only way to reach more individuals and reduce the overall incidence and mortality for this disease. Include Epi proColon test as a CRC screening option in the NCD decision for a segment of the eligible Medicare population will not use another method to be screened.

Less

I am a physician, a licensed pathologist, and have dedicated my 45-year career, including twelve as a Senior Scientist at NIH, to bringing forward new technologies to diagnose and better treat cancer. I am writing today to strongly urge the Centers for Medicare and Medicaid Services (CMS) to immediately enable coverage of a proven, FDA-approved, colon cancer screening test, Epi proColon, that will prevent tens of thousands of American deaths.

[PHI Redacted]

It is high time CMS allowed the taxpayers’ dollars to cover the cost of the test for those who cannot afford it. There is no scientific justification for CMS to withhold it from the public. The science is solid. The FDA has already told us it works by approving it. It is sensitive enough, specific enough and picks up polyps, like mine. It brings people to colonoscopy who otherwise have avoided it. Why is CMS keeping the door closed to cancer screening for America? Do not let cancer win.

In April 2016 FDA approved (PMA) this DNA blood test for colorectal cancer. Since then nearly 200,000 Americans who have not been screened have died of colorectal cancer. It is shocking that CMS still has not initiated the National Coverage Decision (NCD) process to enable Medicare coverage of this blood test and. By improving the screening rate just 10% would save over 8000 American lives each year (American Cancer Society). That is one life saved every hour of every day or 36,000 avoidable deaths while America has waited for CMS to pay for a test. The blood test will save lives. Studies of nearly 10,000 patients prove it. Why are we waiting? Why do we need to see people die unnecessarily when we have the technology in our hands to save these precious lives?

The data is overwhelming that people will accept this blood test even if they have refused all other forms of screening. In study after study nearly 100% of those who have previously refused all screening readily take the blood test and, if positive, most go on to the gold standard, colonoscopy. That means potentially saving thousands of lives by finding cancer early, long before it has spread. Every day that CMS waits to cover the blood test means more deaths that could have been prevented. It is a shameful travesty and a national embarrassment that an agency dedicated to preserving the public health has dragged this out while Americans suffer.

The blood test is designed to be run in CMS approved (CLIA) laboratories throughout the USA. CMS has already established both a Common Procedural Technology (CPT) code and a payment level of $192 but another HHS office has blocked the life-saving test from Americans. Why? There is no moral explanation. We cannot wait while thousands die.

This cancer test is only the beginning. Years of research and development have fueled the pace of technological advancement enabled in HHS programs at the NIH and Human Genome Project. I know for I have been part of it. This progress has been harnessed to better treat cancers through improved screening, recurrence and response monitoring. How can an HHS office prevent Americans from receiving the benefits of our nation’s research? The floodgate must be opened to let proven technology reach the public. It is imperative that HHS agencies are aligned to enable the use of new, approved, safe and effective, technologies to quickly become covered services. Within the decade, many new liquid biopsy methods for blood-based cancer monitoring will change the outcomes but only if appropriately covered. CMS should not use blocking of coverage to discourage scientific development of promising new technologies that save lives and healthcare dollars.

Who among us doesn’t want to save lives from horrible metastatic colorectal cancer and at the same time save millions of health care dollars? It is time, today, to open the door and cover this life saving test. “How wonderful it is that no one has to wait but can start right now to change the world!” (Anne Frank)

Do not wait one minute longer.

Less

Respectfully, this is an egregious ruling. The process you have chosen is the classic equivalent of “pretzel logic.”

Using a one-time sensitivity only (based on one time FIT test), rather than programmatic sensitivity via a microsimulation, such as CISNET, is not how the test will be used. This is clearly developed as an annual and programmatic test. For those who choose a blood test over a colonoscopy or a stool test, as this designed be an annual testing option and this is obvious in their material. So, why did you choose to do this?

Then, the committee shifts with no logical reason to use the specificity of and testing intervals (3 yrs) for testing for stool DNA for comparison. Epi Procolon is intended to be used as an annual test, so why did you choose 3 yrs as the interval ? Very misdirected and misleading as to the intended use. Again, why did you choose this non related comparison option?

Patients need more options, not less. Blood based testing is critical to engaging more at average risk individuals to participate in CRC screening programs. Patients are accustomed to blood draws and want options that don’t require them to have an invasive procedure such as a colonoscopy or to collect stool specimens.

In Kentucky, where our advocacy efforts are centered, there are huge challenges getting persons to participate in CRC screening, including those in Appalachia and communities of colon. This test finally expands the options to include one that is more palatable and likely one that will enhance compliance among those not currently engaged in a screening program.
This is the first of many genomic options for patients and providers based on blood samples so you better start getting it right in your methodology. I recommend starting with a revision of the issues raised above before any final rulings on this ( Epi Procolon) proposed screening test and protocol.

Less

As co-author of several publications regarding the Epi proColon assay for the detection of methylated Septin9 in plasma, I feel compelled to identify and suggest corrections to several misrepresentations contained in the recent Proposed Decision Summary.

The erroneous position in the Proposed Decision Memo on page 38 states “the evidence (direct or indirect) is not sufficient to conclude that screening in Medicare beneficiaries using the Epi proColon test is suitable for prevention or early detection of cancer... The indirect evidence (comparing a blood based screening test to another test that has been shown to improve mortality such as screening fecal immunochemical test) does not show that the blood based screening test is equivalent in the detection of large adenomas or early stage cancer." With respect to direct evidence:

1. In the cited Johnson et al (PLOS 2014) publication (pages 22-23 of the decision memo), data is presented that shows that Epi proColon detects as many positive samples as the automated FIT method used as the comparator. It also confirms that the methods are identical in confirming the absence of CRC.
2. The Proposed Document states that there is clinical utility in using FIT for early CRC detection and disease prevention. The Epi proColon method detected an identical level of early stage cancers (in the cancer sample arm) and detected twice as many small polyps and adenomas in the (non-cancer arm of the trial) comparing Epi proColon and FIT.
3. They concluded that at a sensitivity of 72%, the Epi proColon assay was non-inferior to FIT CRC detection, although at a lower specificity. With negative predictive values of 99.8%, both methods are identical in confirming the absence of CRC. The Septin9 assay did have a higher positivity for subjects who were negative for CRC, and would have resulted in an increased number of follow up colonoscopies.
4. So in fact, the only “harm” would be an increased number of “gold standard “ colonoscopies which are the preferred method (MSTF Guidelines) for CRC screening.
5. Another publication (Liles et al) followed repeatedly non-compliant patients (to FIT or colonoscopy). Within six weeks of the positive blood result, 17/30 had underwent colonoscopy and 10 of the 17 had lesions removed during the colonoscopy.

With respect to indirect evidence:

1. In the Peterse publication (JNCI, 2020), the CISNET modeling effort (which has been used by the USPSTF, CMS and ACS), the data demonstrated that the Epi proColon test with an annual use interval outperformed (reduction in disease incidence and deaths) both Cologuard and FIT which are currently covered services.
2. This same result was also demonstrated in the D’Andrea model (2019).
   Disease modeling is an established comparative methodology for outcomes analysis of colorectal cancer.

Less

At my institution we serve thousands of the medically underserved at student-run health fairs. One of our health fair stations addresses colon cancer screening, and our patients have no reasonable access to endoscopic modalities of screening. Historically we have provided FIT for patients, but the return rate in this population is around 10%, in line with the rate reported in studies of similar cohorts. With the move to Sept9 testing via blood draw, we test >95% of eligible patients, helping close a gap in access to this important screening. This helps us prioritize navigating patients to low-cost colonoscopy for the positive patients.

Less