Friday, July 5, 2024

MolDx Nixes Coverage of Castle Squamous Stratification Test

On July 4, 2024, MolDx released a final LCD L39583 with non-coverage of the Castle Biosciences squamous test.  This non-coverage matched the position of a draft LCD released on June 8, 2023.   The coverage was requested, per an included letter, on July 1, 2020 by Castle.  

Castle share price, which has been around $22 in recent months, slipped from $21 to $17 on Friday (roughly $100M of market cap).   See coverage 7-11 at 360Dx.  This was updated to include Castle comments, e.g. "robust validation data...overlooked by Palmetto" and [Quoting Castle] pointing to two high-quality validation studies addressing MolDx concerns.  [See SIDEBAR on the two "overlooked" papers, ARRON 2024 and WYSONG 2024].

Q&A Document

The final LCD was released along with a question-and-answer article on the public comment period (A59815), which prints as a 128-page PDF for me.  The main MolDx reebuttal section runs 6300 words and is organized on 11 topics.

Conclusions

MolDx's conclusory non coverage remark states:

In conclusion, the current evaluated tests at the time of drafting this policy have not yet demonstrated definitive additional risk stratification beyond that provided by available clinical, pathological, and staging criteria within the NCCN risk framework. Promising evidence is in development for use in selection of patients for adjuvant RT. This contractor will continue to monitor the evidence and may modify coverage based on new information in the pertinent literature and society recommendations.

MolDx was also concerned the populations studied did not include enough persons of color, stating:


Test performance of the 40-GEP has been reported in a population that is predominantly male and essentially all Caucasian, which is representative of the majority of patients with cSCC. Although incidence is highest in non-Hispanic white populations, the risk of metastasis is greater in patients of color. Further studies should include representation of test performance characteristics inclusive of and addressed in patients of color.

I include a PDF-format redline version of Draft LCD and Final LCD in the cloud here.  The LCD has 83 bibliographic citations and the body is 8900 words.

##

AI CORNER

Readers should review the full final LCD and the Q&A article, as linked above.  

Solely for curiousity, I include a  Chat GPT analysis below.



##

Summarize the "response to comments" document.

Below is a concise summary of the 12 points addressed by MolDx in their rebuttal to the non-coverage policy for Castle Bioscience's GEP-40 test for squamous cell carcinoma risk stratification:

  1. Complexity of cSCC and Staging Inaccuracies: MolDx acknowledges that current staging systems for cutaneous squamous cell carcinoma (cSCC) have limitations in accurately predicting outcomes. The GEP-40 test claims to offer improved risk stratification, but its clinical utility and validation need further evidence.

  2. Validation of GEP-40: The GEP-40 test has been validated in cohorts representing the intended use population. However, MolDx emphasizes that this validation should be more comprehensive and consistent with broader clinical evidence.

  3. Focus on Metastasis: MolDx recognizes metastasis as a critical outcome for Medicare beneficiaries. They point out that the GEP-40 test should demonstrate significant impact on patient management and outcomes, especially considering the low metastatic rates in high-risk cSCC groups.

  4. Independent Risk Information: GEP-40 provides additional quantifiable risk information beyond current clinical and pathological criteria. Nonetheless, its added value in improving patient management and outcomes over existing measures needs clearer demonstration.

  5. Risk Stratification Within Stages: The test claims to stratify metastatic risk within stages effectively. However, MolDx notes that its clinical utility must be proven within the context of national guidelines and current risk assessment tools.

  6. Correlation of GEP Results with Staging: MolDx argues that merely correlating GEP results with existing stages or risk factors is insufficient. The test must show independent value and lead to meaningful management changes that improve patient outcomes.

  7. Incorporation into Patient Management: GEP-40 has shown consistent integration into management plans aligning with national guidelines. However, MolDx calls for more robust evidence to support these claims and their impact on clinical practice.

  8. Identifying ART-Eligible Patients: The test helps identify patients with high biological risk of metastasis who may benefit from adjuvant radiation therapy (ART). Yet, MolDx stresses the need for further validation and evidence of its efficacy in special populations like immunosuppressed patients.

  9. Healthcare Cost Savings: Appropriate use of GEP-40 for informing ART decisions can potentially save healthcare costs for Medicare. Nonetheless, MolDx clarifies that cost savings alone do not justify coverage without strong evidence of clinical benefit.

  10. CMS Review and Validation: MolDx disputes claims that CMS's Division of New Technology has validated the GEP-40 test's clinical validity and independence. They assert that coverage decisions are based on medical value and reasonable necessity criteria. [SEE NOTE RE #10]

  11. Criteria for Medical Necessity: The test's medical reasonableness and necessity must be demonstrated within the specific clinical context of cSCC. MolDx emphasizes that each test and disease state requires distinct evidentiary support for coverage.

  12. Overall Evidence and Recommendations: MolDx concludes that the GEP-40 test currently lacks comprehensive evidence of clinical validity and utility. They recommend further studies and peer-reviewed publications to substantiate its claims and impact on patient outcomes.

This summary provides a clear and concise overview of MolDx's rebuttal points, highlighting their key arguments and concerns regarding the GEP-40 test.

[RE #10: BQ notes that MolDx's response to Comment Topic 10, the ADLT status of the test, was an anaphylatic reaction.  MolDx writes in this context, "Statements alluding to this [ADLT] process as one that establishes the medical appropriateness of a test, or that it even considered such evidence are false and wholly without merit."  Ouch.]

##

Which MolDx arguments seem most important [based on Q&A responses]?

The most important and contributory concerns of MolDx leading to the final non-coverage decision for the Castle Bioscience GEP-40 test can be identified as follows:

  1. Lack of Demonstrated Clinical Utility and Validity: MolDx emphasized the need for comprehensive evidence showing that the GEP-40 test improves patient management and outcomes. They pointed out that the test must demonstrate clinical validity and utility beyond current staging and risk assessment tools. This concern is crucial as it directly impacts the perceived value and effectiveness of the test in real-world clinical settings.

  2. Inadequate Evidence of Added Value: MolDx noted that while the GEP-40 test adds quantifiable risk information, its independent value over existing clinical and pathological criteria has not been sufficiently demonstrated. They stressed that the test needs to show a meaningful reclassification of patients that leads to significant changes in clinical management and improved outcomes.

  3. Questionable Impact on Management and Outcomes: MolDx highlighted that the test's proposed management changes based on GEP-40 results have not been validated independently. They questioned whether the test leads to clinically impactful decisions, particularly in patients with low baseline metastatic risk. This concern undermines the argument that GEP-40 significantly enhances clinical decision-making.

  4. Discrepancies in Metastasis Rates and Validation Cohorts: MolDx pointed out inconsistencies between the metastasis rates reported in the GEP-40 validation studies and other independent studies. They raised concerns about the generalizability of the patient populations used to validate the test, suggesting that the test's predictive value might not be as robust in broader clinical practice.

  5. Inconsistencies in Utilization and Guidelines: MolDx mentioned that the incorporation of GEP-40 into clinical practice, as proposed by Castle Biosciences, lacks consistent support from independent experts and national guidelines. They emphasized that the test must align with established guidelines and demonstrate clear benefit in a standardized clinical context.

Overall, the most critical concern was the lack of comprehensive and independent evidence demonstrating that the GEP-40 test provides significant clinical utility and improves patient outcomes. This fundamental issue directly impacted the decision against coverage, as MolDx requires robust proof that any new test meaningfully enhances current medical practices and benefits patients.


##

Now discuss the full Q&A document, including access to all 128 pages of public incoming comments.

Based on the detailed comments and responses in the document, the most significant and contributory concerns raised by MolDx that led to the final non-coverage decision for Castle Bioscience's GEP-40 test for cutaneous squamous cell carcinoma (cSCC) can be summarized and discussed as follows:

  1. Lack of Demonstrated Clinical Utility: One of the primary reasons for non-coverage was the insufficient demonstration of clinical utility of the GEP-40 test. MolDx emphasized that the test needed to show that its use leads to significant improvements in patient management and outcomes. The comments highlight that although the test provides additional risk stratification information, it has not been conclusively shown to impact clinical decision-making in a way that improves patient outcomes.

  2. Inadequate Evidence of Added Value: MolDx pointed out that the GEP-40 test needed to demonstrate definitive value above and beyond existing clinical, pathological, and staging criteria. They critiqued that the test did not consistently show statistically significant, independent risk stratification beyond what is already provided by current staging systems.

  3. Validation Cohorts and Generalizability: MolDx raised concerns about the generalizability of the validation cohorts used for the GEP-40 test. They argued that the cohorts used did not adequately represent the broader population of patients with cSCC. This was particularly important as it questioned the test's applicability and reliability across diverse patient groups, including those treated in community settings versus academic centers.

  4. Independent Risk Information: While the GEP-40 test added quantifiable risk information, MolDx noted that its independent value in improving patient management and outcomes was not clearly demonstrated. They emphasized that merely providing additional risk information is not sufficient unless it leads to actionable changes in clinical practice.

  5. Correlation with Current Risk Assessment Tools: MolDx pointed out that the test's performance needed to be evaluated in combination with current risk assessment tools to demonstrate added improvement. They criticized the lack of robust evidence showing that GEP-40 significantly reclassifies patients within existing staging systems in a clinically meaningful way.

  6. Consistency in Clinical Practice: MolDx highlighted discrepancies in the use of formal staging systems among clinicians, noting that not all physicians consistently use these systems to inform management decisions. This inconsistency undermines the argument that the GEP-40 test provides a necessary improvement over current practices.

  7. Healthcare Cost Savings: Although Castle Biosciences argued that the GEP-40 test could lead to healthcare cost savings by informing ART decisions, MolDx countered that cost savings alone do not justify coverage. They stressed that the primary criterion for coverage is the demonstration of clinical benefit and medical necessity.

  8. Impact on Patient Management: MolDx questioned whether the GEP-40 test leads to meaningful changes in patient management, especially given the low metastatic rates in high-risk cSCC groups. They noted that even if the test stratifies risk, it must show that this stratification translates into improved outcomes through altered management strategies.

  9. CMS Review and Validation: MolDx disputed claims that the CMS Division of New Technology validated the GEP-40 test’s clinical validity and independence. They clarified that the coverage decision relies on demonstrating medical value and reasonable necessity, which was not sufficiently shown for the GEP-40 test.

  10. Appropriateness of Risk Stratification: MolDx emphasized that the GEP-40 test should demonstrate improved risk stratification within the context of current guidelines and risk assessment tools. They highlighted that the test's performance should be validated within a standardized clinical context to ensure it meets the unmet clinical need.

In summary, the most significant concerns from MolDx that led to the non-coverage decision centered around the lack of robust evidence demonstrating the GEP-40 test's clinical utility, added value, and impact on patient outcomes. These concerns underscored the need for more comprehensive validation and proof that the test leads to meaningful improvements in clinical practice and patient management.


##

The above AI analysis is based on the Q&A document.  

Switching to the LCD document.  Next, Chat GPT discusses the actual final LCD.

For those avidly interested in understanding how decisions regarding Medicare coverage for molecular biomarker tests are made, the provided document offers a detailed example, specifically focusing on the non-coverage decision for the Castle Biosciences GEP-40 test for cutaneous squamous cell carcinoma (cSCC). Here’s a TLDR summary of the key points, organized into the coverage decision, the review of evidence, and the analysis of evidence:

Coverage Decision

The Medicare Molecular Genomics Program (MolDx) decided not to cover the GEP-40 test. The primary reason for this decision is the lack of sufficient evidence demonstrating that the test provides additional clinical utility beyond existing risk stratification methods.

Review of Evidence

  1. Clinical Background of cSCC:

    • cSCC is the second most common skin cancer, with increasing incidence.
    • While most cases are treatable, a subset can metastasize, leading to poor outcomes.
  2. Current Risk Stratification and Staging Systems:

    • Existing systems include the National Comprehensive Cancer Network (NCCN) Guidelines, the American Joint Committee on Cancer (AJCC) 8th Edition, and the Brigham and Women’s Hospital (BWH) staging system.
    • These systems aim to categorize tumors into low, high, and very high-risk groups to guide treatment.
  3. Development and Validation of the GEP-40 Test:

    • The GEP-40 test was designed to stratify metastatic risk using gene expression profiling.
    • Validation studies showed the test could categorize patients into low, moderate, and high metastatic risk groups, but these studies had limitations in their design and population.

Analysis of Evidence (Rationale for Determination)

  1. Clinical Utility and Validity:

    • For a new biomarker to be covered, it must demonstrate additional risk stratification and improve patient outcomes within the framework of existing guidelines.
    • The GEP-40 test has not shown definitive clinical utility beyond current staging and risk assessment tools.
  2. Population and Metastatic Rates:

    • The intended use population for GEP-40 is too broad, including many patients unlikely to benefit from the test.
    • The metastatic rates in the test’s validation cohorts are higher than those observed in broader NCCN-validated populations, questioning the generalizability of the results.
  3. Integration with Current Guidelines:

    • The test must work in combination with existing guidelines to reclassify patients in a way that impacts clinical decisions meaningfully.
    • Current analyses do not adequately separate high-risk from very high-risk groups within the NCCN framework, which undermines the test’s proposed utility.
  4. Consistency and Validation of Proposed Frameworks:

    • Proposed management strategies integrating GEP-40 results with current guidelines have not been validated or widely supported outside Castle Biosciences.
    • Most metastases still occur in tumors classified as low or moderate risk by the test, raising concerns about its predictive accuracy.
  5. Representation in Studies:

    • Test performance has predominantly been reported in a demographic not fully representative of all cSCC patients, particularly those of color, who may have higher metastatic risks.

Conclusion

The evidence presented does not sufficiently demonstrate that the GEP-40 test provides additional, reliable risk stratification or improves clinical outcomes compared to existing methods. Consequently, the test has not been granted coverage. MolDx will continue to monitor emerging evidence and may adjust coverage decisions based on new, robust data.

This summary encapsulates the key considerations and rationales behind the non-coverage decision, offering policy experts a clear view of the stringent requirements and rigorous evaluation process employed by Medicare.