Humor: AI Explains Lab Reimbursement in One Graphic

I asked Chat GPT Dalle the following question.

###

I am doing a client deck on intro to coverage and reimbursement for molecular testing. I am a pathologist and policy expert. I would like a Powerpoint cover illustration that shows the complexity of CPT coding, PLA codes, Clinical Laboratory Fee Schedules, Health Technology Assessments, Medicare, MACs, Commercial insurer coverage, Laboratory Benefit Managers, etc.  

###

DJT Rescinds Biden-Era Healthcare AI Policies

HEADER - On Day 1, DJT rescinding a number of policies related to healthcare, particular AI.   Some other policies - like Ozempic under Part D - remain in Biden draft form and will be finalized during the spring.

###

On Day 1, DJT rescinded a number of policies in place by virtue of executive order, sometimes supplemented by additional follow-on policy statements, documents, or goals of the prior administration.

The basic Physician and Hospital Outpatient rules for 2025 were completed November 1, and are in place for 2025, like new rules for unbundling payment for radiophamaceuticals (like AMYVID).

Some Biden-proposed rules are in play, such as the Medicare Avantage annual rulemaking.   That included Part D coverage for GLP drugs like Ozempic, which will be finalized, or not, in March 2025.

Administration and HHS rules for healthcare AI have looming future impact on many things, but immediate impact on the many big labs that are expanding in real-time their AI software services for  clinical genomics and health systems.  Eg January 21, 2025, Tempus releases "olivia" AI enabled healthcare app.

##

There are three early articles on the AI rescission and related topics, all of which are open access, and the STAT newsletter is most detailed.

• Trump rescinds AI safeguards - VERGE.
• Trump inherits unfinished health policies, like Ozempic - AXIOS.
• This M.A. rule also includes pages on AI in Medicare Advantage - see my blog here.
• Trump rescinds AI order, with healthcare insights and links - STAT.

Mario Aguilar's article at STAT includes:

• Linkback to a prior article on what Biden EO meant for health.
• Linkback to an HHS strategic plan for AI in health.
• Linkback to ann inventory of how HHS was using AI.
• Coverage of "model cards" for AI used in Health.  These are almost like food nutrition labels but for software like healthcare machine learning tools.  Link to article about FDA endorsing software "model cards."

Though it's a finalized rule, not a new one, see the early 2024 HHS ONC HTI-1 final rule on certified artificial intelligence in healthcare IT - here.

Outgoing CMS Leader Pens Article: Digital Mental Health, CMS, and FDA Innovation

Outgoing Dr. Meena Seshamani and colleagues pen an article in Stat-Plus, showing ways that CMS was able to innovative and leverage FDA decisions in digital health.  Specifically, digital mental health.  Seshamani was director of the Center for Medicare under Biden.

Find the subscription article here:

https://www.statnews.com/2025/01/17/fda-cms-dmht-code-payments-behavioral-health/

###

###

I'm cautious about summarizing subscription articles, but I'll offer a short one here, considering that this is a CMS-authored article on public health news.  By the way, Dr Seshamani holds an MD from Penn and an economics PhD from Oxford.

###

Chat GPT4 :

The U.S. faces a significant behavioral health crisis due to increasing prevalence of mental health conditions and a shortage of behavioral health providers, leading to long wait times and poor outcomes. To address this, CMS and the FDA have collaborated on a groundbreaking policy under the 2025 Physician Fee Schedule (PFS) to improve access to digital mental health treatments (DMHTs), which are FDA-authorized software devices for conditions such as depression, anxiety, and insomnia.

Key highlights of the policy:

• New Medicare coding and payment for DMHT devices that are part of an ongoing behavioral health treatment plan, effective January 1, 2025.
• Payments include not only the devices but also practitioner services, such as patient education, onboarding, and data review.
• Only DMHTs with FDA authorization and clinical evidence of effectiveness qualify for reimbursement.

The policy aims to:

1. Encourage broader adoption of DMHTs by addressing uneven payer reimbursement, which has historically deterred investment and innovation.
2. Spur new investments in digital mental health technologies by providing predictable payment frameworks.
3. Enable practitioners, including primary care providers, to integrate DMHTs into treatment plans, enhancing accessibility.

The FDA supports developers through programs like the Breakthrough Devices Program and the TAP Pilot, offering guidance and expedited regulatory pathways. The agency is also drafting industry recommendations for clinical considerations in DMHT development.

This joint initiative between CMS and the FDA exemplifies their commitment to leveraging technology and innovative solutions to tackle public health challenges, setting a precedent for future collaborations to expand access to critical health resources.

###

Do you see any implications for clinical genomics, that could be drawn from the novel  approach discussed above for digital health?

###

The CMS and FDA collaboration on digital mental health treatments (DMHTs) offers several implications for the clinical genomics industry, particularly in how innovative technologies may gain reimbursement and drive investment. Here are the key takeaways:

1. Reimbursement Pathways for Digital Tools Could Extend to Genomic Applications

• The establishment of specific coding and payment for DMHTs provides a precedent for reimbursing other advanced digital technologies. In genomics, this could pave the way for similar approaches to digital platforms and AI-driven genomic tools, such as those supporting precision medicine or real-time clinical decision-making.

2. Focus on FDA Authorization and Clinical Evidence

• CMS's requirement for FDA-authorized devices with clinical evidence highlights the importance of demonstrating clinical validity and utility for genomic technologies. This aligns with existing hurdles for genomic test developers, emphasizing the need to engage with the FDA early and pursue programs like the Breakthrough Devices Program to streamline the regulatory pathway.
• Developers of genomic testing software (e.g., tools for analyzing NGS data or predicting risk) could benefit from these programs if they present sufficient evidence of clinical impact.

3. Integration into Broader Treatment Plans

• DMHT reimbursement depends on the device being part of an ongoing behavioral health treatment plan, which suggests a growing CMS preference for technologies that are integrated into holistic care pathways rather than being standalone diagnostics. For genomics, this may mean prioritizing tools that:
  • Link test results to actionable treatment plans.
  • Integrate with EHR systems for seamless clinician workflows.
  • Generate real-time data insights that guide patient management.

4. Opportunities for Investment and Innovation

• The policy aims to encourage investment by reducing uncertainty in reimbursement, a key barrier to innovation. In the genomics space, this could lead to:
  • Renewed interest in digital platforms supporting genomic tests.
  • More robust funding for genomics-adjacent tools, such as digital pathology or AI-driven genomic interpretation, if clear reimbursement pathways emerge.

5. Precedent for Medicare Payment Expansion

• The creation of new Medicare codes for DMHTs demonstrates CMS's willingness to adapt to emerging technologies. This signals potential openness to reimbursing novel genomic applications, such as:
  • Predictive genomic tests (e.g., for mental health, cancer, or rare diseases).
  • AI-powered genomic risk calculators.
  • Digital therapies combining genomic insights with behavior modification programs.

6. Collaborative Models Between Agencies

• The FDA and CMS collaboration could serve as a blueprint for advancing reimbursement in genomics, particularly for companion diagnostics and software-based platforms. Advocacy groups within the genomics industry might push for similar policy frameworks to address unmet needs.

Actionable Takeaways for Genomics Stakeholders:

• Focus on clinical evidence: Strengthen evidence demonstrating the impact of genomic tools on outcomes, not just diagnostics.
• Engage with FDA early: Programs like the Breakthrough Devices Program and Total Product Life Cycle Advisory Program (TAP) can expedite regulatory processes for innovative genomic tools.
• Advocate for reimbursement policies: Push for CMS to adopt coding and payment strategies tailored to cutting-edge genomic applications, leveraging the DMHT model as a precedent.

In summary, the DMHT policy signals a shift toward technology-driven healthcare where evidence-backed, FDA-authorized innovations are rewarded with predictable reimbursement. This creates opportunities for the clinical genomics industry to follow a similar trajectory.

JP Morgan Panel - "What You Won't Hear From the Podium"

The J.P. Morgan life sciences conference appropriately triggers a flood of news coverage - at Genomeweb, Endpoints, Stat, and elsewhere.   I'm on the lookout for coverage of a January 14 panel called, "What You Won't Hear From the Podium."

This panel was co-chaired by Burt Zweigenhaft (Association for Value-Based Cancer Care) and Lisa Gill (JP Morgan).   The five panelists were Shawn Marie Bishop (Akin regulatory law), Marc Samuels (ADVI Health consultancy), John O'Brien (National Pharmaceutical Council), Janet Woodcock (FDA, ret.), and Joe Grogan (DJT White House, USC, Paragon Health Institute).

See Burt's post at Linked In here.  See also Joe Grogan's repost and summary here.

Grogan highlights three points:

1. The DJT administration will be VERY different that the past four years, in healthcare.[*]
2. AI will bring drastic changes, but exactly what, is unknown.
3. To understand federal reimbursement changes, track Paragon Health Institute.

Admittedly, Grogan is the CEO of the Paragon Health Institute.  but, the authoritative Inside Health Policy has linked Paragon with DOGE.

Paragon has current PDF briefings on changes for Medicaid, Medicare, private healthcare, and public health.  These are one-pagers of key points.  The Medicare Initiative at Paragon is led by Demetrios Kouzoukas, one of the top officials for CMS in the first DJT adminisntration.

See also Paragon's 2025 nine-page report on paths forward for Medicare Advantage and for ACOs - here.  

See also Paragon's 2023 20-page report on Medicare as-is, called, "Roadblock to Progress."

##

Outside Paragon, see also an innovative paper by Abe Sutton, likely to be the next head of the Center for Medicare Innovation / CMMI.  In 2024, he published an important proposal for better and less "gamed" risk adjustment in Medicare Advantage, leveraging machine learning in practical ways.  "Inferred Risk: Reforming Medicare Risk Scores to Create a Fairer System,  Sutton & Drapos, Health Affairs.  (AI summary here.)

###

[*] In past weeks, I published several blogs where I used AI to mine and extract long interviews and speeches with DJT, RFK Jr, Vivek Ramaswamy, and others on Republican health policy.  On January 17, WSJ reported Bill Gates was "impressed" by DJT's interest in health issues.

###

AI CORNER

I gave Chat GPT the four Paragon 1-pagers and the 9- and 20-pagers.

###

First, here is a deeper dive on the 20-page Paragon report, Medicare, Roadblock to Innovation

###

Here are summaries of the Paragon Health Institute's policy documents, focusing on health policy for the 2025 presidential administration. I’ve emphasized actionable insights, with added detail for the 9-pager and 20-pager, while assessing the creative content of the one-pagers.

---

One-Pagers

1. Medicaid Reform​

   • Highlights concerns about Medicaid’s growth (87 million enrollees by 2023) and misaligned priorities from the Affordable Care Act (ACA), favoring able-bodied adults over vulnerable populations.
   • Recommendations:
     • Equalize federal reimbursement rates across populations.
     • Shift able-bodied adults toward employment and private coverage.
     • Combat fraud and encourage state-level innovation.

2. Medicare Reform​

   • Addresses Medicare's insolvency risks and inefficient government-centered policies.
   • Key proposals:
     • Expand Medicare Advantage (MA) flexibility.
     • Focus funding on care quality, reducing overpayments and improper payments.
     • Promote price transparency and empower providers to focus on patient care.

3. Private Health Insurance Reform​

   • Critiques ACA subsidies and pushes for more market-driven solutions:
     • Expand flexible, affordable private insurance options.
     • Enhance health savings accounts (HSAs) for cost-sharing.
     • Improve price transparency and reduce regulatory barriers for clinicians.

4. Public Health and Pandemic Response​

   • Criticizes public health missteps during COVID-19, advocating for:
     • Fewer mandates, more individual autonomy.
     • Refocusing agencies like the CDC on infectious disease control, away from politicized topics like social justice.
     • Streamlining regulations to foster innovation in treatments and emergency preparedness.

---

Nine-Pager: Two Paths for Medicare’s Future​ 

• Overview: This report contrasts Medicare Advantage (MA) with Accountable Care Organizations (ACOs) as alternatives to traditional Medicare (TM).
• Key Insights:
  • MA: Provides full capitation, incentivizes cost containment through competition, and integrates benefits like Part D drug coverage. However, inflated benchmarks (108% of TM costs) and "coding intensity" raise taxpayer costs.
  • ACOs: Focus on shared savings and quality metrics for TM enrollees but face limited flexibility and participation bias.
  • Policy Challenges:
    • Risk adjustment discrepancies favor MA over ACOs.
    • Quality metrics often add administrative burden without clear patient benefits.
    • Structural inefficiencies in ACO benchmarks disincentivize long-term savings.
  • Outlook:
    • The report suggests competitive bidding for both MA and ACOs to provide better cost-control incentives across Medicare.

---

Twenty-Pager: Medicare as a Roadblock to Innovation

• Core Argument: Medicare stifles health care innovation through regulatory complexity, inconsistent policies, and market distortions.
• Key Barriers:
  • Uncertain Coverage Policies: Lack of clear criteria for innovation adoption hinders investments in new technologies.
  • Flawed Value Assessment: Medicare’s reliance on administratively set prices undermines efficient resource allocation.
  • Regulatory Delays: Programs like Coverage with Evidence Development (CED) trap innovations in "purgatory" with incomplete data collection.
  • Impact on Private Sector: Medicare policies spill over to private markets, raising costs and slowing innovation.
• Recommendations:
  • Streamline coverage and payment decisions with transparent benchmarks.
  • Reduce reliance on centralized decision-making and empower consumer-driven choices.
  • Adopt policies like the defunct Medicare Coverage of Innovative Technologies (MCIT) program to fast-track innovation integration.

---

Analysis for Biotech and Genomics Experts

1. Medicaid, Medicare, and Private Insurance Reforms (four 1 pagers):

   • Proposals to reduce fraud and expand patient choice could indirectly benefit biotech innovators by reallocating public funds toward high-impact care.

2. Medicare’s Innovation Roadblocks:

   • The 20-pager identifies CED inefficiencies and reimbursement unpredictability as barriers to genomics testing adoption in programs like Medicare. 
   • Simplified pathways like MCIT would better align with the needs of biotech companies.

3. ACOs vs. MA:

   • The 9-pager hints at ACOs’ potential role in genomics-focused precision medicine, given their emphasis on early detection and coordinated, targeted care. 
   • However, Medicare Advantage's flexibility and bundled payments may offer a mix of opportunities for diagnostics and advanced therapies.

---

Critique of One-Pagers:

• The one-pagers summarize familiar conservative policy themes like market-driven reform, deregulation, and fraud reduction.  They enunciate priorities, but lack the depth seen in the longer documents.

Spunky: Gottlieb and Sharfstein Debate FDA vs CMS vs Money

 How much, if at all, should the government tie regulatory approvals (FDA) to payment (CMS)?

The usual industry answer is, NOT AT ALL.

Former FDA administators or deputy administrators Scott Gottlieb and Josh Sharfstein weigh in.   Audience often chuckles.  It's an entertaining video.

One Gottlieb comment, 'I had the impression that CMS didn't usually have enough expertise in clinical trial design to give thoughtful input..." (3m20s)

Find the video here:

https://www.youtube.com/watch?v=OY1jdCIPPcQ

See other new talks of similar interest at the UCSF-Stanford "CERSI" regulatory science institute web site.   Here's one from CERSI with 4 differerent FDA commissioners.  Here's one from CERSI on FDA AI.

AI Summary:

The FDA debate featuring Scott Gottlieb, former FDA Commissioner, and Josh Sharfstein, former FDA Deputy Commissioner, centered on whether there should be a firewall between FDA regulatory policy and drug costs. Here are the key points from their arguments:

---

Scott Gottlieb: Pro-Firewall

• FDA's Role and Expertise: FDA should focus on evaluating drug safety, efficacy, and risk-benefit, not on costs or reimbursement decisions. Adding payer input during the approval process risks making clinical trials longer, more complex, and expensive.
• Challenges with CMS Involvement: Past efforts to integrate CMS into trial design were largely unsuccessful. CMS lacked the expertise in clinical trial design, leading to impractical expectations and unrealistic coverage evidence development (CED) requirements.
• Separation of Roles: Coverage decisions are inherently political and subjective, often influenced by policymakers, which could compromise the FDA’s objective regulatory framework.
• In contrast, specific yes/no FDA decisions landing on the Comissioner's desk "didn't ever happen in my entire experience" (6m30s)
• Regulatory Uncertainty: Introducing cost considerations into FDA approval could create ambiguity and politicize the process, undermining trust in FDA’s standards.
• Alternative Mechanisms: Real-world data collection post-market and payer-driven negotiations are better suited to address cost concerns without impacting the regulatory process.

---

Josh Sharfstein: Anti-Firewall

• FDA as a Public Health Agency: Sharfstein argued that high drug costs hinder public access to medications, contradicting the FDA’s mission to promote health and prevent illness.
• Policy Proposals Integrating Cost:
  1. Encouraging Generic Competition: FDA could prioritize approving generics for high-cost drugs, improve standards for complex generics, and act against price-gouging.
  2. Addressing Patent Gaming: More transparency and regulatory action against anti-competitive practices that delay generics could reduce costs.
  3. Reforming Accelerated Approvals: Strengthening post-approval study requirements and capping payments for certain drugs could ensure value without overburdening patients or Medicare.
  4. Transparency with Payers: Sharing FDA data could enable value-based pricing and broader patient access.
• Partnership with CMS: While acknowledging challenges, Sharfstein advocated for voluntary collaboration between FDA and CMS, particularly in early reimbursement decisions for devices or drugs, to streamline processes and reduce inefficiencies.

---

Debate Outcome

The audience vote was close, slightly favoring the anti-firewall position (51%). While Gottlieb defended the current separation as essential for preserving FDA’s integrity, Sharfstein argued for innovative approaches to integrate cost considerations while maintaining FDA’s focus on public health.

###

BONUS

Here's an AI Summary of the AI talk cited above.  

Russ Altman moderates; Rahul Adora from OpenAI.  Panelists, Mark Taisey, Amgen, Janet Woodcock, fmr FDA, Liang Zhao, UCSF.

TL;DR Summary

A panel discussion at the UCSF-Stanford CERSI explored the integration of generative AI into regulatory science. Participants included industry leaders, academic researchers, and former FDA officials. The focus was on understanding the benefits, risks, and future applications of AI in regulatory workflows. Key themes included:

1. Current Uses of AI:

   • Automating regulatory processes, such as creating clinical study reports and quality summaries, reducing timelines significantly (e.g., from 26 weeks to 8-12 weeks for Amgen’s applications).
   • Exploring co-pilot tools for FDA reviewers to enhance knowledge management, consistency, and efficiency.

2. Challenges and Risks:

   • Concerns over data privacy and security, especially when AI relies on external servers or cloud-based platforms.
   • Risks of AI hallucination and compounded errors in autonomous workflows, requiring rigorous quality checks.
   • Barriers to structured data adoption, including reliance on legacy systems and resistance to change within regulatory agencies.

3. Opportunities:

   • Using AI to reduce human workload on "busy work" and streamline decision-making.
   • Enhancing global regulatory collaboration via cloud-based platforms like Accumulus.
   • Creating structured data systems to eliminate redundant documentation and expedite drug approvals.

4. Future Vision:

   • Transitioning from text-heavy documentation to structured data systems with real-time cloud integration.
   • Leveraging AI for strategic innovations while ensuring human oversight for critical judgments.
   • Expanding public data availability to foster innovation while addressing confidentiality issues.

Takeaway:

AI offers transformative potential for regulatory science, but success requires addressing privacy concerns, ensuring accuracy, and fostering collaboration between stakeholders. A balance between automation and human judgment remains essential

Follow-up Blog: Why Comment on New Coverage Rules for Medicare Advantage (Jan 27, 2025)

In December 2024, I highlighted some potentially faulty new rules that CMS is proposing for Medicare Advantage plans, relative to LCDs.   With the comment deadline approaching on January 27, 2025, I'd like to highlight the issues again and I encourage individuals and lab associations to comment, even if briefly.

IN A NUTSHELL

CMS wants to ensure Medicare Advantage plans operate under water-tight rules so that their coverage matches Fee for Service coverage.   However, by focusing solely on LCD text, and not billing articles or the MolDx Registry, CMS may create new problems.

WHAT'S HAPPENING

We're in the 60 day comment period for annual Medicare Advantage rules.  Labs I work with, labs I hear at conferences, and news articles, all complain about burdensome Medicare Advantage denials.

CMS is well-intentioned, and wants to esnure that Medicare Advantage plans follow LCD/NCD rules.  However, the devil is in the details.

Find the proposal and online comments link here - 89 FR 99340.  

WHAT CMS SAYS

CMS revisits perplexing and longstanding difficulties in getting Medicare Advantage plans, which are numerous and variable, to cover LCD-NCD services.   CMS previously visited this topic in April 2023 and a memo in February 2024.

In Section U of the December 2024 proposal, CMS reiterates that M.A. plans must cover LCD NCD services.  However - perhaps because M.A. plans had misused billing articles for denials - CMS states firmly that M.A. plans need only cover the TEXT of the LCD or NCD.   M.A. plans are not to find coverage rules in related documents any more (like billing articles).   CMS states that LCDs will always have clear coverage rules, and that any uncertain text would have been fixed by public comments.

WHY THIS IS A PROBLEM

Let me focus on the MolDx program, which handles the lion's share of CMS MoPath billings, and which I understand best.   

MolDx now usually writes "foundational" LCDs, with broad guidance for coverage.  For example, the transplant genomic testing policy says that such tests, like donor organ circulating DNA tests, may be medically necessary if they are validated and pass a MolDx tech assessment.   However, you can't tell what tests are covered for what transplant types and how often, from the LCD.  This is intentional - MolDx refers you to billing articles or more likely, its "Palmetto DEX Registry," an entirely separate, copyrighted, database not mirrored anywhere at CMS.   

(In fact, a MolDx notice  informs readers that updated coverage may NOT be found in LCD articles, but INSTEAD, over at the DEX database.)

click to enlarge  https://www.palmettogba.com/moldx

For example, LCD L38671 dates from 2021, and Natera had a press release in 2025 that "single lung transplants" were now covered for its  particular brand of test.   This is an obvious example of fee for service LCD coverage, impossible to tell from the 2021 LCD.  It's only available by lookup in the DEX copyright database (see figure above).

When limited to the text of the LCD itself, as CMS proposes, you get only generic statements like this:  "Clinical validity (CV) of any analytes (or expression profiles) measured must be established through a study published in the peer-reviewed literature for the intended use of the test in the intended population."  While that's "true," it's impossible to tell what is covered, for whom, or at what intervals, from the sentence just quoted.

Exactly the same points could be made for the MolDx Minimal Residual Disease testing policy, updated in 2021, but under constant on-the-fly changes in actual coverage, through various announcements external to the LCD itself. 

WHAT IS NEEDED

CMS needs to be informed about the above situations, and that, in many cases, it is IMPOSSIBLE for each different reader to conclude what is covered "by the text of the LCD alone."  The comment page is here and click the green box by January 27 to comment.

####

MY MODEL COMMENT

I haven't finalized my own comment, but here's a draft.

Comment on Section U, Internal Coverage Criteria (§422.101):

I am commenting as a pathologist and full-time Medicare policy expert. While I appreciate CMS's effort to clarify the use of internal coverage criteria under §422.101, I am concerned that the proposed language at §422.101(b)(6)(i)(A) and related text (e.g., 99457 col. 3 and 99458 col. 1) may severely misalign with the practical realities of Medicare coverage policies, particularly for the Molecular Diagnostic Services (MolDx) program.

MolDx LCDs frequently provide only general guidance for coverage. For example, transplant test LCD L38671 includes broad statements such as, "Clinical validity (CV) of any analytes (or expression profiles) measured must be established through a study published in the peer-reviewed literature for the intended use of the test in the intended population." In practice, actual coverage details are continuously updated in supplemental resources like the Palmetto DEX™ Diagnostics Exchange registry, which MolDx explicitly refers to as the authoritative source for test-specific coverage determinations [FN1].

While CMS assumes LCDs undergo rigorous review and provide clarity and specificity, foundational MolDx LCDs such as L38671 (genomic transplant rejection tests) and L38814 (minimal residual disease testing) remain intentionally broad. MolDx directs stakeholders to supplemental resources, including LCD articles and the DEX registry, to operationalize coverage determinations. These external resources are integral to understanding and applying coverage policy. [FN1, FN2, FN3]

Requiring Medicare Advantage (MA) plans to rely solely on explicit LCD text, as proposed, risks confusion and misinterpretation. For instance, the generic language in MolDx LCDs—e.g., “clinical validity must be established”—cannot alone guide readers at MA plans to determine which tests are covered, for whom, or under what conditions. Without reference to supplemental resources, MA plans would struggle to align with the dynamic and nuanced nature of MolDx policies.

Conclusion: I urge CMS to explicitly acknowledge the role of supplemental resources like the DEX registry in operationalizing coverage policies. Prohibiting MA plans from referencing these essential tools would undermine the intent and functionality of the MolDx program, and reverse the intentions of CMS's new rulemaking.  Given that MolDx publicly and deliberately directs stakeholders to these resources as the main record of its evolving coverage decisions, CMS should revise the proposed language to accommodate their use.

FN1

https://www.discoveriesinhealthpolicy.com/2025/01/follow-up-blog-why-comment-on-new.html

FN2

https://www.discoveriesinhealthpolicy.com/2025/01/decibio-reports-on-mrd-coverage.html

FN3

https://www.natera.com/company/news/natera-announces-medicare-coverage-for-prospera-in-single-lung-transplant-recipients/

##

Addendum 1/21/2025.  Another example of coverage by press release, NOT by revision of the LCD, is Guardant's new coverage for "surveillance" of colon cancer recurrence.   Guardant had had 1X testing to assess completeness of CRC surgery.   Now, it can be used on initially-negative patient to track them under surveillance for relapse.  But there's no change to the 2021 LCD for MRD, L38814.

https://finance.yahoo.com/news/guardant-health-receives-medicare-coverage-130500233.html

Stat Plus: Abe Sutton May Return to Trump Administration, Lead CMMI at CMS

Stat Plus reports that, in the new CMS headed by Dr. Oz, DJT veteran Abe Sutton may be picked to head the Center for Medicare Innovation (CMMI).

Abe Sutton, who's crammed McKinsey, White House, Harvard Law School, and Rubicon Advisors into the last several years, was said to be a pivotal figure in a big, cross-agency renal program rolled out by the prior administration in 2019.

Here's the new subscription article at Stat Plus, by Rachel Cohrs Zhang.

• https://www.statnews.com/2025/01/13/former-trump-white-house-hhs-aide-abe-sutton-likely-to-lead-medicare-medicaid-innovation-center/

Here's my 2019 article on the coordinated wide roll-out of a new renal plan which Sutton was involved in.

• https://www.discoveriesinhealthpolicy.com/2019/08/perspective-coodinate-rollout-of-new.html

Here's an article in 2024 in Health Affairs by Sutton on reforming Medicare Advantage risk adjustment to avoid gaming.

Here's an article on Oz's developing CMS staff, from WaPo.

## 

##

About Adminstration, but not about Sutton

Also at Stat, Gottlieb and Verma discuss the new administration.

At a UCSF-Stanford conference, Gottlieb and Sharfstein debate FDA vs CMS.

U Penn Study Finds Weak Results for FDA PMA Opioid Risk Test

A subscription article at Genomeweb on January 9 highlights:

FDA-Approved Opioid Use Disorder Risk Test Little Better Than Coin Toss, Study Finds

By Forest Ray, here.

I haven't studied the test so I don't have a personal viewpoint on it.  However, some of the fact patterns are interesting.

• The test was approved by FDA with a prominent favorable press release from the FDA - December 19, 2023.
• The test has the more sophisticated type of FDA device approval, a PMA approval, P230032B. (37pp).  Package insert (18pp).
• I believe there was an FDA Ad Comm as well, along the way. See MedPage Today, October 2022.
• By April 2024, the AvertD test saw negative press, e.g. at WBGH PBS here, or at Washington Post here.
• The new paper at JAMA Network is Davis et al, here.

###

HHS OIG Releases Annual Required Lab Spending Report: CY2023 (December 2024)

On December 20, 2024, the OIG released its mandatory annual report on CMS spending on the Clinical Laboratory Fee Schedule.   The report is required by the PAMA legislation of 2014.

Find the report here:

https://oig.hhs.gov/reports/all/2024/total-medicare-part-b-spending-on-lab-tests-decreased-in-2023-driven-in-part-by-less-spending-on-covid-19-tests/

The 12-page report is useful because it includes hospital outpatient lab tests, as well as Part B CLFS tests.  (Many CMS Part B reesources don't include hospital lab tests.)   For the year-on-year analysis,  OIG found that COVID spending fell from $1.7B in 2022 to $751M in 2023, a 56% decrease and a drop of nearly $1B.

In addition, by 2023, CMS had finally closed town the explosion of spending on sequencing code 81408, a $2000 nonspecific sequencing test that had colossal payments with no controls for a while at the Novitas and FCSO MACs (esp. 2019, 2020.)   Genetic spending fell from $1.9B in 2021 to $1.4B in 2022.  (They define "genetics" broadly, e.g., Cologuard).

Kvetch - Why not upload a link for full data?

OIG publishes a data table (page 7) of the top 25 tests by dollar spend.  Only a few of the top 25 are genetic tests.   For my money, since OIG has the data, it could ALSO simply post a link to the entire spreadsheet (>1000 tests) for the public.  It would cost OIG nothing and cost the public nothing.

Top 25

click to enlarge

The top genetic test (#4 of all tests) is 81528, Cologuard, $508, voume up 13%, spend $301M.

The next genetic test (#9) is 81455, >50 tumor genes, uup 59%, $145M.   (I believe this code is only used outside of MolDx states).

The next genetic test (#14) (55-74 genes, Guardant CDx) is0242U, $5000, up 31% YonY, $105M.   

The fourth genetic test (#17) is Oncotype Dx, 81519, $3873, up 2%, at $95M.  

DeciBio Reports on MRD Coverage; Natera Updates on Plans

 Via Linked-In, Amal Thommil of DeciBio reports on the status of MRD coverage.   

Take-home lesson - the top three insurers providing MRD coverage are MolDx, MolDx, and MolDx.  (Old joke).

Find his Linked In post here:

https://www.linkedin.com/posts/amalthommil_im-a-bit-late-for-this-update-i-apparently-activity-7282519172910067713-aEdH/ 

Here's a rough screen shot; see original above.

I believe the Louisiana BCBS coverage is related to a state biomarker bill.  Note that the circles denote tissue type by color, and name-brand by name.   (E.g. yellow circle = bladder = Natera).

See also a follow-on post by Amal [Jan12], here.    See comments by Amal on Natera using WGS rather than exome for MRD [Jan 17] here.

###  NATERA

During JP Morgan 2025, Naterea announces a webinar on its "Innovation Roadmap."

See Linked In essay by Julie Deschenes here.   Go direct to January 15 webinar info here.  Go driect to Natera/Investors.  

Red Letter Day: Novitas Finalizes Massive Oncology LCD

Header:  Novitas MAC finalizes a massive LCD on oncology testing, proposed in Summer 2023 and delayed since Summer 2024.

###

In  June 2023, Novitas issued a massive and unexpected level of revisions to its LCD for oncology testing.   There were protests to CMS that this final LCD  different so much from its publicly-commented draft, that it was improper.

In July 2023, Novitas withdrew that "final" LCD and reissued it verbatim  as a new, "draft" LCD for public comment.   Of particular note, about a dozen branded tests were listed as non-covered.   Of particular impact in my view, the LCD as a whole was hugely difficult to read, repetitive, and poorly written.

The final version was "due" per CMS rules in July 2024, but Novitas issued that CMS had granted a finalization delay.

On January 9, 2025, Novitas releases a final LCD (L39365), which will be effective in 45 days.

DRAFT

https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=39667&ver=9&stateRegion=all&contractorNumber=all&proposedStatus=F&sortBy=noticeStart&bc=11

FINAL

https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=39365&ver=131&stateRegion=all&contractorNumber=all&proposedStatus=F&sortBy=noticeStart&bc=11

BRUCE ZIP FILE

https://drive.google.com/file/d/1020FmaYQy_sOMAhcPLvWtNFq6Qe0TFc7/view?usp=sharing

ZIP includes draft, final, and redline.

At Genomeweb here.  They summarize that the LCD rescinds coverage for, "Castle Biosciences' DecisionDx-SCC test; Pacific Edge Diagnostics' Cxbladder Detect, Enhanced Cxbladder Detect, Cxbladder Monitor, and Cxbladder Triage assays; Interpace Biosciences' PancraGen; Clinical Genomics' Colvera; and the University of Pittsburgh Medical Center's ThyroSeq Cancer Risk Classifier and PancreaSeq Genomic Classifier."

Novitas' explanation of "changes" (quote)

The provision of coverage for molecular testing endorsed by selected knowledgebases (e.g., NCCN, OncoKb, and ClinGen) has been removed. The knowledgebases will continue to serve as evidence resources, where applicable. As a result, the final LCD addresses 10 molecular tests, and the title has been changed to reflect the change in policy scope. 

• Following review of additional evidence, limited coverage has been granted to the UroVysion fluorescence in situ hybridaztion (FISH) test. 
• The following tests have been removed from the [non coverage] policy; DecisionDx Melanoma, Enhanced Cxbladder Triage and Cxbladder Resolve. 
• Considering the lab’s location, DecisionDx-Melanoma claims are not processed within this jurisdiction; therefore, its coverage determination is outside of the scope of this LCD, and all text referencing DecisionDx-Melanoma has been removed. 
• Enhanced Cxbladder Triage and Cxbladder Resolve are removed since they are not yet commercially available. 
• Additional information has been added to the summary and analysis of evidence sections in response to comments and supporting literature received.

#####

#####

The Coverage Guidance / Indications    section (first section) is not concise and clear but very long and subjective, inclues numerous definitions, etc.   (See why this is important in context of proposed, new, Medicare Advantage rules here.)

For me, the draft has 37702 words, 75 pages, 230 citations.   

For me, the final has  33905 words, 64 pages, 470 citations.

Articles

Scroll to the bottom of the Final LCD webpage and find two billing and coding articles (A59125, A58917) and a Response to Comments article (A59856).  I haven't assessed the billing/coding articles but my memory is the proposed billing and coding article was hugely long with a vast number of subgroups, and there seem to be only a few subgroups now.

##

In May 2024, MolDx published an important article updating and showing how it interprets language about pathologist test orders in the modern context of genomics.  Here. 

For comparison, here is the new Novitas language on the same topic:

• Treating physician is the physician who is treating the beneficiary, that is, the physician who furnishes a consultation or treats a beneficiary for a specific medical problem and who then uses the results in the management of the beneficiary's specific medical problem. Tests not ordered by the physician who is treating the beneficiary are not reasonable and necessary. Nonphysician practitioners that are enrolled in the program, who are operating within the scope of their authority under State law and within the scope of their Medicare statutory benefit, are also considered for this purpose, as treating and managing a beneficiary’s specific medical problem. (see CFR, Title 42, Volume 2, Chapter IV, Part 410.32(a))

###

Back in October 2023, I used AI to summarize all 21 presentations at the public comment meeting.

Outside the LCD, see a draconian Novitas  policy a few months ago on new PLA codes.

AI Corner

Chat GPT compares draft and final.

##

Comparing the draft and final versions of LCD 39365 reveals that both remain quite verbose and detailed, though the final version appears to have undergone some reorganization and minor editing to improve clarity. Here are the key observations:

Improvements in the Final LCD:

1. Streamlining: Some sections in the final version are more concise compared to the draft, focusing more directly on actionable points. However, the overall document still includes extensive background, definitions, and technical details that might overwhelm readers unfamiliar with the subject matter.

2. Specific Focus: The final version seems to narrow its focus on certain tests and their evidence. For example, UroVysion is discussed with clearer clinical indications, which helps make that section more targeted.

3. Additional Evidence: The final document includes more comprehensive summaries of evidence for specific tests, like the Cxbladder family, which were either absent or less detailed in the draft. This aligns with its intent to base coverage decisions on actionable and well-documented evidence.

Persistent Issues:

1. Verbose Nature: Both versions are dense, making it challenging for readers to extract key points quickly. Even in the final version, lengthy sections of background and technical definitions may obscure the coverage determinations.

2. Repetitiveness: Definitions and concepts, such as clinical utility, clinical validity, and test categories, are repeated in multiple sections, which adds to the bulk without necessarily aiding comprehension.

3. Complexity: The use of highly technical language and exhaustive references to external guidelines and databases like NCCN, ClinGen, and OncoKB may make the document less accessible to general stakeholders, including clinicians without specialized genomic expertise.

4. [References to ClinGen etc were ONLY in draft, not in Final- BQ]

Is It Still Confusing?

Yes, the final LCD retains a level of complexity and verbosity that can confuse readers. While the reorganization and added specificity are steps in the right direction, the document still lacks a user-friendly structure or clear summaries to aid quick understanding, particularly for clinicians needing actionable guidance.

##

Addendum

References to ClinGen, NCCN, and OncoKB that were present in the draft LCD appear to have been entirely removed in the final LCD.

Draft LCD:

The draft contained detailed references to ClinGen, NCCN, and OncoKB as evidence-based sources for evaluating genetic tests' clinical actionability and utility. These sections outlined criteria for determining whether specific tests meet the medical necessity requirements based on these organizations' guidelines.

Final LCD:

The final LCD does not mention ClinGen, NCCN, or OncoKB at all. Instead, it focuses on specific tests like UroVysion and Cxbladder, with evidence drawn directly from the literature or clinical studies without citing these external knowledge bases.

This change suggests a deliberate shift in approach, possibly to simplify the document or avoid reliance on external organizations' frameworks, which might have been seen as too complex or non-essential for the final coverage policy.

##

Response to Comments

The Response to Comments document for LCD 39365 shows that Novitas Solutions adopted a mixed approach in addressing public feedback. Below is an analysis of their responses:

Strengths in Novitas' Approach:

1. Acknowledgement of Comments: Novitas consistently acknowledged all submitted comments, showing that they reviewed them and incorporated changes where appropriate.

   • Example: Removing references to knowledge bases (e.g., ClinGen, NCCN, OncoKB) in the final LCD was a direct response to significant stakeholder concerns about outsourcing decision-making to unaccountable external sources.

2. Specific Revisions Based on Feedback:

   • Histologic/Cytologic Requirement: Flow cytometry was added as an acceptable method for establishing a "substantiated suspicion of cancer" in response to stakeholder concerns about diagnostic flexibility.
   • Limitations Adjusted: For example, they removed overly restrictive provisions like the "one-time germline testing" rule and clarified the allowance for repeat molecular testing when clinically appropriate.

3. Clarity of Justifications: They provided detailed explanations and cited CMS guidelines or NCD precedence to support their decisions, lending credibility to their responses.

Weaknesses and Potential Gaps:

1. Perception of Dismissiveness: While Novitas made several significant changes, there are instances where responses seem to brush off stakeholder concerns by reiterating their initial rationale rather than substantively engaging with the issues raised.

   • Example: Concerns about the restrictive nature of evidence requirements (e.g., analytical validity, clinical utility) were acknowledged but largely dismissed, with no substantial revision to these requirements.

2. Insufficient Action on Complex Issues:

   • The request for clearer integration of NCD 90.2 (NGS for solid tumors) with the LCD was not fully addressed, leading to lingering ambiguity about how NGS for hematologic malignancies will be handled.
   • Comments requesting a maintained, up-to-date database of covered tests were not implemented, even though such a resource could simplify navigation for stakeholders.

3. Changes Without New Supporting Evidence:

   • The removal of coverage for specific tests (e.g., Cxbladder, Colvera) relied heavily on Novitas’ claim of insufficient evidence without offering updated evidence reviews or detailed counter-arguments to address comments.

Overall Assessment:

Novitas made meaningful changes in response to key stakeholder comments, demonstrating a willingness to revise their policies and enhance flexibility where justified. However, their approach was sometimes dismissive or overly reliant on existing guidelines without addressing the practical complexities raised by stakeholders. The final LCD is less controversial than the draft but retains some of the verbosity and complexity that led to the initial concerns.

Blog Highlight: "AI Predicts ACLA Win in the LDT Lawsuit."

Most people see this blog via daily news feeds, so let me highlight again a fun blog issued December 25.   "AI Predicts ACLA Win in LDT Case."

At that time, final court documents were filed (by ACLA, AMP, and Government) in the FDA LDT federal case in Texas.

The blog had three parts:

(1)  I asked Chat GPT to read the documents and predict the winner.  It summarized the documents and predicted  FDA to lose, ACLA/AMP to win.

(2)  More adventurously, I asked Chat GPT to go ahead and write a mock up of the judge's opinion, which won't be known until March 2025.  It did so.

(3) Pivoting to a humorous note, I then asked Chat GPT to re write the mock up judge's opinion in different fictional voices - Columbo, comedian Lewis Black, Charles Dickens, and others.  (Links to sidebars).

Here's the original December 25 blog:

https://www.discoveriesinhealthpolicy.com/2024/12/chat-gpt-4o-reviews-final-fda-ldt-court.html

Journal Watch: Bloomberg Sees Fall-off in Equities Analysts; A Link to Genomics?

Two interesting articles related to Bloomberg this week.   

On January 8, 2025, Rao et al. discuss the job cuts for equities analysts that are occurring on Wall Street.   Professional analysts are pared back, forced to cover more industries or companies more thinly, or dropping coverage.   Here.    The Bloomberg authors link to a SSRN public article by Nathan Dong of Boston College, "Can AI Replace Stock Analysts?"  SSRN here.   

• Some analyst functions may be replaced by AI; 
• While freelance deep-dive analysts are filling in with self-employment via personal subscriptions (eg Substack).   
• Cutbacks in analysts could result in "undervaluations" or other effects, say Rao et al.

###

Both traditional biotechnology (drugs) and genomics are highly volatile markets. However, biotech volatility often stems from tangible, random events, such as unexpectedly positive or negative phase 2 trial results. In contrast, genomic volatility can be more perplexing. Take the Covid-era equity and VC bubble (2021–2022) as an example.  In a couple years, Invitae's stock soared from $15 to $50, only to crash to near zero, even as the company’s revenue and growth remained relatively stable.

The genomics sector includes a diverse range of public companies like Natera, Guardant, Veracyte, Castle, Exact, GeneDx, and others. If the thesis of Rao et al. holds true—that public market coverage is becoming increasingly thin and less granular—then genomics might present new and unique opportunities for informed investors. Those with deep expertise and a nuanced understanding of these markets (including factors like LCDs, NCDs, coding policies, and Medicare Advantage behaviors) may find underappreciated opportunities for both buying and selling.

AI Corner

The last two paragraphs are an AI rewrite that is clearer than the text I typed myself.

##

As for the weirdo corners of genomics policy, see a nerdy blog this morning on a new cancer LCD. Or odd MAC non coverage of PLA Codes.

##

AI Corner 2

Chat GPT reads my blog, above, and gives me feedback, below.

##

Your blog piece highlights an important intersection of finance and genomics, framing it within a broader industry trend of reduced equity analyst coverage and the implications for highly specialized fields like genomics. Here's how your insights could be further refined and emphasized:

Key Points to Emphasize

1. Distinction Between Biotech and Genomics Volatility:

   • You effectively differentiate biotech's event-driven volatility from genomics' complexity-driven volatility. This insight is a cornerstone of your argument, as it underlines why genomics requires deeper expertise that AI or generalized analysts cannot yet provide. Expanding on real-world examples, like the Invitae case, strengthens this point.

2. AI’s Limitations in Genomics:

   • While AI is gaining ground in financial analysis, genomics operates in a regulatory and scientific labyrinth where human expertise remains irreplaceable. Medicare policies, coding nuances, and payer behaviors are not just complex—they're dynamic and require on-the-ground expertise to interpret their implications on market valuations.

3. Opportunities Amid Reduced Analyst Coverage:

   • The thinning of equity analyst ranks creates gaps that can be exploited by those with specialized knowledge. This scenario could lead to undervaluations or missed signals, which represent opportunities for investors attuned to the idiosyncrasies of the genomics sector.

Suggestions for Refinement

• Strengthen the Call to Action: Conclude with a direct appeal to investors or companies, emphasizing how specialized expertise, like yours, can bridge the growing analytical gaps in the genomics industry.

• Expand on AI's Role: Discuss the types of tasks AI is adept at in financial analysis (e.g., pattern recognition, earnings prediction) versus its struggles in interpreting Medicare LCDs/NCDs or payer-specific nuances, which require deep domain expertise.

• Connect Freelance Analysts to Genomics:

  • Highlight how the rise of freelance analysts or Substack-style publications could benefit the genomics sector, given its need for detailed, expert-driven insights. Position yourself as an exemplar of this emerging model.

Expanded Conclusion Example

"In an era where Wall Street thins its analytical ranks, the genomics sector represents both a challenge and an opportunity. Its complexity defies automation, requiring not only technical acumen but also a mastery of its arcane regulatory and reimbursement frameworks. For those willing to engage deeply, this landscape offers unique possibilities for identifying both undervaluations and hidden risks. As we navigate this evolving terrain, one thing is clear: in genomics, the value of true human expertise has never been greater."

Would you like to refine any particular section further or add more examples?

News Updates: CMS Prices MRD; NIH Names 2 Screening Tests; FDA Releases Flood of Guidances.

Flurry of news this week.

MRD Pricing at CMS

Adaptive Technologies issues a press release that its clonoSEQ test, PLA 0364U, is priced at $2007.  (This was via the gapfill process).   MolDx episode pricing for multiple tests updated to $8029.

VANGUARD Moving Forward at NCI (MCED RCT)

Two tests selected by NCI for its VANGUARD study of multi cancer screening.

• ClearNote Health announces that NCI selected its "Avantec" multi-cancer screening test as one of the two tests.
• Guardant announces that its SHIELD test is the second test selected for Vanguard. 

Read more about the Vanguard study at Cancer.gov (NIH/NCI) here.   NCI writes, "These two assays will not be compared to each other but will be compared to cancers detected in the Control Arm. The companies are providing the assays without cost to the study participants."

At Genomeweb here.

###

FDA Releases Guidance Flurry

As tallied by Endpoints News, FDA just released "dozens" of guidance and draft guidance documents, in the waning days of the Biden administration.   Endpoints here.

Draft guidance for AI in medical devices here.  See the FDA's own press release.

Draft guidance for tissue biopsies in clinical trials here.  Coverage at Genomeweb here.

Guidance on skin color and pulse oximeters here.

Guidance on accelerated approvals for drug trials that must be "underway."

Guidance on pre determined change plans (which may be important for FDA regulation of lab tests) here.

On a Personal Note: Best Books of 2024 and "AI History Tourism"

No Medicare or Genomics news from New Year's Eve...but...

I'm releasing my tenth-annual best books and films list, 2024 edition.

https://bqwebpage.blogspot.com/2025/01/best-books-and-films-of-2024.html

##

If of interest, a quirky use of AI paired with a history tour I took. 

I visited the museum for a 1920s  evangelist, Aimee Semple MacPherson, here in Los Angeles.  In her day, she was one of the best-known Americans.   Chat GPT helped me condense a short, informative essay that records my visit:

Short Essay (Public)

https://bqwebpage.blogspot.com/2025/01/aimee-semple-macpherson-chat-gpt.html

Research Session (A record of my research)

I said, Chat GPT "helped me condense."  

Condense what?  

For the curious, I'll leave a link to my lengthy Q&A with Chat GPT over 30 or 40 minutes as I was putting together my thoughts.   It shows be chatting with AI in peak "nerd mode," but it became the “input” that created the shorter blog above.

https://bqwebpage.blogspot.com/2025/01/aimee-semple-macpherson-los-angeles.html