DOGRA CMAA

After reviewing the CMAA proposal alongside the Dogra et al. analysis, I’d say:

Where CMAA Helps (in a narrow sense)

• Recognition of algorithmic services without physician work. CMAA is designed for cases where an algorithm produces a clinically meaningful output (risk score, index, probability) but no physician interpretation is embedded. That’s a real coding gap today, especially for “autonomous” AI outputs.

• Descriptor structure. CMAA descriptors force applicants to specify discipline, inputs, result type, and Appendix S category. That helps standardize how algorithmic services are presented, and at least acknowledges the augmentative vs. autonomous taxonomy.

• FDA tie-in. Requiring FDA authorization and linking codes to specific proprietary algorithms (via Appendix X registry) ensures these aren’t speculative codes.

Where CMAA Needs Work (relative to Dogra’s insights)

• Granularity of value capture. Dogra stresses that generalist AI (radiology, genomics, digital pathology) creates value across multiple tasks, modalities, and autonomy levels — and evolves under FDA’s PCCP framework. 

• CMAA, by contrast, is still static and siloed: one code = one algorithm = one output. That doesn’t solve the problem of dynamic AI that shifts roles and autonomy over time.

• Valuation mechanics. Dogra highlights the fundamental problem that CMS’s practice expense model can’t handle SaaS costs (licenses, cloud, maintenance). 

• Like other AMA codes, CMAA codes don’t tackle valuation at all; they just create a home for descriptors. That leaves the RVU and PE assignment problem entirely untouched, and an issue for CMS and other payors.

• Coverage and payment relativity. Dogra underscores that generalist AI needs composite RVU strategies, cross-setting consistency, and outcomes-linked payment options. 

• CMAA is silent as it’s a coding container, not a reimbursement framework.

• Adaptability and PCCPs. Dogra emphasizes the need for codes that remain valid as algorithms adapt under FDA’s predetermined change control plans. 

• CMAA, however, could require new codes for material changes (inputs added, autonomy level shifts, etc.). 

  That reintroduces the “coding whack-a-mole” problem Dogra is trying to get us past.

• Risk-based models. Dogra asks how SaaS/AI integrates into APMs and chronic disease management. CMAA codes don’t touch this; they are still FFS-centric artifacts.

My Take

CMAA is more a taxonomy exercise than a solution to the real reimbursement issues. It might help with cataloging FDA-cleared algorithms that don’t fit Cat I or III.  But others will  need to the ball on valuation, relativity, coverage, or adaptability.  

---

   

AMA PLA and TAB 94

 In the past four years, there have been several examples of DIGITAL PATHOLOGY codes (like whole slide imaging and ML-AI on either H&E or IHC) coming through the PLA system.

However, in 2025, no more WSI-PLA codes have appeared.   As of August 2025, AMA is offering stakeholders a chance to download documents for "TAB 94" fo the September 2025 AMA CPT meeting.  This will discuss major and important proposed changes for digital pathology, and concurrently for other types of AI such as in Radiology and advanced imaging.

This documents outlines some of the key issues, while respecting the AMA rules about confidentiality of new proposals (which AMA lets you download only after disclosure agreements).

ACTION POINT "TAB 94"

Go to AMA CPT website for CPT editorial meetings.

https://www.ama-assn.org/member-benefits/events/cpt-editorial-panel-meeting

See the AGENDA for September 2025 meeting.

https://www.ama-assn.org/system/files/cpt-panel-september-2025-agenda.pdf

https://cptsmartapp.ama-assn.org/ipdashboard

At the AMA CPT SMART APP, be sure to click on top for the INTERESTED PARTY PORTAL.

Now scroll down to the agenda list and click forward a couple screens to see TAB 94.

Go to TAB 94 and request permission to view those documents. When I did so, the web site granted me instant access to 2 documents stored under Tab 94 "Clinically meaningful algorithmic analyses" CMAA.

ORIENTATION TO THE AMA CPT ISSUES

THE PLA CODING WORLD

PLA codes go back to about 2016.   

PLA codes stem from a PAMA 2014 requirement that CMS make rapidly quarterly codes for ADLT tests.  AMA decided rapidly that it, the AMA, wanted to create these codes.  Since regular CPT codes take a year, and these codes must be quarterly, AMA had to craft new rules and processes for PLA codes.

AMA decided to take any proprietary lab test, since AMA couldn't make ADLT policy decisions.  This gives us the broad and unorganized quarterly coding system for PLA codes we have today.  By mid 2025, there are 600 PLA codes.

PLA CODES MEET WHOLE SLIDE IMAGING

One of the first whole slide imaging WSI codes was 0108U, TissueCypher, Cernostics.  Later another WSI PLA code, 0261U (HalioScore, acquired by Veracyte) was crosswalked to 0108U.   

Then, 0108U was acquired by Castle and became an ADLT, which also doubled the price of 0261U (due to the just-mentioned pricing crosswalk).   But later, CMS restored the price of 0261U back downward to its original crosswalk price.  Whew.

WSI CODES (BEFORE THEY STOPPED)

Existing slide-based PLA codes include 0108U, 0220U, 0261U, 0295U, 0376U, 0414U, 0418U, 0512U, 0513U.  513U was created a year ago, July 1, 2024, and effective since October 1, 2024.  All the codes 376U-513U are $706 at CMS.

Note that CMS classified the WSI codes as CLFS tests and priced them by CLFS methods.

The last WSI code, I think, was July 2024.   

Several WSI since then have been applied for, but all "died" in the PLA committee and no WSI codes have been created this year.

WHAT'S UP IN 2025

AMA is planning to create a new category of "dry lab" codes for algorithms.  In this proposal, one new CPT section will hold cardiology, radiology, or pathology codes of this type.  Indeed, this topic is Agenda Tab 94 at the September 18-20 AMA CPT meeting in Chicago.

TALK BY ZACH HOCHSTETLER

At the NextGenDx conference in DC on August 18, 2025,, Zach Hochstetler of AMA CPT discussed what is going on.   You did not need to sign any AMA confidentiality agreement to hear his talk. 

A DEEPER EDUCATION IN WHAT'S GOING ON

APPENDIX S

https://www.ama-assn.org/practice-management/cpt/cpt-appendix-s-ai-taxonomy-medical-services-procedures

A number of years ago, AMA CPT created a digital medicine coding committee.   

https://www.ama-assn.org/practice-management/digital-health/digital-medicine-payment-advisory-group

https://www.acr.org/News-and-Publications/ama-cpt-panel-considers-ai-related-codes

I think its output value has been variable.  It created Appendix S of the CPT book, which is a one page guide to digital coding plans.   Appendix S defines digital services as ASSISTIVE (detects or flags certain cases only).   AUGMENTED provides data that contributes to diagnosis or action but requires physician interpretation.   For example HEARTFLOW is radiology software that performs very sophisticated math but requires a human to finalize.  AUTONOMOUS - provides a final result that does not require physician interpretation.   AMA example is a retinal scan that outputs a retinal diagnosis.  I think most MAAA tests would be here ("Based on 21 RNA, your Genomic Health score is 15.")

AMA CONCERNS (in my view)

• AMA wants to stay up to date and relevant in an AI world.  AMA needs to innovate the coding system for new services.
• AMA wants to avoid having too much flood into the CPT as "physician services" because they deflates/devalues the RVU system by adding RVUs.  A code word here "sustainability" = avoid making new RVU-based services that will deflate the RVU system and make it unsustainable.

WHAT's UP FOR SEPTEMBER (TAB 94)

The Digital Medicine Payment Advisory Committee DMPAG proposes an entirely new section of the CPT.  It will hold algorithmic codes of all types, such as cardiology, radiology, and lab medicine.   They will be proprietary like PLA.  They will have minimal review like AMA.

Amer Coll Radiol writes,

The Panel’s Digital Medicine Coding Committee (DMPAG) is considering a foundational coding solution to develop a framework for a new category of CPT codes under the Clinically Meaningful Algorithmic Analyses pathway [CMAA].

The new CPT category would codify algorithmic analyses of clinically relevant patient data (e.g., biophysical signals, imaging data, lab results) to produce clinically meaningful output or conclusions that impact patient care.

Services in CMAA must be FDA authorized.

ISSUES FOR LAB INDUSTRY

1. Are we happy with PLA as it is?  Or not?
2. If AI Digital Pathology tests are CLIA LDT tests, why put some in PLA and some in CMAA?
3. Will payers view proposed CMAA as a payable category at all?
4. How do AMA policy leaders view AI codes coming into the RVU system?
5. How should the lab industry respond to the AMA's proposed requirement to put AI or digital pathology lab codes in CMAA category (where they must be FDA approved) and simultaneously disallow them in the PLA category (where they were placed in the past)..
6. Does the lab industry want digital pathology codes placed in the new CMAA category- along with radiology and cardiology tests?   Or does the lab industry want to avoid that, because the lab industry has unique issues (such as CLIA rather than FDA authorization).

STAKEHOLDERS

Both special companies (e.g. Artera AI) and large labs (e.g. TEMPUS) are working on digital pathology.

Association stakeholders might include (e.g.) ACLA, Adavmed, even PHRMA (for companion diagnostics access).

SEPTEMBER

One can go to the AMA CPT website, download the agenda for September 2025, and get a copy of the two documents filed as Tab 94.   

NERD NOTE

The CMAA says the tests may be either "Augmentive" or "Autonymous."  However, AMA also says that CMAA tests do not require physician interpretation, while Augmentive category does require physician interpretation (see Appendix S and Tab 94)..

##

##

##

See a review of 174 radiology software products.

https://www.linkedin.com/posts/kickyvanleeuwen_ai-in-radiology-173-products-and-evidence-ugcPost-7355517722912333828-CCJ1/ 

See a review of 1247 FDA Ai medical devices.

https://www.linkedin.com/posts/campbell-arnold_1247-fda-cleared-ai-medical-devices-and-activity-7361385023934980097-3N1D/  

Accoding to NY Times, new CMS plans will reduce the AMA's input to RVU levels.

https://www.nytimes.com/2025/07/21/upshot/primary-care-doctors-salaries.html?

##

Blog discusses the rising and falling price of code 0261U.

https://www.discoveriesinhealthpolicy.com/2023/06/and-then-miracle-occurs-veracyte-test.html

MolDx Clinical Lab Scientists Palmetto

 Summary

 

Provide molecular diagnostic expertise to suggest improvements on the standardization of processes for provider coverage determinations. Focus is reviewing molecular, genetic, and esoteric tests for unique characteristics. Provide guidance on updates and improvements to identifiers and development of processes.

Description

 

Logistics:

• This position is full-time work from home.

• You must have high-speed (non-satellite) internet service and a private home office to work from home.

What You'll Do:

• Reviewing/assigning/maintaining unique identifiers for registered laboratory tests and associated processes.  Including validating test registrations include industry-accepted laboratory terminology and are compatible with the data model input requirements for inclusion in the clinical catalog. 

• Compare new test registrations to currently identified tests and assign new identifiers as needed. Review currently identified items for accuracy and utility in the clinical catalog. Respond appropriately to test registration and associated questions. 

• Research emerging technologies such as next-generation sequencing of the human genome, micro-fluidics, and nanotechnology, point of care molecular diagnostics testing, array comparative genomic hybridization (aCGH), etc.

• Make recommendations to Management regarding emerging technologies regarding strategy, ramifications, and incorporation of said technologies into various Advanced Diagnostic applications; market acceptance of the technologies; etc.

• Other duties as assigned.

To Qualify for This Position, You'll Need:

• Required Education: Bachelor's degree in a scientific-related field OR Four years of additional job-related work experience in molecular lab, or two years of job-related experience in a molecular lab plus a technical degree and related certifications in medical technology.

• Required Work Experience: Five years' experience in a molecular lab plus two years' experience in a clinical lab (may be concurrent)

• Required Skills and Abilities: Experience in the area of molecular diagnostics, genetics, genomics, or related field. Excellent verbal and communication skills, including the ability to communicate technical information to all levels of staff, management, and end-users.

• Required Software and Tools: Microsoft Office.

What We Prefer You Have:

• Three years of experience in clinical molecular or genetics lab.

• Experience running & creating clinical lab test.

• Experience developing new tests and/or test procedures.

What We Can Do for You:

We understand the value of a diverse and inclusive workplace and strive to be an employer where employees across all spectrums have the opportunity to develop their skills, advance their careers and contribute their unique abilities to the growth of our company.

Our comprehensive benefits package includes:

• 401(k) retirement savings plan with company match

• Fantastic health plans and free vision coverage

• Life insurance

• Paid annual leave — the longer you work here, the more you earn

• Nine paid holidays

• On-site cafeterias and fitness centers in significant locations

• Wellness programs and a healthy lifestyle premium discount

• Tuition assistance

• Service Recognition

What to Expect Next:

After submitting your application, our recruiting team members will review your resume to ensure you meet the qualifications. This may include a brief telephone interview or email communication with our recruiter to verify resume specifics and salary requirements. Management will conduct interviews with the most qualified candidates

Esophageal Tests (AI)

 Chat GPT 5 (it used its slow answer mode, 3 min 40 sec)

Seveal genomics companies in the US make LDT genomic tests for esophageal dysplasia prognosis or severity. What are they. They tend to have limited insurance coverage.

##  This is a direct AI answer.  Do note quote without verifying.

##

Here are the main U.S. players with LDTs aimed at Barrett’s esophagus (BE) / esophageal dysplasia risk or severity—and, as you noted, most have thin/patchy coverage:

• Castle Biosciences — TissueCypher® Barrett’s Esophagus test (PLA 0108U). An AI-driven tissue systems pathology assay on FFPE biopsies to predict progression risk (NDBE/IND/LGD → HGD/EAC). It has CMS ADLT status (payment category) but coverage remains inconsistent across payers. (ir.castlebiosciences.com, Nasdaq)

• Previse (Capsulomics) — ESOPREDICT® (PLA 0398U). A DNA methylation classifier on FFPE biopsies that reports risk of progression to HGD/EAC; company was acquired by Castle in 2025. Many commercial policies still list it as investigational/non-covered. (AAPC, Previse, BCBS Michigan)

• Lucid Diagnostics — EsoGuard® (PLA 0114U). A non-endoscopic esophageal DNA methylation test (balloon collection + methylation panel) for detection across the BE–dysplasia–EAC spectrum; commercial coverage just began at a regional BCBS plan (Mar 2025), and MolDX has scheduled a multi-MAC CAC meeting on Sep 4, 2025 for LCD reconsideration (i.e., no MolDX coverage yet). (PR Newswire, Lucid Diagnostic Investor Relations)

• Interpace Diagnostics — BarreGEN®. A genomic instability / mutational load assay (PathFinderTG platform) to stratify BE progression risk. Interpace announced it stopped accepting BarreGEN specimens May 30, 2025; historically coverage was very limited. (Interpace Biosciences®, Inc., barregen.com)

• Cytogenetic “genomic” adjuncts (FISH-based) from reference labs — e.g., NeoGenomics’ NeoSITE™ BE FISH and Genesis/Pathnostics BE FISH panels (probes such as MYC, p16/CDKN2A, HER2, 20q13), marketed to help identify higher-risk BE from brushings or biopsy. These are available but typically not covered as risk classifiers by major plans. (NeoGenomics, Inc., Genesis Labs)

Why coverage is “limited”

• MolDX LCD L39256 (Upper GI molecular testing) remains a non-coverage policy, now under reconsideration (CAC 9/4/2025). Until/unless that changes, Medicare coverage in MolDX regions is generally no for molecular BE detection/risk tests like EsoGuard/ESOPREDICT. (CMS, Palmetto GBA)

• Multiple commercial payers explicitly classify these tests (TissueCypher, ESOPREDICT, EsoGuard) as investigational/not medically necessary in current policies. (providencehealthplan.com, BlueShieldCA, Blue Cross Blue Shield of Rhode Island)

If you want, I can map who covers what by MAC/plan (MolDX vs. NGS/JN/JK/JL, plus top commercial policies) and tie each test to its current code & price point so you’ve got an at-a-glance grid for payer conversations.

###

###

I think the EsoGUARD company LUCID makes a device that is FDA cleared and a LDT test that is not FDA reviewed.   Can you pull the names of each.

You're absolutely right about the dual nature of Lucid Diagnostics' offerings. Here's the breakdown:

---

Names of Each Product

1. FDA-cleared device

• EsoCheck™ Cell Collection Device: This is a swallowable balloon capsule catheter used to collect surface cells from the esophagus during a brief, non‑endoscopic office procedure. It is FDA 510(k) cleared for the collection and retrieval of surface esophageal cells in adults (and adolescents aged 12 and older, per updated indications) (sec.gov).

2. Laboratory-Developed Test (LDT)

• EsoGuard® Esophageal DNA Test: This is an LDT performed on the samples collected by EsoCheck. It is a bisulfite-converted next‑generation sequencing (NGS) DNA methylation assay, analyzing 31 methylation sites across two genes (VIM and CCNA1). It is not FDA reviewed or approved, instead commercialized under CLIA/CAP certification as a Laboratory Developed Test (sec.gov).

---

Summary Table

Product Name	Type	FDA Status
EsoCheck™	Cell Collection Device	FDA 510(k) cleared
EsoGuard®	DNA Methylation Assay (LDT)	LDT (no FDA review)

---

So indeed, the EsoCheck™ device is FDA-cleared, while the EsoGuard® test is offered as an LDT under regulatory enforcement discretion.

 

The Raw Numbers You Need for MRD Statistics

My manual answer at top.  An AI re-written answer at bottom.

##

Dear BRUCE,

How many patients do you need for a successful MRD study at Moldx?

##

Dear (JOHN),

Good questions.  This area is confusing to many people, but I think I've developed a good sense of what MOLDX is looking for, and some insights into the "moats" which are bigger than one might guess.   Total population size (55, 79, 101) is not the only driver.

01 LCD

Start with the LCD.  Basically, you can get coverage if you have data for:

1) Post op check for completeness.  You have a breast or colon resection, and at 6 weeks you check for ongoing tumor DNA.

2) If the patient is negative post op, surveillance, e.g. quarterly for several years.    For small resected cancers the recurrence rate is likely around 10% per year.

3) Drug response  Serial testing during drug therapy (esp. immuno-oncology).  Rising DNA is bad, low DNA is good.

The LCD says that the MRD test must detect earlier than a standard of care test, and, with at least similar accuracy.   Earlier doesn't mean 2 days; it's not defined but is 2+ months, often 4 months or so.

02   Comparison for mental framing

Before going into MRD, consider colon cancer screening.   These are usually 10,000 patient studies.  The reason is that the cancer rate is around 1%, and FDA wants you to accumulate circa 100 cases (60+ let's say).   This is because they want precise statistics around sensitivity and specificity of the test (assuming a colonoscopy gold standard).   They want to know that sensitivity is (say) 80% plus or minus 2%, not 80% plus or minus 15% due to small numbers and noise.

With colon screening, the pivotal data is the 100 cancer patients, not the 9900 excessive no-cancer patients.  Similar for MRD, the "relapsed" headcount is key to the performance statistics, and you might need 200 MRD patients to get 20 relapses.

03  Data for MRD success

Typically, you want something like 15 or more relapsed patients.   If the relapse rate is 10% per year, then you need 150 patients for a 10% relapse of 15 patients in the first year.   

Also, you want reasonable error bars around the sensitivity.   That means you need enough relapsed cases (say, maybe 15, but more is better).  And, crucially, you need fairly frequent checks against a gold standard (say, CT).   Let's say you want 5-6 blood tests and 4-5 or more CT tests over the observation period.   

If you have 1 blood test and 1 CT test a year later, it's a statisical disaster as there's no way to say the lead time for detection is, say, 3 months plus or minus one month with so little data.

So:  #1 Total number of patients, and #2 Relapse patients (e.g. at a 10% rate), and #3 frequency of blood sampling, AND ALSO, #4 frequency of standard of care CT test.   This is why data is often summarized as "125 patients and 550 total blood samples".   They are flagging there were about 5 blood samples per patient.  If you are missing any one of those parameters (#1 #2 #3 #4) the statistics flop.

04  MOAT

The moat is wider than you might guess.  If you use NEW data, you must start the trial, enroll patients, and follow up with minimal study loss over two years.  While getting them back for serial testing every couple months going into a second year in the future.   Then allow another year for publication and moldx review.   

Let's say you use BIOARCHIVE data.  Then you are at the mercy of how many patients the archive collected, how many frozen plasma samples, how many timed and matched CT scans for the gold standard.   

05 A GOOD EXAMPLE

MolDx recently approved a brast cancer MRD which is PATHLIGHT from SAGA DIAGNOSTICS (Univ. Lund).

https://www.discoveriesinhealthpolicy.com/2025/08/moldx-greenlights-pathlight-new-mrd.html

See the article by ELLIOTT 2025, which summarizes all their data in one place.   

See also the ASCO announcement from NATERA that they had perfected a genome-based MRD test.  i don't have it in front of me, but i think they maybe had something like 500 patients and 3000 blood samples.   That second number contributes to the precision of the statistical data of the MRD test.

##

##

##

AI RE WRITE

Here’s a polished rewrite in a more professional style while keeping the structure and logic intact:

---

The Raw Numbers Behind MRD Study Design

Dear (John),

You’ve raised an excellent question. This area can be confusing, but over time I’ve developed a clear sense of what MolDx expects for minimal residual disease (MRD) coverage decisions. Importantly, the barriers—or “moats”—to entry are broader and deeper than they first appear. Patient sample size (e.g., 55, 79, or 101) is not the sole determinant of success.

---

1. Coverage Criteria from the LCD

MolDx’s local coverage determination (LCD) outlines three main use cases:

1. Postoperative clearance – e.g., testing for circulating tumor DNA six weeks after resection of breast or colon cancer.

2. Surveillance in MRD-negative patients – serial testing (e.g., quarterly) to monitor for recurrence, with recurrence rates in small resected cancers often around 10% per year.

3. Therapy response monitoring – repeated testing during treatment, particularly in immuno-oncology; declining DNA levels indicate response, rising levels suggest resistance.

The LCD requires that the MRD test detect recurrence earlier than standard-of-care methods, with comparable or superior accuracy. “Earlier” is not defined as a few days, but rather on the scale of 2–4 months ahead of conventional imaging.

---

2. Colon Cancer Screening as a Mental Model

A useful comparison is colon cancer screening trials. These require ~10,000 patients because cancer prevalence is ~1%. To evaluate sensitivity and specificity with narrow confidence intervals, FDA generally requires ~100 cancer cases. With fewer cases, confidence intervals become too wide (e.g., 80% sensitivity ±15% vs. ±2%).

The same principle applies to MRD: the number of relapse cases, not total enrollment, drives statistical precision. For example, with a 10% relapse rate, a study of 200 MRD patients may yield only 20 relapse events—just enough to anchor meaningful performance statistics.

---

3. Data Elements for MRD Success

A credible MRD study design typically requires:

• Adequate relapsed patients: ~15 as a minimum, ideally more. With a 10% annual relapse rate, 150 patients are needed to achieve 15 events in year one.

• Reasonable error bounds: sufficient relapse cases are necessary to narrow sensitivity and specificity estimates.

• Frequent blood sampling: ideally 5–6 blood draws per patient over the study period.

• Frequent gold-standard imaging: 4–5 CT scans (or equivalent) to enable valid comparisons.

Without this frequency, the statistics collapse. For example, one blood draw and one CT scan a year later cannot support meaningful lead-time estimates. This is why results are often reported as “125 patients with 550 blood draws”—highlighting the density of longitudinal data as much as cohort size.

---

4. Why the “Moat” is Wider Than It Appears

Two common approaches illustrate the barriers:

• Prospective trials: require patient enrollment, repeated blood draws and imaging over two years, high retention, and then an additional year for publication and MolDx review.

• Use of biobanked specimens: depends on what the archive actually contains (number of patients, matched plasma, and contemporaneous CT scans). Missing elements can cripple statistical validity.

Thus, both pathways demand substantial time, resources, and infrastructure—much more than a simple headcount might suggest.

---

5. Examples

• Saga Diagnostics (PathLight): Recently, MolDx approved a breast cancer MRD assay from Saga Diagnostics (University of Lund). See Elliott, 2025 for a concise summary of their dataset.

• Natera: In its ASCO announcement, Natera highlighted development of a genome-based MRD test supported by ~500 patients and ~3,000 blood samples. The large number of serial blood draws directly strengthens the precision of their statistical claims.

---

In summary: Success depends not only on total patient numbers but also on relapse-event counts, sampling frequency, and the robustness of standard-of-care comparisons. Missing any of these dimensions risks undermining the statistical case for coverage.

---

Would you like me to also condense this into a one-paragraph “payor-facing” version (like an executive summary) so you could drop it into decks or client communications?

Patent Moats in Cardio Analysis

 The following is direct AI output (Chat GPT 5) and SHOULD NOT BE TAKEN LITERALLY.

It is provided NOT AS FACT but only as an example of UNEDITED AI reasoning.

###

The potential for CT scanner manufacturers (Siemens, GE, Philips, Canon, etc.) to embed FFR-like functionality directly into their imaging workstations represents a serious strategic and IP-adjacent threat to HeartFlow’s business model, especially in the context of Medicare billing via CPT code 75580 (FFRCT).

Below is an in-depth discussion of the threat landscape, reimbursement dynamics, and patent workarounds relevant to this issue.

---

💣 Threat Landscape: OEMs Embedding Local FFR Estimation

What’s happening?

Companies like Siemens Healthineers and Canon Medical have reportedly been exploring on-device or local workstation-based blood flow analysis tools that can extract hemodynamic data from CCTA scans—just like HeartFlow’s FFRCT—but without cloud-based modeling or HeartFlow’s proprietary AI.

These tools aim to:

• Reduce turnaround time (point-of-care analysis)

• Eliminate external data transfer to HeartFlow’s servers

• Bypass HeartFlow’s platform entirely—and associated per-use fees

This approach would allow radiology or cardiology groups to retain control of the entire workflow, from imaging to billing, potentially capturing the ~$900 Medicare fee under CPT 75580.

---

💸 Billing & Reimbursement Advantage for OEM-Based FFR

• CPT 75580 is a public, nationally priced code (approx. $900 under Medicare Part B) for FFRCT derived from CCTA.

• The code is carrier-priced but generally well-accepted.

• Radiology groups using embedded tools could bill 75580 directly, capturing the full technical and professional fee.

🟡 HeartFlow’s model, by contrast, is dependent on:

• Image upload to cloud servers

• Software-as-a-service pricing (per-case)

• Third-party positioning (HeartFlow does not own the scanner or patient relationship)

Thus, any validated in-scanner solution threatens HeartFlow’s revenue per case, even if technically inferior.

---

🧬 Patent Moat and Possible Workarounds

HeartFlow has hundreds of patents covering:

• CFD (computational fluid dynamics) modeling of coronary blood flow

• AI-driven 3D reconstruction of coronary anatomy

• Fusion of anatomic and physiologic data

• Specific algorithms for FFR estimation from CCTA

However, scanner manufacturers might avoid infringement by:

• Using machine learning–based surrogate models that estimate pressure drop without explicit CFD simulation

• Building simplified geometric flow approximations not covered by HeartFlow’s patents

• Avoiding key claims like "fractional flow reserve" and instead offering alternative metrics (e.g., “relative pressure index” or “contrast transit times”)

• Bypassing patented cloud workflows with on-premise execution

These tactics might allow “FFR-equivalent” results that are:

• Inferior, but “good enough” for many clinical decisions

• Positioned as workflow-integrated tools (e.g., Philips IntelliSpace or Canon’s work-in-progress AI modules)

• Covered by Medicare under existing codes like 75580, assuming FDA clearance

---

🔍 Regulatory & Software Risk Factors

HeartFlow’s own disclosures note:

• Software defects have occurred, some of which were reported to the FDA

• Reliability risks exist in a clinical context

• Radiologists may prefer vendor-native tools with full integration into PACS and scanner software

Even if scanner-based tools are less accurate, their workflow simplicity, faster turnaround, and direct billing advantage could tempt buyers away from HeartFlow.

---

⚖️ Strategic Summary for Investors

Factor	HeartFlow	OEMs with Embedded FFR
Accuracy	High (peer-reviewed, guideline-backed)	Likely lower, less validated
Workflow	Cloud-based, offsite	Fully embedded, point-of-care
Billing	HeartFlow licenses or partners with providers who bill 75580	Radiology group bills 75580 directly
Moat	Deep patent portfolio, guideline status	Potential to work around patents
Risk	Software defect history, dependence on external adoption	Must secure FDA clearance and validation

---

🔐 Conclusion: A Defensible But Not Impregnable Moat

HeartFlow’s moat is strong—grounded in clinical validation, guideline integration, and a robust patent portfolio—but it is not immune to disruption from:

• Workstation-native solutions

• Alternative algorithms avoiding core patent claims

• Reimbursement capture by imaging providers

To mitigate this, HeartFlow must:

• Accelerate adoption of Plaque Analysis and Roadmap tools to diversify revenue

• Continue lobbying to ensure CPT code 75580 stays linked to best-validated platforms

• Expand into OEM partnerships or offer white-label versions of its platform

• Maintain a clear clinical superiority narrative and long-term outcomes data

Would you like a visual risk matrix or a side-by-side valuation sensitivity model based on possible market share erosion from embedded FFR entrants?

###

How might a company engineer around Heartflow's ffR patents? How would they approach such a strategy and build it out?

###

Engineering around HeartFlow’s FFRCT patents is a technically and strategically sophisticated process. It would likely be attempted by large imaging OEMs (like Siemens, GE, Philips, or Canon) or well-funded startups. Below is an outline of how such a company might approach the strategy, and the methods they would use to reduce infringement risk while developing a viable commercial alternative.

---

🔬 Step-by-Step Strategy to Engineer Around HeartFlow's FFRCT Patents

1. Patent Landscape Analysis

• Conduct an exhaustive review of HeartFlow’s granted patents and pending applications (U.S. and globally), with a focus on:

  • Claims involving computational fluid dynamics (CFD)

  • Claims involving 3D modeling and mesh creation

  • Claims involving cloud-based data transfer and processing

  • Claims that reference "fractional flow reserve" specifically

• Use IP analytics firms to identify:

  • Expiration dates

  • Jurisdictional differences

  • Claim scope

  • Opportunities for design-arounds or claim re-interpretation

---

🧠 Technical Methods to Avoid Direct Infringement

A. Avoiding CFD

HeartFlow’s key early patents rely heavily on reduced-order CFD modeling of coronary blood flow. A competitor could avoid this by using:

• Machine learning surrogate models trained on historical CFD results

• Empirical flow estimations based on statistical modeling of geometry, vessel tapering, and contrast attenuation

• Simplified physics approximations (e.g., pressure drop using Poiseuille’s law or lumped-parameter models)

This allows software to generate FFR-like outputs without literally running CFD, thereby dodging the specific language in HeartFlow’s claims.

---

B. Skipping "FFR" Terminology

Instead of computing or marketing results as “FFR” (which may be referenced in patent claims or raise FDA scrutiny), alternative outputs could include:

• "Relative perfusion index"

• "Virtual ischemia score"

• "Contrast-based flow index"
  These avoid the need to invoke HeartFlow's core term while still providing clinical value.

---

C. On-Scanner, Local Processing

HeartFlow’s workflow is heavily cloud-centric, including:

• Uploading raw CCTA DICOM files

• Server-side segmentation and modeling

• Return of processed images and reports

Competitors could avoid infringement by offering:

• On-device, workstation-based tools embedded in the CT scanner’s image post-processing software

• Edge computing implementations to reduce cloud dependency

• Fully offline workflows that do not mimic the sequence of HeartFlow’s SaaS process

---

D. Anatomical vs Physiological Modeling

If HeartFlow’s core patents cover the fusion of anatomic and physiologic data, a workaround could:

• Focus solely on anatomic risk scoring (like plaque volume, curvature, bifurcation index)

• Or use contrast timing analysis (like CT perfusion) rather than simulated flow modeling

---

⚖️ Regulatory Strategy

Companies trying to compete with HeartFlow must still navigate the FDA:

• HeartFlow's products are De Novo cleared with strong clinical validation

• A workaround tool would need:

  • Either 510(k) clearance with a new predicate

  • Or its own De Novo petition if substantially different

• Labeling the tool as “adjunctive” rather than diagnostic may reduce regulatory burden

---

💰 Business Strategy Considerations

Reimbursement Targeting

• CPT 75580 is modality-agnostic: it pays for "FFRCT derived from CT angiography" regardless of specific vendor

• A competitor would build a clinical case to justify:

  • Use of CPT 75580 for their version of "FFRCT" or surrogate

  • Or propose a new CPT code or CMS review if the product is different enough

Bundling with Imaging Hardware

• Imaging OEMs could bundle the tool with new CT scanners to gain early traction

• Or license a third-party tool, rebranded, to offer a native-like experience

---

📊 Realistic Outcome Scenarios

Strategy	Legal Risk	Clinical Risk	Reimbursement Potential	Feasibility
Use ML surrogate model	Medium	Moderate	High if labeled as FFRCT	High
On-scanner flow analysis	Low (patent)	Medium	High	Medium
Contrast perfusion timing	Very Low	High (less validated)	Lower	High
License expired/expiring patents	Low	Low	High	Long-term only

---

🧩 Final Thoughts

HeartFlow’s patents are strong, but they are not unbreachable. A well-capitalized competitor could:

• Work around specific claims (e.g., CFD, cloud usage, "FFR" terminology)

• Offer simplified or alternate metrics

• Bundle directly into CT workflow

• Capture the full CPT 75580 fee without paying a per-use license to HeartFlow

For investors or strategic observers, it’s critical to track:

• OEM software roadmap announcements

• FDA 510(k) filings referencing “CT-derived hemodynamic metrics”

• Patent expirations or lawsuits involving workarounds

Would you like me to locate and review a few recent FDA 510(k) summaries or pre-market filings for competitive FFR-like tools?

###

 The above is direct AI output (Chat GPT 5) and SHOULD NOT BE TAKEN LITERALLY.

It is provided NOT AS FACT but only as an example of UNEDITED AI reasoning.

Roger Klein dot com

 https://rogerdklein.com/

8 4 2025

https://www.foxnews.com/video/6161670175001#sp=show-clips