Monday, April 28, 2025

Stat Plus Runs Detailed Profile of Chris Klomp, New Head, Center for Medicare

 

Kudos to Mario Aguilar, journalist at STAT PLUS, for his very detailed profile of Chris Klomp, the new head of the Center for Medicare.

##

When thinking of CMS, in the lab industry we think about NCDs and we think about the Clinical Laboratory Fee Schedule.   

  • The NCD group is part of the Office of Clinical Standards and Quality, where the Coverage and Analysis ("NCD") group report to the CMS Chief Medical Office.   
  • For the CLFS - and all sorts of fee schedules and operational rules - pivot to the Center for Medicare.   

Chris Klomp replaced the prior Center for Medicare chief, Meena Seshamani, on April 21, 2025.

##

Per the STAT PLUS article, Klomp was CEO of health IT company "Collective Medical," which raised $50M before its exit by acquisition.   (He thereafter had engagements such as board seats for Maven Clinic and Nomi Health.)  

Per STAT, Klomp's former associates and other experts say he has big plans for his role, both on the Medicare Advantage side and on the fee for service side.   As head of Center for Medicare, he will report to Dr. Oz, who is head of CMS as a whole.

##

I've asked Chat GPT for a paragraph of two on Klomp.

Chris Klomp has recently been appointed as the Deputy Administrator and Director of the Center for Medicare at the Centers for Medicare & Medicaid Services (CMS), reporting directly to Administrator Dr. Mehmet Oz. In this role, Klomp oversees the Medicare program, which provides health coverage to approximately 68 million Americans and manages an annual budget of around $1 trillion.​

Prior to his CMS appointment, Klomp was the CEO of Collective Medical, a healthcare technology company focused on real-time care coordination, which was acquired by PointClickCare in 2020 . He also served as a senior advisor at PointClickCare and he has held board positions at organizations such as Nomi Health, Maven Clinic, and InnovaCare Health. 

Background.  Klomp's background includes experience in private equity at Bain Capital and consulting at Bain & Company. He holds a BA in Economics and English from Brigham Young University and an MBA from Stanford Graduate School of Business, where he was an Arjay Miller Scholar .​

Chat GPT summarizes several public sources in the business press as: 

Chris Klomp brings a leadership style deeply rooted in data-driven innovation and a commitment to improving care for vulnerable populations. As the former CEO of Collective Medical, he emphasized the importance of real-time care coordination to enhance patient outcomes. His approach at Collective Medical focused on integrating data to close information gaps across the healthcare continuum, particularly benefiting high-risk patients.  Given his background, Klomp may aim to modernize the program through better data sharing and technological advancements.


##

A sidebar to the Stat article cites CBO data that Medicare is now solvent to 2052

 

CMS Releases 2025 Gapfill Prices for Comment; MAC Pricing Ratios

Header - CMS has released gapfill proposed prices for about 34 codes.   These are codes that failed to be priced in Fall 2024 by crosswalk, so they are transferred to the MACs for pricing during 2025.  Comment will be open for 60 days (June 28).  

The Monday April 28 version had typos; the Tuesday April 29 version corrects them, according to a CMS email.

##

Find it here:

https://www.cms.gov/medicare/payment/fee-schedules/clinical-laboratory-fee-schedule-clfs/annual-public-meetings

On this webpage, scroll down to "Meeting Notice, Agenda, Other Materials" and then scroll down to "CLFS Gapfill Determinations" and then find "CY2025 preliminary Determinations."

Typos

These have been corrected as of 04/29, per CMS email.

A Couple Views

Some of the prices were not too surprising.  

  • For example, 0485U, the CARIS ASSURE test which includes CHIP testing of WBCs, has a MolDx price of $3649, the range for complex LBX comprehensive tests.   
    • However, it might also suggest there's not too much financial bonus for adding workflow for CHIP testing, something that may be increasingly important (Caris/Magee 2025).  
    • Prices are subject to revision in September after the comment period.   
  • 0426U, an ultra-rapid genome for pediatric cases, comes in at $7582, the value CMS has used before for a regular rapid genome (0094U).
Better Formatting

CMS formatted the spreadsheet with full code names next to the code number, which is helpful.  On the other hand, the spreadsheet is "locked," so you have to copy paste into a new Excel to manipulate it.

Rationales

Tab 2 is "rationales."  For every code, Novitas has the same rationale (that it used a combination of reference data to price).   

All the other columns of rationales are the same, for each row, but vary, between rows.  These appear to be MolDx rationales, which were copied (like MolDx prices) by NGS MAC.

Differences Among MACs

The MolDx price is always the same for its participating MACs and drives the CMS Median.

NGS MAC prices (e.g. New York State) always match the MolDx price (as do its rationales).   

Novitas/FCSO prices do not usually match MolDx, but are usually within a low multiple.   Sometimes the multiple is 2.2X (0486U) 3X (0440U) or as high as 11X (0441U).  Rarely the Novitas price is half the MolDx price (0498U, 0501U).   

I've uploaded an Excel to Google Docs open access, which shows the ratios of MAC pricing to one another line by line.  Here
Screen shot of cloud excel of MAC ratios


Friday, April 25, 2025

NYTimes Says: New England Journal Gets "Vague Threatening" Prosecutor Letter

 On April 25, 2025, New York Times reported that New England Journal of Medicine received a letter described in the article as "vaguely threatening" - from a federal prosecutor.

https://www.nytimes.com/2025/04/25/health/nejm-prosecutor-letter.html


The news may have been first reported by Stat Plus:

https://www.statnews.com/2025/04/23/new-england-journal-of-medicine-us-attorney-letter-scientific-journals/

###

I made the point in March, that nearly every Op Ed in places like JAMA, NEJM, Annals of Internal Medicine,since November, has been negative about the administration (and by extension, the majority party on the Hill.)   Concurrently, AMA has gotten nowhere in its battle against the most dramatic price cut ever in physician Medicare pay levels (see hte same blog).   It seemed to me there was a dotted line between publishing 20 negative JAMA op eds in a row against Republicans, and then, surprise, AMA lobbyists get a cold shoulder on the Hill when they come asking for a pay raise to Republicans.   

 

Thursday, April 24, 2025

Novitas LCD for Oncology - Specific Tests - Is Finally "FINALED"

Completeing a several-year saga, the Novitas LCD for oncology testing is finally "finalized" - and effective as of today, April 24, 2025.

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

The take-home lesson is, it took Novitas three years (spring 2022-spring 2025, about 36 months) to non-cover several LDT tests, which could have been done, A to Z, in about six months.

Interpace to nix its Pancragen testing now.

#####

The LCD was released in January 2025 as "final,' but put "on hold" for the "Trump Administration to review" according to a press release by one company involved.

The LCD was started as a proposal in 2022, finalized in 2023.  But CMS put the radically-changed LCD on hold, and it entered a new comment period in summer 2023.   That versiin should have been "finalized" in summer 2024, but Novitas issued a delay notice.   One source commented that besides the LCD appeals, court actions had been filed along the way.

##

The LCD has some quirks.  First, it states that the term "genomic" shall not be used, because it is too confusing.  It uses the word "genetic" to describe any kind of DNA-RNA test.   

The LCD was proposed as a broad LCD with many topics, and extensive references to outside sources (like NCCN) for coverage and non-coverage.   The final LCD is vastly restricted in spoke and basically reviews about 10 named proprietary (branded) tests, many of which you've likely never heard of, and almost none of which are covered.

The authors write, for example, "Based on these factors, there is insufficient evidence to determine the clinical utility for DecisionDx-SCC."  (This test appears to be non-covered also under MolDx LCD L39583.)  However, the corresponding code is listed as an ADLT status (0315U) by CMS, part of which status requires a determination of coverage.

Tuesday, April 22, 2025

CMS Asks for Ways to Simplify Its Regulations

 This week, CMS releases a remarkable 3-page document requesting suggestions how Medicare can "unlease prosperity" by de-regulating its overburdened providers.

Throwing the 14-day rule overboard without a life jacket, comes to mind.   

Find the document at this link.  Suggestions until June 10, 2025.   AI discussion includes 14-day-rule and, at bottom, long needed revisions to oncology NCD 90.2.

https://www.cms.gov/files/document/unleashing-prosperity-through-deregulation-medicare-program-request-information.pdf

Consider in context of HHS's announced reorg and simplification of the entire HHS hierarchy of agencides, while downsizing from ~80,000 to ~60,000 staff.   Here.

Decisions on regulatory reform in areas like prior auth or 14 day rule will bubble up to Chris Klomp, the new head of Center for Medicare (here).

Though not specific to this RFI, see OMB on rules for deregulation here.

###

AI CORNER

###

For readers in medtech policy (and genomics) study this RFI and explain what it means, how likely it is to effect change, and what should stakeholders suggest.   What nuances will make the suggests rise up in the mass of responses this will generate.   For the lab industry, what should they suggest?   I can think of 14 day rule, or some bizarrely complicated and confusing LCD instructions.   Another simplification would be making the MolDx program national, allowing labs to submit claims under one clear rule set.    Have at it.  Thanks.

##

This new CMS Request for Information (RFI)—titled Unleashing Prosperity Through Deregulation of the Medicare Program—is a wide-open invitation for stakeholders to recommend simplifications, eliminations, or automations of Medicare regulatory requirements, under the banner of Executive Order 14192, which mandates that for every new rule issued, 10 older regulations must be repealed.

Below is an analysis tailored to medtech policy professionals, especially those in genomics, diagnostics, and lab services:


⚙️ What the RFI Means

CMS is:

  • Soliciting specific feedback on which regulations, guidance documents, forms, or processes are excessive, duplicative, or burdensome.

  • Looking to identify regulatory redundancies, documentation overload, and non-value-added compliance burdens—especially those that could be removed without affecting patient safety.

  • Triggered by a White House executive order, the RFI has political urgency and real potential to shape upcoming deregulatory policy shifts, especially if CMS receives clear, well-documented feedback.


๐Ÿ“ˆ Likelihood of Affecting Change

Moderate to high—if responses are:

  • Specific (e.g., cite exact forms, LCDs, or regulatory language),

  • Replicable across multiple stakeholders (e.g., same complaint from 50 labs = strong signal),

  • Politically consistent with the deregulation agenda (e.g., cost-saving without new federal spending),

  • Tied to economic productivity (especially if tied to ROI, job creation, or innovation).

Stakeholders that submit data-backed, shovel-ready, or easily implementable proposals will rise above the crowd.


๐Ÿงฌ For Genomics and Lab Industry: What Should You Recommend?

Here are high-value lab-related suggestions CMS should hear:


1. ๐Ÿ“† Fix or Eliminate the 14-Day Rule

  • Problem: The 14-day rule (aka DOS rule) limits labs from billing Medicare for tests ordered within 14 days of a hospital discharge.

  • Burden: Creates billing chaos, stifles access to complex molecular tests, delays care, and generates huge denial rates and appeals.

  • Solution: Repeal it or standardize Date of Service as Date of Sample Collection, as long as patient is not inpatient at time of test.


2. ๐Ÿงพ Simplify or Standardize LCDs Across MACs

  • Problem: Labs must navigate wildly divergent LCD policies and molecular edit rules depending on the MAC region (e.g., Palmetto vs. Noridian).

  • Burden: National labs must maintain dozens of billing workflows and software updates per state.

  • Solution: Nationalize MolDx or implement a “National Genomics Policy Repository” across all MACs for identical tests.

  • Alternative: Create a CMS Molecular Policy Hub as a single source of truth for LCDs, billing edits, coverage criteria.


3. ๐Ÿ” Simplify the ‘Partial Reporting’ Dilemma

  • Problem: LCDs often deny payment if a test panel can’t fully report all biomarkers due to specimen quality or assay thresholds.

  • Burden: Labs are penalized for partial yet clinically valuable results.

  • Solution: Clarify that partial results meeting analytical validity still count as ‘reportable’ and are reimbursable.


4. ๐Ÿงฎ Remove Redundant Prior Auth/Documentation Layers

  • Problem: MACs and contractors often demand clinical records already used by ordering physicians or require labs to become compliance police.

  • Burden: Labs must chase down EHR data, wasting time and risking HIPAA compliance.

  • Solution: Permit attestation by the ordering physician and eliminate duplicative documentation demands when the same info is already in the EHR or certified ordering platform.


5. ๐Ÿ“Š Simplify Reporting Burdens for Lab Test Utilization

  • Problem: Some MACs require labs to report utilization or test-specific data beyond CLFS rules.

  • Burden: Adds layers of custom Excel or manual tracking.

  • Solution: Centralize all reporting in PECOS or PAMA frameworks, and eliminate redundant field-level reporting.


๐Ÿง  How to Make Your Comments Stand Out

  1. Use CMS language:

    • Quote the RFI back to them (e.g., "This policy introduces duplicative burden under the administrative section of 42 CFR...").

    • Mention CMS form numbers or OMB numbers if possible.

  2. Provide real-world case studies:

    • A claim denied due to LCD ambiguity.

    • A lab that lost millions due to DOS rule denials.

    • Delays in patient care due to MolDx inconsistency.

  3. Bring allies:

    • A coordinated comment from 5+ stakeholders carries 10x more weight than a single comment.

  4. Use economic framing:

    • “This policy costs us $400,000 per year to administer. Removing it would allow us to hire 3 FTEs or invest in expanded testing.”

  5. Connect to care improvement:

    • “This change would speed up cancer diagnosis by 5 days and reduce unneeded procedures.”


๐Ÿ”š TL;DR Summary

This RFI is a real opportunity for labs, genomics innovators, and medtech stakeholders to suggest clear, impactful deregulation. CMS seeks concrete proposals to reduce burdens—especially around documentation, duplicative regulations, and administrative inefficiencies.

Your response should focus on:

  • 14-day rule repeal

  • Nationalization or simplification of MolDx

  • LCD interpretation consistency

  • Simplified reporting + data integrity

  • Economic and care-quality framing

Would you like help drafting a comment for submission? Or want this as a slide deck summary for internal review?


###
They could also simplify NCD 90.2, which targets advanced cancer patients for 1 NGS test, when today they may need multiple, including serial testings that illudidate checkpoint inhibitor response.

##

Great addition—and spot on for this RFI.

Let’s unpack that for strategic effect, and shape it into a targeted recommendation CMS could take seriously, especially if submitted by stakeholders in oncology, genomics, or molecular diagnostics.


๐ŸŽฏ Specific Suggestion for RFI: Simplify and Modernize NCD 90.2

Current Problem

  • NCD 90.2 restricts coverage to a single FDA-approved or FDA-cleared NGS test for advanced cancer, at initial diagnosis.

  • This excludes serial testing, resistance profiling, and longitudinal monitoring, even when:

    • New tissue is sampled (e.g. progression biopsy),

    • Liquid biopsy is needed due to inaccessible tissue,

    • Tumor biology evolves (e.g. loss of PD-L1, new MET/ALK fusions),

    • Treatment hinges on current biomarker status (e.g. MSI, TMB, ctDNA, emerging markers).


๐Ÿ“‰ Why It's a Problem

  • Clinically outdated: Oncology practice has shifted toward iterative, longitudinal testing, especially in immuno-oncology and targeted therapy.

  • Restricts patient care: Prevents coverage of essential tests that inform:

    • Checkpoint inhibitor response (e.g., blood-based TMB),

    • Therapy resistance (e.g., EGFR T790M),

    • MRD monitoring or recurrence detection via ctDNA.

  • Misaligned with FDA science: FDA has cleared multiple tests for different settings (e.g., Guardant360 for monitoring and profiling), yet Medicare will pay only once under this NCD.


๐Ÿ”ง Proposed Fix for CMS (in line with EO 14192)

Revise NCD 90.2 to allow coverage of additional FDA-approved NGS tests when clinically indicated, such as:

  • A different specimen (liquid vs. tissue),

  • A different clinical timepoint (e.g. progression),

  • A different clinical objective (e.g., MRD, immunotherapy biomarker testing),

  • A new cancer diagnosis (e.g., multiple primaries).

Alternatively:

Convert NCD 90.2 into a “baseline” national policy and delegate additional NGS uses to MAC discretion (LCDs or claim-by-claim medical review), as CMS does for nearly every other diagnostic test.


๐Ÿง  Policy Leverage Points

  • NCD 90.2 is ripe for amendment: It hasn’t been revised since 2020 and is misaligned with real-world cancer care in 2025.

  • Lung cancer care in particular now often requires 2+ NGS tests during a patient's disease course (initial tumor test, repeat biopsy, plasma test).

  • MolDx and MACs already approve many NGS uses under LCDs—allowing local flexibility in place of a rigid national barrier is logical and deregulatory.


✍️ How to Make This Suggestion Stand Out

  • Pair with clinical use case examples:

    • “At our cancer center, 35% of advanced NSCLC patients need a second NGS test within 9 months of diagnosis.”

  • Highlight FDA test availability:

    • “CMS is blocking access to tests already cleared by FDA for monitoring or serial profiling.”

  • Frame as deregulatory + pro-innovation:

    • “CMS could deregulate by delegating these additional test uses to MACs, rather than enforcing a one-size-fits-all policy.”

  • Avoid suggesting costly new benefits—instead, argue this is a modernization and alignment, not an expansion.


Would you like this packaged as a short written comment for CMS submission or added into a stakeholder briefing deck or memo? I can help with either.

YES

๐Ÿ“ CMS Comment for RFI Submission

Topic: Modernizing NCD 90.2 and Other Regulatory Burdens in Genomic Testing

To CMS,

Thank you for the opportunity to provide input on regulatory simplification under Executive Order 14192.

As a stakeholder in precision medicine and genomic diagnostics, I urge CMS to consider the following reforms that would meaningfully reduce burden and improve care:

  1. Revise NCD 90.2 to reflect current oncology practice.
    The current policy restricts Medicare coverage to a single FDA-approved NGS test for advanced cancer at the time of initial diagnosis. This is no longer sufficient. Patients frequently require multiple rounds of testing—including liquid biopsies, resistance profiling, and monitoring of response to immunotherapies. These are all uses supported by FDA-cleared tests and clinical guidelines.
    We recommend allowing additional NGS testing in the following scenarios:

    • New specimens (e.g., progression biopsy, liquid biopsy)

    • New clinical indications (e.g., immunotherapy response)

    • Longitudinal monitoring (e.g., ctDNA or MRD)

    • Diagnosis of a new primary cancer

  2. Simplify the 14-Day Rule (Date of Service regulation).
    The current DOS rule prevents labs from billing for tests ordered within 14 days of hospital discharge, even when the lab performed testing post-discharge. This policy causes confusion and delays. CMS should consider repealing or modernizing this rule, setting DOS as the date of specimen collection unless the patient is currently an inpatient.

  3. Create a national framework for MolDx policies.
    Today, LCDs and billing edits for molecular tests vary across MACs. This creates confusion, denials, and inefficiencies for national labs. CMS should unify coverage policies for molecular diagnostics across jurisdictions or centralize the MolDx program nationally.

These changes would reduce administrative complexity, align policy with science, and ensure Medicare beneficiaries have access to state-of-the-art genomic testing throughout their care journey.

Respectfully,
[Your Name / Organization (optional)]


๐Ÿ“„ Stakeholder Briefing Memo: 

RFI on Medicare Deregulation (April 2025)

Subject: CMS Request for Information – Strategic Opportunities for Regulatory Reform in Genomics and Lab Services

Deadline to Comment: June 10, 2025


๐Ÿ” Background

CMS has issued an open Request for Information (RFI) to identify regulations that may be repealed or simplified under Executive Order 14192, which mandates the repeal of 10 regulations for every new one issued. This is a rare opportunity to help shape Medicare policy—especially for high-impact sectors like precision oncology, genomics, and laboratory testing.


๐ŸŽฏ Key Issues for Genomic and Lab Stakeholders

1. NCD 90.2 – One-Time NGS Limit for Advanced Cancer

  • Problem: Current policy allows only one FDA-approved NGS test per advanced cancer diagnosis.

  • Real-world gap: Patients may require serial tests to monitor progression, identify resistance mutations, or evaluate checkpoint inhibitor response.

  • Ask: Revise NCD 90.2 to allow multiple NGS tests under defined clinical criteria (e.g., new tissue, different modality, new indication).

2. 14-Day Rule (Date of Service Regulation)

  • Problem: Prevents hospital-initiated tests from being billed post-discharge, even when the lab performs and reports after discharge.

  • Burden: Causes confusion, denials, and billing errors.

  • Ask: Repeal or modernize the DOS rule, using the specimen collection date when the patient is no longer inpatient.

3. Inconsistent LCDs and MolDx Rules

  • Problem: Labs must comply with dozens of conflicting LCDs and MAC edit rules depending on location.

  • Solution: Nationalize the MolDx program or standardize molecular billing policies to reduce complexity.

4. Denials for 'Partial Reports'

  • Problem: Labs may receive denials if even one biomarker in a multi-marker panel cannot be reported (e.g., low DNA yield).

  • Ask: Permit reimbursement for partially reportable tests when clinical value is preserved.


๐Ÿงญ Strategy for Engagement

  • Coordinate multi-party submissions (e.g., labs, clinicians, payors).

  • Use real-world cases and economic framing (e.g., cost of denials, delay to treatment).

  • Suggest devolution of rigid NCDs to MAC discretion, aligning with EO goals to streamline burdens.


Monday, April 21, 2025

AI Corner: AI Reviews the Transcript in the Supreme Court USPSTF Case (Braidwood)

 

AI SUMMARY:

SCOTUS Takes Up Braidwood v. Becerra: Will the Preventive Care Mandate Survive?

On April 21, 2025, the Supreme Court heard oral arguments in Braidwood v. Becerra, a pivotal case challenging the Affordable Care Act’s requirement that private insurers cover preventive services recommended by the U.S. Preventive Services Task Force (USPSTF) without cost-sharing. At stake is whether the USPSTF’s role violates the Constitution’s Appointments Clause, potentially unraveling a core part of the ACA’s preventive coverage framework. Below is an in-depth breakdown of the oral arguments, key constitutional questions, and a prediction on how the Court may rule.

###

AI DISCUSSION

###

Here is a summary and legal analysis of the April 21, 2025 Supreme Court oral arguments in Braidwood Management v. Becerra, as requested—tailored for an audience of health policy attorneys.


Case Overview

At issue is whether Section 2713 of the Affordable Care Act (ACA), which mandates that private insurers cover services recommended by the U.S. Preventive Services Task Force (USPSTF) without cost-sharing, violates the Appointments Clause of the Constitution. The key constitutional question: Are USPSTF members valid "officers of the United States," and if so, are they properly appointed and supervised under Article II?


Petitioners’ Argument (Government / HHS)

Hashim Mooppan, representing the government, defended the ACA’s reliance on USPSTF recommendations as constitutionally sound because:

  • USPSTF members are inferior officers, not principal officers.

  • The Secretary of HHS has adequate supervisory control, notably:

    • At-will removal power (critical in prior SCOTUS decisions like Free Enterprise Fund and Edmond).

    • Rulemaking authority over the effective timing of USPSTF recommendations.

    • Power to require pre-approval of Task Force recommendations under the agency’s rulemaking authority (e.g., under §300gg–92).

  • The term “independent”, used in ACA statute 299b-4(a)(6), refers to independent scientific judgment—not immunity from oversight or removal.

Constitutional avoidance was heavily invoked: the Court should interpret “independence” in a way that avoids finding the statute unconstitutional.


Respondents’ Argument (Braidwood / Religious Employer)

Jonathan Mitchell, arguing for Braidwood, countered:

  • USPSTF members issue binding mandates that immediately affect national insurance coverage and must be treated as principal officers.

  • These officers lack presidential appointment and Senate confirmation, in violation of Article II.

  • There is insufficient supervision: the Secretary of HHS cannot direct USPSTF recommendations, only remove members after the fact.

  • The statute’s use of “independent” implies the Task Force was designed to be free from political oversight, bolstering the separation from executive control.

  • The ACA grants USPSTF powers akin to legislating policy, which is impermissible for a body that is neither elected nor appointed via constitutional channels.


Key Moments in the Argument

  • Justice Gorsuch and Justice Barrett pushed hard on the issue of who actually appoints USPSTF members—Secretary or Director of AHRQ—and whether the Reorganization Act truly transfers appointment power to the Secretary.

  • Justice Alito expressed skepticism that someone removable at-will could meaningfully be called “independent.”

  • Justice Jackson emphasized the presumption of at-will removal unless a statute says otherwise, seeming to support the government’s reading.

  • Chief Justice Roberts and Justice Kavanaugh were visibly concerned about the statutory mechanism that allows USPSTF recommendations to bind the public while only being “reviewable” via back-end removal or timing delays.


Legal Stakes

The case tests the validity of insurance mandates across millions of health plans, potentially stripping cost-free coverage from services like PrEP (HIV prevention), cancer screenings, and immunizations if the statutory structure collapses.

It also has broader implications:

  • Could restrict Congress's use of external expert panels as reference points in legislation.

  • Could heighten scrutiny on similar panels like ACIP (immunization) and HRSA (women’s and children’s health).


Prediction

While difficult to call definitively, the majority may lean toward a compromise that preserves the ACA mandate but narrows the interpretation of “independent” and requires stronger oversight protocols from HHS.

Several justices (Roberts, Jackson, Kagan, Sotomayor) appear open to a functional supervision test, wherein the Secretary's ability to delay, veto, or remove USPSTF members suffices to satisfy Article II.

However, Justices Gorsuch, Alito, and Barrett may vote to strike down the mandate or require USPSTF members to be presidentially appointed and Senate-confirmed—or recommend severance of the USPSTF coverage linkage altogether.

My AI prediction: 6-3 or 5-4 to uphold the ACA structure, relying on constitutional avoidance and finding sufficient supervision by the Secretary under the Reorganization Act and statutory rulemaking authority.


 Were there any humorous moments?

##

Yes, there were a couple of light, even humorous moments during the April 21, 2025 oral argument in Braidwood v. Becerra. Here are two literal excerpts from the transcript that captured those lighter tones:


1. Justice Gorsuch, with classic dry delivery:

“I understand you think it's easy. Counsel always thinks it's easy.”
(p. 12)

This came after government attorney Hashim Mooppan insisted that determining the Secretary’s appointment authority was straightforward. Gorsuch gently pushed back, highlighting that what seems “easy” to counsel may not be so in court.


2. Shared laughter over Collins precedent:

JUSTICE ALITO: “Well, maybe that's a little bit unfair. But, I mean — maybe I was wrong in Collins. But explain to me —”
(Laughter.)
(p. 16)

Here, Alito jokingly undercut his own prior opinion in Collins, provoking audible laughter in the courtroom. It was a rare moment of self-deprecation and levity amid otherwise weighty constitutional debate.


These moments won’t rival a comedy club, but for Supreme Court oral argument, they definitely count as whimsical highlights.



##

You write that SCOTUS may "preserve the ACA mandate" but "require stronger oversight from HHS."   But SCOTUS has to render a decision now, how can that include an "if" clause such as "if" HHS makes satisfactory future protocols for oversight?

###

Great and very sharp point.

You're right—the Supreme Court must rule on the constitutionality of the statute as it stands, not on contingent future actions. The Court cannot condition its decision on HHS implementing better oversight protocols later, because:

  • It’s not a legislative body and doesn’t issue advisory opinions.

  • It cannot defer constitutional validity to hypothetical future agency conduct.

  • It must decide whether the current statutory framework meets Article II requirements.

So when I wrote that SCOTUS may preserve the ACA mandate but require stronger oversight from HHS, the more accurate formulation is this:

SCOTUS may uphold the ACA mandate by ruling that, under current law, the Secretary of HHS already has constitutionally sufficient oversight powers—such as at-will removal, rulemaking control, and power to delay or block USPSTF recommendations.

The Justices probing the removal and appointment structure (especially Gorsuch, Barrett, and Kavanaugh) seemed to suggest they could accept the status quo if they interpret the statute as already granting the Secretary these powers. That’s not a request for new oversight, but a narrowing or clarifying statutory construction to avoid striking down the mandate.

So the “if” is really in the Court’s interpretive reasoning:

If we read the statute to grant HHS sufficient supervision, then it’s constitutional.

This would likely be framed as a constitutional-avoidance holding: the Court would interpret the statute in a way that avoids a separation-of-powers violation, even if the statute’s language about “independence” is awkward.

Let me know if you want help rephrasing that section of your blog or opinion memo.


##
Fun Fact (BQ).  The parties from Braidwood originally asked to be exempted from a few benefits for religious reasons, and, the Texas judge granted them that (religious exemption) which hasn't been challenged in court and they still have it.   (What does that mean?  Does their BCBS plan subtract 1/1000 or 1/5000 of their monthly cost?)  

##
Original prompt (BQ): CHAT GPT 4o

On April 21, 2025, SCOTUS heard oral arguments in the BRAIDWOOD case.  At issue is the validity of the ACA statute that requires private insurers to cover (and without copay) services endorsed by the USPSTF.   Is the USPSTF a legally possible reference body for such insurance mandates.   Attached is the transcript of the oral argument.   Read the transcript, describe and discuss for an audience of health policy attorneys.   Include your prediction of the outcome of the case.

Friday, April 18, 2025

CMS Announces Summer 2025 Lab Pricing Meetings (June 27; July 23-24)

CMS has announced plans for this summer's new lab test pricing meetings.  The public comment meeting will be June 27, 2025, 10am-4pm ET.  It will  be a hybrid meeting (in person + zoom).  

The complementary Expert Advisors panel will meet in hybrid fashion on July 23-24, 10am-4pm ET.  (They move fast and likely finish early).  

See Fed Reg here:

https://www.federalregister.gov/public-inspection/2025-06756/hearings-meetings-proceedings-etc-medicare-program-new-and-reconsidered-clinical-diagnostic

and here:

https://www.federalregister.gov/public-inspection/2025-06758/hearings-meetings-proceedings-etc-medicare-program-medicare-advisory-panel-on-clinical-diagnostic

Also track the meeting announcement section on this CMS webpage [may be updated slower]:

https://www.cms.gov/medicare/payment/fee-schedules/clinical-laboratory-fee-schedule-clfs/annual-public-meetings

And the expert advisors web page (see "Meetings")

https://www.cms.gov/medicare/payment/fee-schedules/clinical-laboratory-fee-schedule-clfs/clfs-advisory-panel

Notes

  • See registration dates in the notices (for the public meeting, register May 1 -29; submit presentations by May 29).
  • Historically, even ten years ago, anyone wanting to enter CMS and is NOT a US national had special elaborate weeks-in-advance procedures.  FYI only.
  • I had noted that HHS instructed agencies to minimize public comment wherever possible, and, the PAMA act for lab pricing only requires CMS to "consult" an expert panel, not specifically to hold a public meeting. 
    • It's nice to see the public meeting is planned as in the past.
  • For this national meeting, thanks to CMS for starting at 10 ET / 7 PT rather than 8 ET which is 5am PT.
  • Check CMS for updated PPT templates.



Thursday, April 17, 2025

GeneDx Buys Fabric Genomics for at Least $33M

On February 16, 2025, GeneDx announced it was acquiring Fabric Genomics, an advanced clinical bioinformatics company, for at least $33M.   The valuation may go as high as $51M with benchmarks.

See news at:


I can't speak to the total capital previously invested in Fabric, but a valuation of $51M for a clinical bioinformatics company is notable.   (Publicly held SOPHiA has a market cap of $180M).

The acquisition will support GeneDx clinical services in areas like urgent NICU genomics, newborn screening, and international expansion.  GeneDx has a market cap around $2.7B.  Yahoo Finance lists the 52-week trading range as $9-$115.   

**
AI CORNER
**
Chat GPT gives a concise summary of Fabric Genomics.  (Provided as-is.)
___

Fabric Genomics, founded in 2009 and headquartered in Oakland, California, is a clinical bioinformatics company specializing in AI-driven genomic analysis and interpretation. Its platform enables rapid and accurate analysis of whole genomes, exomes, and gene panels, facilitating clinical decision support for applications such as rare disease diagnosis, hereditary risk screening, and oncology.

GEM and ACE

The company's technology, including tools like GEM and ACE, integrates deep genetic, phenotypic, and clinical data to prioritize disease-causing variants, achieving diagnostic yields up to 50% —which is significantly higher than the industry average. 

Clientele

Fabric Genomics serves a global clientele, including children's hospitals, hospital systems, commercial labs, and national genome initiatives, and has been instrumental in projects like the UK's 100,000 Genomes Project. 

Thesis

By leveraging explainable AI and scalable cloud-based solutions, Fabric Genomics has streamlined genomic workflows, reducing clinical review times and enhancing the scalability of genetic testing programs.


##

Compare to SOPHiA?

##
(Provided as-is.)

Fabric Genomics and SOPHiA GENETICS are both leaders in clinical genomics informatics, but they differ in scope, platform strategy, and business model.

Fabric Genomics

  • Focus: Primarily on rare disease diagnostics, neonatal ICU, hereditary risk, and whole genome/exome analysis.

  • Strength: AI-driven variant prioritization and clinical decision support, especially through its tools like GEM and ACE, known for explainability and high diagnostic yields.

  • Clients: Pediatric hospitals, academic medical centers, and national genome projects (e.g., UK 100,000 Genomes).

  • Business Model: Licensing cloud-based software for use in hospital labs and reference labs—platform-as-a-service for in-house clinical interpretation.

  • Scale: More niche and focused, with a leaner commercial footprint than SOPHiA.

SOPHiA GENETICS

  • Focus: Broader, covering oncology, inherited disorders, and infectious disease, with a strong presence in tumor profiling.

  • Strength: Cloud-native platform called the SOPHiA DDM™ (Data-Driven Medicine) platform, integrating genomics, radiomics, and clinical data into multimodal analysis.

  • Clients: Large hospital systems globally, including in Europe, Middle East, and the U.S., with over 800 customers.

  • Business Model: Software-as-a-service (SaaS) model—customers upload NGS data to SOPHiA’s cloud platform, which returns interpreted results.

  • Scale: Publicly traded (Nasdaq: SOPH), larger in revenue and global footprint, with strong oncology and pharma partnerships.

Key Differences

  • SOPHiA is a much larger, publicly traded firm with extensive oncology capabilities and pharma collaborations.

  • Fabric is smaller and focused more on rare diseases and deep AI-assisted variant interpretation, especially for high-throughput genomic programs.

In short: Fabric Genomics excels in deep clinical AI for rare disease diagnostics, while SOPHiA GENETICS offers a broader multimodal analytics platform with strong global and oncology reach.

Natera and BillionToOne Win MolDx Coverage

As many readers know, MolDx issues "foundational" or very broad LCDs, and MolDx controls actual coverage and indications by later technological reviews which occur on a test-by-test (company-by-company) basis.

Updated coverage may appear in billing articles attached to LCDs.  But MolDx has notified the public a couple years ago that updates may appear, NOT in such LCD/billing articles, but in its separate "DEX REGISTRY" online database.  (Here, 5/12/23 & here.)

Often, the first notification is not there, at MolDx, but in press releases.  Here are two new examples. 

NATERA - LUNG CANCER MRD COVERAGE

On February 25, 2025, Natera issued a press release that its ctDNA test, Signatera, had garnered coverage from MolDx for surveillance for relapse or recurrence in Stage 1-2-3 lung cancer.  Find it here:

https://www.natera.com/company/news/natera-announces-medicare-coverage-of-signatera-for-surveillance-in-lung-cancer/

Natera's announcement includes citations to 3 supporting publications, the largest study with 108 patients and 378 plasma timepoints (of which 10% or about 10 with recurrent disease, which is usually the "N" that drives the study size).

MolDx also covers MRD-type testing for relapse/response decisions in immune checkpoint therapies.  See a new review on that space, Vega et al.

BILLION TO ONE - LBx CGP

On April 16, 2025, BillionToOne announced Medicare (MolDx) coverage for its Northstar Select test, which is a liquid biopsy CGP test for actionable tumor genes.  It is an 84-gene test, including 19 copy number amlifications and 9 fusions.  It is covered for all advanced stage solid tumors.   Find it here:

https://www.prnewswire.com/news-releases/billiontoone-announces-medicare-coverage-for-northstar-select-302430353.html


From MedTechDive: Update on FDA LDT

The district court decision on the FDA LDT case came out several weeks ago, with a flurry of rapid news and opinion.   

Now that people have had time to reflect, on April 16, MedTechDive revisits the topic. Read a number of updated interviews with experts.  Find it here:

https://www.medtechdive.com/news/lab-developed-tests-what-now-analysis/745464/


## 

Separately, see a deep dive subscription article by Adam Bonislawski at 360Dx, which is also an updated on FDA-LDT and new expert opinions.

##

Re MedTech Dive - I'd mention two things - 

  • The FDA (the DOJ) has 60 days to file a "notice" of appeal.
    • If they do that, it's just one sentence, and keeps the case alive for possible, further, later motions.  
    • So if the filing doesn't occur, the case is dead.
    • BUT, if it DOES get filed, I wouldn't give that filing too much weight.  It could be a bookmark in a book nobody picks up again.
  • Let's give a little attention to what didn't happen - the FDA and HHS didn't make a peep of complaint about the decision.  
    • They could have, if they had wanted to.  Just sayin'.




Separately - Rubrum Advising's Nancy Stade has a rather complex discussion of the LDT ruling and related regulatory law concepts - here.  (See Stade's school of thought about "LDT" and with an asterisk - "LDT * " - which I don't fully understand.)


Wednesday, April 16, 2025

AMA Releases Spring Quarter PLA Applications

On the AMA PLA page here, find the Spring PLA PDF agenda here.   (See "Calendar" at bottom of blog).

There are 49 items, although a few are deletions or simply administrative changes (like owner-laboratory).

Ask for a comment packet immediately (see instructions on PDF) and submit comments by April 22.

Tempus applies for four codes (HRD, Immune Profile Score, whole slide imaging for biomarkers (lung & endometrial).  


##

CALENDAR

Spring 2025 PLA calendar

March 11, 2025: Application submission deadline

April 15, 2025: Public agenda posted to website

April 22, 2025: Interested party comment request deadline

April 24, 2025: PLA-TAG consideration completed

May 1, 2025: Panel vote

July 1, 2025: New and deleted codes publication date

Oct. 1, 2025: New and deleted codes effective date

CPT® 2026: New and deleted codes publication

CMS Releases Proposed Inpatient Rule FY2026

On April 15, 2025, CMS released the final Part C/Part D rule for 2026, with the total length cut in half from draft to final (many topics dropped - here).  That rule was a "Biden" proposal and a "Trump" final.

##

On April 11, 2025, CMS released the inpatient proposed rule for CY2026.  Find the CMS fact sheet here.  Find the inspection copy here, and look for the typeset Fed Reg copy on April 30.    

It's Still Long - Equity for Measures Dropped

It's still long under the new adminstration; the inspection copy (typescript) runs 1361pp.   The term "equity" appears about 60 times, but often in context of proposed dropping of hospital quality measures containing the term "equity."  I had no matches on the term "artificial intelligence."  (There is one on "Machine learning," in an NTAP discussion).

There is a discussion of developing national cost to charge ratios (inspection, 217ff).  

The SEP-1 sepsis measure has been somewhat controversial but remains in the 2026-27-28 measures.

New Tech Applications

Add on new tech payments (NTAP) beging at E, page 221 (inspection) and run to page 485.   Sunsetting and continuing NTAP are at page 243 (inspection), table II.E-01.A and .B.   There were 19 new applications under the normal pathway (page 247, inspection).  14 are carried forward into the current discussion (starting (a) AUCATZYL, page 249 inspection).   

There's at least one diagnostic test, TRIVERITY (sepsis), page 473ff (inspection).  Alternate pathway NTAPs begin at page 481 (inspection).  

##

AI CORNER

##

For a CMS policy audience, here's a structured summary of the FY 2026 New Technology Add-On Payment (NTAP) applications and policy context under the Inpatient Prospective Payment System (IPPS), as outlined in the FY2026 Proposed Rule (CMS-1833-P):


1. NTAP Program Purpose and Criteria

CMS offers NTAPs under 42 CFR 412.87 to provide temporary additional payments for eligible new medical services and technologies that:

  • Are new (not substantially similar to existing technologies)

  • Are costly, such that the MS-DRG payment is inadequate

  • Offer substantial clinical improvement over existing treatments

Alternate pathways are available for:

  • FDA-designated Breakthrough Devices

  • Qualified Infectious Disease Products (QIDPs)

  • Limited Population Antibacterial Drug (LPAD) pathway products

Add-on payments are not budget neutral and are typically up to 65% or 75% of the cost above the standard MS-DRG payment, depending on technology type.


2. New Application Volume and Process

For FY2026, CMS received:

  • 19 traditional pathway applications

    • 14 applications are reviewed in the proposed rule

    • 2 were ineligible (no FDA marketing authorization by May 1, 2025)

    • 3 were withdrawn

CMS has also detailed the process improvements, including:


3. Summary of Continuing, New, and Discontinued NTAP Technologies

A. Continuing NTAPs in FY2026 (selected technologies)

(See Tables II.E-01.A and II.E-01.B)

Examples include:

  • CASGEVY™ and LYFGENIA™: gene therapies for sickle cell disease, with NTAP amounts up to $2.3 million

  • TriClip™ G4 (mitral valve repair): $26,000

  • EPKINLY™ and COLUMVI™ (B-cell lymphomas): $6,504

B. Discontinued NTAPs

(Table II.E-02)

Technologies are discontinued if the 3-year anniversary of market entry occurs before April 1, 2026. Discontinued examples include:

  • CYTALUX® (ovarian indication)

  • GORE® TAG® Thoracic Branch Endoprosthesis

  • iFuse Bedrock Granite

Note: CERAMENT® G was approved for one indication in a prior year but now has a new FY2026 application for a different clinical use.


4. Highlights from FY2026 Traditional Pathway Applications

Detailed summaries are included for each technology in the Proposed Rule and MEARIS. Example:

  • AUCATZYL® (obecabtagene autoleucel): A CAR-T cell therapy for relapsed/refractory B-ALL. FDA approved November 2024. Uses new ICD-10-PCS codes XW0338A and XW0438A.

Other applicants include novel biologics, implantable devices, and diagnostics, each evaluated against the newness, cost, and clinical improvement criteria.


5. Strategic Implications for Stakeholders

  • Hospitals should ensure they code eligible discharges properly to capture NTAP.

  • Manufacturers must align FDA timelines with CMS deadlines (May 1), submit cost modeling data, and clearly differentiate their technology.

  • Policy teams should monitor MEARIS for transparency in applications and rulemaking trends, particularly for high-cost therapies (e.g., gene therapy NTAPs exceeding $1M).


Highlight a few examples from the NTAP applications.

###

In the FY2026 CMS IPPS proposed rule, several of the 14 reviewed NTAP applications stand out for their clinical novelty, technological sophistication, or potential policy significance. Among the most striking is CASGEVY™ (exagamglogene autotemcel), a gene therapy approved for sickle cell disease (SCD), which carries a proposed NTAP of $1.65 million. CASGEVY™ uses CRISPR-based gene editing to reprogram the patient’s own stem cells—a dramatic shift from conventional transfusion-based SCD management. 

Closely related is LYFGENIA™ (lovotibeglogene autotemcel), another cell-based gene therapy for SCD, with an even higher proposed NTAP of $2.325 million, underscoring the emergence of ultra-high-cost genetic interventions and CMS’s evolving role in bridging access through temporary add-on payments.

Equally innovative is the HEPZATO™ KIT, which combines melphalan chemotherapy with a hepatic delivery system that isolates liver circulation during infusion, enabling high-dose chemotherapy targeted to liver tumors. The proposed NTAP of $118,625 reflects both the technical complexity and the potential to extend life in metastatic disease settings with few alternatives. 

Another standout is the EVOQUE™ Tricuspid Valve Replacement System, one of the first devices to target tricuspid valve regurgitation—a historically undertreated condition. With a proposed NTAP of $31,850, EVOQUE™ signals CMS’s growing recognition of structural heart innovations beyond the aortic and mitral domains.

Finally, the Paradise™ Ultrasound Renal Denervation System represents a novel application of ultrasound to modulate sympathetic nerves in the renal artery, offering an interventional therapy for patients with resistant hypertension. The device is seeking a $14,950 NTAP, and its inclusion highlights CMS’s increasing engagement with neuromodulatory devices that cross disciplinary boundaries between cardiology and nephrology. Together, these several applications reflect the growing diversity and sophistication of NTAP submissions, and the policy challenge of integrating frontier therapies into a prospective payment model.

Tuesday, April 15, 2025

Friends of Cancer Research: From RECIST to AI-RECIST

 Friends of Cancer Research has identified multiple key operational problems in precision oncology, and help stakeholders craft solutions.

Here's a new effort of considerable interest - bringing AI to RECIST - the detection of cancer relapse.

Find the home page here:

https://friendsofcancerresearch.org/ai-recist/

See also a 3 page mini summary deck:

https://friendsofcancerresearch.org/wp-content/uploads/ai.RECIST-Project-Slides.pdf

Better measurement (consistency and precision) of imaging relapse is CRITICAL to molecular MRD development, as imaging is taken as the gold standard against which molecular MRD must out-perform.   The more precise and tight the error bars are, around imaging, the easier and quicker it is to show better performance for liquid biopsy.



###

They summarize as follows:

What is the unmet need and why does it matter?

Tumor response metrics are used to determine the efficacy of cancer therapies in solid tumor clinical trials. These measurements rely on standardized and unbiased criteria through the Response Evaluation Criteria in Solid Tumors (RECIST) performed by expert human readers. RECIST-based assessments provide a systematic approach to objective tumor measurements at defined timepoints, but their implementation faces several challenges, including investigator bias, subjectivity in lesion selection, and variability in measurements across clinical sites and radiologists. Artificial intelligence (AI)-driven tumor measurement tools have the potential to address these challenges, reducing variability, increasing efficiency, and improving measurement accuracy.


How are we helping to find solutions?

Friends created a research partnership to evaluate AI-driven tumor measurement tools alongside human-reader RECIST assessments. 

Key objectives: 

  • Assess AI tool agreement – Can AI-based tools provide consistent tumor measurements?
  • Compare variability among AI tools and human assessments – How well do AI-driven measurements align with RECIST-based readings by human readers? 
  • Explore AI’s impact on efficiency – Can AI tools reduce variability and streamline clinical trials? 

How does this impact patients?

Blinded Independent Central Review (BICR) is used in clinical trials to ensure accurate tumor assessments. Regulators often require BICR to minimize bias by blinding human readers to patient and treatment details when evaluating imaging-based endpoints like progression-free survival and objective response rate. However, BICR is resource-intensive, potentially prolonging trial timelines, delaying treatment decisions, and increasing costs. These delays may limit patient access to new therapies and, in some cases, may require additional imaging or adjudication when discrepancies occur between local and central assessments. AI-driven tumor measurement tools have the potential to streamline this process by ensuring consistent, unbiased verification of local assessments, reducing review time, and improving trial efficiency without compromising data integrity. By enhancing the speed and reliability of tumor measurements, AI could accelerate clinical trial progress and improve patient access to effective treatments.


###


Friday, April 11, 2025

AI Watch: Google Publishes 68-page Booklet on Prompt Engineering (For Using AI)

Google publishes a 68-page booklet on the science and art of "prompt engineering," that is, interacting with AI to accomplish goals.

https://www.kaggle.com/whitepaper-prompt-engineering

Google has an auto podcast (19 min) about the book here:

https://www.youtube.com/watch?v=F_hJ2Ey4BNc

  • Note that the booklet is designed for a particular interface to Google Gemini, and, it is written for pretty advanced users.   
  • BUT there is still much to be gleaned for less-advanced users, as you can look over the shoulders of professionals and see how experts interact with AI via prompts.


 See an extensive "AI CORNER" below.

###

AI CORNER

Here’s an AI review of the 2025 Google Guide to Prompt Engineering by Lee Boonstra, tailored for your genomic policy blog readers who also follow AI developments:


๐Ÿ” Worth Your Download?

Yes—if you use, tune, or even think about large language models (LLMs) in your work.
This is Google’s internal playbook for how to write prompts that actually work. It’s presented accessibly but backed by deep product and engineering expertise from teams working on Gemini and Vertex AI.


๐Ÿ“˜ What the Paper Covers

This 68-page guide serves as both a primer and a cookbook for crafting prompts that optimize performance, reliability, and safety in LLMs. It walks you through:

  • Prompt formats (zero-shot, few-shot, chain-of-thought, system/role/contextual prompting)

  • Sampling controls (temperature, top-K, top-P)

  • Advanced strategies like:

    • Step-back prompting

    • Tree of Thought

    • ReAct prompting (LLM + tools = baby agents)

    • Self-consistency voting

    • Automatic prompt engineering


๐Ÿง  Highlights & Takeaways

  • Prompting is engineering, not guessing. Each example (whether for classifying a movie review or writing Python code) comes with clear configurations and an explanation of why it works.

  • LLMs are stochastic, not deterministic. You’ll get repeatable, better output when you configure sampling controls like temperature, top-k, and top-p in combination—not in isolation.

  • Chain-of-Thought and Step-Back prompting are now table stakes for complex reasoning tasks.

  • Few-shot examples aren't just filler—they’re essential for guiding structure, logic, and tone.

  • Best practices include using positive instructions over constraints, setting token limits smartly, and mixing up class order in classification prompts to reduce overfitting.

  • Debugging code with LLMs? The guide walks through real examples, from bash-to-python translations to catching bugs using self-generated logs and prompts.


๐Ÿงช Why It’s Relevant to Genomics and Policy

Even if you're not writing code:

  • Prompt engineering controls reproducibility and reduces hallucinations, two themes familiar to diagnostics labs.

  • The system/role/contextual prompts mirror clinical decision pathways, which could be useful when working with AI to parse literature, generate SOPs, or synthesize regulatory language.

  • As LLMs are deployed in diagnostics, good prompt design becomes a compliance issue, not just a UX feature.


๐Ÿ“Ž For Power Users and Curious Analysts Alike

Whether you’re validating a new clinical NLP tool, training a chatbot for patient triage, or using GPT to summarize MolDx LCDs, this guide shows you how to fine-tune your input for consistent, policy-aligned output.


Bottom Line:

Google’s Prompt Engineering Guide is the best structured, most readable, and most example-rich guide to LLM prompting as of 2025.
You’ll come away with actual copy-pasteable strategies and a mental model of how prompts, model settings, and outputs interact.

Let me know if you'd like a visual cheat sheet or a one-page summary handout for internal staff or clients.

###

OK. Let's assume you use AI 30 minute  a day in your work, for the past year, so you're pretty experienced, but self taught.    What are 10 things you will find both NEW and also USEFUL in this book?

###

Great framing. Assuming I’m an experienced, non-professional AI user (e.g., using ChatGPT, Gemini, Claude, or Perplexity 30–60 minutes daily for tasks like research, email drafts, code help, writing assistance, brainstorming), here are ten concepts from the 2025 Google Prompt Engineering Guide that would be both new to me and valuable:


๐Ÿ”Ÿ Ten Valuable New Concepts

  1. Step-Back Prompting
    ๐Ÿ”„ Instead of solving a task directly, first prompt the model to consider general principles or patterns, then use that response to guide the specific task.

    Why it’s useful: Helps with tasks that require reasoning or “thinking outside the box.” Great for strategy, synthesis, or creative writing.

  2. Self-Consistency Prompting
    ๐Ÿ—ณ️ Send the same Chain-of-Thought prompt multiple times at high temperature, then “vote” on the most common final answer.

    Why it’s useful: Makes your answers more robust and reduces hallucinations in complex reasoning tasks—especially for ambiguous queries.

  3. Tree-of-Thought Prompting (ToT)
    ๐ŸŒณ Instead of a linear Chain-of-Thought, this method explores multiple reasoning branches before converging on a solution.

    Why it’s useful: Helps in complex problem-solving, planning, or design work where multiple valid paths exist.

  4. ReAct Prompting (Reason + Act)
    ๐Ÿค– Combines internal reasoning with external tool use—like search APIs or code execution—to improve real-world answers.

    Why it’s useful: Gives you a basic agent-like workflow without needing to build an agent. Perfect for market research or current events queries.

  5. System / Role / Contextual Prompting as Distinct Modes
    ๐Ÿงฉ Clear separation of three prompt layers: system (what the model should do), role (who it's acting as), and context (the data it needs).

    Why it’s useful: Lets you tune tone and style (e.g., make the model act like a physician or a 5th-grade teacher), and also better structure your multi-part prompts.

  6. Prompt Variables for Reusability
    ๐Ÿง  Use {city}, {disease}, {product} etc. inside prompts so they can be reused dynamically across tasks or in apps.

    Why it’s useful: Makes it easy to automate AI workflows or build prompt libraries—saves time in daily use.

  7. Output Control via Sampling Configuration
    ๐ŸŽ›️ Combining temperature, top-K, top-P, and max tokens effectively—not just using one slider at a time.

    Why it’s useful: You gain much better control over creativity vs. reliability (e.g., using temperature 0 + top-P 0.9 for deterministic classification tasks).

  8. JSON Prompting for Safer, Structured Output /   JavaScript Object Notation. 
    ๐Ÿงพ Ask the model to return its answers in a strict JSON schema—even showing the format in the prompt.

    Why it’s useful: Helps with automation and reduces hallucination—critical if you're using AI for data extraction or to pass into other tools.

  9. Automatic Prompt Engineering (APE)
    ๐Ÿคฏ Use the model itself to generate better prompts, then score and select the best-performing one.

    Why it’s useful: You can "meta-prompt" your way to better performance in novel or high-risk tasks (e.g., regulatory writing, hypothesis generation).

  10. Role Prompting with Tone/Style Modifiers
    ๐ŸŽญ Assign roles and style at once (e.g., “You are a humorous travel guide who speaks informally.”)

Why it’s useful: Saves you from endlessly rewriting tone—great for brand voice, presentations, or teaching materials.


###
Let's say you work in reimbursement consulting for investors, genomisc labs, and write white papers in the field.   ???

###

Perfect—thanks for that context. Here are ten things from the Google Prompt Engineering Guide that would be new and valuable specifically for a genomics policy consultant like you, especially in fundraising strategy, investor due diligence, and white paper development:


๐Ÿ”Ÿ Prompt Engineering Insights for Genomics Industry Consulting

  1. System/Role/Context Prompt Layering
    ๐Ÿงฉ Explicit separation of “what the model should do,” “who it’s acting as,” and “what background data it has.”

    Use case: Draft a report where the model behaves like an oncology investor relations officer or a regulatory reviewer—shifting tone, vocabulary, and depth accordingly.

  2. Step-Back Prompting for Landscape Synthesis
    ๐Ÿ”„ Ask the model to summarize general trends (e.g., "What are the components of a successful MRD test launch?") before writing your actual white paper content.

    Use case: Improves strategic framing in pitch decks and executive summaries—forces a broader view before specifics.

  3. Chain-of-Thought Prompting for Claim Support
    ๐Ÿง  Walks the model through intermediate reasoning steps, showing how it links data, citations, or regulatory events.

    Use case: Helps when you’re comparing payor coverage criteria for LDTs vs. PMA tests, or modeling reimbursement timelines.

  4. Self-Consistency Sampling for Investment Theses
    ๐Ÿ—ณ️ Run a CoT prompt 5+ times at high temperature, then find the dominant investment thesis or forecast.

    Use case: Reduce bias or cherry-picking when generating multiple perspectives for a strategic options memo.

  5. Prompt Variables for Reusable Due Diligence Templates
    ๐Ÿ” Write prompts with placeholders like {company}, {assay}, {FDA status} to generate fast turnaround analyses.

    Use case: Quickly switch focus from a ctDNA MRD company to a methylation early detection firm without rewriting core frameworks.

  6. Few-Shot Prompting with Counterexamples
    ๐Ÿ“Š When evaluating pitch language, regulatory claims, or clinical utility arguments, provide “good” and “bad” versions.

    Use case: Great for training junior team members or creating investor training materials—e.g., what not to say in an MRD pitch.

  7. ReAct Prompting for Competitive Intelligence
    ๐ŸŒ Combine LLM reasoning with external tools like search to assess, say, “recent PMA approvals in liquid biopsy.”

    Use case: Prototype lightweight research agents that triangulate public FDA, CMS, and investor data in one query loop.

  8. Structured JSON Outputs for Investment Memos  / JavaScript Object Notation. 
    ๐Ÿงพ Prompt the LLM to output a JSON with structured fields like {clinical utility}, {reimbursement}, {competitive landscape}.

    Use case: Automates memo population and lets you export data directly into Excel or investor briefing templates.

  9. Instruction-Over-Constraint Framing
    Instead of “Don’t talk about reimbursement,” say “Summarize investor risks without addressing reimbursement issues.”

    Use case: Better control when shaping nuanced LLM output (e.g., evaluating MRD platforms where coverage is still pending).

  10. Multimodal Prompt Planning
    ๐ŸŽฅ Anticipates combining text prompts with graphics, tables, or datasets (e.g., inputting a Kaplan-Meier curve or coverage policy PDF).

Use case: As LLMs integrate visual inputs, you’ll prompt them to summarize data slides or parse a Medicare LCD directly.


Let me know if you’d like a tailored prompt library for genomics consultants, or a slide summarizing these for your next investor pitch workshop.

##

Prompt library.

##



 Absolutely. Here’s a Tailored Prompt Library for Genomics Consultants, based on the 10 insights we discussed. These are real-world, reusable prompt templates designed for your work in:

  • ๐Ÿงฌ Genomics policy

  • ๐Ÿ’ฐ Fundraising strategy

  • ๐Ÿ“Š Investor due diligence

  • ๐Ÿ“„ White paper development


๐Ÿง  Prompt Library: Genomics Consulting Edition

Each entry includes:

  • Use Case

  • Prompt Template

  • Tips for Use


1. Positioning a Company for Investors (System + Role Prompting)

Use Case: Write a compelling intro to a pitch deck or white paper.

Prompt:

Act as a biotech investor relations officer. Write a one-paragraph summary introducing {company_name}, a genomics company specializing in {assay_type}. The summary should emphasize clinical need, competitive edge, and addressable market.

Tip: Replace {company_name} and {assay_type} with client details.


2. Fundraising Framing via Step-Back Prompting

Use Case: Start a white paper with strong strategic framing.

Prompt:

Before analyzing {company_name}, summarize the top 3 unmet needs or bottlenecks in the field of {domain}, such as MRD testing, ctDNA, or methylation detection. Then use that to frame the company’s solution.

Tip: Great for executive summaries or slide 2 of a deck.


3. Coverage and Reimbursement Reasoning (Chain of Thought)

Use Case: Generate a walkthrough of reimbursement risks.

Prompt:

You are a Medicare reimbursement expert. A genomic test has FDA approval but limited Medicare LCD coverage. Think step-by-step about the hurdles this test will face for nationwide reimbursement and how a company might overcome them.

Tip: Use when planning payor strategy timelines.


4. Compare Companies with Self-Consistency Sampling

Use Case: Generate multiple investment perspectives, then pick the best.

Prompt:

Compare Guardant Health and Personalis in the MRD space. What are the key strategic differences in their approach to clinical evidence, FDA approval, and payer strategy? Let’s think step-by-step.

Tip: Run 3–5 times and choose the most consistent or compelling answer.


5. Automated Due Diligence Memos with Prompt Variables

Use Case: Quickly switch from company to company.

Prompt:

Summarize the clinical value, commercial traction, and regulatory status of {company_name}, which offers a {assay_type} assay targeting {clinical_indication}. Focus on unmet need, FDA status, and payor engagement.

Tip: Plug into Airtable, Notion, or Google Sheets for batch outputs.


6. Messaging Calibration (Few-Shot Prompting with Good/Bad Examples)

Use Case: Train teams or refine pitch language.

Prompt:

Below are examples of how to describe a genomic test to payers. Which is stronger, and why? Then write a better version:
Example A: "We believe our test is helpful."
Example B: "Our test demonstrates 95% sensitivity for recurrence detection in early-stage CRC patients, with peer-reviewed validation."

Tip: Add your own examples to personalize.


7. Market Landscape via ReAct Prompting

Use Case: Combine LLM + real-time search for competitive intelligence.

Prompt:

How many FDA-approved liquid biopsy MRD tests exist in the U.S. market as of 2025? Search for public announcements and approvals, then summarize key players and test types.

Tip: For tools like Perplexity, Claude with browsing, or LangChain agents.


8. Structured Memo Output (JSON Prompting)

Use Case: Auto-populate due diligence templates.

Prompt:

*Summarize {company_name} in this format:

{
  "clinical_utility": "",
  "regulatory_status": "",
  "reimbursement_outlook": "",
  "competitive_advantage": "",
  "investor_risks": ""
}
```*

**Tip**: Outputs can be dropped into Airtable, Excel, or merged into slides.

---

### 9. **Instruction Over Constraint Prompting (Tone Shaping)**  
**Use Case**: Avoid ambiguity or over-hallucination.

**Prompt**:
> *Write a three-paragraph section for an investor memo about {company_name}. Emphasize product-market fit and FDA status. Do not discuss reimbursement strategy.*

**Tip**: Emphasize what *to* include rather than just what *not* to.

---

### 10. **Image or Slide Interpretation (Future Multimodal Prompt)**  
**Use Case**: Explain a chart or K-M curve for a genomics test.

**Prompt**:
> *Here is a Kaplan-Meier survival curve from a clinical trial of {company_name}’s MRD test in NSCLC. Summarize what it shows about test performance and how it supports the clinical utility claim.*

**Tip**: Will be useful in GPT-4 Vision, Gemini 1.5, and multimodal models.

---

Would you like this as a **Google Doc**, **Markdown**, or **Notion database**?

Or, I can add **a section with prompts tailored for slide decks** and **investor FAQs** next.


####
For users of SCITE.AI, see an 18 page guide:
https://www.researchsolutions.com/ebook-aiprompts-academic