Sunday, August 24, 2025

Decibio: Case Studies on Molecular Test Adoption (#2, Oncotype Dx)

Consultancy Decibio is providing a series of case studies (online at Linked In) for adoption of major molecular tests.

See Case #2 here (Oncotype Dx).

https://www.linkedin.com/posts/nacarmegan_case-study-2-oncotype-dx-activity-7364379534630010880-RERl/

https://www.linkedin.com/feed/update/urn:li:activity:7364355655379423232/

Decibio writes,  ManĂ© Mikayelyan and Nila Venkat analyze five case studies: BRCA1/2, Oncotype DX, Cologuard, NIPT, and FIT to illustrate how adoption unfolded across different applications. They trace specific timelines, highlight inflection points, and uncover the structural levers that moved tests from early innovation to widespread use.



Consultancy MA360i Releases Report on Minimal Residual Disease Industry & Policy

The consultancy MA360i has released a report on everything connected to minimal residual disease - MRD.   

Designed for both commercial and policy teams, the report will bring together the status of MRD across solid tumor types, summaries of payor policies to date, summary of major guidelines (NCCN, ESMO), and identification of evidence gaps.

Contact MA360i for more.  They also have a downloadable 2-page summary of key findings - here.

https://ma360i.com/evidence-assessment-coverage-implications/



Friday, August 22, 2025

ARPA-H-Update: Adaptive Treatment Selection in Precision Oncology; Funding?

Back in March 2024, I wrote a blog about ARPA-H launching a new program called ADAPT (2024 blog here).  

I noted then that Medicare patients were laboring under the outdated NCD 90.2, which assumes that major genomics on a cancer will only be done once - one and done.

From Genomeweb and elsewhere, here are some updates on progress with ADAPT, eighteen months later.

###

See a new subscription-based update at Precision Medicine Online by Alison Kanski - here.

###

See the ARPA-H home page for ADAPT, here.

See ARPA-H press release, May 2025, here.

See a listing of awardees with brief program descriptions, here.

See a press release from UNC Lineberger here. $28M.

See a press release from Inova here.

See a press release from Stanford here.

ARPA-H leadership spans two administrations - Acting Director Jason Roos PhD, who has held several HHS positions, came to ARPA-H in early 2024 and is still in the leadership role in August 2025.

FUNDING?

ARPA-H had a budget of $1.5B for FY2024 (here).  

For the FY2026 budget proposal, several organizations may be merged (ASPR, ARPA-H, becoming ASHF), so it is more difficult to tease out an equivalent future budget for ARPA-H.  However, I see a remark that ARPA-H will have a programmatic decrease of -$555M down to $945M total (p. 90).  

The new theme areas for ARPA-H are proposed as, 'Chronic Disease," "America-Made Manufacturing," and "Healthy Well being," "Healthcare efficiency and transparency," and "American Leadership."  

###

AI CORNER

###

Using only the public links, I asked Chat GPT5 to draft an article for expert readers.

I have not proofed for details - so take it just as an example of AI writing style.

###

 Here’s a structured ~1,500-word in-depth expert briefing on ARPA-H’s ADAPT program, written for a health policy and oncology expert audience.


Expert Briefing: ARPA-H’s ADAPT Program Targets Tumor Evolution with $142M Initiative

By [Your Name] – August 2025


Introduction

In May 2025, the Advanced Research Projects Agency for Health (ARPA-H) announced one of its most ambitious oncology initiatives to date: the ADvanced Analysis for Precision cancer Therapy (ADAPT) program. With up to $142 million in milestone-based funding, ADAPT aims to transform cancer care from static, biomarker-guided interventions into dynamic, continuously adaptive treatment frameworks that reflect the real-time biology of evolving tumors.

The program is emblematic of ARPA-H’s mandate to fund high-risk, high-reward research modeled after DARPA. Its ambition is not incremental improvement but rather a paradigm shift in precision oncology, particularly for patients with metastatic cancers where therapy resistance is inevitable and outcomes remain poor.


Why ADAPT Now?

Metastatic cancers remain one of the largest sources of mortality and cost in U.S. healthcare, accounting for more than 600,000 deaths annually. Nearly all metastatic tumors eventually develop resistance to systemic therapy. Traditional oncology practice rarely integrates real-time biological monitoring into decision-making: molecular profiling is typically performed once, prior to treatment, leaving clinicians blind to subsequent tumor adaptations.

The scientific case for ADAPT rests on several trends:

  • Tumor evolution is predictable: Recent advances in systems biology and mathematical modeling suggest resistant traits can often be anticipated.

  • Technology has matured: Next-generation sequencing, liquid biopsies, advanced imaging, and AI-driven multi-modal data fusion are now feasible at clinical scale.

  • Trial design is shifting: Adaptive clinical trial models are increasingly validated in other disease areas, demonstrating feasibility of “mid-course corrections.”

ARPA-H aims to integrate these strands into a single program that addresses the core challenge: how to continuously match therapy to a tumor’s moving biological target.


Program Architecture

ADAPT is organized into three technical areas (TAs), with 10 funded teams spanning academia, industry, and integrated cancer centers.

  1. Therapy Recommendation Techniques (TA1)

    • Develop AI models, biomarkers, and computational frameworks that interpret longitudinal, multi-modal tumor data.

    • Goal: Identify resistant traits early, predict drug response, and recommend therapy switches.

  2. Evolutionary Clinical Trial Design (TA2)

    • Test adaptive trial frameworks in breast, lung, and colon cancer.

    • Goal: Operationalize TA1’s predictions into real-world patient care.

  3. Treatment and Analysis Platform (TA3)

    • Create a secure, cloud-based data ecosystem for real-time collaboration.

    • Goal: Provide shared infrastructure for clinicians, data scientists, and regulators.

Funding is milestone-contingent: continuation depends on aggressive progress toward predictive accuracy, trial enrollment, and clinical utility.


Technical Area 1: Therapy Recommendation Techniques

The five TA1 awardees reflect a cross-section of AI, computational biology, and biomarker development expertise.

  • Arizona State UniversityTumor Ecology & Evolutionary Modeling

    • Focus: Map tumor resistance traits including gene activity, oncogenic state, immune evasion, and cellular evolvability.

    • Strategy: Borrow ecological models (e.g., predator-prey dynamics) to predict when resistance traits emerge and proliferate.

    • Significance: May provide a framework for “forecasting” tumor resistance before it is clinically evident.

  • Stanford UniversityInterpretable AI for Longitudinal Multimodal Data

    • Investigators: Olivier Gevaert, PhD, and Andrew Gentles, PhD.

    • Focus: Build machine learning models that integrate pathology, radiology, molecular, and clinical data.

    • Significance: Unlike “black box” AI, Stanford emphasizes interpretable models, ensuring clinical decision-makers understand why a therapy switch is recommended.

    • Quote: “This is the first adaptive trial where computational analysis is an intrinsic part of patient care,” said Gentles.

  • UC San DiegoDynamic Biomarkers for Adaptive Therapy

    • Focus: Construct biomarkers that evolve in real-time alongside tumor biology.

    • Method: Continuous updating of predictive models using new biopsy and circulating tumor DNA (ctDNA) data.

    • Significance: Provides an infrastructure for serial biomarker re-interpretation, addressing the current bottleneck of static companion diagnostics.

  • MITMachine Learning for Therapy Optimization

    • Focus: Develop algorithms that synthesize genomic, imaging, and EHR data to recommend therapy sequences.

    • Significance: Bridges computational oncology with clinical informatics, potentially making ADAPT insights usable within existing health system data pipelines.

  • Brigham & Women’s HospitalFoundational Models for Clinical-Genomic Data Fusion

    • Focus: Build multi-modal foundational models that integrate genomic, clinical, and treatment history data.

    • Role: Provide shared analytic tools to other TA1 teams, acting as a computational backbone.

Collectively, TA1 represents an attempt to bring machine learning into oncology practice at the same level of rigor as radiology or pathology.


Technical Area 2: Evolutionary Clinical Trials

Perhaps the most ambitious element of ADAPT is its commitment to evolutionary trial design: trials that adapt therapy while patients are on treatment, guided by TA1-generated biomarkers.

  • UNC Lineberger Comprehensive Cancer Center (Breast Cancer)

    • Award: Up to $28M.

    • Trial: EVOLVE (Evolutionary Clinical Trial for Novel Biomarker-Driven Therapies).

    • Enrollment: Up to 700 patients with metastatic breast cancer.

    • Methods: Serial biopsies, ctDNA monitoring, advanced imaging, and EMR integration.

    • Investigators: Lisa Carey, MD; Charles Perou, PhD; Ian Krop, MD, PhD; Eric Winer, MD; Antonio Wolff, MD.

    • Significance: Aims to generate predictive biomarkers where few exist today in breast cancer, particularly for resistance to endocrine and HER2-targeted therapies.

  • Beckman Research Institute of City of Hope (Lung Cancer)

    • Focus: Build a biomap of immunotherapy resistance mechanisms in non-small cell lung cancer.

    • Strategy: Validate new biomarker-driven therapies in near real-time, using immunogenomic profiling.

    • Significance: Tackles the growing clinical challenge of immune checkpoint inhibitor resistance.

  • UT MD Anderson Cancer Center (Colon Cancer)

    • Trial: Two-phase umbrella design for colon cancer.

    • Focus: Rapidly identify emergent resistant traits, integrate biomarker-guided drug assignments.

    • Collaborators: Inova Schar Cancer Institute among others.

    • Quote: “This is perhaps the most ambitious collaborative effort in precision oncology,” said Tim Cannon, MD, Inova.

    • Significance: Could create a new paradigm for adaptive colorectal trials, historically dominated by static trial designs.

Together, these trials test ADAPT’s central hypothesis: can continuously updated biomarker insights translate into longer survival and reduced toxicity in metastatic disease?


Technical Area 3: Treatment and Analysis Platform

Data infrastructure is the linchpin of ADAPT. Without a secure, interoperable, and scalable system, the adaptive framework would be impossible.

  • DNAnexusTreatment and Analysis Platform (TAP)

    • Focus: Build a cloud-based data ecosystem linking clinicians, computational scientists, and trialists.

    • Features: Secure data sharing, regulatory compliance, real-time integration of TA1 biomarkers with TA2 trials.

    • Significance: Leverages DNAnexus’s commercial expertise in hosting genomics pipelines for scale.

  • Washington University in St. LouisMAX System Integration

    • Focus: Extend TAP with Multi-modal Analysis with XNAT (MAX) tools.

    • Role: Provide advanced capabilities in managing imaging and clinical data, expanding TAP beyond genomics.

The TAP is also intended to generate public goods: anonymized aggregate datasets and algorithms will be made publicly accessible, creating an open science dimension to ADAPT.


Policy Implications

From a policy perspective, ADAPT raises several important issues:

  • Coverage & Reimbursement – Adaptive trials with mid-course treatment changes may challenge existing reimbursement frameworks (e.g., CMS coverage for companion diagnostics is tied to fixed, pre-treatment tests). ADAPT could pressure CMS and FDA to reconsider evidentiary standards.

  • Data Governance – With real-time cloud-based analysis, questions of HIPAA compliance, patient consent, and data security become central.

  • Clinical Guidelines – If ADAPT succeeds, NCCN guidelines may eventually need to integrate “adaptive biomarker monitoring” as a standard of care.

  • Regulatory Science – FDA’s role in regulating “evolving biomarkers” remains undefined; ADAPT could provide the first large-scale test case.


Outlook

Patient enrollment is expected within 12 months, with early results anticipated in 2–3 years. The program’s duration is set at six years, aligning with ARPA-H’s focus on achievable but transformative milestones.

The long-term goal is not simply better clinical trials, but a permanent change in how oncology is practiced:

  • from static biomarkers to dynamic, real-time monitoring,

  • from rigid protocols to adaptive trial frameworks,

  • and from siloed data to a shared national platform.

If successful, ADAPT could establish a blueprint for adaptive care models in other diseases, from infectious disease to neurodegeneration.


Conclusion

The ADAPT program represents ARPA-H’s most ambitious foray into oncology, combining the computational sophistication of Silicon Valley, the clinical trial expertise of NCI-designated centers, and the data infrastructure of leading biotech firms.

By aligning incentives through milestone-based funding, ARPA-H is forcing a level of cross-disciplinary integration rarely achieved in oncology. Whether ADAPT fulfills its promise will depend on whether predictive models can be translated into real-world trial outcomes – a challenge that has eluded precision oncology for decades.

For now, ADAPT stands as a test of ARPA-H’s core proposition: that disruptive, high-risk investments can accelerate U.S. biomedical innovation in ways incremental NIH funding cannot.


Word Count: ~1,540


 ?

Thursday, August 21, 2025

AdvaMed, Others, Urge Coverage of Breakthrough Devices

 AdvaMed and other groups have released a public letter asking CMS to restore rapid coverage of FDA-approved "Breakthrough Devices."   Similar coverage, called MCIT, was finalized by the first Trump admnistration but deleted by the Biden administration.

Find the information here:

  • AdvaMed press release here.
  • MedTechDive coverage here.
    • Also, MedTechDive on the breadth of the BT program here.
  • The actual PDF letter here.
    • Also refers to a legislative approach, S 1717.
    • The Senate bill could allow statutory changes not available to CMS alone.



Wednesday, August 20, 2025

14 Day Rule Part Two: 14 Day Rule Becomes a Dramatic Novel (AI)

Previous:
 14 Day Rule Part One:
   Where Did It Come From?   A Human vs AI Blog

      https://www.discoveriesinhealthpolicy.com/2025/08/medicares-14-day-rule-where-did-it-come.html

Here:
14 Day Rule Part Two:
   The Dramatic Novelization

Here’s a treatment for a historical novel on the “birth” of the 14-day rule—structured as if it were a sweeping, character-driven saga set in Washington, DC, 2005–2007. 

I’ve leaned into drama, irony, and bureaucratic comedy, with plenty of fictional characters (some archetypes, some satirical) to animate the story.


See also a fictional New Yorker article about the history of the 14 Day Rule here.


Novelistic Treatment: Fourteen Days


Setting

Washington, DC, 2005–2007. The backdrop is the Department of Health & Human Services, Centers for Medicare & Medicaid Services (CMS) headquarters in Baltimore, Capitol Hill offices, K Street lobbying firms, and the sterile fluorescent-lit conference rooms where policy gets hammered out.

The mood: part The West Wing, part Catch-22, part Kafka, with touches of Armando Iannucci satire.


Main Characters

  • Eleanor Whitcomb (The Architect)
    A seasoned CMS policy lawyer, idealistic but weary. She believes in harmonizing rules, yet fears that labs are finding too many cracks in the system.

  • Dr. Martin Kellner (The Scientist)
    A molecular pathologist at Johns Hopkins, brilliant but impatient with bureaucratic delay. He sees firsthand how patients are harmed when hospitals hoard specimens under bundling rules.

  • Ralph “Bud” Crenshaw (The Lobbyist)
    A gregarious former Hill staffer now lobbying for a coalition of independent labs. He believes every Medicare reg is negotiable with the right dinner and scotch.

  • Sister Margaret O’Connell (The Ethicist)
    A nun and hospital compliance officer, who insists that hospitals must be fairly reimbursed but also that labs deserve recognition for lifesaving tests. She quotes Aquinas in committee meetings.

  • Victor Sloan (The Politician)
    A mid-level Senate staffer whose boss chairs the Finance Committee. Victor knows nothing about genomics but everything about which MAC is headquartered in his boss’s state.

  • Carla Espinoza (The Patient Advocate)
    A fiery Latina organizer, whose husband’s cancer recurrence went undetected because a genomic test couldn’t be billed under Medicare’s DOS rules. She storms every meeting demanding clarity.

  • The “Reg Writers” (Comic Chorus)
    Three faceless CMS analysts—Janet, Theo, and Raj—who sit in a basement office drafting and redrafting the Federal Register notice, always bickering about commas, CFR cross-references, and whether “archived” means 30 or 31 days.


Act I: The Gathering Storm

    1. CMS realizes its “date of service” manual instructions (DOS = collection date) are wreaking havoc. Hospitals are bundling claims, labs are screaming, and patients are caught in the crossfire.

  • Eleanor convenes a “listening session.” Around the table: Martin’s fury, Bud’s schmoozing, Sister Margaret’s theology, Carla’s tears, and Victor’s political warnings. The meeting collapses in acrimony.

  • The Reg Writers downstairs joke: “Why don’t we just flip a coin and call it the 14-and-a-half-day rule?”


Act II: Negotiation and Breakdown

  • Bud pulls strings, arranging late-night bar meetings between lab execs and Hill staffers. Deals are promised, none stick.

  • Martin tries to demonstrate a new tumor recurrence assay, but the CMS projector fails; no one sees the science.

  • Sister Margaret testifies to a congressional subcommittee, insisting the hospital’s moral duty is to care for patients, not hoard billing privileges. The press ignores her.

  • Carla organizes a march of cancer survivors outside CMS HQ, chanting “We need a Date! We need a Date!”

  • Eleanor drafts a proposed rule (August 22, 2006). Victor pressures her: “Pick a number. Congress won’t understand nuance.” Eleanor sighs, and with the stroke of a pen, 14 days is born.


Act III: The 14-Day Rule Emerges

  • The proposed rule lands in the Federal Register. Pandemonium ensues:

    • Labs say 14 days is too long.

    • Hospitals say 14 days is too short.

    • Patients say 14 days is arbitrary.

    • The Reg Writers argue about whether “14” should be spelled out (“fourteen”).

  • Public comment sessions become theatrical: Carla brings a coffin-shaped cake with “RIP Innovation.” Bud hands out cigars labeled “DOS Exception.”

  • Eleanor faces sleepless nights, caught between CMS lawyers citing bundling statutes and her conscience whispering patients deserve better.


Climax: December 1, 2006

  • After a bruising year of drafts, comments, and Hill interventions, the final rule is published. The DOS for archived specimens remains tied to 30 days, but for tests ordered 14+ days after discharge, the DOS shifts to the date of performance.

  • At the release meeting, the Reg Writers cheer ironically. Bud orders champagne. Martin mutters it’s progress but not enough. Carla vows to keep fighting.

  • Eleanor closes her binder, exhausted. “We’ve created something that will outlive us all. God help the labs who must explain it.”


Epilogue

  • 2007–2017. The 14-day rule mutates, carving out exceptions for molecular pathology, sparking new lobbying wars.

  • Characters scatter: Bud to a think tank, Sister Margaret to Rome, Martin to industry, Eleanor to academia.

  • Carla narrates the final page: “It was never about fourteen days. It was about who controlled the future of medicine—hospitals, labs, or the people who needed the tests. The fight goes on.”


Tone and Themes

  • Comedy of Bureaucracy: Endless meetings, wordsmithing of Federal Register text, political horse-trading.

  • Moral Stakes: Patients waiting for life-saving tests.

  • Irony: A world-changing policy born not from science or ethics, but from an arbitrary compromise number—14.


Would you like me to write the opening chapter in full novelistic style (with dialogue, setting, inner thoughts), perhaps the chaotic first “listening session” at CMS where all these characters collide? That would give you a taste of how the treatment translates into narrative.

###

###

Great — here’s a novelistic opening chapter for Fourteen Days, set at the first CMS listening session in 2005–2006. It introduces the key fictional characters in scene, with dialogue, inner thought, and tone.


Chapter One: The Listening Session

The conference room on the third floor of CMS headquarters in Baltimore was too small for the occasion. Fluorescent lights hummed overhead, flickering slightly. The carpet smelled faintly of disinfectant, like a hospital that had been scrubbed at 5 a.m. and forgotten by noon. On the table sat stacks of briefing binders with a bold header:

DATE OF SERVICE: DEFINITIONS AND POLICY OPTIONS.

Eleanor Whitcomb adjusted her glasses and surveyed the mismatched delegation she had summoned. She was fifty-two, with the air of a woman who had spent too many years parsing commas in federal regulations. Her job, officially, was Deputy Director for Payment Policy. Unofficially, it was cat herder of the American laboratory economy.

“Thank you for coming,” she began, voice even. “Our purpose today is to discuss unintended consequences of the current Date of Service rules for laboratory tests.”

Dr. Martin Kellner leaned forward before she could finish. Tall, silver-haired, with an oncologist’s sharp intensity, he jabbed the table with a forefinger.

“Unintended? Eleanor, my patient needed a molecular assay for cancer recurrence. The hospital sat on her tissue block, and because of your rules, the lab couldn’t bill. She died waiting for clarity. That’s not unintended. That’s a failure.”

The room fell silent.

From the corner, Ralph “Bud” Crenshaw cleared his throat. He wore a silk tie patterned with little microscopes. He was a lobbyist’s lobbyist: tan in February, Rolex flashing whenever he gestured.

“Doctor, with respect, nobody’s disputing patient stories. But the issue is billing mechanics. Claims bundling, DRG overlap, carrier interpretations—Eleanor’s folks can’t fix cancer. They can fix policy. Let’s not lose sight of the prize.” He winked at Eleanor.

Carla Espinoza slammed a folder onto the table. Mid-thirties, fire in her voice, she carried herself like someone used to shouting at rallies.

“The prize? My husband couldn’t get a test that might have given him two more years. All because Medicare thinks a date on a claim form matters more than a life. You all talk about ‘bundling’ like it’s accounting. For me, it’s a funeral.”

Sister Margaret O’Connell, in her starched habit, spoke softly but cut through the noise. “The principle at stake,” she said, “is justice. The hospital bears costs, yes. The labs perform miracles, yes. The patient deserves truth, most of all. Aquinas said: Law without justice is a perversion of law.

Bud groaned. “Here we go, Sister, quoting saints.”

Martin snapped back. “Better saints than lobbyists.”

At the far end, Victor Sloan, the Senate staffer, tapped his pen. He looked about twelve years old, though he was thirty. “Look,” he said, “my boss doesn’t care about Aquinas or billing systems. He cares about voters. And voters don’t understand why the same test is paid one way in Florida and another in Alaska. You need a simple rule.”

“Simple,” Eleanor repeated, rubbing her temples. “Yes. Simple.”

Behind her, in the corner, the Reg Writers—Janet, Theo, and Raj—typed furiously on a shared laptop, whispering.

“What if we just… pick a number?” Theo muttered. “Ten days? Fifteen?”

“Fourteen has a nice ring,” Raj said. “It’s biblical.”

Janet rolled her eyes. “You two are children. This is the Federal Register, not bingo night.”

Eleanor caught fragments of their bickering and almost laughed. Almost. Instead, she straightened her papers.

“Ladies and gentlemen,” she said, “this agency needs a national standard. One that avoids abuse, protects hospitals, allows labs to bill fairly, and does not disadvantage patients. That’s why we’re here.”

Carla folded her arms. Martin seethed. Bud checked his watch. Sister Margaret prayed silently. Victor texted his boss. The Reg Writers kept typing.

Eleanor looked around the table and knew, with sudden clarity, that whatever rule emerged from this chaos, it would be despised by everyone.

And that, perversely, was how she would know it might survive.


Would you like me to continue with Chapter Two, perhaps the late-night K Street bar scene where Bud (the lobbyist) woos Hill staffers into supporting the idea of a fixed “waiting period,” setting the stage for the infamous 14 days?

###

###

Chapter Two: K Street Arithmetic

The rain had been threatening all afternoon and finally committed around eight, slanting across K Street like a curtain being pulled. Inside The Laffer Curve—a bar that sold itself as a “policy tavern” with framed CBO charts on the walls—Bud Crenshaw had staked out the back booth. It was the same booth he’d used for a deficit deal in 2002 and a durable medical equipment fiasco in 2004. He swore the cushions still smelled faintly of appropriations.

“Over here,” he called, lifting his bourbon as Victor Sloan shook water from his jacket. A junior from Energy & Commerce trailed Victor, cheeks flushed from the cold. “You brought help.”

“This is Liv,” Victor said. “Smart, dangerous, and underpaid.”

Liv grinned, slid in, and set a legal pad on the table. “I’ve been told to observe and never tweet.”

“Then you’ll go far,” Bud said. “Drink?”

Before anyone could answer, two more arrived. Denise Park, a calm, surgical vice president from the Hospital Association, hair swept into a precise twist; and Nadia Ben-Ari, a coalition lead for the independent labs, her scarf still spangled with rain. Their eyes met like old adversaries at a chessboard.

“Folks,” Bud said, “this is the supper club. Only rule: we don’t pretend this is about anything but power and paperwork.”

“Patients?” Nadia asked.

“Patients are why paperwork matters,” Bud said quickly. “Sit.”

They settled, and the bartender—a veteran who knew the difference between a CR and an RFA—dropped off a round. On the wall behind them, a framed headline read: PAY-FOR GAP CLOSES HOURS BEFORE VOTE.

“Where are we?” Victor asked.

“Lost in the maze,” Bud said. “CMS wants a standard date of service. They’ve got the thirty-day ‘archive’ carve-out, but it doesn’t deal with post-discharge testing. Labs are getting strangled by hospital bundling. Hospitals are worried about unbundling the world.”

Denise folded her hands. “Hospitals are worried about fair payment for comprehensive care. You can’t yank pieces out of DRGs and pretend the economics still hold.”

“And labs are worried about not getting paid at all,” Nadia said. “A tissue block sits for two weeks, someone finally orders a test for a treatment decision—boom, claim hits a brick wall because the DOS gets dragged back to the inpatient date. Tell me that’s rational.”

Victor rubbed his eyes. “I need something a Senator can explain in twelve seconds, preferably six.”

Liv clicked her pen. “A waiting period,” she said. “Define a waiting period that signals: ‘This test is not part of the inpatient stay.’”

Denise shook her head. “A waiting period invites games. Collect the specimen Tuesday, order the test on Wednesday, call it separate—”

“Which is why it can’t be one day,” Bud cut in. “Look, we’ve tried polite letters. We’ve tried PowerPoints that made grown men cry. Nothing travels on Capitol Hill like a number. We pick a number that no one loves and everyone can live with.”

Nadia eyed him. “Seven?”

“Seven is a week,” Denise said. “Too easy to schedule around. Also sounds cute. Cute gets abused.”

“Twenty-one?” Nadia countered.

“Too long for oncology,” Liv said, surprising herself with the confidence in her voice. “Physicians don’t plan adjuvant therapy on Senate calendars.”

Bud leaned forward. “Fourteen. Two Sundays. It feels like we waited. It’s defensible.”

Silence. Victor ran it through the internal algorithm known only to staffers: constituent appetite x donor tolerance x op-ed mockery risk.

“Fourteen,” he said slowly, “I can sell. It’s common sense wrapped in an even number.”

Denise raised an eyebrow. “And the mechanics?”

Bud slid a stack of napkins into the center like he was dealing poker. “We attach conditions.”

Liv wrote while Nadia dictated, teeth worrying the cap of her pen between items:

  1. Ordered by the patient’s physician at least fourteen days after discharge.

  2. Could not reasonably have been ordered during the hospitalization.

  3. Hospital procedure wasn’t performed for the purpose of collecting the specimen.

Denise frowned. “Purpose language is slippery.”

“Then we ladder it to clinical appropriateness,” Liv said, scribbling. “Medically inappropriate to collect the sample other than during the procedure.

  1. The test is reasonable and medically necessary.

“Of course,” Denise said. “That one’s a rosary bead in every paragraph.”

Bud tapped the last blank square. “And we decide what the new date of service is. ‘From storage’ gave us headaches. ‘Date of performance’ telegraphs that this is a post-hospital event.”

Victor nodded. “OMB will want the logic chain clean. Discharge—wait fourteen—new order—new DOS equals performance date.”

Nadia took a sip of her gin and stared at the napkin like it was a newborn. “Hospitals will scream.”

“They’ll complain,” Denise corrected. “Screaming is for radiologists.”

“We can sweeten,” Bud said. “A carve-out for rare situations—fresh tissue chemo sensitivity. Live cells, weeks of culture. Not inpatient-relevant. You get to say ‘patient-centric’ three times on cable.”

Denise weighed it. “You’ll put in language to avoid Swiss-cheesing DRGs?”

“Intent language,” Bud said. “We are very concerned about unbundling.” He smiled. “CMS loves being ‘very concerned.’”

A server leaned in with a basket of fries. The napkin migrated, dodging grease like a sail trimming to wind. Rain hammered the windows. Somewhere near the bar a retired actuary told a story about a 1999 conversion factor and everyone groaned on cue.

Victor’s phone buzzed. He glanced, held it to his chest. “My boss wants guardrails. He doesn’t want MSNBC saying we let labs upcode.”

“They can’t ‘upcode’ a lab test,” Nadia said. “They can misbill it, but that’s not—”

“He means headlines,” Liv translated. “We add documentation expectations to the Program Integrity Manual. Physician order. Discharge date. Clinical rationale. It reassures auditors.”

“Fine,” Bud said. “We hand CMS the script.”

They worked the napkin like it was a truce map. Denise swapped “could not reasonably have been ordered” for “would have been medically inappropriate to order,” then swapped it back after Nadia pointed out that “ordering” was the verb that mattered to claim edits. Liv drew arrows between conditions and wrote ‘calendar days’ in the margin.

“Why ‘calendar’?” Victor asked.

“Because if anyone says ‘business days,’ we’ll still be arguing on Presidents’ Day,” Liv said. “And you don’t want a hospital attorney explaining bank holidays to a judge.”

Bud laughed, warm and brief. “You’re going to terrify me into hiring you.”

Denise leaned back. “Hospitals need one more comfort. The test must not guide treatment provided during the hospital stay.”

Nadia considered, then nodded. “Add it. It’s true. These genomic panels don’t decide whether the surgeon closes with staples.”

Liv wrote the fifth condition in neat block letters and underlined it twice. The napkin now read like a CFR fever dream:

  • 14 days post-discharge

  • Couldn’t reasonably be ordered inpatient

  • Collection not the purpose of the hospital procedure

  • Medically necessary

  • Does not guide treatment during the hospital stay

  • DOS = date of performance

“Eleanor won’t swallow this whole,” Victor said. “She’ll want to move a clause or two.”

“She can move the furniture,” Bud said. “Not the foundation.”

They sat with it, the way people sit with a decision they suspect will outlast them. The bartender passed by, glanced down, and shook his head. “Don’t write law on napkins,” he said. “Napkins get lost.”

“Federal Register will find it,” Bud said. “It always does.”


At 11:37 p.m., outside under an awning dripping like a metronome, Bud called Eleanor. She picked up on the third ring, voice low, office quiet.

“What did you build?” she asked.

“A bridge, maybe a trap,” Bud said. “Fourteen days. With conditions that look like common sense dressed for church.”

He read the list. Eleanor didn’t interrupt, which meant she was thinking ten moves ahead: claim edits, manual language, inevitable abuse, and the sentence that would be quoted back at her in litigation five years hence.

“‘Date of performance’ instead of ‘obtained from storage’?” she asked finally.

“It’s cleaner,” Bud said. “Also more truthful. The event happens when the test is done. That’s the day the money meets the road.”

Silence. Then: “OMB will ask why fourteen.”

“We’ll say two Sundays,” Bud said. “And the hospitals will agree it’s not seven. And oncology will agree it’s not twenty-one.”

Eleanor sighed. “I hate that this is how policy gets made.”

“By people in the rain?” Bud asked.

“By numbers that sound reasonable,” she said. “Send me your napkin.”

“It’s, ah, greasy.”

“Scan it. The grease is the legislative history.”

She hung up. Bud pocketed the napkin as if it were currency.


Across town, the Reg Writers sat in a diner that never closed. Janet stirred coffee, Theo ate eggs, Raj highlighted a printout of the manual section that would eventually become their cross to bear.

“Calendar days,” Janet said, circling the phrase on a draft. “Yes. Saves us from business day purgatory.”

Theo frowned. “Hyphenate ‘post-discharge’?”

“Not in the CFR section header,” Raj said without looking up. “We keep it clean.”

They spoke like mechanics around an engine: not loving the machine, but responsible for whether it started in the morning. Theo flipped the napkin scan on his phone and cringed at the ketchup halo around DOS = date of performance.

“Do you ever think,” he asked, “what happens to the people on the other side of these commas?”

Janet nodded once. “Every day,” she said. “It’s why the commas matter.”

Raj tapped the fifth condition. “This one—‘results do not guide treatment during the hospital stay’—that’s our firewall against unbundling.”

“It’s also true,” Janet said. “Truth is good policy from time to time.”

Theo smiled. “Write that in the preamble and see if OMB lets it through.”

They laughed, tired and gentle, and went back to work.


By midnight the rain had thinned to a mist. Victor and Liv shared a cab. She cradled her legal pad like a relic.

“You did well,” Victor said. “You found the words that make fourteen feel like a policy.”

“It’s just arithmetic,” she said.

“It’s politics pretending to be arithmetic,” he replied. “But it’ll do.”

Back in the booth, Denise and Nadia shook hands. Not warm, not cold—like adversaries who had agreed to a ruleset before the next round.

Outside, Bud pinned the napkin to the corkboard above his desk in the K Street office. A grease stain darkened the corner; a number sat in the middle, circled twice.

He sat down, opened an email, and began to translate the napkin into sentences that could survive lawyers, commenters, and history. Fourteen wasn’t a principle. Fourteen wasn’t science. Fourteen was a bridge over a river that wouldn’t stop rising.

He hit send.

And somewhere in Baltimore, a new paragraph began writing itself into the future.

[STOP]


##
See also a fictional New Yorker article (8000 words!) about the 14 Day Rule.


14 Day Rule Part One: Where Did It Come From? (Compare Human & AI Blog)

14 Day Rule Part One:
   Where Did It Come From?   (Compare Human & AI Blogs)

14 Day Rule Part Two:
   The Dramatic Novelization!
   https://www.discoveriesinhealthpolicy.com/2025/08/14-day-rule-part-two-dramatic-novel-ai.html


14 Day Rule Part One:
   Where Did It Come From? [human]  

###

It's hard enough to explain Medicare's awkward "14 day rule."   Among other things, it bundles payment for genomic testing on inpatient cancers unless the test is ordered at least 14 days after inpatient discharge.

Explaining the in's and out's is a 30 minute class.  Let's ask a different question: Where did it come from?  My human blog is below, and an AI blog for comparison, at bottom.

###

Negotiated Rulemaking Proposed

In 1997, the lab industry was concerned that different Part B MACs (aka "carriers") had very different local policies for lab tests and even different definitions, or no definitions, of basic terms.  Section 4554 of the Balanced Budget Act of 1997 brought CMS into "negotiated rulemaking" with the lab industry.  The goal was that a lab, like Quest, could submit the same lab claim in Florida and Alaska and it would be processed the same way.

Negotiated Rulemaking Conducted

See a CMS webpage here.  A multi-year process resulted in the creation of 23 NCDs for common lab tests.  Published on November 23, 2001, these are now almost 25 years old.

The LCDs are basic - covering urine culture, HIV testing for monitoring, prothrombin time, thyroid testing, etc.  Here

Along the way, CMS published some definitions that are not NCDs.  For example, CMS defined the "date of service" as the date of specimen collection.

Date of Service and "Archive"

Of course, it doesn't make sense for the date of service to be the "date of specimen collection" forever.  What if you go back to a paraffin block after 2 years?  If the DOS was the date of specimen collection, it would be unbillable (you can't submit 2-year-old claims).  So CMS said, if the specimen is stored more than 30 days, the DOS would be the day it's pulled from the archive.

This was simplified to a 30-day rule (it didn't matter if you literally put it in an archive or not, or what an "archive" was.)

DOS Enters Regulation

CMS has a lot of definitions and guidance in its manuals.  But if they want to really depend on something, CMS may put the statement into a legally binding regulation.   This happened for Lab DOS in fall 2006, effective CY 2007.

On Tuesday, August 22, 2006, at Fed Reg 71:49065, CMS proposed "to add a new regulation 414.414 to address concerns regarding the date of service of a clinical laboratory test that uses a stored specimen."   (Reference to 2001 and 2002 notices that were less official.)

Critically, where the money meets the road, CMS stated that:

  • The date of service of a test may affect payment because, if the date of service falls during an inpatient stay or on a day on which the beneficiary had an outpatient procedure, payment for the laboratory test usually is bundled with the hospital service.

CMS proposed a 14 day exception issuing a new date of service (the date of test performance) if the test was ordered by the physician 14 days after an inpatient discharge or a hospital outpatient day procedure discharge.

On Friday, December 1, 2006, 71:69705-706, CMS finalized the rule and updated the regulation position to 514.410.  It never becomes clear exactly how 14 days was chosen.  The discussion on December 1 is fairly detailed, becuase (as I recall) lab stakeholders meet with Congress, with CMS, etc, about concerns.   Trivia point - the 14 day rule date of service was shifted from "date from archive" to "date of performance" between August and December.

##

In 2007, the rule was extended to apply to pathology specimens as well.

In about 2014, most tests other than "molecular pathology tests" were bundled to an inpatient or outpatient stay and not paid separately.   "Molecular pathology" was implemented to mean, human DNA or RNA.

In about 2017, CMS made an exception for molecular pathology tests, which could now be billed "by the lab that performed the test" rather than only the hospital were the specimen was collected.   

See the calendar 2006 policy-making here.


###

AI CORNER

###

I gave Chat GPT 5 this blog, and the federal rule (2006), and asked it to write it up.

###

Here’s a polished expert-level blog article draft that integrates your notes with the regulatory history documented in the August 22, 2006 proposed rule and December 1, 2006 final rule from the uploaded Federal Register.


Medicare’s Date of Service Rule and the Origins of the 14-Day Rule [AI]

It’s hard enough to explain Medicare’s awkward “14-day rule.” In practice, it means that most lab tests tied to an inpatient admission or outpatient hospital encounter are bundled into the hospital’s payment unless the ordering physician waits at least 14 days after discharge. For molecular tests in oncology, this policy has had decades of downstream consequences.

But instead of another tutorial on how the 14-day rule works, let’s ask: Where did it come from?


Negotiated Rulemaking: Setting the Stage

The story begins with Section 4554 of the Balanced Budget Act of 1997. At that time, clinical labs faced a patchwork of inconsistent local policies across Medicare carriers. A single test might be reimbursed in one jurisdiction but denied in another. To fix this, Congress required CMS to engage in “negotiated rulemaking” with industry stakeholders.

The process culminated in 23 National Coverage Determinations (NCDs) for common lab tests, published on November 23, 2001. These were intended to harmonize coverage rules nationally for routine services like urine cultures, prothrombin time, and thyroid testing.

Along the way, CMS also formalized definitions. Most importantly, the “date of service” (DOS) for a laboratory test was defined as the date of specimen collection.


The Archival Problem

That rule worked well—until it didn’t. Consider this scenario:

  • A tissue block from 2004 is retrieved in 2006 for a new test.

  • If the DOS were the original collection date, the claim would be unbillable—Medicare does not accept 2-year-old claims.

To solve this, CMS introduced the 30-day archive rule. If a specimen was stored for more than 30 days, the DOS shifted from the original collection date to the date the specimen was pulled from storage. This avoided impossible billing situations, even if the concept of “archiving” was never fully defined.


The Leap Into Regulation

By 2006, CMS wanted a more formal framework. On August 22, 2006, CMS proposed adding new regulation §414.410 to define lab DOS, focusing specifically on the problem of tests performed on stored specimens.

The agency explicitly acknowledged the payment consequences:

“The date of service of a test may affect payment because, if the date of service falls during an inpatient stay or on a day on which the beneficiary had an outpatient procedure, payment for the laboratory test usually is bundled with the hospital service.”

To address stakeholder concerns, CMS proposed a new exception. The DOS could shift to the date of test performance if:

  1. The test was ordered by the patient’s physician at least 14 days after hospital discharge.

  2. The test could not reasonably have been ordered during the hospitalization.

  3. The hospital procedure was performed for reasons other than specimen collection.

  4. The test was medically necessary.


Finalization and the Birth of the 14-Day Rule

On December 1, 2006, CMS finalized the rule, moving the operative section to §414.510. Several key points are worth noting:

  • The final rule shifted the exception from an “archive” framing to a broader “date of performance” exception.

  • The 14-day interval was retained despite industry objections that it was arbitrary. CMS rejected requests for shorter intervals, citing risks of unbundling hospital care.

  • A carve-out was added for chemotherapy sensitivity tests, where fresh tissue was required, and results were rarely relevant to inpatient treatment decisions.

The rule became effective for CY 2007.


Aftermath and Evolution

  • 2007: The DOS/14-day framework was extended to pathology specimens.

  • 2014: CMS clarified that almost all lab tests remained bundled, except molecular pathology (human DNA/RNA) tests.

  • 2017: CMS issued a major revision: molecular pathology tests could now be billed by the performing laboratory, not the hospital where the specimen was collected.


Why It Still Matters

For molecular oncology, the 14-day rule continues to shape billing workflows, hospital-lab relationships, and even test ordering patterns. What began as a regulatory fix for billing archived specimens morphed into one of Medicare’s most enduring—and most confusing—laboratory payment rules.

In short, the 14-day rule was not a deliberate policy to slow innovation. It was a side effect of CMS trying to reconcile specimen storage rules, hospital bundling, and the fear of unbundling abuses. Nearly twenty years later, labs and hospitals are still living with the consequences.


 

PLA Codes, CPT Codes: Next CPT Meeting September 18-20, Chicago

The next AMA CPT Editorial Panel Meeting will be held in Chicago, September  18-20, 2025.   

It's hybrid (in person registration vs online zoom registration) - and both are free.

Find it here:

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

###

At that link, you can also find the Agenda,

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

Pathology issues had a comment deadline of July 25.  You can still request materials to review, though, til September 12.

Diagnostics Field

Prostate biopsy codes for the surgical procedure are up for revision (55xxx).

Under genomic sequencing procedures, look for new codes for renal conditions, immune conditions, and Alport syndrome.  

There is one application for Administrative MAAA code (bladder) and one deletion.

There are remote device codes for monitoring oncology therapy.

There are a number of codes with the term, "Augmentative AI" in areas like radiology and cardiology and dermatology.

BIG NEWS FOR AI PATHOLOGY ("CMAA")

There is a "discussion only session" (tab 94) on "clinically meaningful algorithmic analyses" or CMAA that do not require physician work.   (AMA has been discussing where to put "dry lab” meaning AI of whole slide imaging, and this session may be relevant.)

PLA CODES

Codes under review by the PLA committee are posted here:

https://www.ama-assn.org/system/files/august-2025-pla-public-agenda.pdf

By September, the PLA committee will have finished its reviews and edits and these codes will be voted on (briefly) at the Chicago CPT.

More About CMAA 

Several years ago, AMA created Appendix S, which provides a taxonomy for AI-assisted services.  Specifically, it defines "Assistive" services, "Augmentative" services (both of which are signed off or signed out by a physician), and "Autonomous" services, requiring no physician sign-out.  (The autonomous category has 3 sublevels, Levels I,II,III).

As I understand it, at the September AMA, the digital workgroup will present CMAA, Clinically Meaningful Algorithmic Analyses.   (aka, Tab 94).  I believe this could develop into an entirely new coding section which would have subchapters for cardiology tests, radiology tests, laboratory tests.  If that happened, it would be anomalous to have "dry lab" algorithmic codes in the PLA section and they would be located instead in CMAA (with radiology, etc).   This is all TBD but it is my impressioin so far.  I think CMAA slide diagnostics would be "autonomous" usually, no physician sign out.  Of course, if they did require physician sign-out, they would not be eligible to be PLA codes, anyway.

The "Agenda" PDF has a link that takes you to the AMA website manager for tabs and comments.  Tab 94 includes two documents, be sure and review both.


Tuesday, August 19, 2025

Congressional Budget Office: More Medicare Fee Sequestration

Given all the news, you'd be excused for meeting an AP article on Medicare budget cuts due to sequestration due to budget deficits.

See an August 15, 2025, AP article here.

The article builds on a May 20, 2025, letter from Congressional Budget Office to Repr. Boyle, Committee on the Budget.

##

Would a $2.3T budget deficit triggger more PAYGO sequestration?  Yes, a $2.3B ten year deficit would add $230B a year to the PAYGo scorecard.  "OMB...would be required to issue a sequestration order...to reduce spending by $230B in FY2026."

How would that affect Medicare?  Medicare reductions are limited to 4%, or $45B in 2026.  The other $185B would be sequestered from other federal direct spending.  Rising to $75B in 2023.



Saturday, August 16, 2025

NCI MATCH Final Concordance Study: High Accuracy Among 26 Labs

 NCI and Clinical Cancer Research publish a large-scale concordance study of tumor comprehensive genomic profiling across 26 NGS laboratories.   Results were very good.

See a blog by Howard McLeod here.

See the full paper by Zane et al., released August 15, here.

###
I have not seen the full paper, but I'll include the remarks and the abstract below.

####

McLeod

Howard McLeod, Center Director at Center for Precision Medicine and Functional Genomics, shared a post on LinkedIn:

  • “The National Cancer Institute (NCI) MATCH study gave an opportunity to assess concordance of comprehensive genomic profiling across 11 commercial and 14 academic laboratories (with the NCI central lab being the standard). 
  • As many of us have seen in practice, high concordance for SNV, insert, deletion, and copy number variants in therapeutically relevant driver genes for NGS across the labs (not as good for hybridization capture assays). 
  • Gives confidence for using the NGS lab that gives you the genes you need, at the turn around time that is needed.”
  • Title: A Concordance Study among 26 NGS Laboratories Participating in the NCI Molecular Analysis for Therapy Choice Clinical Trial Available to Purchase
####
Zane

Purpose:
NCI selected a network of Clinical Laboratory Improvement Amendments–certified laboratories performing routine next-generation sequencing (NGS) tumor testing to identify patients for the NCI Molecular Analysis for Therapy Choice (NCI-MATCH) trial. This large network provided a unique opportunity to compare variant detection and reporting between a wide range of testing platforms.

Experimental Design:
Twenty-eight NGS assays from 26 laboratories within the NCI-MATCH Network, including the NCI-MATCH central laboratory (CL) and 11 commercial and 14 academic designated laboratories (DL), were used for this study. DNA from eight cell lines and two clinical samples were sequenced. Pairwise comparisons in variant detection and reporting between each DL and CL were performed for single-nucleotide variant, insertion and deletion, and copy-number variant classes.

Results:
We observed high concordance in variant detection between CL and DL for single-nucleotide variants and insertions and deletions [average positive agreement (APA) > 95.4% for all pairwise comparisons] but lower concordance for variant reporting after analysis pipeline filtering. 

We observed much higher agreement between CL and assays using amplification as the target enrichment method (84.2% < APA ≤ 95.7%, average APA = 88.7%) than other assays using hybridization capture (69.7% < APA ≤ 93.8%, average APA = 77.4%) due to blacklisting of actionable variants in low complexity regions. 

For copy-number variant reporting, we observed high agreement (APA > 82%) except between CL and two assays (APA = 76.9% and 71.4%) due to differences in estimation of copy numbers. Notably, for all variants, differences in variant interpretation also contributed to reporting discrepancies.

Conclusions:
This study indicates that different NGS tumor profiling tests currently in widespread clinical use achieve high concordance between assays in variant detection. For variant reporting, observed discrepancies are mainly introduced during the bioinformatic analysis.

Friday, August 15, 2025

AMA CPT - Detailed Deck on PLA Code Rules and Processes

For PLA Codes, find the AMA's 36-slide deck and FAQ at links provided.

###

There are now over 600 AMA CPT PLA codes (proprietary laboratory analyses), and more are issued quarterly.

PLA codes are issued both for single-lab proprietary tests, and for distributed FDA approved or cleared tests.

Unfortunately, there is no master categorization of PLA codes: they are issued serially.  So there's no index to which are genomic sequencing procedures, which are minimal residual disease, which are pharmacogenetic, and so on.   

AMA CPT Speaker at Conference

On August 18, 2025, in Washington, hear two talks about lab codes to kick off the Next Generation Dx Summit, reimbursement track.  On that Monday, Zach Hochstetler of AMA CPT Editorial & Regulatory Services, describes "Critical Updates to Lab CPT Codes."   He's followed by Victoria Pratt (Indiana School of medicine and AMA PLA committee), on "PLA Codes and the CLFS Rate Setting Process."

Find agenda and registration here.

###

There's a home page for all things PLA at the AMA CPT - here.  

I was looking for some updates and ran across a detailed 36-page slide deck by AMA on PLA, as well as a detailed FAQ.

The 36-page deck is here.  The FAQ is here.


sample slide


###

I noted in a July blog that, by inspecting PLA codes released over the last several quarters, they seemed to stop issuing digital-only lab PLA codes (blog here).  For example, you could track some that were being applied-for but not finalized.  I didn't see any direct confirmation of that in the AMA instructions.

Monday, August 11, 2025

Brief Blog: RFK Jr Requests Journal Retraction - Finding the Links

You may have heard that HHS has asked a top journal to retract a recent paper.  Find the links below.

The original study (Andersson) mined the records of 1.2M persons vaccinated over 24 years for certain risk/benefit issues.

  • See an August 11, 2025, Reuters article on the journal retraction issue, by Erman & Rigby, here.
  • See the August 1, 2025, Op Ed from HHS leadership here.
    • See an August 7, 2025, entry point to additional debate at TrialSiteNews here.
  • See the original Andersson et al. at Ann Intern Med here.   See July coverage at Stat here.

Ted Cho and Brian Miller: CMS and the (Innovation) Valley of Death

Last winter, Brian Miller and Ted Cho published a 15-page white paper on the Innovation Valley of Death.  Here, here.  American Enterprise Institute; Cho is at UCSF, Miller at Hopkins.

In July, Miller announced a two-part series on CMS & Valley of Death, followed by FDA & Valley of Death.   Linked In here.  Find Part 1, Coverage Reform, online at Hoover Institution here.

The authors' key takeaways are three:

  1. Delays in Access to Innovation: Differing regulatory standards between FDA approval and Centers for Medicare and Medicaid Services (CMS) coverage delay access to the life-changing innovations in medical devices and health technology over the past several decades, with a median 5.7-year lag.
  2. Prior Regulatory Reform Efforts Have Lagged: For 30 years, policymakers and regulators have introduced advisory committees, new coverage pathways, and other process improvements, but access remains limited—underscoring the failure of these reforms.
  3. Urgent Need for CMS Reform: Policymakers and CMS can provide clarity of coverage principles, improve local coverage determination processes, and expand transparency to facilitate access to innovation.
What Do I Think?

I like the paper, the body of which is only about 8 pages.  It gives a good history and overview.   It notes the decades of attempts at improvement and reform, and it notes the stasis of the once-lauded MEDCAC process.  

I'd add that when CMS does hold a MEDCAC, the voting questions are almost always so vague that the panel wastes a lot of time, inconclusively, debating what is meant.  There are a host of reasons that innovation gets imperiled, one of the worst being the years of delay before truly novel NCDs or LCDs appear.   I track the CMS policy website weekly and truly new LCDs are very, very rare.   

As far as CED goes, I see little evidence of benefit, and a new complex (byzantine) CED proposal for renal denervation for hypertension looks very messy to me (blog).   CMS does only 4 NCDs per year, yet if its TCET plan was fully implemented, it would be doing 12 (4 NCDs, 4 TCET NCDs, and 4 TCET NCDs coming up for re-appraisal.)  No way.

Forgotten but Important:  Death by Coding
Discussions of reimbursement and innovation generally go in this direction - coverage policy.  However, there is also the less-remarked on area of "death by coding."   While I haven't check these two areas lately, for many years, there was one HCPCS code for "insulin pump," one HCPCS code for "sleep apnea pump."   No matter if you developed one that was twice as clinically effective, twice as high compliance, twice as important new features, there was no  incentive - there was one code, one price for the field.   This also can be an important disincentive to innovation.

###

AI Corner

###

Here's an AI summary (Chat GPT 5).

The Hoover Institution brief “Crossing the Valley of Death, Part 1: The Innovation Imperative and Medicare Coverage Reform” describes the persistent gap between FDA approval of new medical technologies and Medicare coverage sufficient to ensure patient access—a delay averaging about 5.7 years. This “Valley of Death” arises because innovators face two sequential regulatory hurdles: FDA’s safety and efficacy review, followed by CMS’s coverage determination under its “reasonable and necessary” standard. The agencies operate under different statutory mandates, resulting in misaligned processes and timelines that slow adoption of innovations and reduce their potential impact on patient care.

The authors frame the case for reform as both an innovation imperative and a moral obligation. While 45 percent of Medicare beneficiaries experience impairments in activities of daily living and over 6,000 rare diseases lack treatments, the United States has a strong history of medical innovation—producing more than 1,200 new drugs in the past 60 years and transforming care for conditions such as HIV and heart disease. Despite these successes, many conditions still impose enormous human and economic costs, such as insulin-dependent diabetes, which affects millions and drives billions in health care expenditures. The authors argue that U.S. policy should focus on removing inefficiencies that hinder timely adoption of proven innovations.

Medicare coverage reform has been attempted before. In 1999, under Administrator Nancy-Ann DeParle, CMS restructured its coverage analysis group and created the Medicare Evidence Development & Coverage Advisory Committee (MEDCAC). In 2021, the Trump administration introduced the Medicare Coverage of Innovative Technologies (MCIT) rule, which granted automatic four-year coverage for FDA-designated breakthrough devices. The Biden administration repealed MCIT in 2022 and replaced it with the Transitional Coverage for Emerging Technologies (TCET) pathway, which accepts only five devices per year and excludes diagnostics. In 2023, legislation (H.R. 1691) was introduced to restore MCIT-like transitional coverage.

The current coverage process includes National Coverage Determinations (NCDs), Local Coverage Determinations (LCDs) by Medicare Administrative Contractors (MACs), and Coverage with Evidence Development (CED).   

(BQ: See Joe Grogan's WSJ Op Ed, nix CED, from June 2025.)  (See also a WSJ op ed, nix CED for amyloid drugs, August 2025.)

MEDCAC’s activity has declined sharply in recent years, and the number of NCDs has dropped from a peak of 20 in 2003 to just four in 2022. CED, intended to accelerate coverage while gathering evidence, often stretches for years, drains resources, and rarely leads to full coverage—only three of 26 CED programs since 2005 have transitioned to routine coverage. The authors recommend a moratorium on new CED programs until CMS establishes defined milestones, completion criteria, and timelines. They also criticize CMS’s increasingly skeptical stance toward drugs granted FDA accelerated approval, noting that states like Oregon have sought to exclude such drugs from Medicaid coverage.

A central policy concern is the lack of a statutory definition for “reasonable and necessary.” CMS currently defines this administratively as safe and effective, not experimental or investigational, and appropriate for the Medicare population. Without a national definition, MACs apply their own interpretations, leading to inconsistent access based on geography. The authors recommend codifying broad evaluative principles in rulemaking, clarifying the division of responsibility between NCDs and LCDs, and maintaining local flexibility for positive customization.

Finally, the report proposes improvements to the LCD process, including imposing transparent timeframes for key steps, enhancing CAC membership with MEDCAC experts, posting CAC membership and technical questions at least 30 days before meetings [*], and holding annual CMS public meetings with MACs to review policy and operational lessons. The authors conclude that innovation is both an economic and moral imperative, and that streamlining, standardizing, and making coverage decisions more transparent can reduce inequities, improve access, and provide innovators with greater certainty, all in a way that should attract bipartisan support.

##

* Noridian has posted a meeting on molecular cancer tests, but by 8/11 offered no meaningful agenda for the 9/4 meeting.

HHS to Nix the LDT Final Rule

 At Linked In, Karin Hughes is one of the first to report that Executive Order #12866 will be used to reverse the FDA LDT rule.  It will formally retract the FDA LDT rule, which lost in federal court in Texas back in May 2025.

It looks to me like the document was "received August 6" but may be in regulatory review, it may be titled but not released.  It will be interesting if it just briefly retracts the LDT rule or if it elaborates on a new agency direction.  

https://www.reginfo.gov/public/do/eoDetails?rrid=1044563

The title is, Medical Devices; Laboratory Developed Tests; Rescission, and the status seems to be "pending."

The "meetings" tab shows no related meetings.

##

EO12866 is not new; it represents a 1993 regulatory framework.   The new document would, I think, be a new document or rule issued under the authority of 12866.  I think FDA Law Blog also indicates the rule is written but under review as of August 11.

https://www.reginfo.gov/public/jsp/EO/eoDashboard.myjsp



See comments at Linked In by Christine Bump here.


Prior Authorization in Fee for Service Medicare: More Follow-up

On July 8, 2025, I had a blog discussing the CMMI plan to introduce Prior Authorization into Medicare Part B.   Prior-auth companies would bid for the right to interpose their Prior Auth software between the provider and the MAC.   I noted (and you had to dig for this) that the Prior Auth companies would be paid based on charges avoided (ie, the volume of denials they could generate.)

I also noted that the PA computers were supposed to simply implement the rules for existing LCDs/NCDs, but that some of the topics to be authorized actually had no meaningful guidance at CMS.

Some follow-up a month later.

At Linked In, Brian Miller highlights the potential problems, pointing us to a new Op Ed by Donald Berwick and Andrew Ducas.

See the STAT article by Berwick & Ducas here.

See also an opinion piece by Jessica Hall at Marketwatch here.


Kerry Hannon, at Yahoo Finance, August 17 here.    

Highmark on real time POC Prior Auth (collab with Abridge, here.)    


Note that if you look at sources like CMMI leader Abe Sutton's linked in channel, there are also lots of parties weighing with strong endorsements of WISER.   

Saturday, August 9, 2025

MolDx Greenlights "Pathlight," the New MRD Test from Saga Diagnostics

On August 5, 2025, we highlighted Decibio's latest update on MRD coverage (here).  I mentioned late-breaking coverage by MolDx of the SAGA "Pathlight" MRD test.  Here's some more.

###

  • SAGA Diagnostics is a spin-out from Lund University in Sweden.  See an article at Lund University, June 2025, here.
  • See a July 30 press release from SAGA about its MolDx coverage here.  
    • The coverage is for all breast cancer subtypes and  recurring monitoring (surveillance) for "up to six years."
    • See coverage at Genomeweb.
  • See a Linked In article by Achyut Saroj on August 8, here.  
    • The Linked In article includes an embedded Elliott article.
  • See the full article by Elliott et al. in Clinical Cancer Research here.  
    • While the webpage is dated April 14, internal text says editorial approval was February 7 and online publication February 9.  
    • February would have been the earliest date for a MolDx coverage submission, which requires MRD publication(s).
  • The executive chairman is Roopom Banerjee.  He holds a Master's from the Harvard Kennedy School and  has been on the board since 2021.

###

MolDx DEX Registry lists the lab as Morrisville, NC.  It lists Pathlight (their only test entry on MolDx as of now) as $3652 and "covered."  I suspect that is for an initial test plus set-up genomics, but I can't be sure.   

  • dPCR vs NGS and CMS Policy
    • Since the recurring test is digital PCR rather than NGS, I think it falls outside of restrictions of CMS NGS NCD 90.2.
  • The test is described as genome-informed.  
    • Natera recently got MolDx coverage for a genome-informed Signatera test.  So this is a trend.

As of August 9, the MolDx DEX descriptor is:

  • The test is a dPCR-based, tumor-informed liquid biopsy assay that is designed to detect circulating tumor DNA (ctDNA) in the blood for the purpose of detecting molecular residual disease (MRD), monitoring treatment response, or detecting cancer recurrence.  
  • The test entails whole genome sequencing of a patient's tumor specimen, which is then used to generate a customized multiplex dPCR assay (fingerprint) for up to 16 structural variants (SVs).  This unique patient fingerprint is then used to detect the presence or absence of ctDNA. 
  • The test is intended to be used for patients diagnosed with solid tumor malignancies, initially focused in breast cancer (all subtypes), for multiple purposes, including neoadjuvant treatment monitoring, post-surgical MRD detection, adjuvant treatment monitoring, and post-treatment surveillance for recurrence.

###

The DEX text includes "adjuvant treatment monitoring," and Elliott et al. has some data on this, but the press release on coverage doesn't seem to directly include "adjuvant treatment monitoring" (or for what chemotherapy classes).

###

On July 19, I published a popular blog on a new proposed MolDx LCD about DCIS breast cancer test coverage (non-coverage), suggesting it might be used as a "teaching case study" for how MolDx thinks.   

You can't say the same for the Pathlight coverage, "how MolDx thinks."  MolDx releases no info on its thought process other than (1) the original, four-year-old LCD and the (2) binary status of "covered" for Pathlight at DEX registry..  

However, you can on your own put together coverage statements in the press release from the company, with a detailed view of the Elliott et al. publication.  This lets you infer what MolDx found acceptable.

  • Analytical performance accurately classified 217 tumor and 217 control cases.  
  • Breast-cancer retrospective performance had 100 cases, across several BC types, and found ctDNA ahead of known distal recurrence 19/19 recurrences.  (This includes postsurgical positives).

##


###

###

AI CORNER

###

Google Notebook LM writes a single 400 word article based on the links provided.

Direct AI output.  Do not quote without checking details against original sources.

###

SAGA Diagnostics has launched Pathlight MRD, an ultrasensitive liquid biopsy test designed to transform cancer care by enabling early detection of cancer recurrence. Originating from research at Lund University in Sweden, this groundbreaking test is now commercially available in the U.S. and has received Medicare coverage.

Unlike many other circulating tumor DNA (ctDNA) assays that primarily target single-nucleotide variants (SNVs), Pathlight focuses on structural variants (SVs). SVs are stable, highly tumor-specific genomic changes that reflect underlying genomic instability and provide a clearer, earlier indication of disease recurrence. This approach allows for precise molecular residual disease (MRD) tracking, especially in ER-positive (ER+) low-shedding tumors, where other assays often fall short. The assay uses whole genome sequencing (WGS) of tumor tissue to create a customized "fingerprint" of up to 16 structural variants for each patient, which is then used in digital PCR (dPCR) to detect ctDNA in blood samples.

Clinical validation, primarily through the cTdna evaluation in eaRly breAst canCER (TRACER) study published in Clinical Cancer Research, demonstrated remarkable performance. Pathlight achieved 100% sensitivity and 100% specificity in detecting early breast cancer recurrence. It provides a median lead time of 417 days (or 13.7 months) before recurrence is detected by standard clinical methods, including imaging. The test also boasts a high baseline detection rate of 96% across all early-stage breast cancer subtypes, notably 94% in ER+ disease. Furthermore, clearance of ctDNA during neoadjuvant therapy was associated with improved outcomes, enhancing residual cancer burden prognostication.

The Pathlight MRD test is now covered for Medicare beneficiaries for recurrence monitoring in the surveillance setting for up to six years in patients with stage II-III breast cancer, including all subtypes (HR+/HER2-, HER2+, and TNBC). This coverage decision, issued by Palmetto GBA’s MolDX Program, reflects a rigorous review of Pathlight’s analytical and clinical validity. The currently listed cost on the MolDX DEX Registry is $3652. This development aims to empower patients and clinicians with ultra-sensitive tools for more personalized and informed treatment decisions, moving towards smarter surveillance and long-term monitoring programs.

###

Is the company a "spin-out?"

Yes, SAGA Diagnostics is a spinout company from Lund University in Sweden.

Specifically:

  • It was founded in 2016 from research conducted at Lund University.
  • SAGA Diagnostics is described as a "portfolio company of LU Ventures" and "a brilliant example of how research from Lund University can reach the world and improve healthcare," having started as a spinout with support from LU Ventures and LU Innovation.
  • While its origins are in Sweden, SAGA's headquarters and CLIA-certified laboratory are located in Research Triangle Park, North Carolina, U.S..
###
Summarize Elliott et al; do not just repeat the abstract, we have that.

The "Ultrasensitive Detection and Monitoring of Circulating Tumor DNA Using Structural Variants in Early-Stage Breast Cancer" study, published in Clinical Cancer Research by Elliott et al., describes the development and validation of a novel circulating tumor DNA (ctDNA) assay, Pathlight, that utilizes structural variants (SVs) for molecular residual disease (MRD) detection in early-stage breast cancer (EBC).

The study addresses the need for improved biomarkers in cancer diagnosis, treatment selection, and monitoring, moving beyond invasive tissue biopsies to liquid biopsies. Unlike many existing ctDNA assays that primarily target single-nucleotide variants (SNVs), this study highlights the potential of SVs, which are stable, highly tumor-specific genomic changes reflecting underlying genomic instability. The Pathlight assay workflow involves whole genome sequencing (WGS) of tumor tissue to generate a personalized "fingerprint" of up to 16 SVs per patient, followed by multiplex digital PCR (dPCR) analysis of plasma-derived cell-free DNA (cfDNA) to detect ctDNA.

Analytical validation demonstrated the assay's ultrasensitivity, achieving a 95% limit of detection (LoD95) of 0.00052% (5 parts per million) with standard cfDNA inputs, and even lower (0.00011% or 1.1 ppm) with higher inputs. It also showed 100% analytical specificity across 217 cancer-free donor samples and 217 genomic DNA samples, testing 5,268 SVs.

In a retrospective clinical evaluation of 100 EBC patients (28 TNBC, 33 ER+, 39 HER2+) undergoing neoadjuvant systemic therapy, the assay successfully generated fingerprints for 95% of participants. Key findings include:

  • High baseline ctDNA detection: 96% of evaluable baseline samples showed ctDNA, with similar rates across all receptor subtypes, including 94% in ER+ disease where other assays often struggle.
  • Prognostic value of on-treatment dynamics: Clearance of ctDNA prior to cycle 2 of neoadjuvant therapy was associated with improved outcomes, and persistent ctDNA detection enhanced prognostication when combined with residual cancer burden (RCB) status.
  • Early detection of recurrence: ctDNA was detected prior to distant recurrence in 100% of cases (19/19) with evaluable samples, providing a median lead time of 417 days (13.7 months) before clinical detection by standard methods. Any ctDNA detection after surgery or during follow-up was strongly associated with recurrence.

The study concludes that this novel SV-based approach offers sensitive and specific ctDNA detection for treatment monitoring and prognostication in EBC. The uniqueness of SV breakpoints minimizes false positives and the selection of amplified SVs enhances sensitivity. While acknowledging limitations such as its retrospective nature, the study supports prospective trials to evaluate ctDNA-guided treatment strategies.