Evaluating Evidence is Subjective: In JAMA This Week

Header: JAMA Health Forum (open access) has a lead article, "The Subjective Interpretation of the Medical Evidence," by Bauchner and Ioannidis (BU and Stanford, respectively.)

##

https://jamanetwork.com/journals/jama-health-forum/fullarticle/2816950

Opening paragraph:

Experts often subjectively disagree on how they interpret the same evidence and what recommendations they derive from it. Meticulous processes to resolve diverging views in guideline development efforts, for example, may not remove subjectivity. Even the most prestigious organizations sometimes have different guideline recommendations. Subjective disagreements can be common, extreme, and unsettling when evidence is limited and rapidly evolving—as in many questions related to COVID-19. However, subjectivity exists, and differences ensue even for common diseases where evidence has accrued and been evaluated for decades. For example, the American College of Physicians, the American Cancer Society, and the US Preventive Services Task Force (USPSTF) vary on when to initiate screening for colorectal cancer and the preferred screening methods. Breast cancer and depression screening recommendations have been debated for decades.

Closing:

Therefore, the public should be reminded that uncertainty—and even the extent of uncertainty—is common. This realization is not paralyzing; medical decisions can still be made with the best available evidence, once values, preferences, and biases are carefully vetted.

##

Cited by JAMA alongside this article, include:

• NCCN Guidelines for accelerated approval cancer drugs.https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2811820
• What do we know about LDCT lung cancer screening after 20 years?https://pubs.rsna.org/doi/10.1148/rycan.2020190058
• Bauchner had a separate article on the same topic, in JAMA Open Network 2023, here. 
• It was tied to articles about hype in abstracts about research, even when scoped down to NIH funded studies.

###

AI CORNER with Chat GPT 4

AI summarizes 7 takeaways of Bauchner & Ioannidis.

1. Subjective Disagreements in Medical Guidelines: Experts and guideline committees often subjectively interpret the same medical evidence differently, leading to divergent recommendations even for well-studied conditions such as colorectal cancer and breast cancer screening.
2. Specialist vs. Generalist Perspectives: Specialists and generalists may value different aspects of evidence, affecting guidelines. For instance, the USPSTF's recommendations may differ from those of specialist societies due to different evaluation criteria and emphasis on certain types of evidence, like randomized clinical trials.
3. Impact of Legal and Policy Considerations: Recommendations may be influenced by their potential impact on insurance coverage and policy, with entities like the USPSTF being particularly aware of the implications of their guidance under laws like the Affordable Care Act.
4. Methodological Differences in Guideline Development: The approach to developing guidelines, including how evidence is synthesized and evaluated, varies significantly among organizations, contributing to inconsistencies. Techniques like the GRADE approach aim for reproducibility but still encounter variability.
5. Financial and Ideological Influences: Financial support from the industry and allegiance bias among guideline committees and advocacy groups can shape recommendations, with contentious areas such as COVID-19 and cancer screening being particularly susceptible.
6. Role of Subjectivity in Weak Evidence: In areas where evidence is limited, subjective opinions heavily influence recommendations, highlighting the challenge in balancing emerging evidence with existing treatment options, such as in the management of obesity in adolescents.
7. Misinformation and Public Trust: The subjective interpretation of medical evidence extends beyond professional circles to media and public discourse, contributing to misinformation, polarized views, and a decline in trust in medical science. This underscores the need for clear communication about the uncertainty and quality of evidence behind medical news and guidelines.

GPT4 illustrates Millar 2023 on Hype (Cited in Bauchner 2023 JAMA Open Netw)

Brief Blog: CMS Creates a New ADLT Test (Naveris NavDx, H&N Cancer)

Header: After creating just 4 ADLT tests in 2022/2023, CMS opens the first quarter 2024 with a new ADLT test, from Naveris, for Head & Neck cancer.

##

Background

Just last week, I wrote an update on the history of ADLT tests, a special pricing category created by Congress in 2014.  Iin a decade, there were only about 15 ADLT tests created.  ADLT's must be sole source tests that are either FDA approved (like Foundation Medicine F1) or covered tests of the MAAA type that provide "unique clinical information."  The update blog included at bottom an elaborate chart of all things ADLT.   Here.   Read about the elaborate application process for ADLTs, here.

New ADLT Code

On March 18, 2024, CMS created a new ADLT, code 0356U, NavDx from Naveris.  PDF here.  Classification as ADLT removes 0356Ufrom the current year gapfill process.  (Gapfill had been requested by Naveris last summer, endorsed by the CMS expert panel, and decided by CMS).

The price the first nine months starting April 1 will be $1800.  After that, the price will reset annually to the median of NavDx commercial payor prices.

Ins and Outs

According to my December 2023 blog, at the time Naveris announced LCD coverage under the MolDx MRD LCD, MolDx's online directory "Dex" listed the locally-set price at $900.  Here.  That would have been the result of a spreadsheet and rulebook that MolDx calls its "EPM, Equitable Pricing Model."  The price I see in DEX today (March 25, 2024) is $827. (Click to enlarge).   I believe that on April 1, MolDx will delist that local price and instead say "See CPT code" which means the CMS CLFS fee schedule which will in turn display the $1800 just-set ADLT price.

click to enlarge

DEX: Not Just a Code Directory

Note that the DEX page includes not merely a description of the test, but a summary of its Medicare-covered use(s).   Here, the test is covered as an MRD test both for surveillance (if it recurs) and to detect if MRD shows "residual" cancer, presumably (I think) at a timepoint that is shortly post-surgical or post-{XRT-chemo}. 

DEX: Oncology (oropharyngeal and anal), cell-free DNA, droplet digital PCR to profile the fragmentation pattern of tumor tissue modified viral (TTMV) HPV DNA using 17 DNA biomarkers, weighted fragment size distribution algorithmic analysis, whole blood, algorithm reported as a prognostic TTMV risk score for cancer recurrence. (NavDx(r), Naveris Inc.) The intended use of the test is for monitoring patients with previously diagnosed HPV-driven head and neck cancers for recurrent and or residual HPV-driven oropharyngeal cancer.

##

ddPCR:  Big Impact on CMS Policy!

This is one of relatively few tests for MRD that use "droplet digital PCR."  Because it does not use next generation sequencing, MolDx isn't bound by the outdated rules against serial testing of the CMS NGS NCD (NCD 90.2).   I discussed how outdated 90.2 is, in the context of a new massive ARPA-H grant, recently here.

Here's the AMA PLA code text:

Oncology (oropharyngeal) evaluation of 17 DNA biomarkers using

droplet digital PCR (ddPCR), cell-free DNA, algorithm reported as a

prognostic risk score for cancer recurrence.

###

• News sources for ADLT at Genomeweb.  
• Naveris ADLT press release here.
• Coverage at Genomeweb last fall re the then-new coverage at CMS here.  
• Naveris recently announced a large-scale H&N cancer registry, here.   

For anyone who thinks there is a quick one-and-done pathway to Coverage, see the extensive listing of Naveris clinical accuracy publications here.

Very Brief Blog: FY2025: Budget in Brief for HHS

Header:  Congress just closed the FY2024 budget authorization (due last fall.)  Now, proposals for the FY2025 budget for this coming October are circulating.  What the Administration plans for HHS divisions.

##

The Biden administration, looking ahead to the budget year October 2024 to September 2025, foresees $1.7T in HHS mandatory budget authority and $130B in discretionary spending.  Mandatory spending is, e.g., Medicare patient benefits.

Find it here:

https://www.hhs.gov/about/budget/fy2025/index.html

The "Budget in Brief" is a formidable 182 pages.

https://www.hhs.gov/sites/default/files/fy-2025-budget-in-brief.pdf

The budget for CMS is here:

https://www.cms.gov/about-cms/performance-budget/current

At CMS, see in particular the FY2025 327 page PDF:

https://www.cms.gov/files/document/fy2025-cms-congressional-justification-estimates-appropriations-committees.pdf

The CMS budget opens with an org chart updated to March 1, 2024.  The Center for Medicare is budgeted for 820 staff, up from 780 (p 228 actual pdf page).   The CMMI is listed with a 0 staff on this page but 626 on page 325.   On page 36, CMS notes that its operations budget has increased by 3% while there was a 23% loss of purchasing power due to inflation.   Budget for parts A and B (p. 37) is about $900M.   

I noted recently that ARPA-H is worth tracking closely for its groundbreaking $400M cancer biomarker discovery program, here.   FInd the ARPA budget page here, including the new 2025 budget plan here (51pp).  The molecular evolution program is at ARPA-H page 35, where it is called the ADAPT program (ADvanced Analysis for Precision Cancer Therapy).

##

60-page OIG budget here.

Resources: The 3-Hour Congressional FDA LDT Hearing on March 21, 2024

On March 21, 2024, the House Energy & Commerce committee held a 3-hour hearing on the FDA's plan to regulate LDTs.  Here are some resources, including a Zip file with all major documents.

###

The three hour session is online at YouTube here:

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

(The YouTube program notes give access to an online YT auto transcript).

The House home page for the hearing is here:

https://energycommerce.house.gov/events/health-subcommittee-hearing-evaluating-approaches-to-diagnostic-test-regulation-and-the-impact-of-the-fda-s-proposed-rule

(Click on Hearing Announcement, Hearing Memo, and View Agenda).

Presentations included 

• Susan Van Meter, President, ACLA
• Zach Rothstein JD, AdvaMed Dx
• Donald S. Karcher MD, President, CAP
• Jeff ALlen PhD, President & CEO, Friends of Cancer Research
• Dara L. Aisner MD PhD, "Academic Coalition for Effective Laboratory Tests" (academic LDTs)

An excellent rapid online summary was posted by Joe Lennerz MD PhD of BostonGene, at Linked In.

https://www.linkedin.com/feed/update/urn:li:activity:7176512644672036864/

Note that Lennerz's article includes an 11-slide summary that includes each speaker's main points ("Summary of Witness Testimony").

See subscription reporting at Genomeweb here.

####

AI CORNER - TESTMONY IN BRIEF

Here is a Chat GPT summary of the 3 adverse comments, ACLA, CAP, Acad LDTs:

https://docs.google.com/document/d/1c6_NstX3exeYDmoEjQ1Z1CLx4V1qxZapI7-MBCf3HcU/edit?usp=sharing

Here is a Chat GPT summary of the 2 favorable comments, FOCR, AdvaMed:

https://docs.google.com/document/d/1N3_UFFyJ_WZq8za_Tnq4ZRku5_vam7mTyd-2t7q1u60/edit?usp=sharing

####

AI CORNER - FULL 3 HR AUTO SUMMARY, AUTO TRANSCRIPT

Find a Google doc with an Otter.ai Auto-summary and next, the 60-page full Auto-transcript here:

https://docs.google.com/document/d/16yoaYZE2cpdq6pzpRYSDq_eXjzfEkgdkvIL98WTxaEk/edit?usp=sharing

Direct  shared access to Otter.ai auto transcript here:

https://otter.ai/u/H9L-Um_RvSARI_g_BNIEcNdROn8?utm_source=copy_url

###

EASY CLOUD ZIP FILE SUPER BUNDLE

I have included a ZIP file with all FDALDT testimony and reference documents, plus a 62 page auto summary and auto transcript.  Open access to the ZIP file here:

https://drive.google.com/file/d/1UZW0Vz7MIzlZIYcHo2g2QMOgQaHAmvT-/view?usp=sharing

##

OTHER

Chairman Cathy McMorris Rodgers(R-WA) gave opening remarks.

https://energycommerce.house.gov/posts/chair-rodgers-opening-remarks-on-the-regulation-of-diagnostic-tests

Subcommittee chair Brett Gurhrie (R-KY) gave opening remarks:

https://energycommerce.house.gov/posts/subcommittee-chair-guthrie-opening-remarks-on-the-regulation-of-diagnostic-tests

AI CORNER - Summary of "Both Opening Remarks"

Summary of Opening Remarks

Subcommittee Chair Guthrie's Opening Remarks on the Regulation of Diagnostic Tests

Subcommittee Chair Brett Guthrie opened the hearing by addressing concerns over the FDA's proposed rule to regulate laboratory developed tests (LDTs) as medical devices, potentially impacting access to reliable diagnostic tests for children and patients with rare diseases. He outlined the historical context of diagnostic test regulation, highlighting the split oversight between the FDA and the Centers for Medicare and Medicaid Services, and the 1988 Clinical Laboratory Improvement Amendments (CLIA) which aimed at ensuring test accuracy and reliability.

Guthrie emphasized the significance of LDTs in various medical and public health applications, including cancer treatment and drug trafficking monitoring. He criticized the FDA's attempts to reform LDT regulation, arguing that the proposed rule would limit labs' flexibility to modify tests, potentially leading to higher costs and stifling innovation without necessarily enhancing patient protection. He expressed concern over the rule's unintended consequences, such as increased consolidation among testing providers and adverse impacts on personalized medicine and disadvantaged populations. Guthrie called for Congressional engagement with the public health community to address diagnostic testing challenges without compromising innovation.

Chair Rodgers's Opening Remarks on the Regulation of Diagnostic Tests

Chair Cathy McMorris Rodgers critiqued the FDA's proposed rule on LDT regulation for potentially stifling biomedical innovation and delaying patient care. She stressed the importance of LDTs, especially for rare diseases and certain cancers, and voiced concerns over the rule's impact on laboratory operations and costs. Rodgers detailed the burdensome nature of the proposed regulations, predicting significant financial strain on labs and restricted access to tests for vulnerable populations.

She contrasted the FDA's estimate of the number of LDTs affected by the rule with the agency's capacity to review premarket submissions, suggesting the rule would lead to fewer diagnoses, higher costs, and delays in care. Rodgers called for a reevaluation of the rule and expressed interest in exploring legislative alternatives to the proposed regulatory approach, emphasizing the need to maintain America's leadership in biomedical innovation and ensure timely patient access to necessary diagnostic tests.

CMS Reveals: One ADLT Application in Three Fails

"ADLT" is a narrow, special class of lab test with special pricing.  About 15 of 23 applications for this status have succeeded.  About 1 in 3 goes belly-up.

##

What are ADLTs?

Created by the 2014 PAMA legislation, ADLT means advanced diagnostic laboratory test.  However, it's not "advanced" in  a general sense.  It's a very specifically defined category.  Applicable tests skip the crosswalk-gapfill pricing process and are paid list price, and then, an annually reset market survey price.

See an ADLT CMS web page:

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

Two Types of ADLTs

ADLTs are always sole-source tests.  Meeting that rule, they may be (A) multiple analyate tests with algorithms that are covered by Medicare, and are clinically unique; or (B) FDA cleared or approved tests.

There are about 15 ADLT tests created over five years, via an application to CMS and CMS approval. (See web page above).

What's New: The Pass Fail Rate Revealed

You can't tell how many applications for ADLTs that CMS has received.  At least 15, since 15 succeeded.  But 15 out of how many?   20?  30?  40?   No way to know. (*).   

At the American Clinical Laboratory Association meeting, March 14, 2024 in DC, CMS official Jason Bennett spoke about the process.  He's director of the "Technology, Coding, and Pricing Group" at CMS.  He described the ADLT procsss and remarked, there have been about 22 or 23 total applications.  That means the failure rate is about 1 in 3.   The most common cause of failure is not proving the test provides unique clinically valuable information.   The sole source lab requirement can also trip up some applicants.

Timetable Starts to Slip

CMS promised to turn around the ADLT application, which merely reviews a few rules, in a month or so.  However, lately, some applications have taken over a quarter.  Bennett mentioned this and attributed it to short staffing and budget caps.

###

(*) In 2023 I requested copies of all the applicaitons (turns out, there are 22 or 23) via FOIA.  I haven't got this yet.   PAMA price information is shielded from FOIA, but as CMS remarked in PAMA rulemaking, ADLT applications are not (but some internal biz info could be blacked off).

Nerd Note

PAMA statute only required ADLT tests to be multi analyte, algorithmic tests, covered by Medicare. 

During ADLT rulemaking, CMS on its own added the requirement that an ADLT MAAA test must be clinically unique.   (That factor doesn't apply that to the FDA-approved ADLT tests).  

ADLT-ology

The table below (click to enlarge) shows ADLT codes, whether FDA approved, year of first price, current price (01/2024), price delta if any, and CMS Pt B payments in CY2022 (data from 9/2023).

click to enlarge

CY2022 Pt B payments for ADLT codes were $265M, of which 64% went to FDA-approved ADLT codes.   The three largest ADLT codes were over $50M and comprised 76% of the dollar volume for all ADLT codes.

(This concentration of most ADLT spending in the top 3 codes mirrors the concentration of most PPLA code spending in the top several codes and most Category III spending in the top several codes). See my August 2023 blog.

FDA Publishes Position and Plans on AI Regulation. And Plans for Internal AI, Too.

On March 15, 2024, FDA publishes a 7-page position paper on its plans for regulating AI, including a discussion of how different parts of the agency will work together. ALSO includes a vision for internal  use of AI at FDA (Califf blog).

##

See the agency's cross-departmental home page for AI, including the new 7p white paper and a blog by the Comimissioner - here.  The blog includes discussion of internal AI at the FDA.

• See a news article at ITN (Imaging Technology News) on the new white paper, here.  
• See the FDA's home page for device AI here.  See also the FDA "Digital Health Center of Excellence," DH-CoE.

• ##
• See an article in JD Supra about FDA and device-AI regulation, highly critical of the FDA's framework.  Dated 3/15/2024, but not dependent on the FDA's new memo.  Here.
• See an article on European AI healthcare regulation, dated 3/14/2024.  Here.
• See a Politico article on medical AI from February 2024, "The sheriff is missing in action."  Here.
• See a 2022 article in STAT+ on then-current issues for AI at FDA, here.

AI Corner

Here's a Chat GPT4 review of the 7 page document...

In a 7-page paper published by the Food and Drug Administration (FDA) in March 2024, the FDA lays out plans for collaborative efforts among its various centers to address the challenges and opportunities presented by Artificial Intelligence (AI) in the development and regulation of medical products. The FDA aims to foster responsible and ethical innovation in AI applications within the healthcare and medical product sectors while ensuring public health and safety. The document outlines the FDA's approach and initiatives in this regard, highlighting four main areas of focus:

Fostering Collaboration to Safeguard Public Health: The FDA seeks to engage with developers, patient groups, academia, global regulators, and other stakeholders to develop a patient-centered regulatory approach emphasizing collaboration and health equity. Efforts include soliciting input on critical aspects of AI in medical products and promoting educational initiatives.

Advancing the Development of Regulatory Approaches that Support Innovation: The document describes the FDA's intention to develop policies that provide clarity and predictability for AI's use in the medical product lifecycle. This includes monitoring trends, addressing knowledge gaps, and issuing guidance on AI in medical product development.

Promoting the Development of Harmonized Standards, Guidelines, Best Practices, and Tools: The FDA is committed to ensuring safety and effectiveness standards for AI-enabled medical products, including developing considerations for their safe and ethical use, promoting best practices, and developing frameworks for quality assurance.

Supporting Research Related to the Evaluation and Monitoring of AI Performance: The FDA plans to support demonstration projects that explore various aspects of AI in medical product development, including bias, health inequities, and performance monitoring.

The document emphasizes the FDA's commitment to the responsible and ethical use of AI in medical product development, underlining its mission to ensure access to safe and effective medical products. It acknowledges the rapid evolution of AI and indicates that regulatory approaches may adapt as the field progresses.

Take-Home Lessons:

Commitment to Ethical and Responsible Innovation: The FDA is focused on fostering responsible and ethical innovation in AI, recognizing its potential to revolutionize healthcare and medical products.

Collaboration and Engagement: Collaborative partnerships with a wide range of stakeholders are crucial to developing patient-centered and equitable regulatory approaches.

Adaptability and Forward-Thinking: Given the rapid advancements in AI, the FDA acknowledges the need for adaptable and forward-thinking regulatory approaches that can evolve with the technology.

Safety, Effectiveness, and Equity: Ensuring the safety, effectiveness, and equitable use of AI in medical products is a primary goal, with specific actions planned to address these aspects.

Definitiveness vs. Aspirational Nature:

The document outlines definite actions and intentions by the FDA to address the integration of AI in medical product development and regulation. However, it also carries an aspirational tone, acknowledging the evolving nature of AI and indicating that the FDA's approaches will adapt over time. The FDA positions itself as both a competent authority ensuring public health and safety and as a facilitator of innovation, aiming not to be an arbitrary barrier but to support the development and deployment of AI in a responsible and ethical manner.

And the Blog, Too

Dr. Robert M. Califf, the Commissioner of Food and Drugs at the FDA, in his blog post titled "Harnessing the Potential of Artificial Intelligence," discusses the FDA's proactive approach towards understanding and integrating Artificial Intelligence (AI) into its regulatory and operational framework. This blog complements the detailed white paper on the subject by providing a broader overview of the FDA's engagement with AI and emphasizing its importance in revolutionizing medical products and food safety.

Key Takeaways from Dr. Califf's Blog: Includes Internal AI Usage at FDA

1. Long-term Engagement with AI: The FDA has been anticipating the challenges and potentials of AI for years, indicating a strategic and forward-looking approach towards emerging technologies.
2. Impact on Medical Products and Food: AI is recognized for its potential to enable significant advancements in the development of safer, more effective medical products and more nutritious food. Dr. Califf highlights the extensive experience the FDA has with AI, noting over 300 submissions for drugs and biological products with AI components, and more than 700 for AI-enabled devices since 1995.
3. Facilitation of Internal Operations of FDA: The FDA is also exploring AI technologies to improve its internal operations and regulatory processes. This could lead to more streamlined workflows and faster delivery of high-quality medical products to patients.
4. Operational Efficiency of FDA: AI can enhance productivity and efficiency within the FDA by accelerating data processing and analysis, automating repetitive tasks, and allowing staff to focus on more complex and meaningful activities.
5. Commitment to Responsible AI Use: The blog reaffirms the FDA's commitment to promoting the responsible and ethical development, deployment, use, and maintenance of medical products developed with or incorporating AI, ensuring they are safe and effective.

##

Here's the blog quote on FDA, Internal AI, and efficiencies:

The FDA is also exploring the use of AI technologies to facilitate our internal operations and regulatory processes, which could benefit both agency experts and the public by streamlining workflows and facilitating high quality, novel medical products more quickly reaching the patients who need them.  

At its most basic, AI can strengthen our operational systems and bring increased productivity, opportunity, and efficiency to our work, helping us process and analyze complex data faster, including data from medical imaging or digital health technologies, for example. 

 

We can free up staff by automating repetitive administrative functions and enable them to focus on more complex meaningful activities to weigh the evidence and arrive at better decisions. Our workforce should also have more time to explain those decisions to the public and learned intermediaries in the biomedical and clinical world.

FDA LDT ON HILL: House Holds Hearing on LDT Regulation: March 21, 2024

Header:  House E&C committee holds hearing on FDA LDT on March 21, 2024.  Pair with Sen. Cassidy's request for input on FDA LDT, open til April 3.

##

The House Energy & Commerce committee will hold a hearing on FDA LDT issues, on March 21, 2024 at 10 am ET.  See details in press release. Expect to hear from ACLA, CAP, AdvaMed, Friends of Cancer Research.

https://energycommerce.house.gov/posts/chairs-rodgers-and-guthrie-announce-health-subcommittee-hearing-on-regulation-of-diagnostic-tests

In parallel, Sen. Cassidy (R) has an open request for comments on the Senate side; open til April 3.   More here:

https://www.discoveriesinhealthpolicy.com/2024/03/senate-requests-information-asap-on-lab.html

FDA LDT ON HILL: Senate Requests Information ASAP on Lab Regulation

 Senator Cassidy has requested public input on regulation of laboratory tests.  The deadline for written documents is April 3, but I've also heard there may be a hearing as early as March 21.  Stay tuned.

###

Update; House holds hearing on March 21 - here.

###

See the Senate HELP RFI press release here:

https://www.help.senate.gov/ranking/newsroom/press/ranking-member-cassidy-seeks-information-from-stakeholders-on-regulation-of-clinical-tests

See the full link and a detailed question list. See coverage at Genomeweb.

Here's a summary (AI):

U.S. Senator Bill Cassidy, ranking member of the Senate HELP Committee, has initiated a request for stakeholder feedback on reforming the regulation of clinical tests in the United States. Highlighting the critical role of clinical diagnostics in healthcare and the lack of significant regulatory updates since 1976, Cassidy criticizes the FDA's recent attempts at unilateral regulatory reform as overreach. He seeks insights to guide Congress in modernizing regulations to encourage innovation and maintain safety and effectiveness in diagnostics, including the use of AI and the expansion of test settings. Cassidy's request emphasizes the need for a balanced approach to oversight. Feedback is due by April 3, 2024.

NEJM: A Quartet of Papers about Non Invasive CRC Screening

 NEJM offers a quartet of papers on new tests for non invasive Colorectal Cancer screening.

• Chung et al. report a 10,000 patient study with specificity of about 90% and cancer sensitivity of about 83%-87% depending on definition.  
• Imperiale et al. report a next-generation of a stool test, 20,000 patients, with sensitivity and specificity both in the 90% range.
• Carethers of UCSD provides an op ed. 
• Lo of Chinese Univ Hong Kong provides a brief overview of cell-free DNA detection methods.

NYT here.  AP here.  Genomeweb here.   Medpage here.

##

AI Corner (GPT4): Summarizing the OpEd

[Sources like USPSTF also remark, like Carethers, that "the best CRC screening test is the one that gets completed."]

In his NEJM Op Ed, Dr Carethers discusses the evolution and effectiveness of noninvasive colorectal cancer screening methods. Historically, guaiac-based tests, despite their limitations such as poor sensitivity and the need for dietary adjustments, demonstrated benefits like early cancer detection and reduced mortality. The introduction of Fecal Immunochemical Testing (FIT) improved sensitivity and specificity without requiring diet changes, although annual testing is still recommended. The addition of multitarget stool DNA markers to FIT, as seen in the DeeP-C Study, further increased sensitivity for detecting colorectal cancer and advanced adenomas, albeit with a reduction in specificity and an increase in false positives, recommending a three-year testing interval.

The article highlights two promising studies: the BLUE-C Study, which evaluated a next-generation multitarget stool DNA test showing preserved sensitivity for colorectal cancer and improved specificity, and the ECLIPSE study, which explored a blood-based "liquid biopsy" showing notable sensitivity and specificity for colorectal cancer detection. These advancements aim to enhance screening adherence and effectiveness, especially given the below-target screening rates in the pre-COVID-19 era.

The op-ed emphasizes the critical role of screening in saving lives and the importance of selecting the most effective and accessible tests to increase screening participation and adherence. The best screening test is described as the one that gets completed, underlining the importance of ease of use, cost-effectiveness, and appropriate testing intervals to improve screening rates and reduce colorectal cancer mortality.

(I asked GPT4 for, "an interesting illustration, be sure it is interesting.")

##

In a sidebar, I asked GPT4 to adopt the voice of a journalist and summarize both studies.  I needed to ask for a re-write, which it also did.  Here.

##

NEJM text on Cologuard Tech:

The next-generation [stool] test incorporated a new molecular panel (including the methylated DNA markers ceramide synthase 4 gene [LASS4], leucine-rich repeat-containing protein 4 gene [LRRC4], serine–threonine protein phosphatase 2A 56-kDa regulatory subunit gamma isoform gene [PPP2R5C], and the reference marker zinc finger DHHC-type containing 1 gene [ZDHHC1], while retaining fecal hemoglobin). [from Imperiale]  

Per text in its NCD, Cologuard today is run as: Cologuard tests for two DNA methylation markers [NDRG4, BMP3], seven point mutations on K-ras [codons 12 and 13], quantitative DNA [β-actin], and fecal hemoglobin.  

##

AI-shortened ECLIPSE method, based on the long technical appendix.   

Plasma from each subject was pooled and sent for cfDNA extraction, which was then divided into methylated and unmethylated fractions based on methylation states. Unique dual barcodes were applied to DNA molecules for identification, with subsequent PCR amplification and enrichment targeting about 1Mb of the human genome.

The analysis distinguished between methylated and unmethylated cfDNA, focusing on specific genomic regions. A neoplasia detection algorithm, the Shield Test, evaluated thousands of features to identify cancer-specific signals, incorporating epigenetic changes and somatic mutations. Results were based on methylation-based tumor fraction regression (TFR) and an integrated cfDNA score, determining the sample's status as either positive (abnormal) or negative (normal).

The algorithms were trained on a diverse sample set, including healthy individuals, CRC-negative donors, and CRC patients, refining the detection of tumor-derived cfDNA. This involved quantifying aberrant methylation, assessing molecule counts, and employing logistic regression models to predict the presence of tumor-derived molecules based on methylation status, fragmentation patterns, and detected somatic mutations.

FDA Guidance for Early Alzheimer's Disease: Key Roles of Biomarkers

Recently, both LA Times (February 14) and NY Times (March 4) have had articles on the debates about how to define the use of biomarkers in early Alzheimer's disease, and at what point the term "Alzheimer's Disease" comes into play for early cases.  

FDA releases for comment a new 11-page draft guidance on early diagnosis of Alzheimer's by biomarkers and subtle cognitive testing.  This is in the spotlight because of the belief that only early intervention will be effective (supported also by subgroups in existing trials.)  The topic also comes to the fore because advances in proteomics make plasma biomarkers practical for the first time as opposed to PET scans or CSF tests.  

• Article on this shift, here.  
• See Ashton 2024 in JAMA, data on phospho-tau 217 in plasma.   
• pTau217 is newly in breakthrough-review at FDA.   
• See an excellent webinar on pTau217 sponsored by ALZPATH, here.

Find the FDA guidance here:

https://www.fda.gov/media/134605/download

Wisely, I think, FDA breaks early Alzheimer development into 3 stages, culminating in the earliest  symptoms.

• Stage 1 is no cognitive symptoms, not even on subtle testing, but Alzheimer biomarkers like tau and amyloid are rising.
• Stage 2 has biomarkers plus subtle neuropsych changes or subtle/mild complaints.
• Stage 3 has biomarkers, probably stronger signals, and detectable functional impairment.   

FDA calls these "stages of Early AD" so they're using the D word (disease) but clearly defining it as biomarker only (#1) or biomarker and subtle specialized testing changes (#2). 

I won't belabor it here, but if you're directly interested in this field, the guidance has many complex and  subtle points to make such as regarding functional and cognitive scales as endpoints at different stages, and how this shapes approval processes.  

Comment within 90 days.  Early coverage at RAPS here.

The guidance revises a 2018 draft that was not finalized.  Coverage at Genomeweb.

##

CMS Tie-In

Read about the CMS "GUIDE" demo care program for enhanced integrated care for Alzheimer patients in FFS Medicare.  Here.

##

See an article in Bloomberg, "Do you really want to know?" if you'll get Alzheimer's.

##

AI CORNER

From Chat GPT 4.

Let's open with the definitions of Stages 1, 2, and 3 for early Alzheimer's disease (AD) as provided in the FDA's guidance. These stages are identified based on the progression of the disease, marked by pathophysiological changes, and the emergence of clinical symptoms and functional impairment.

Stage Definitions:

Stage 1:

• Characterized by the presence of pathophysiological changes indicative of AD, without any clinical impact.
• Patients are asymptomatic with no subjective complaints, functional impairment, or detectable abnormalities on sensitive neuropsychological measures.
• The diagnosis is primarily based on biomarker evidence, demonstrating the characteristic pathophysiological changes of AD.

Stage 2:

• Patients exhibit the characteristic pathophysiological changes of AD and may have subtle detectable abnormalities on sensitive neuropsychological measures or subjective complaints of mild cognitive symptoms.
• However, there is no functional impairment. This stage can be considered a transitional phase where slight cognitive symptoms first become apparent.

The transition to Stage 3 is signaled by the emergence of subtle functional impairments.

Stage 3:

• Characterized by more apparent pathophysiological changes of AD, more evident abnormalities on sensitive neuropsychological measures, and mild but detectable functional impairment.
• The functional impairment at this stage is not severe enough to warrant a diagnosis of overt dementia. This stage roughly corresponds to the syndrome of "mild cognitive impairment."
• It is noted that the term "mild cognitive impairment" could also encompass patients in late Stage 2 or early Stage 4, highlighting the continuum of the disease progression.

These stages reflect the evolution of AD from its pathophysiological onset, through initial clinical manifestations, to the point where mild functional impairments begin to emerge, but before overt dementia develops. The FDA emphasizes the importance of accurately identifying these stages in clinical trials to ensure appropriate enrollment and the selection of relevant outcome measures that align with the specific stage of disease progression.

And Overview...

The FDA's draft guidance for developing drugs for the treatment of early Alzheimer's disease (AD) provides a comprehensive framework aimed at guiding sponsors through the clinical development process for drugs targeting stages of sporadic AD that occur before the onset of overt dementia (i.e., Stages 1 through 3). This guidance emphasizes the importance of considering the disease continuum and ensuring that clinical trials are appropriately designed to evaluate potential treatments in the early stages of AD, where intervention might halt or reverse the disease process.

Specific Recommendations for Designing Clinical Trials:

Diagnostic Criteria for Early AD:

• Enrollment should be based on biologically based diagnostic criteria that reflect the pathophysiological changes of AD. This is crucial for ensuring that trial participants truly have AD, minimizing the inclusion of individuals who do not have the disease.
• The guidance outlines the staging of AD based on pathophysiological changes and clinical impact, ranging from Stage 1 (asymptomatic but with biomarker evidence of AD) to Stage 3 (mild but detectable functional impairment, not severe enough to warrant a diagnosis of overt dementia).

Outcome Measures:

• Clinical trials should include both clinical outcome assessments and biomarkers. The primary endpoint selection should aim to measure a clinically meaningful change. Direct measures of clinical benefit or validated surrogate endpoints may support traditional approval, while surrogate endpoints that predict clinical benefit may support an accelerated approval.

• For stages with detectable cognitive and functional impairments (Stages 3 and higher), a co-primary endpoint approach assessing both cognitive and functional (or global) measures is recommended.
• In early stages (Stages 1, 2, and early 3), FDA may consider other approaches, including endpoints based on cognitive assessments or surrogate endpoints, which may allow for shorter trial durations.

Stage-specific Considerations:

• Stage 1: Focus on the effect on the characteristic pathophysiological changes of AD as demonstrated by various biomarkers, considering the challenge of demonstrating a clinically meaningful benefit due to the lack of clinical impairment at baseline.
• Stage 2: Trials may need sufficient duration to evaluate clinical measures assessing cognition and function. A persuasive effect on cognition, supported by effects on pathophysiological changes, may provide adequate support for marketing approval.
• Stage 3: Demonstrating a favorable effect on observed impairments in both cognition and daily functioning is important. An integrated scale assessing effects on daily function and cognition could serve as a single primary efficacy outcome measure.

This proposed guidance may serve as a cornerstone for discussions among the FDA, pharmaceutical sponsors, the scientific community, and the public about drug development for early AD, emphasizing the need for early intervention and precise diagnosis in clinical trials.

In Writing: My Prediction for FDA LDT, 2024-2031

 It was big news in the past week that FDA has sents its final LDT rule to the White House for final sign-off and publication.  

https://www.discoveriesinhealthpolicy.com/2024/03/very-brief-blog-fda-sends-final-ldt.html

I'm putting my official multi year policy forecast in writing.  Here.  Click to enlarge.

click to enlarge

Discussion.

First, for simplicity, I'm leaving the initial FDA timelines as published in Fall 2023, but I think it is very likely (80%) then will all advance by one year.  Regardless, the court case will start immediately.

Second, I think there is a 95% (really, 100%) chance this will tie up in court for several years.   Allow 2024 for federal court, 2025 for appeals court, and 2026 if not easily 2027 for Supreme Court.

Third, at Supreme Court, while I am not an attorney, I give a 50% chance the regulation will be tossed out as agency overreach and lack of clear Congressional text governing LDTs.   I give a 10% chance that in 2026 or 2027, it will undergo further legal circling like being remanded to a lower court.  

This leaves a 40% chance of the regulation being turned back on and its engine revved in 2027 or so.

Fourth, I predict, as in my initial white paper in October 2023, that the rule will have to bog down in several years of delays and deferrals if not worse.   By 2029 or 2030 this will lead people to dust off VALID or other alternatives.

## 

Offline.

Other possibilties include a 2nd Trump administration that might reverse the rule in a 90-day rulemaking (or un-rulemaking) cycle.  Or might (more subtly) just decline to defend it assertively in the courts, raising the chance the FDA reg loses in court.

White House, ARPA-H Launch New Cancer Program "ADAPT"

Header: In a press release March 7, 2024, the Biden administration announces its "ADAPT" program, via ARPA-H, which is new programs in precision oncology and adaptive oncology.  The total budget is $400M.

##

See the press release here.

See the ADAPT home page at ARPA, here.

Here's a solicitation for a "Proposer's Day," here.  And a solicitation website here.  See a 25 page PDF solicitation details document here.

See coverage at Genomeweb here.  Adjacent news, Zephyr AI raises $111M for AI-guided precision medicine.

The project results should be open license, open source.  To me, this might be problematic.  Let's say you get $20M to develop something, and are successful.  You get your $20M repaying your demonstrated costs.  Now it takes $10M to replicate and confirm the project and go to FDA.  How do you get the $10M to finish the investment if there is no ROI?  (I don't know).

Program Should Include Rebooting Outdated NCD 90.2, Medicare and NGS

My Medicare Takeaway:  This massive evolution toward newer molecular tests and tje assessment of tumor evolution should be the nail-in-the-coffin of the now-ridiculous NCD 90.2, which forbids the tumor genome to be assessed more than once in a Medicare patient.   

The only MAC system that has allowed serial monitoring, the MolDx MAC, has had to use well-meaning but truly goofy and bizarre workarounds stacked on each other, like calling each test  for minimal residual disease or tumor evolution, "four assays over one year are one test," to fit the one test rule.  

Making Medicare coverage look like the high school game "Twister."  Let's stop it now.  Fix the outdated and warped "one test rule" in NCD 90.2.

##

AI CORNER

The deepest document, clicking through the various press releases and websites, is the 25 page PDF solicitation discription.   Here's an independent analysis by Chat GPT4:

Advanced Research Projects Agency for Health (ARPA-H) Solicitation for ADvanced Analysis for Precision Cancer Therapy (ADAPT)

The Advanced Research Projects Agency for Health (ARPA-H) has announced a solicitation for proposals under its ADvanced Analysis for Precision cancer Therapy (ADAPT) program. The initiative aims to revolutionize cancer care by developing adaptive strategies for treating cancer as it evolves. The program focuses on leveraging novel clinical trial designs, advanced tumor biology measurement, and analysis technologies to match each patient's evolving cancer with the most effective therapy. The goal is to improve survival for patients with metastatic cancer by facilitating a continuous and interconnected learning process that rapidly informs clinical practice. Proposals may address individual Technical Areas (TAs) or a combination thereof, spanning therapy recommendation techniques, evolutionary clinical trials, and a treatment & analysis platform.

Key Takeaways:

1. Program Objective: Accelerate better health outcomes through innovative research on adaptive cancer treatment strategies.
2. ADAPT Program: Aims to create an adaptive treatment platform to match treatments with tumor changes, improving survival rates for metastatic cancer patients.
3. Eligibility: Open to various organizations capable of tackling the program's challenges across its three interconnected TAs.
4. Technical Areas (TAs): Includes 
1. Therapy Recommendation Techniques (TA1), 
2. Evolutionary Clinical Trial (TA2), and 
3. Treatment & Analysis Platform (TA3).
5. Proposal Submission: Proposals should address the program's structure, integration, and metrics, adhering to specific content and format requirements.
6. Equity and Accessibility: Proposals must include a plan to ensure equitable health care access, reflecting the diversity of the U.S. population.
7. Funding and Awards: Multiple awards anticipated, with the scope and budget aligned with the complexity of proposed solutions.
8. Data and Software Standards: Emphasis on open-source technology, interoperability, and adherence to existing health IT standards.
9. Intellectual Property: Encourages alignment with open-source regimes to promote a sustainable ecosystem for translational oncology.
10. Collaboration: Expected among performers to share information freely, work on common standards, and ensure the program's goals are met.

This summary encapsulates the main aspects of the ADAPT program solicitation, targeting innovations in cancer treatment through adaptive strategies and collaborative research efforts.

User

Can you provide a summary and 5  takeaways for each of the programs, being tA1, TA2, and TA3.    

ChatGPT

TA1: Therapy Recommendation Techniques Summary

TA1 focuses on developing data-driven methods to identify resistant cancer traits, discover new predictive biomarkers of drug response, and guide optimized treatments based on a tumor's genetic and phenotypic traits. It encompasses three sub-areas: Multi-Modal Data Fusion (TA1.1), Resistant Trait Modeling (TA1.2), and Biomarkers that Predict Drug Response (TA1.3). The aim is to leverage comprehensive assessments of tumor biology, including genetic, protein-level data, and clinical data, to build predictive biomarkers for personalized cancer treatment strategies.

Key Takeaways for TA1:

Objective: To enhance the precision of cancer treatment by identifying tumor resistance traits and developing predictive biomarkers for drug response.

Multi-Modal Data Fusion (TA1.1): Focuses on integrating various data types to provide a more accurate and comprehensive understanding of tumor data.

Resistant Trait Modeling (TA1.2): Aims to increase the precision and number of identified resistant traits through advanced tumor evolution models.

Biomarkers for Drug Response (TA1.3): Seeks to develop and test multi-modal biomarkers that will be implemented within an evolutionary clinical trial.

Collaborative Approach: Requires collaboration among systems biologists, data scientists, and clinicians to analyze and interpret tumor biology and clinical data for therapy optimization.

TA2: Evolutionary Clinical Trial Summary

TA2 is designed to revolutionize cancer clinical trials by introducing an evolutionary clinical trial framework that adjusts drug treatments based on how the tumor evolves during therapy. It includes three sub-areas: New Tumor Measurement Technologies (TA2.1), Evolutionary Trial Protocol (TA2.2), and Evaluation of TA1 Biomarkers (TA2.3). The goal is to improve cancer treatment by integrating advanced tumor analysis methods to identify emerging resistant traits, enabling clinicians to adapt treatments effectively.

Key Takeaways for TA2:

Innovative Design: Proposes a new type of clinical trial that dynamically adjusts treatments in response to changes in tumor biology.

Tumor Measurement Technologies (TA2.1): Emphasizes the collection of diverse temporal tumor measurements for resistance trait identification and biomarker development.

Evolutionary Trial Protocol (TA2.2): Develops a modular, open-source protocol design capable of responding to tumor changes in near-real-time.

Biomarker Evaluation (TA2.3): Focuses on testing and integrating predictive biomarkers into the evolutionary trial to guide therapy selection as the tumor evolves.

Patient-Centric Approach: Aims to significantly improve patient care by personalizing treatment strategies based on individual tumor characteristics.

TA3: Treatment & Analysis Platform Summary

TA3 aims to create an open-source collaboration space for developing, analyzing, and sharing data models and trial protocols between researchers and clinicians. This platform will facilitate the real-time availability of harmonized multi-modal and patient data, supporting adaptive therapy recommendation techniques. Key focus areas include building a collaboration ecosystem, creating a comprehensive data lake, facilitating data processing, promoting data standardization and linkage, and developing open APIs and toolsets.

Key Takeaways for TA3:

Collaborative Ecosystem: Establishes a platform to unite clinicians and scientists, enhancing access to data and therapy regimen analysis.

Data Lake Creation: Maintains a centralized repository for various data types, improving data accessibility and management.

Data Processing Facilitation: Develops pipelines for rapid tumor biology measurement data processing, including quality control and data curation.

Standardization and Open APIs: Implements data standardization and linkage processes, alongside developing open APIs for broad data access and sharing.

User-Friendly Tools: Aims to create intuitive, AI-powered dashboards and data exploration tools, making biomedical data easily accessible and usable for the scientific and clinical communities.

AI Corner 2

I don't know how accurate this is but I asked CHAT to compare ARPA and NIH NCATS, here.

Journal Club: Tsimberidou et al, Practice Gaps for Genomics in Cancer Care

 A new paper on March 5, 2024, from Tsimberidou, Sireci, Dumanois, and Pritchard:   "Strategies to Address the Clinical Practice Gaps Affecting the Implementation of Personalized Medicine in Cancer Care."   Find it open access at J Clin Oncol / Practice.

https://ascopubs.org/doi/pdf/10.1200/OP.23.00601

Issues discussed include:

• Biopsy referral
• Biospecimen collection (procedures)
• Biospecimen evaluation (pathology)
• Biomarker test orders
• Biomarker performance (false negatives, etc)
• TAT and test reporting; therapy moves ahead w/o result
• Treatment decision fit to results

Each has a calculated attrition value, which is discussed.  The role of liquid biopsy vs needle biopsy is discussed.

Regarding biospecimen collection and adequacy, pair this paper with FINTELMANN 2023, here, which looks at the tissue collection issues from a radiologist's viewpoint.  great paper.

##

For Tsimberidou's 2023 article in Nat Rev Clin Onc on tumor boards, here.   She's also a coauthor on 2023's consensus paper, Lockwood et al., on cell free DNA assays (AMP/CAP), here.

See a discussion of the new paper by Daryl Pritchard of PMC at Linked In.

Timeline: NCD Request to Final Change, 2.5 Years (!)

Header: How long did it take to get an NCD changed?  2.5 years.

##

In October 2021, the American Society of Hematology and other major groups requested CMS to expand coverage of stem cell transplantation in myelodysplastic syndrome (MDS), and remove a CED requirement dating back to 2010.   

https://www.cms.gov/Medicare/Coverage/DeterminationProcess/downloads/id312.pdf

The 9-page letter is a good example of a clear request.

CMS initiated public review in June 2023, about 18 months later.   A draft NCD was produced in December 2023, and finalized in March 2023.   

https://www.cms.gov/medicare-coverage-database/view/ncacal-tracking-sheet.aspx?ncaid=312&=

CMS made a number of detailed changes based on the December 2023 comment period, generally accepting proposals for broader coverage (some of them nuanced).

Kimberly Long was the lead analyst and Dr James Rolllins the lead medical director.

AI Corner: Can AI Read the FOCR Poster on HRD Harmonization?

Header: FOCR summarized interim results of its HRD harmonization workstream as a poster.  I found the poster hard to get my head around.  Can AI read the poster better than me?

###

For a couple years, Friends of Cancer Research FOCR has been running a  major project on HRD harmonization  - homologous recombination deficiency testing.   Today I got a long email from them reporting results, and, linking to a PDF "poster presentation" to summarize results from two experiments.  The first was an "in silico" study, the second was a "patient sample" study.   

See the FOCR HRD home page here:

https://friendsofcancerresearch.org/hrd/#ProjectOutcomes

• I clip a copy of the new FOCR email here:
•    https://brucedocumentblog.blogspot.com/2024/03/3-6-2024-focr-hrd-email.html
• The data was provided by a link to a cloud PDF of a poster presentation (dated 2022).
•    https://friendsofcancerresearch.org/wp-content/uploads/20221002-AACR-OvCa-Poster_FINAL.pdf

I found the poster challenging:

However, one take home lesson is on the right, that whether one used percent aggreement or percent disagreement, these were in the 70-80% range.  You can rate agreement data this way; it's hard to say "right or wrong" as there isn't a gold standard.

Here are three AI readings of the poster:

CHAT GPT 4

The scientific poster presents findings from the Friends of Cancer Research HRD Harmonization Project, which analyzed 13 assays for measuring Homologous Recombination Deficiency (HRD) in 90 high-grade serous ovarian tumors. Key findings include:

• The median HRD positivity rate was consistent with prior publications, 49% for in silico analysis and 53% for patient sample analysis.
• Inter-assay agreement on HR status was variable, with altered BRCA1/2 samples more uniformly called HRD.
• Positive/negative agreement varied across assays, with modest to high levels of agreement.
• The influence of BRCA1 and BRCA2 alterations on HRD calls was significant, indicating the importance of these genes in HRD status determination.

The conclusion emphasizes the importance of understanding assay agreement to inform assay interpretation and improve alignment of HRD scores, facilitating better treatment decisions. Future steps include further analysis on the impact of various factors on HRD call concordance and hosting a public meeting to discuss findings and recommendations.

###

CLAUDE.AI

Here are the key findings and take-home lessons from the scientific poster on the HRD (Homologous Recombination Deficiency) harmonization project:

Key Findings:

• There was significant variability in HRD positivity rates across different assays, ranging from 23-74% (median 53%) for patient samples and 9-67% (median 49%) for in silico samples.
• Agreement on HRD status calls between assay pairs was variable, with modest to high levels of positive and negative percent agreement.
• Samples with BRCA1/2 mutations were more uniformly called HRD positive across assays compared to wild-type BRCA1/2 samples.
• Inclusion of different factors (e.g., genomic scars like LOH, mutational signatures, non-BRCA1/2 gene mutations) in assay algorithms did not strongly explain the variability in HRD calls for in silico samples.

Take-Home Lessons:

Significant inter-assay variability exists in determining HRD status, which can impact treatment decisions for drugs like PARP inhibitors.

• BRCA1/2 mutation status strongly influences HRD calls, with higher agreement for mutation-positive samples.
• Understanding the sources of variability (e.g., algorithms, biomarkers used) is crucial for improving alignment of HRD scores and assay interpretation.
• Further research is needed to examine the impact of clinical factors, sample quality, and specific gene alterations on HRD call concordance.
• Harmonization efforts and recommendations from this project will inform future use of HRD assays to improve treatment selection for patients.

In summary, while HRD testing can guide targeted therapy, there are significant inter-assay differences that need to be addressed through harmonization and understanding the key drivers of variability for optimal clinical implementation.

###
PERPLEXITY

Key Findings from the HRD Harmonization Study:

• Study Objective: The research analyzed 13 independently performed assays to measure Homologous Recombination Deficiency (HRD) using 90 high-grade serous ovarian tumors.
• Participants: The study involved 16 organizations representing 18 HRD assays, including commercial and academic developers.
• Methodology: Assay developers analyzed TCGA ovarian cancer samples in silico and patient samples to measure HR status and contributing factors for each sample.

Results:

1. In Silico Sample Results: Variability in HR status determination was observed across assays, with correlations among continuous HR scores varying substantially. The median HRD positivity rate was consistent with prior publications.
2. Patient Sample Results: Inter-assay agreement on HR status calls was variable, influenced by the inclusion of factors related to causes versus consequences in HRD score determination. Samples with altered BRCA1/2 showed higher Positive Predictive Agreement (PPA) and lower Negative Predictive Agreement (NPA).

Conclusions:

• Understanding the agreement among assays is crucial for interpreting results accurately and making informed treatment decisions.
• Future research should focus on examining the impact of clinical and sample factors on HRD call concordance and provide recommendations for the future use of HRD assays.

Take-Home Lessons:

• Interpretation Challenges: Variability in HR status determination across assays highlights the need for standardization to improve clinical interpretation.
• BRCA1/2 Influence: Samples with altered BRCA1/2 genes demonstrated higher agreement levels, emphasizing the impact of these genes on HRD calls.
• Clinical Implications: Enhancing assay alignment and understanding factors influencing HRD scores can aid in guiding appropriate treatment decisions for patients.

###

What do I think?

Only CLAUDE noted that gross percent positive ranged from around 10-20% for one test to 60-70% for the highest test.  To me, that sounds like an important (if frightening) fact.  Imagine if one doctor sends samples to a lab that calls 10% positive (1 in 10), and another doctor sends samples to a lab that calls 70% positive (7 in 10).

However, those numbers define the highest highest and lowest extremes, over the range of tests, the average agreement was better circa 75%.   

CHAT GPT and PERPLEXITY (but not Claude) noted that 13 tests (labs) were tested, which is key helpful fact for me to understand what was done.

###

Definitions and Explanations of HRD

What if you don't know HRD from a hole in the ground?  Separate from the poster, I asked Perplexity to define HRD, show the testing used, and explain why a doctor needs it. Note, in the response, it footnoted 5 sources.  It should be open access:

https://www.perplexity.ai/search/Provide-a-definition-t3xFrB8bQzGZaEHdWyYQwA

###  

Perplexity Explains HRD to Us:

Definition of HRD (Homologous Recombination Deficiency) and Testing Methods

HRD, or Homologous Recombination Deficiency, is a cellular phenotype characterized by the inability of a cell to effectively repair DNA double-strand breaks using the homologous recombination repair (HRR) pathway. This deficiency can lead to genomic instability and contribute to cancerous tumor growth 

Molecular Steps or Methods Used to Test for HRD:

1. Genomic Scars Analysis: HRD testing involves assessing genomic scars that indirectly measure HRD, reflecting permanent genomic changes caused by DNA repair deficiency. This includes:
1. Loss of heterozygosity (gLOH): Permanent loss of one parent's allele copy of a gene at a specific part.
2. Number of telomeric imbalances (TAI): Regions in a chromosome with allelic imbalance.
3. Large-scale transitions (LST): Genomic alterations involving chromosome breakages
2. Comprehensive Genomic Profiling (CGP): CGP using Next-Generation Sequencing (NGS) can assess causal genes and genomic scarring together in one assay. 
1. It helps identify rare mutations and other relevant genomic signatures in various cancers like ovarian, breast, prostate, and pancreatic cancers

PARP Inhibitor Testing: HRD testing can guide treatment decisions, especially for PARP inhibitors. If a tumor is positive for HRD, it may respond well to PARP inhibitors, which block the enzyme's action in cancer cells, preventing them from repairing their damaged DNA and causing cell death

Importance of HRD Testing for Cancer Patients:

1. Treatment Personalization: HRD testing provides crucial information for personalizing treatment options for cancer patients, guiding the selection of therapies like PARP inhibitors or platinum-based chemotherapy based on the HRD status of the tumor
2. Prognostic Insights: It offers insights into the disease course and helps identify variants that may cause HRD, aiding in predicting outcomes and planning comprehensive treatment strategies
3. Family Risk Assessment: By detecting germline mutations of BRCA1/2, HRD testing can also provide predispositional insights into family members' risk of developing ovarian cancer and other cancers

In summary, HRD testing plays a vital role in oncology by enabling personalized treatment decisions, offering prognostic insights, and assessing familial cancer risks based on genetic mutations and genomic instability markers.

##

Not AI.  

See HRD at plain old Wikipedia, here:

https://en.wikipedia.org/wiki/Homologous_recombination

It's helpful to see again what "homologous recombination" is, to start with (HR), before factoringin a 'deficiency" of it (HR-D)..

###

Friends of Cancer Research - Note on "External Control Studies"

 I was using the Friends of Cancer Research website today and ran across an article, though from 2019, that's still a good reference.   It discusses "external control arm" studies, such as when contemporary external controls are used as a reference, or when administrative controls (based on computer records) are used as a reference.

I am sure this can be good sometimes, but experts (and me) have seen cases from time to time where external control arms produce comparisons of effect that are simply unbelievable.   (An example was a prediabetes YMCA study as a single arm study with adminstrative billing records of other Medicare patients used as  controls. Alva 2017).

FOCR has a 2019 article on this topic.  I gather it was reprinted with permssion from Pink Sheet.  Worth reading if you are considering a one-arm study relying for comparison, on propensity-adjusted controls.  Pros and Cons discussed.

https://friendsofcancerresearch.org/news/pink-sheet-external-control-arms-better-than-single-arm-studies-but-no-replacement-for-randomization/

CHAI: Coalition for Health AI Accelerates, Names Leadership

See a March 4, 2024, press release that the COALITION FOR HEALTH AI or CHAI has launched and has named its board of directors and CEO.   Brian Anderson MD will be the first CEO, and the Board will be chaired by John Halamka MD.  See press release:

https://www.coalitionforhealthai.org/updates/march-4th-2024

They also held a webinar kick off event on March 5, led by FDA Commissioner Dr Califf and Jennifer Goldsack, who is CEO of the Digital Medicine Society.   It looks like they will post an archive video when it is ready:

https://www.coalitionforhealthai.org/events/