THE SCIENTIST - 2025 - exercise and Parkinson's disease, an update.
FORBES 2008 - Andy Grove's Last Stand - Retirement and Parkinson Disease.
https://www.forbes.com/forbes/2008/0128/070.html
##
Though unrelated to Parkinson's, describing an event in late 1980s, see also 'My first and last interview with Andy Grove". 2016 (year of death).
https://www.forbes.com/sites/jboyd/2016/03/22/i-only-met-intels-andy-grove-once-and-once-was-enough/
Chat GPT 5.
Guardant Investor Day 20250924. Deck, transcript. Notes from UBS, Nephron.
###
Here is a detailed, expert-level synthesis of Guardant Health’s 2025 Analyst & Investor Day, combining insights from the company’s slides, transcript, and analyst reports (Nephron, UBS).
2028 Targets Raised: Guardant now projects $2.2B sales by 2028 (up from ~$2.0B prior), driven primarily by Oncology ($1.4B), Screening ($0.5B), and Biopharma/Data ($0.3B).
Breakeven Pulled Forward: Free cash flow breakeven is now expected in 4Q27, one year earlier than prior guidance.
Margins: Management targets 65–70% gross margins by 2028, an expansion from today’s ~63–64%.
Cash Position: As of 2Q25, Guardant held ~$629M in cash vs. $1.1B in converts outstanding. Screening burn remains heavy (~$200M annually in 2025–26), but Oncology growth is expected to offset losses by 2027–28.
Core Business Expansion:
Market TAM increased from $3B to $10B, with blood expected to shift from ~1/3 to ~2/3 penetration due to repeat testing.
Guardant360 ASPs: Upgraded by CMS from $3,500 → $5,000 in 2024; forward ASP assumption lifted to $3,300 by 2028.
Tissue Testing: ASP assumption raised from $2,000 → $3,000, reflecting better reimbursement and broader adoption.
Competitive Landscape:
Guardant volumes +20% y/y in 2Q25, but competitors are growing faster: Tempus (+26%) and Caris (+22% overall; +56% in blood).
Guardant leverages its Smart Platform with AI/epigenomics for higher resolution profiling, aiming to differentiate beyond mutation detection.
Current Leadership: Guardant is #1 in tissue-free MRD with Reveal (>50K tests in 2025; ~$600 ASP). But it significantly trails Natera’s Signatera, approaching $800M in 2025 sales.
Reveal Ultra (tissue-informed): Launching to match market demand for tumor-informed approaches. Sensitivity promised to 1 part per million (ppm).
Coverage Expansion: Already reimbursed in CRC surveillance; submissions filed for breast cancer and IO monitoring.
Clinical Data: Upcoming PEGASUS study (ESMO 2025) will be among the first large prospective trials validating tissue-free MRD utility.
Strategic Tension: Guardant continues to assert tissue-free as the long-term winner, but Ultra signals concession that payors/physicians favor tumor-informed assays today.
Shield CRC: FDA approved with Medicare ADLT pricing at $1,495. On track to be one of the most successful Dx launches outside COVID.
Multi-Cancer Detection (MCD): Starting Oct 2025, Shield will report MCD results alongside CRC if physicians opt-in and patients consent to data sharing. This generates real-world evidence while billing Medicare under CRC coverage.
Commercial Strategy:
Partnership with Quest Diagnostics (launching Q1 2026) expands access.
ACS guidelines expected by YE25; USPSTF delayed to 2027–28. Guardant is working on interim alternative quality metrics.
Screening revenues still targeted at >$500M in 2028.
Investor Concerns: Screening remains a high-burn franchise ($200M+ annually through 2026) with uncertain scaling profitability.
>1M patient samples and 200+ petabytes of data amassed; expected to hit 1 exabyte within 2 years.
Guardant’s InfinityAI learning engine leverages multimodal data (genomics + epigenomics + real-world evidence) for:
Faster product iteration and Smart Apps on Guardant360.
Biopharma partnerships for novel signature discovery and trial support.
Biopharma/Data projected at ~$300M revenue by 2028.
Innovation Flywheel: Smart Platform (genomics + epigenomics) → InfinityAI → new apps & products. Analogous to successive “S-curves” of electricity or semiconductors.
Execution vs. Competition:
Guardant has leadership in liquid biopsy, but competitors (Tempus, Caris, Natera) are scaling aggressively.
Heavy Screening burn raises questions about long-term profitability despite raised targets.
Investor Sentiment:
UBS: Raised PT to $80 (Buy), citing upside from oncology ASPs, Shield MCD launch, and breakeven pulled forward.
Nephron: Maintains Sell (PT $35), cautious on competition, product scaling, and cash burn.
Oncology (therapy selection) remains Guardant’s most credible and sustainable growth driver, underpinned by ASP expansion, repeat testing, and Smart Apps.
MRD is a two-front war: Guardant dominates tissue-free, but Natera’s tumor-informed lead is formidable. The Reveal Ultra launch both hedges risk and complicates Guardant’s “tissue-free superiority” narrative.
Screening (Shield) is strategically essential but financially contentious: differentiation via MCD and Quest partnership could create a moat vs. CRC-only players, but burn levels remain high.
Data/AI moat is perhaps Guardant’s deepest long-term asset, with InfinityAI + >1M samples forming a defensible barrier to entry.
The Bull vs. Bear divide is sharp: UBS sees upside in execution and differentiated innovation; Nephron stresses cash burn, competitive pressure, and balance sheet risk.
FDA databases give you overlapping but distinct views into device approval.
Here's the basic homepage for a PMA device, the IDYLLA (Biocartis) MSI CDx test:
https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?ID=P250005
You get the device, the applicant, the P-number (P250005), the time under review (2/25 to 8/25), the product code (SFL), the approval order (labeling).
https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfCLIA/Detail.cfm?ID=41847
https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPCD/classification.cfm?id=5488
Click on the SSED (Summary of Data), 45 pp.
https://www.accessdata.fda.gov/cdrh_docs/pdf25/P250005B.pdf
https://www.accessdata.fda.gov/cdrh_docs/pdf25/P250005D.pdf
https://www.accessdata.fda.gov/cdrh_docs/pdf25/P250005A.pdf
When fda approves new lab products, including lab tests with whole slide imaging and machine learning, fda classifies them under CLIA, such as "CLIA high complexity test." What are the definitions, border zones, and examples between CLIA low complexity and high complexity? You are writing for experts but can be concise.
Here’s a concise expert-level overview of CLIA complexity categories, their border zones, and examples, especially as they apply to FDA-approved lab tests (including digital pathology / AI products):
Under CLIA (42 CFR 493.17), each laboratory test method is scored across seven criteria (yes/no or scaled questions). The scores are summed; ≤12 points = “moderate complexity” and >12 points = “high complexity.” Tests waived under CLIA are outside this scoring (they are considered “low complexity” by definition).
Scoring Criteria
Knowledge required (lay vs. medical technologist vs. pathologist)
Training & experience (minutes vs. months of training)
Reagents & materials prep (ready-to-use vs. manual)
Characteristics of operational steps (simple vs. multi-step w/ judgment)
Calibration, QC, and maintenance requirements
Troubleshooting & interpretation (automatic vs. operator judgment)
Result interpretation (objective readout vs. subjective pattern recognition)
Waived (Low Complexity)
Definition: CLIA-waived tests are “so simple and accurate as to render the likelihood of erroneous results negligible.”
Examples: Urine dipstick, fingerstick glucose meters, rapid antigen COVID tests (CLIA-waived version), fecal occult blood cards.
Borderline issues: When waived tests are used outside their IFU (e.g., alternate specimen types), they revert to high complexity under CLIA.
Moderate Complexity
Definition: Score ≤12. Requires trained staff but does not require a pathologist’s interpretive judgment.
Examples: Most automated chemistry panels, hematology analyzers, many immunoassay platforms (e.g., ELISA on a closed system).
Border zones: A test may be moderate if readout is fully automated with internal QC and minimal operator interpretation.
High Complexity
Definition: Score >12. Typically requires technical judgment or interpretive skill.
Examples: PCR tests requiring manual extraction, NGS workflows, cytogenetics, manual microscopy (Gram stain interpretation), digital pathology WSI with AI where pathologist review is part of workflow.
Why FDA calls AI/WSI tests “high complexity”: They require oversight by a qualified pathologist and involve interpretive decision-making even if software provides a score or heatmap.
Automated ML Classifiers:
If the AI provides a binary call with clear QC, and no pathologist review is required, classification could theoretically be moderate complexity.
But if a pathologist must review digital slides and integrate AI output, it remains high complexity due to interpretive judgment.
Image Analysis Add-Ons (e.g., HER2 quantification):
Fully automated readers that produce discrete scores may be moderate.
Systems that generate heatmaps or probability scores requiring interpretation are scored high.
Molecular Tests on Closed Platforms:
Example: Cepheid GeneXpert assays are typically moderate because operator steps are minimal.
Same PCR test run manually (open-tube PCR with manual result read) would be high complexity.
Would you like me to include a visual table summarizing the scoring criteria with typical examples (chemistry, PCR, AI slide analysis) so you can drop it directly into a slide deck?
Direct AI output from Chat GPT 5. Do not quote without verification. Prompt at bottom.
The AI article draws quite a bit from a 2004 CAP TODAY article on this "obscure history."
https://www.captodayonline.com/Archives/feature_stories/0304Spiritofthelaw.html
###
Short answer: yes—there’s a paper trail. The fight over whether pathologists could bill a “professional component” (PC) for every clinical lab test began in the late-1970s/early-1980s and led—through litigation, statute, and rulemaking—to today’s narrow definition in 42 CFR 415.130.
What kicked it off (1970s → 1980):
Pathologists in many regions were billing patients/insurers under Part B for PC on clinical pathology (oversight, protocols, QA/QC, availability to consult) even when they didn’t personally touch each specimen. HCFA (now CMS) moved to make treatment uniform and, in a March 11, 1980 notice, barred direct Part B billing for the PC—adopting the “New England” model where the hospital pays the pathologist and rolls costs into Part A. Pathologists (CAP and the Arkansas Society) sued immediately, and a federal court in Little Rock enjoined HCFA’s 1980 policy, recognizing the PC as “professional services performed for the benefit of patients.” (Cap Today)
Congress steps in (1982 TEFRA §108) & HCFA regulations (1983):
To resolve the stalemate, TEFRA §108 (1982) told HHS to distinguish:
(A) physician services personally rendered for an individual patient (Part B), versus
(B) professional services for the general benefit of hospital patients (Part A, paid via the hospital). HCFA implemented this in 1983 rules on “Payment for Physician Services Furnished in Hospitals,” establishing the Part A/Part B split for hospital-based physicians (including pathologists). (Cap Today)
Prospective payment flips incentives (1984–1987):
Once DRGs/OPPS replaced reasonable cost, hospitals no longer recovered whatever they paid for Part A “general benefit” pathology services, fueling recurring PC disputes in private insurance and state courts through the 1990s. A key federal case, Central States v. Pathology Laboratories of Arkansas (7th Cir. 1995), described PC activities (setting protocols, calibrating, supervising, consulting) and upheld PC billing under the plan’s terms (private insurance context). (Justia Law)
Codification as 42 CFR Part 415 and the narrow definition (1995–1999):
In December 8, 1995, HHS reorganized physician-in-provider rules into Part 415; this is the source note for § 415.130 (“Conditions for payment: physician pathology services”). The rulemaking history shows § 415.130’s lineage beginning at 60 FR 63178 (Dec. 8, 1995), later amended Nov. 2, 1999 (64 FR 59442) and in subsequent years. The section narrows what may be paid under the physician fee schedule to: (1) surgical pathology; (2) listed cytology/hematology/blood bank services requiring a physician; and (3) clinical consultations—i.e., not a blanket per-test PC. (eCFR)
The “TC bundling” fight & hospital-only payment (1999–2008):
Through the 1999 PFS final rule, HCFA announced it would pay only the hospital for the technical component (TC) of physician pathology for hospital patients—ending independent lab TC billing for hospital in/outpatients. Congress repeatedly delayed this via BIPA/MMA/TAX RELIEF extensions for “covered hospitals,” but the policy ultimately took effect for services after 12/31/2007, with CMS transmittals implementing edits in early 2007–2008. These actions are cross-referenced directly to § 415.130(d). (Centers for Medicare & Medicaid Services)
A clear through-line to the present § 415.130:
Litigation (1980) blocked HCFA’s attempt to ban PC billing outright, prompting TEFRA §108 (1982) and 1983 rules that drew the Part A vs Part B line for hospital-based physicians. (Cap Today)
1995 reorganization placed these payment conditions into Part 415, and § 415.130 evolved via amendments (1999 and later) to spell out a narrow, enumerated list of pathology services payable under the PFS—surgical path, certain cytology/hematology/blood bank services, and consultations—not a per-test “PC for everything.” (eCFR)
1999–2008 rulemaking and transmittals then sealed hospital-only payment of TC for hospital patients, closing a major billing avenue and reinforcing the narrow scope of payable physician pathology services. (Centers for Medicare & Medicaid Services)
Helpful primary sources if you want to footnote:
CAP Today’s historical explainer walking through 1965→1987, the 1980 notice, the Arkansas injunction, TEFRA §108, and 1983 regs. (Cap Today)
42 CFR 415.130 text and source notes (showing 60 FR 63178 (1995) origin; 64 FR 59442 (1999) amendment). (eCFR)
1999 PFS final rule and CMS transmittals implementing TC bundling for hospital patients and tying it to § 415.130(d). (GovInfo)
Central States v. Pathology Labs of Arkansas (7th Cir. 1995) for the federal appellate description of PC activities (useful for context; private-plan case). (Justia Law)
If you want, I can assemble a one-page timeline with pinpoint FR cites (TEFRA §108 text, 1983 preamble language, 1995/1999 CFR notes, and the 2007–08 transmittals).
###
42 CFR 415.130 DEFINITIONS - Pathologist Service
Note that "ordinarily require performance by a physician" = >515
###
Prompt
Ligation about pathologist fees, a bit of statute, and CFR rulemaking in the 1970s led to the creation of 42 CFR 415.130, which defines physician pathology services. The context was that pathologists sued CMS in the 1970s demanding they could get a professional (perhaps $3?) for every clin chem test run in their lab (!). CMS said no, and the point of 42 CFR 415.130 is to narrowly, not broadly, define the scope of services of a pathologist. It's pretty obscure, but do you have any information on the background of this debate and the creation of 415.130?
These new hashtag#CMAA codes would reimburse directly for AI that “produce actionable, clinically relevant insights that directly impact patient care.”
And this marks a dramatic change in the world of AI’s future in healthcare.
AI was previously reimbursed in “remote monitoring codes” (think 99454) that reimbursed AI for data collection - an administrative function.
These new codes allow healthcare systems to now be reimbursed for AI for into clinical insights and intelligence as it relates to support clinical care and patient outcomes.
A dramatic difference in the reimbursement of AI to billed by hospital systems in support of clinical functions, insights and intelligence.
Now although I support reimbursement for AI in healthcare - the inherent question for me - is why can we not bill these same codes for nursing?
We have been told for decades that it would be too complex, too impossible to unbundle nursing workflows and services to be billable from the current design - but the exact definition of the new CMAA codes align to the deliverables of nursing. So the imperative and time is now to also redefine CPTcodes for nursing.
What I do not want to see happen - is when nursing leadership asks for more nurses - they are again told there is no way to pay for more nurses (as our profession is a cost - no billable reimbursement for nurses) - but that you can have more AI (because AI is billable).
AI cannot operate without humans in the loop in clinical care - and much of those humans in the loop in healthcare will be nurses.
If the AMA can create an entire new class of codes to bill for AI - they can find a way to create a class of CPT codes to bill for nursing services.
The time is now to act and make sure nursing is not left invisible and written out of these coming changes.
hashtag#NursesOnLinkedIn hashtag#HealthcareOnLinkedIn hashtag#CPTCodes hashtag#AIHealthcare American Academy of Ambulatory Care Nursing (AAACN) Commission for Nurse Reimbursement hashtag#Nurses hashtag#FutureOfNursing hashtag#Reimbursement