Care pathway, site of service, and what is already in the room — ctDNA MRD assay validation

1. Summary

This domain's "care pathway" is a laboratory workflow, not a clinical site of care — the dossier's own adaptation note establishes that, and this report inherits it. Tumor-informed ctDNA minimal-residual-disease (MRD) testing runs exclusively as a CLIA-regulated high-complexity laboratory-developed test (LDT), a status that a federal court and FDA itself, not just CLIA's own categorization criteria, reaffirmed in 2025 when FDA's 2024 rule asserting device authority over LDTs was vacated and then formally rescinded [1][6]. The specimen workflow that any reference material or QC product has to sit inside is two-part and sequential — archived surgical tumor tissue first, a whole-blood draw second — per the design of the largest US real-world Signatera study located this pass [8]. Turnaround time is not a solved problem in this workflow: two independent published reviews name it, alongside assay standardization and cost, as a live barrier to integrating tumor-informed ctDNA MRD testing into routine oncology practice [9][10]. Staffing "already in the room" is set by the same CLIA personnel regulations that set the site of service, not by anything specific to ctDNA testing [3][4][5].

2. What Changed

_Baseline (v1). No prior version; this establishes the starting point for future diffs._

3. Details

Regulatory floor: CLIA governs the room, not FDA

CLIA's own test-categorization standard scores every laboratory test system on seven criteria of knowledge, training, and process complexity; systems scoring above 12 are categorized as high complexity [1]. A tumor-informed ctDNA MRD assay — patient-specific panel design from whole-tumor sequencing, followed by error-suppressed detection of a handful of variant molecules in plasma — sits at the specialized-knowledge, specialized-training end of every one of those criteria as CLIA itself defines them [1]; no such assay is offered as an FDA-cleared or FDA-authorized in vitro diagnostic today. That distinction sharpened rather than blurred in 2025. FDA's May 2024 final rule had amended the definition of "in vitro diagnostic products" at 21 CFR 809.3 to add "including when the manufacturer of these products is a laboratory," asserting device authority over LDTs for the first time [6]. On March 31, 2025 a federal district court vacated that rule, and on September 19, 2025 FDA published a final rule stating that it "reverts to the text of the regulation as it existed prior to the effective date of the May 2024 final rule," removing the disputed clause [6]. The practical effect for this domain: the laboratories that run tumor-informed ctDNA MRD assays, and by extension the reference-material buyers this dossier is built around, answer to CLIA and to no FDA premarket or site-inspection authority over the assay itself [1][6].

Staffing already required in that room

CLIA's personnel standards for high-complexity testing are the staffing floor any reference- material buyer or CLIA lab is already operating under, independent of whether it ever buys a third-party control panel. So is the daily control-procedure requirement: CLIA requires that "for each test system, the laboratory is responsible for having control procedures that monitor the accuracy and precision of the complete analytic process," including a control run at least once each day patient specimens are tested [2] — an obligation the laboratory carries whether its control material is bought from a reference-material vendor or built in-house. The laboratory director must hold a current state license where required and meet one of several credential pathways — a physician board-certified in pathology, or a doctoral-degree holder with defined coursework, certification, and experience requirements [3]. Each individual actually performing high-complexity testing must likewise hold a current state license where required and meet a defined education/training threshold — a doctoral, master's, bachelor's, or (with documented additional training) associate degree in a laboratory or clinical science [4]. Testing personnel are made explicitly responsible for "specimen processing, test performance and… reporting test results," for following the laboratory's own quality-control policies, and for either correcting or escalating any problem that could affect a result [5]. None of this is ctDNA-specific — it is the general high-complexity-testing personnel floor CLIA sets for every molecular test a lab runs, this one included.

Bench workflow: tissue first, then serial plasma

The BESPOKE CRC study — Natera's own prospective, US-based, real-world observational study of Signatera use, 1,788 patients enrolled — required that enrolled patients have "residual formalin-fixed paraffin-embedded (FFPE) tissue available" and "provide FFPE and whole blood samples" [8]. That two-part requirement (archived surgical tissue to build the patient-specific variant panel, then blood for serial cell-free-DNA testing) is the bench sequence this domain's reference material has to validate against. Patients in BESPOKE were "selected by their healthcare provider to receive the SIGNATERA test according to the current evidence-informed schedule as part of their routine practice" [8] — ordering happens at the point of oncology care, but the analytic work does not; nothing in this record locates the sequencing or panel-design step anywhere but a centralized laboratory that returns a result to the ordering clinician. The CIRCULATE-Japan ALTAIR trial's own eligibility criteria confirm the same two-step logic from a different, completed, results-bearing study: enrollment required patients who "tested positive for ctDNA using Signatera™ by an analysis of the latest blood samples collected within 3 months prior to enrollment" [7] — a serial-plasma design layered on top of an already-established tumor-derived panel, not a single blood draw.

Turnaround time: the room's actual clock

A 2026 narrative review of ctDNA-defined MRD in gastrointestinal cancers names "assay standardization, false-positive calls, turnaround time, reimbursement, and the need to distinguish prognostic validity from proven clinical utility" as the field's key barriers, in that order [9]. A separate 2025 review of colorectal-cancer MRD testing states the mechanism directly: "The tumor-informed MRD test has high sensitivity but requires tumor sequencing that takes longer to process, whereas the tumor-naive MRD test has a shorter turnaround time but a lower sensitivity" [10] — turnaround time and sensitivity trade directly against each other across the two assay architectures this domain's reference material would have to serve. Both reviews are independent of each other and of the assay manufacturers, and both name turnaround time as a live constraint rather than a solved parameter, so this is stated plainly rather than carrying a marker. A Japan Society of Clinical Oncology position paper adds, from outside the US market this dossier is scoped to, that "low awareness among healthcare professionals, high costs, and lack of standardized assays and comprehensive evidence" remain barriers to implementation more broadly [11] — read narrowly here as one non-US professional society's framing, not folded into the two US-relevant reviews above.

4. Sources

[1] Standard: Test categorization, 42 CFR 493.17 — eCFR, Title 42 (issue 2026-08-13; accessed 2026-09-01). 42 CFR 493.17 — https://www.ecfr.gov/current/title-42/section-493.17 [federal-registry] [2] Standard: Control procedures, 42 CFR 493.1256 — eCFR, Title 42 (issue 2026-08-13; accessed 2026-09-01). 42 CFR 493.1256 — https://www.ecfr.gov/current/title-42/section-493.1256 [federal-registry] [3] Standard; Laboratory director qualifications, 42 CFR 493.1443 — eCFR, Title 42 (issue 2026-08-13; accessed 2026-09-01). 42 CFR 493.1443 — https://www.ecfr.gov/current/title-42/section-493.1443 [federal-registry] [4] Standard; Testing personnel qualifications, 42 CFR 493.1489 — eCFR, Title 42 (issue 2026-08-13; accessed 2026-09-01). 42 CFR 493.1489 — https://www.ecfr.gov/current/title-42/section-493.1489 [federal-registry] [5] Standard; Testing personnel responsibilities, 42 CFR 493.1495 — eCFR, Title 42 (issue 2026-08-13; accessed 2026-09-01). 42 CFR 493.1495 — https://www.ecfr.gov/current/title-42/section-493.1495 [federal-registry] [6] Regulation Identification Number 0910-AJ05, Medical Devices; Laboratory Developed Tests; Implementation of Vacatur — Federal Register, Food and Drug Administration, HHS (published 2025-09-19; accessed 2026-09-01). FR Doc No. 2025-18239; 90 FR 45134 — https://www.federalregister.gov/documents/2025/09/19/2025-18239/regulation-identification-number-0910-aj05-medical-devices-laboratory-developed-tests-implementation [federal-registry] [7] ALTAIR (CIRCULATE-Japan) protocol record — ClinicalTrials.gov, sponsor National Cancer Center Hospital East (record last updated 2026-01-21; accessed 2026-09-01). NCT04457297 — https://clinicaltrials.gov/study/NCT04457297 [trial-registry] [8] BESPOKE Study of ctDNA Guided Therapy in Colorectal Cancer, protocol record — ClinicalTrials.gov, sponsor Natera, Inc. (record last updated 2023-12-12; accessed 2026-09-01). NCT04264702 — https://clinicaltrials.gov/study/NCT04264702 [trial-registry] [9] Practical integration of ctDNA-defined minimal residual disease in gastrointestinal cancers: testing windows and evidence-aligned management frameworks — Clinical & Translational Oncology (published 2026-05-24; accessed 2026-09-01). PMID 42177726 — https://pubmed.ncbi.nlm.nih.gov/42177726/ [peer-reviewed] [10] The promise of molecular residual disease (MRD) testing for patients with colorectal cancer — Clinical Advances in Hematology & Oncology (published 2025-06, day not given in the PubMed record; accessed 2026-09-01). PMID 40590865 — https://pubmed.ncbi.nlm.nih.gov/40590865/ [peer-reviewed] [11] Japan Society of Clinical Oncology position paper on appropriate clinical use of molecular residual disease (MRD) testing — International Journal of Clinical Oncology (published 2025-04, day not given in the PubMed record; accessed 2026-09-01). PMID 39920551 — https://pubmed.ncbi.nlm.nih.gov/39920551/ [guideline]

5. Sourcing & Gaps

Well established: That tumor-informed ctDNA MRD testing sits entirely inside CLIA's high-complexity-testing framework with no FDA premarket or site-inspection layer above it — CLIA's own categorization criteria [1] and the federal government's own 2025 reversal of the contrary rule [6] are two independent official sources agreeing on the same operative fact, so this is stated plainly. That turnaround time is an active, unresolved constraint on workflow integration — two independent, non-affiliated peer-reviewed reviews agree [9][10], stated plainly.

Thin: Each individual CLIA staffing requirement (director qualifications [3], testing-personnel qualifications [4], testing-personnel responsibilities [5]) rests on that one CFR section alone — [single-source] each, though all three are official regulatory text rather than a directional account. The BESPOKE bench-workflow description rests on the sponsor's own registry record, with no results yet posted — [single-source · sponsor-reported] [8]. The ALTAIR eligibility-criterion detail rests on the one registry record consulted for it — [single-source] [7], distinct from the already-published ALTAIR outcome data the dossier's own §4a treats separately. The JSCO implementation-barriers list rests on one source that is also outside this dossier's US scope — [single-source] [11], read narrowly and not merged with the two US-relevant reviews.

Rescoped from class 3: "Would oncology laboratories and clinicians integrate tumor-informed ctDNA MRD testing smoothly into existing workflow?" is unanswerable as asked. Its twin, answered above: what do published reviews and registered study protocols actually name as the workflow constraints and bench sequence for this testing, right now? [7][8][9][10][11].

Out of scope: The CAP Molecular Pathology checklist and ISO 17034 commutable-reference-material standard — both sit behind a member or purchase paywall this pass did not attempt to cross, the same disposition the dossier's own §4b and §9 already record. fda.gov's own guidance documents on LDTs — not attempted, per this report's default scope note on intermittent bot mitigation; the Federal Register document [6] was reachable and used instead. CPT/HCPCS billing site-of-service differentials — the dossier's §6 already states reimbursement is not applicable to this RUO-reagent domain and treats the downstream assay's own coding separately; not re-derived here.

Not searched vs. not found: The MolDX/Palmetto Local Coverage Determination pages describing MRD specimen-type and testing-window requirements were sought but returned an HTTP 403 to a direct fetch; recorded as not retrieved, not as absent — no equivalent identifier-bearing mirror of that specific LCD text was located this pass. An institutional send-out laboratory's own test-guide page (naming Natera's San Carlos, California and Austin, Texas CLIA sites and a 3–4-week turnaround from sample receipt) was fetched and read, but is not cited above because it carries no resolvable identifier beyond its URL under this fleet's sourcing rule; recorded as found but undocumentable in this format, not as untrue. A bench-level hour-by-hour SOP for any single reference laboratory was not searched — it is proprietary operating detail, not the kind of fact a public primary source is expected to publish.

[inference] The two-part specimen requirement (archived FFPE tissue plus a subsequent blood draw) [8] plausibly compounds with tumor-sequencing turnaround time [10] to make the first result in a course of care slower than any subsequent one — no single retrieved source states that composition directly, so it is flagged here as this report's own synthesis rather than written into section 3 as a finding.

6. Claim Candidates