{ "schema": "auturnal.deep-research-review.v2", "slug": "liquid-biopsy-screening", "title": "Finding a Rare Signal: The Statistical Architecture of Blood-Based Cancer Detection", "category": "Health and measurement", "dek": "Rare-disease screening is governed by denominators as much as by assay sensitivity.", "english": "Rare-disease screening is governed by denominators as much as by assay sensitivity. Blood-based cancer studies illustrate the distinction between case-control discrimination, prospective feasibility and clinical benefit. This review derives prevalence-dependent predictive values and a downstream-workup budget, then identifies the evidence required to move from a detectable signal to an improved population outcome.", "abstract": "Rare-disease screening is governed by denominators as much as by assay sensitivity. Blood-based cancer studies illustrate the distinction between case-control discrimination, prospective feasibility and clinical benefit. This review derives prevalence-dependent predictive values and a downstream-workup budget, then identifies the evidence required to move from a detectable signal to an improved population outcome.", "refs": [ "ef-06-1", "ef-06-2", "ef-06-3" ], "sections": [ [ "A positive signal is the beginning of a pathway", "A blood test can indicate a molecular abnormality without locating a treatable cancer. The full pathway includes confirmatory imaging, tissue diagnosis, treatment and follow-up. Each transition changes the relevant denominator. Evaluating the assay alone can miss the burden placed on people whose positive result does not lead to a cancer diagnosis. The central question is whether the complete pathway improves a specified outcome in the intended screening population.", [] ], [ "Different designs answer different questions", "Cohen and colleagues evaluated CancerSEEK in a case-control setting. Lennon and colleagues studied the feasibility of integrating a multicancer blood test with PET-CT in a prospective cohort. Chung and colleagues evaluated a cell-free DNA blood test for colorectal cancer screening in ECLIPSE. Their designs address discrimination, implementation and screening accuracy in distinct ways. They cannot be ranked by a single headline sensitivity without preserving population, cancer spectrum and verification procedures.", [ "ef-06-1", "ef-06-2", "ef-06-3" ] ], [ "Prevalence changes the meaning of accuracy", "Sensitivity is conditional on disease, and specificity is conditional on its absence. A reader usually wants the reverse conditional probability: the chance of disease after a positive test. Bayes' rule connects them through prevalence. Even a small false-positive rate can dominate a rare condition because the non-diseased population is much larger. A high area under a receiver-operating curve does not remove this arithmetic or determine the appropriate operating threshold.", [] ], [ "The detectable spectrum matters", "A study enriched for advanced disease can overstate sensitivity relevant to early screening. Tumor burden, shedding, anatomical location and biological subtype can change the signal available to the assay. Stage-specific reporting helps, but it does not eliminate differences in specimen collection or case ascertainment. Prospective sampling before diagnostic workup is valuable because it reduces the opportunity for case selection to shape the apparent performance.", [] ], [ "Verification can introduce a hidden bias", "If only positive tests receive definitive follow-up, some false negatives remain unknown. If follow-up intensity differs across groups, apparent accuracy partly measures the verification process. A study must specify how disease status is established among both positive and negative tests, and how interval cancers are counted. Excluding unresolved cases without showing their outcomes can improve reported metrics while making the screening pathway less interpretable.", [] ], [ "Earlier detection does not prove longer life", "Lead time can lengthen measured survival from diagnosis without postponing death. Overdiagnosis can add cancers that would not have caused symptoms during a person's lifetime. These mechanisms mean that detection rate and post-diagnosis survival are insufficient substitutes for population benefit. A benefit claim needs an endpoint and design capable of addressing these alternatives, with harms and competing risks assessed over a meaningful period.", [] ], [ "Multiple cancers create a routing problem", "A multicancer assay must do more than flag an abnormal specimen. Incorrect tissue-of-origin predictions can generate a sequence of inconclusive investigations. Combining per-cancer sensitivities also requires specifying the mix of cancers in the population. A weighted average can change when the target population changes even if every conditional assay property remains fixed. Routing accuracy, unresolved positives and time to diagnostic closure belong beside conventional assay metrics.", [] ], [ "A useful comparison preserves the whole denominator", "A future evaluation should account for everyone entering the pathway: invitations, participation, valid tests, positives, completed workups, diagnoses and later outcomes. Attrition is not simply missing paperwork; it can correlate with risk and access to care. Decision analyses should distinguish resource use from health outcomes and make their assumptions explicit. This review does not recommend an individual screening choice or claim that a constructed predictive-value calculation establishes clinical utility.", [] ], [ "The intended use defines the relevant control group", "A diagnostic study assembled from known cancer cases and healthy volunteers estimates a different operating condition from testing people before disease is recognized. Symptomatic controls, benign conditions and the intended age distribution can change the score distribution. A proposed evaluation would specify whether the test is used for triage, population screening or recurrence monitoring, and would fix the threshold before the independent cohort is unblinded. The number of positive tests should be linked to the number and consequences of downstream investigations. Sensitivity averaged over known cases cannot by itself establish useful early detection in an asymptomatic population.", [] ], [ "Outcome benefit requires a separate causal bridge", "A test can advance the date of diagnosis without extending life, and it can detect lesions that would not have caused harm during follow-up. The review's predictive-value arithmetic addresses neither issue. A clinical-benefit claim would require a design that compares meaningful outcomes under clearly specified diagnostic and treatment pathways. Stage distribution can be an intermediate signal, but it is not identical to reduced mortality or improved quality of life. The evidence chain should keep analytical detection, clinical discrimination and outcome benefit as distinct propositions. A successful analytical assay can justify further evaluation without satisfying the final proposition.", [] ], [ "A model that can be challenged", "The four equations below are a compact analytical model, not a joint fit to all the cited papers. Their interpretation is as follows. p is disease prevalence, Se sensitivity and Sp specificity in a defined population; PPV is conditional on a positive result. Expected positives among N screened people, assuming common conditional performance and complete testing. Expected false negatives; the formula does not model future disease arising after the screening episode. If each positive triggers an average w downstream procedures, W is a simple workload expectation, not a harm or benefit estimate.\nThe observational target is positive predictive value and completed diagnostic pathways. The model deliberately isolates a tractable mechanism; its purpose is to reveal which measurements would distinguish that mechanism from an alternative. Parameter values in the figures are declared illustrative inputs. A curve produced by those values is a mathematical consequence, not an estimate of an unmeasured biological or physical quantity.", [] ], [ "A discriminating validation design", "The primary comparison should test performance and downstream burden at the prespecified operating threshold. The unit of independent replication is screened participant; repeated readings from the same unit improve precision but do not create independent replication. A strong validation would hold out a prospectively enrolled screening population before choosing thresholds or fitting nuisance corrections. In particular, disease spectrum and incomplete verification must be measured, constrained or varied in a sensitivity analysis. Report both the principal estimate and how it changes when that assumption is relaxed. A model that succeeds only after choosing the most favorable exclusion rule has answered a different question from one evaluated with a locked rule.", [] ], [ "Worked numerical analysis", "In an illustrative population of 100,000 with prevalence 0.5%, sensitivity 80% and specificity 99%, the expected counts are 400 true positives and 995 false positives. PPV is about 28.7%. Raising specificity to 99.9% while holding the other assumptions fixed gives 99.5 expected false positives and PPV about 80.1%. Fractional expected counts are legitimate population averages, not literal partial people. The figures vary prevalence and specificity; none of these values is attributed to a cited assay.\nThese calculations can be reconstructed from the companion Python file and its parameter table. Axis labels retain the model's units, and the caption distinguishes the quantities varied from those held fixed. The numerical examples were selected to expose a dependency, not to reproduce the appearance of a published figure. A useful independent check is to evaluate a limiting case and then recompute an intermediate point without the plotting code. Agreement validates the arithmetic within the declared model; it does not establish that nature follows that model.", [] ], [ "Evidence, transfer and next steps", "The scope of the conclusion is limited by disease spectrum and incomplete verification. Transfer to a prospectively enrolled screening population requires evidence beyond fitting the original observations. A future replication should preserve raw measurements, calibration records, exclusions and a machine-readable link between each result and the corresponding analysis. The evidence table below identifies what each selected primary study contributes. The comparison is a focused narrative synthesis, not a systematic review or a claim that all subsequent evidence has been surveyed. The proposed study has not been performed here. Its value is that a negative result would constrain an explicit explanation rather than merely lower a benchmark score.", [] ] ], "equations": [ { "tex": "\\mathrm{PPV}=\\frac{p\\,Se}{p\\,Se+(1-p)(1-Sp)}", "explanation": "p is disease prevalence, Se sensitivity and Sp specificity in a defined population; PPV is conditional on a positive result." }, { "tex": "N_+=N\\{p\\,Se+(1-p)(1-Sp)\\}", "explanation": "Expected positives among N screened people, assuming common conditional performance and complete testing." }, { "tex": "N_{FN}=Np(1-Se)", "explanation": "Expected false negatives; the formula does not model future disease arising after the screening episode." }, { "tex": "W=N_+w", "explanation": "If each positive triggers an average w downstream procedures, W is a simple workload expectation, not a harm or benefit estimate." } ], "table": { "headers": [ "Evidence or design choice", "What it identifies", "What remains unresolved" ], "rows": [ [ "Source 1", "CancerSEEK evaluated combined circulating protein and DNA information.", "Case-control performance does not establish population benefit." ], [ "Source 2", "A prospective study examined integration of blood testing and PET-CT.", "Feasibility does not establish reduced cancer mortality." ], [ "Source 3", "ECLIPSE evaluated blood-test accuracy for colorectal cancer screening.", "A single-round accuracy study has bounded disease and follow-up scope." ], [ "Primary observable", "positive predictive value and completed diagnostic pathways", "disease spectrum and incomplete verification" ], [ "Independent test", "performance and downstream burden at the prespecified operating threshold", "Requires new measurements" ], [ "Transfer set", "a prospectively enrolled screening population", "Not validated by the constructed numerical example" ] ] }, "protocol": [ "Specify positive predictive value and completed diagnostic pathways before fitting.", "Retain independent identifiers for each screened participant.", "Separate a prospectively enrolled screening population from model development.", "Measure or bound disease spectrum and incomplete verification.", "Record units, calibration and the observation operator.", "Compare the primary model with a named rival explanation.", "Prespecify the test: performance and downstream burden at the prespecified operating threshold.", "Archive raw data, exclusions, parameter files and executable analysis." ], "limitations": [ "Case-control performance does not establish population benefit.", "Feasibility does not establish reduced cancer mortality.", "A single-round accuracy study has bounded disease and follow-up scope.", "Companion figures are constructed analytical examples, not new experimental results.", "Transfer to a prospectively enrolled screening population has not been demonstrated by this review." ], "evidence": [ { "source": "ef-06-1", "result": "CancerSEEK evaluated combined circulating protein and DNA information.", "boundary": "Case-control performance does not establish population benefit.", "locator": "Abstract / selected primary-source sections; see linked bibliographic record." }, { "source": "ef-06-2", "result": "A prospective study examined integration of blood testing and PET-CT.", "boundary": "Feasibility does not establish reduced cancer mortality.", "locator": "Abstract / selected primary-source sections; see linked bibliographic record." }, { "source": "ef-06-3", "result": "ECLIPSE evaluated blood-test accuracy for colorectal cancer screening.", "boundary": "A single-round accuracy study has bounded disease and follow-up scope.", "locator": "Abstract / selected primary-source sections; see linked bibliographic record." } ], "research_questions": [ "Which endpoints distinguish detection from meaningful clinical benefit?", "How does unresolved diagnostic workup alter net population value?", "Which sampling designs reveal stage-dependent ascertainment bias?" ], "diagram": [ "Screened population", "Blood signal", "Diagnostic workup", "Health outcomes" ], "keywords": [ "Health and measurement", "liquid-biopsy-screening", "measurement", "identifiability" ], "kind": "ai_assisted_research_guide", "author": "Codex (AI)", "original_language": "en", "created_date": "2026-10-22", "published_date": "2026-10-22T13:02:56+00:00", "publication_status": "published_commissioned_review", "version": 5, "previous_version": "/guides/liquid-biopsy-screening/v4/", "analytic": "liquid-biopsy-screening", "peer_reviewed": false, "journal_publication": false, "disclosure": { "authorship": "Source-based research review and original mathematical analysis written by Codex (AI). Cited experiments belong to the cited research teams.", "figures": "Concept cover created with the built-in ImageGen tool. Quantitative figures are explicit constructed analyses, with code and numerical data; they are not measurements from the cited papers.", "review": "This local draft has no independent peer-review or journal-acceptance record." }, "revision_note": "First English local draft; upload and publication deferred at the owner's request.", "review_scope": "A focused narrative analysis of the named primary studies, not a systematic literature search or a claim to comprehensive coverage through 2026.", "section_kinds": [ "author_analysis_or_verification_proposal", "source_based_review", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal", "author_analysis_or_verification_proposal" ], "source_slug": "liquid-biopsy-screening", "source_revision": "41dc5ea7bf1db9868e2ee5b121f7021c823cc018", "source_article_sha256": "aeefa7df89581d22967f65942db2ba04ae17fa8f41ed1dd2cb2b75712d1d5dca", "sources": [ { "id": "ef-06-1", "doi": "10.1126/science.aar3247", "url": "https://doi.org/10.1126/science.aar3247", "title": "Detection and localization of surgically resectable cancers with a multi-analyte blood test", "authors": [ { "given": "Joshua D.", "family": "Cohen" }, { "given": "Lu", "family": "Li" }, { "given": "Yuxuan", "family": "Wang" }, { "given": "Christopher", "family": "Thoburn" }, { "given": "Bahman", "family": "Afsari" }, { "given": "Ludmila", "family": "Danilova" }, { "given": "Christopher", "family": "Douville" }, { "given": "Ammar A.", "family": "Javed" }, { "given": "Fay", "family": "Wong" }, { "given": "Austin", "family": "Mattox" }, { "given": "Ralph H.", "family": "Hruban" }, { "given": "Christopher L.", "family": "Wolfgang" }, { "given": "Michael G.", "family": "Goggins" }, { "given": "Marco", "family": "Dal Molin" }, { "given": "Tian-Li", "family": "Wang" }, { "given": "Richard", "family": "Roden" }, { "given": "Alison P.", "family": "Klein" }, { "given": "Janine", "family": "Ptak" }, { "given": "Lisa", "family": "Dobbyn" }, { "given": "Joy", "family": "Schaefer" }, { "given": "Natalie", "family": "Silliman" }, { "given": "Maria", "family": "Popoli" }, { "given": "Joshua T.", "family": "Vogelstein" }, { "given": "James D.", "family": "Browne" }, { "given": "Robert E.", "family": "Schoen" }, { "given": "Randall E.", "family": "Brand" }, { "given": "Jeanne", "family": "Tie" }, { "given": "Peter", "family": "Gibbs" }, { "given": "Hui-Li", "family": "Wong" }, { "given": "Aaron S.", "family": "Mansfield" }, { "given": "Jin", "family": "Jen" }, { "given": "Samir M.", "family": "Hanash" }, { "given": "Massimo", "family": "Falconi" }, { "given": "Peter J.", "family": "Allen" }, { "given": "Shibin", "family": "Zhou" }, { "given": "Chetan", "family": "Bettegowda" }, { "given": "Luis A.", "family": "Diaz" }, { "given": "Cristian", "family": "Tomasetti" }, { "given": "Kenneth W.", "family": "Kinzler" }, { "given": "Bert", "family": "Vogelstein" }, { "given": "Anne Marie", "family": "Lennon" }, { "given": "Nickolas", "family": "Papadopoulos" } ], "journal": "Science", "year": 2018, "article_slug": "liquid-biopsy-screening", "verified_date": "2026-10-22", "verification": "DOI/title/authors/journal metadata checked against Crossref. Narrative claims were scoped to accessible primary publisher or author records, abstracts and selected text; full-text and supplementary-methods coverage is not exhaustive.", "source_article_slug": "liquid-biopsy-screening" }, { "id": "ef-06-2", "doi": "10.1126/science.abb9601", "url": "https://doi.org/10.1126/science.abb9601", "title": "Feasibility of blood testing combined with PET-CT to screen for cancer and guide intervention", "authors": [ { "given": "Anne Marie", "family": "Lennon" }, { "given": "Adam H.", "family": "Buchanan" }, { "given": "Isaac", "family": "Kinde" }, { "given": "Andrew", "family": "Warren" }, { "given": "Ashley", "family": "Honushefsky" }, { "given": "Ariella T.", "family": "Cohain" }, { "given": "David H.", "family": "Ledbetter" }, { "given": "Fred", "family": "Sanfilippo" }, { "given": "Kathleen", "family": "Sheridan" }, { "given": "Dillenia", "family": "Rosica" }, { "given": "Christian S.", "family": "Adonizio" }, { "given": "Hee Jung", "family": "Hwang" }, { "given": "Kamel", "family": "Lahouel" }, { "given": "Joshua D.", "family": "Cohen" }, { "given": "Christopher", "family": "Douville" }, { "given": "Aalpen A.", "family": "Patel" }, { "given": "Leonardo N.", "family": "Hagmann" }, { "given": "David D.", "family": "Rolston" }, { "given": "Nirav", "family": "Malani" }, { "given": "Shibin", "family": "Zhou" }, { "given": "Chetan", "family": "Bettegowda" }, { "given": "David L.", "family": "Diehl" }, { "given": "Bobbi", "family": "Urban" }, { "given": "Christopher D.", "family": "Still" }, { "given": "Lisa", "family": "Kann" }, { "given": "Julie I.", "family": "Woods" }, { "given": "Zachary M.", "family": "Salvati" }, { "given": "Joseph", "family": "Vadakara" }, { "given": "Rosemary", "family": "Leeming" }, { "given": "Prianka", "family": "Bhattacharya" }, { "given": "Carroll", "family": "Walter" }, { "given": "Alex", "family": "Parker" }, { "given": "Christoph", "family": "Lengauer" }, { "given": "Alison", "family": "Klein" }, { "given": "Cristian", "family": "Tomasetti" }, { "given": "Elliot K.", "family": "Fishman" }, { "given": "Ralph H.", "family": "Hruban" }, { "given": "Kenneth W.", "family": "Kinzler" }, { "given": "Bert", "family": "Vogelstein" }, { "given": "Nickolas", "family": "Papadopoulos" } ], "journal": "Science", "year": 2020, "article_slug": "liquid-biopsy-screening", "verified_date": "2026-10-22", "verification": "DOI/title/authors/journal metadata checked against Crossref. Narrative claims were scoped to accessible primary publisher or author records, abstracts and selected text; full-text and supplementary-methods coverage is not exhaustive.", "source_article_slug": "liquid-biopsy-screening" }, { "id": "ef-06-3", "doi": "10.1056/nejmoa2304714", "url": "https://doi.org/10.1056/nejmoa2304714", "title": "A Cell-free DNA Blood-Based Test for Colorectal Cancer Screening", "authors": [ { "given": "Daniel C.", "family": "Chung" }, { "given": "Darrell M.", "family": "Gray" }, { "given": "Harminder", "family": "Singh" }, { "given": "Rachel B.", "family": "Issaka" }, { "given": "Victoria M.", "family": "Raymond" }, { "given": "Craig", "family": "Eagle" }, { "given": "Sylvia", "family": "Hu" }, { "given": "Darya I.", "family": "Chudova" }, { "given": "AmirAli", "family": "Talasaz" }, { "given": "Joel K.", "family": "Greenson" }, { "given": "Frank A.", "family": "Sinicrope" }, { "given": "Samir", "family": "Gupta" }, { "given": "William M.", "family": "Grady" } ], "journal": "New England Journal of Medicine", "year": 2024, "article_slug": "liquid-biopsy-screening", "verified_date": "2026-10-22", "verification": "DOI/title/authors/journal metadata checked against Crossref. Narrative claims were scoped to accessible primary publisher or author records, abstracts and selected text; full-text and supplementary-methods coverage is not exhaustive.", "source_article_slug": "liquid-biopsy-screening" } ], "original_source_url": "/guides/assets/v9-english-frontiers-articles-liquid-biopsy-screening-json-aeefa7df89581d22.json", "pdf": { "document_id": "6bff3497-3f1f-568b-9c29-31ddf75c8190", "version_id": "84cb38f1-aee8-5739-8ec0-1dc6242121da", "media_type": "application/pdf", "authentication_required": true, "download_audited": true }, "presentation_change": "Complete version-bound translated reading views and selectable authenticated original/translated PDFs; original scientific text and evidence retained.", "source_locale": "en", "source_version_id": "d3a14b14-7053-56e6-8496-e14da1d76654", "translations": [ { "locale": "zh", "translation_id": "68413e63-a3e4-505f-b5e8-869fc0f67383", "translated_content_sha256": "04bed2d8a21065a149931e4105e2ac16dcc79d03745399ea72f6d30f25939ee2", "sha256": "48bc26f02ddb55d29bfd9cf6b63e42e91965b9d4541a458707243b07ea94296a", "byte_length": 966753 }, { "locale": "es", "translation_id": "54a3b1a1-2c81-55bb-b585-a88ba4594dd6", "translated_content_sha256": "413a57de6aa85b6997ee22bcced906ff02d9c22792d303e2b3080700ed2a6633", "sha256": "5b877cbc5b365c4b8de944fcc1d60c16cea16606ee648fa805c909b3a6552777", "byte_length": 736560 }, { "locale": "ja", "translation_id": "247ca61a-a083-5b6a-8ac2-7cdf97eb02f1", "translated_content_sha256": "5a8fba7e61453163e424a3e3c808d4c46f038463b19c6c1a810c14b62a68a771", "sha256": "42fa051bce980be2fc7927b09ac8f23ba2b54e18bcffef187f15d54c3fbf5a47", "byte_length": 934048 }, { "locale": "ar", "translation_id": "4ceda0fe-4121-5bac-8fe6-a4424710bcfd", "translated_content_sha256": "01234d04e38cd9ff0e66a5f88e86610a42d2265abd6337381323801457f6ecc5", "sha256": "08e2df202c1ba23a9f04c67e56e108886cec3d4cde72c7a55af431663a4baff7", "byte_length": 723892 } ] }