How multi-cancer blood tests promise to detect disease earlier, why clinical trials are producing complicated results, and whether finding cancer sooner will ultimately save more lives.
1. The Promise in One Tube of Blood
Cancer screening in modern medicine is divided among tests for individual organs. Depending on age, sex, medical history, and risk factors, adults may be advised to navigate distinct screening protocols: mammography for breast cancer, colonoscopy or fecal immunochemical tests for colorectal cancer, cervical cytology and human papillomavirus testing for cervical cancer, and low-dose computed tomography (CT) for adults with a substantial history of tobacco use. Each procedure evaluates a single anatomical site, demands its own clinical workflow, and carries its own physical preparation.
Routine screening is available for only a subset of cancers. The vast majority of human cancer types, including pancreatic, ovarian, esophageal, liver, stomach, and gallbladder cancers, have no established, routine screening programs for asymptomatic individuals. National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) data show that esophageal cancer is often diagnosed after it has spread beyond the primary site: 32 percent of cases are regional and 39 percent are distant, compared with 19 percent localized [2]. For pancreatic cancer, 51 percent of cases are diagnosed at the distant stage [3]. These SEER summary-stage categories describe anatomical extent and should not be conflated with American Joint Committee on Cancer (AJCC) stages I through IV. One US analysis estimated that, among adults aged 50 and older, 69.8 percent of cancer deaths were from cancer types not addressed by USPSTF screening recommendations after accounting for lung cancer screening eligibility. When the estimate also accounted for the rate of lung cancer screening receipt, the proportion rose to 84.3 percent [1].
Multi-cancer early detection (MCED) blood tests propose an alternative approach to this clinical challenge. Rather than scanning individual organs sequentially through separate imaging modalities or invasive procedures, MCED tests aim to detect biological signals associated with dozens of cancer types from a single peripheral blood sample. Substantial investment from commercial developers, including GRAIL, Exact Sciences, Guardant Health, and Freenome, has accelerated the technological and clinical development of these assays. The core premise is that identifying circulating tumor signals across multiple organ sites before symptoms emerge could expand early-stage surgical and systemic interventions.
In late September 2026, however, the clinical evidence supporting this approach faced two pivotal evaluations. On September 22, 2026, The New England Journal of Medicine published the primary results of the NHS-Galleri trial, a randomized controlled trial evaluating the Galleri test across more than 140,000 participants in England. The trial did not achieve its primary endpoint: it failed to demonstrate a statistically significant reduction in the combined incidence of stage III and stage IV cancers across 12 prespecified tumor types after three rounds of annual screening [4].
The following day, on September 23, an advisory panel convened by the U.S. Food and Drug Administration reviewed GRAIL’s premarket approval application for the same assay. According to the FDA 24-Hour Meeting Summary, the FDA’s Molecular and Clinical Genetics Panel voted unanimously (10–0) that the test demonstrated reasonable assurance of safety, split 6–4 on whether its effectiveness was reasonably assured, and voted 7–2, with one abstention, that its benefits outweighed its risks when used as an adjunct to guideline-recommended screening in adults aged 50 and older [7].
The contrast between a trial that missed its primary endpoint and an FDA advisory panel that voted that the benefit-risk balance was favorable for the proposed indication (a non-binding recommendation, not a regulatory decision) reflects a central question: a test can detect cancer signals in blood without yet demonstrating that screening reduces deaths.
2. How the Test Reads Molecular Signals
Blood contains fragments of DNA, known as cell-free DNA (cfDNA), released as human cells turn over. In healthy physiology, cellular turnover constantly releases fragmented strands of chromatin into peripheral circulation. Cell-free DNA is cleared from circulation relatively quickly, although estimates of its half-life vary with the biological context and measurement method. Cellular uptake, blood-borne nucleases, and clearance pathways in organs including the liver, spleen, and kidneys contribute to its elimination [16].
Malignant tumors also shed genetic material through cellular apoptosis and necrosis, releasing circulating tumor DNA (ctDNA) into the vascular stream. In early-stage disease, circulating tumor DNA can make up only a very small fraction of circulating cell-free DNA, with shedding rates varying widely across cancer types. Attempting to identify early-stage malignancy solely by sequencing somatic gene mutations (such as variants in KRAS or TP53) is complicated by biological background noise: non-malignant hematopoietic stem cells routinely acquire somatic mutations as an individual ages, a phenomenon known as clonal hematopoiesis; when specific diagnostic criteria are met, it is termed clonal hematopoiesis of indeterminate potential (CHIP), generating mutant fragments that circulate in plasma without representing solid tumors [18].
To overcome the limitations of mutation-based sequencing, leading assays such as Galleri evaluate DNA methylation patterns. DNA methylation is an epigenetic mechanism wherein chemical methyl groups attach to cytosine bases in the genome, regulating gene expression without altering the genetic sequence itself. Different human tissues exhibit distinct methylation profiles that reflect their developmental lineage and cellular identity. Cancer can alter DNA methylation patterns, including hypermethylation in some regulatory regions and broader hypomethylation across parts of the genome. These changes vary by tumor type and are among the molecular features that researchers investigate for cancer detection and tissue-of-origin prediction.
Galleri uses targeted DNA methylation sequencing and a machine-learning classifier to assess whether a cancer-associated signal is present [10]. If a signal is detected, the classifier predicts its likely tissue of origin. These are screening outputs, not a confirmed diagnosis or proof that the predicted organ contains a tumor. Published methodological work describes how targeted methylation patterns can support multi-cancer detection and tissue-of-origin prediction across diverse malignant and non-malignant samples [13], generating two sequential outputs:
- A Binary Cancer Signal: The algorithm evaluates DNA methylation patterns to determine whether the observed signal meets a predefined threshold for a cancer-associated result, returning either “Cancer Signal Detected” or “No Cancer Signal Detected.”
- Cancer Signal Origin (CSO) Prediction: If a cancer signal is identified, the classifier predicts the likely anatomical site of the originating malignancy, providing one or two predicted tissue locations (such as pancreas, lung, or upper gastrointestinal tract).
Clinical evaluation requires distinguishing an observational screening test from a diagnostic procedure. An MCED test does not establish a clinical diagnosis; it identifies a molecular signal in circulating blood. A positive result requires appropriate clinical evaluation, which may include diagnostic imaging, laboratory tests, endoscopy, and tissue biopsy when indicated, to determine whether cancer is present and where it originated. An MCED test is a screening tool, not a test of inherited cancer susceptibility. It looks for molecular signals associated with cancer that may already be present; it is not designed to predict whether a currently cancer-free individual will develop cancer in the future. These distinctions are important when interpreting both positive and negative results.
3. The Trial That Complicated the Promise
Until late 2026, clinical evidence for multi-cancer blood tests rested primarily on observational studies, case-control cohorts, and single-arm trials. While these designs characterize test performance in participants with known clinical outcomes, they cannot demonstrate that screening healthy populations alters the trajectory of disease.
A major randomized test of the approach was conducted through the NHS-Galleri trial, established in partnership between NHS England, Queen Mary University of London, and GRAIL. As detailed in The New England Journal of Medicine, the trial consented 142,826 adults aged 50 to 77 across England and randomized 142,250 after baseline blood collection. Of those randomized, 71,122 participants were assigned to the intervention group offered three annual Galleri tests alongside usual care, and 71,128 were assigned to the control group receiving usual care while blood samples were stored without analysis [4].
The primary endpoint of the trial was the combined incidence rate of stage III and stage IV cancers across 12 prespecified cancer types (anal, bladder, bowel, esophageal, head and neck, liver, lung, ovarian, pancreatic, stomach, lymphoma, and myeloma) after three annual screening rounds.
The underlying premise of this endpoint is the concept of “stage shift.” If a screening test successfully identifies tumors when they are localized or less advanced, fewer patients should ultimately be diagnosed with late-stage disease. If screening detects additional early-stage cancers without reducing the incidence of stage III and stage IV diagnoses, detecting more tumors at earlier stages does not, by itself, demonstrate a reduction in the burden of advanced cancer.
The primary analysis published in September 2026 demonstrated that the trial did not achieve its primary endpoint:
- The incidence rate ratio (IRR) for combined stage III and stage IV cancers in the intervention group compared with the control group was 1.03 (95% confidence interval [CI], 0.92 to 1.14; P = 0.63) [4].
- The combined rate of late-stage diagnoses across the prespecified cancer types did not differ significantly between screened participants and those receiving standard care alone.
For stage IV cancers considered separately, the incidence rate ratio was 0.86 (95% CI, 0.74 to 1.00) [4]. This is a potentially important signal, but it is not conclusive: the confidence interval reaches 1.00, and the trial did not meet its primary endpoint. Further follow-up is needed.
The trial did not establish why the combined late-stage endpoint remained unchanged. The study authors noted that further analyses and longer-term follow-up are needed to evaluate the biological factors and clinical pathways influencing late-stage cancer incidence.
The trial’s primary outcome had important policy implications. In November 2020, NHS England announced a Galleri trial involving 165,000 people, with the possibility of expanding the programme to around one million participants if the results were positive [5, 6]. In 2024, after reviewing preliminary data from the first year, NHS England said the evidence was not compelling enough to proceed directly to the large-scale pilot while awaiting the final trial results [6]. Published in September 2026, the final results did not demonstrate a statistically significant reduction in the combined incidence of stage III and stage IV cancers across the 12 prespecified cancer types [4]. Further follow-up, including cancer-mortality outcomes, will be needed to inform decisions about population-wide screening.
4. Finding Cancer vs. Saving Lives
A laboratory test can perform reliably under analytical standards without demonstrating that its widespread use saves lives. Evaluating an MCED test requires examining three distinct, non-interchangeable properties:
- Sensitivity: True Positives / (True Positives + False Negatives)
- Specificity: True Negatives / (True Negatives + False Positives)
- Positive Predictive Value (PPV): True Positives / (True Positives + False Positives)
In a peer-reviewed publication in Nature Medicine, investigators reported 12-month performance results from PATHFINDER 2, a prospective interventional study of asymptomatic adults aged 50 and older in the United States. Of 35,350 participants in the analyzable cohort, 32,007 had completed 12-month cancer-status assessments and formed the performance-analysis set [9].
The study reported the following performance parameters within its defined cohort [9]:
| Parameter | PATHFINDER 2 Reported Result | Interpretation within Study Scope |
|---|---|---|
| Test Positivity Rate | 0.9% | About 9 in 1,000 participants received a positive result (287 of 32,007); the confirmed cancer detection rate was 0.54% (173 of 32,007). |
| Specificity | 99.6% | High specificity was consistent with the vast majority of participants without cancer receiving negative results. |
| Positive Predictive Value (PPV) | 60.3% | About six out of ten positive signals (173 of 287) were confirmed as cancer within the 12-month evaluation window. |
| Episode Sensitivity (All Cancers) | 39.3% | The test detected roughly four out of ten cancers that surfaced among participants during the study year. |
| Sensitivity (12 Selected Cancers) | 69.8% | Higher sensitivity was observed among 12 prespecified cancers accounting for major US cancer mortality. |
| Cancer Signal Origin Accuracy | 91.3% | Among true-positive cases, the predicted primary site matched clinical findings in roughly nine out of ten cases. |
Conflating specificity with overall diagnostic accuracy is an analytical error. A specificity of 99.6 percent indicates that the test rarely produces a false signal in a participant without cancer; it does not indicate that the test detects 99.6 percent of all tumors.
Furthermore, PATHFINDER 2 demonstrated that test sensitivity was substantially lower for localized stage I and II cancers than for advanced stage III and IV cancers. In the study publication, the authors noted that early-stage tumors shed less circulating DNA into peripheral blood, which complicates detection at localized stages across several tumor types [9].
These performance characteristics carry three direct clinical consequences:
The Diagnostic Workup (False Positives)
In PATHFINDER 2, the positive predictive value was 60.3 percent, meaning that roughly 40 percent of individuals who received a positive test did not have cancer confirmed within the 12-month evaluation window. This does not establish that those participants were cancer-free indefinitely. Positive predictive value depends directly on disease prevalence in the screened population. When an asymptomatic person receives an unconfirmed cancer signal, further clinical evaluation is generally recommended to determine whether cancer is present and, if so, where it originated. In PATHFINDER 2, the median time to diagnostic resolution after a positive result was 48 days [9]. This figure describes the study population and should not be treated as a guaranteed timeline for clinical practice. Follow-up investigations typically include cross-sectional CT imaging, magnetic resonance imaging (MRI), positron emission tomography (PET), endoscopy, or tissue biopsies. For individuals whose diagnostic workup reveals no malignancy, the process involves financial expense, procedural risks from invasive investigations, and psychological distress.
False Reassurance (False Negatives)
With an overall 12-month episode sensitivity of 39.3 percent in PATHFINDER 2, a negative test result does not establish the absence of cancer. The FDA’s review materials emphasize that a negative Galleri result does not rule out cancer and that individuals should continue guideline-recommended screening for individual cancer types [7]. Multi-cancer blood tests should therefore be considered alongside established screening methods, not as replacements for them. If someone interprets it as comprehensive clearance and skips recommended mammography, colorectal, or cervical screening, the decision could delay diagnosis of a cancer that routine screening might otherwise detect.
Overdiagnosis and Lead-Time Bias
A potential concern with cancer screening is overdiagnosis: detecting a real, histologically confirmed tumor that would never have progressed to cause symptoms or death during the person’s lifetime. The extent of overdiagnosis associated with multi-cancer early detection tests remains uncertain and requires further study. As described in screening research by H. Gilbert Welch and William C. Black [14], cancers exhibit variable growth rates:
- Highly aggressive cancers may arise and advance rapidly between annual screening cycles.
- Moderately growing cancers may be detected early enough to alter clinical outcomes.
- Indolent, slow-growing cancers shed detectable biomarkers but may not threaten life.
Detecting indolent cancers can lead to clinical treatments, including surgery, radiation, and pharmacotherapy, that carry physical toxicities without extending survival.
Additionally, extending the interval between diagnosis and death does not automatically prove life extension. Under lead-time bias, diagnosing a tumor earlier in its natural history moves the date of diagnosis forward without altering the date of death.
The following diagram illustrates the concept of lead-time bias, showing how earlier detection can lengthen the measured survival interval without necessarily delaying the time of death:
Hypothetical Conceptual Illustration of Lead-Time Bias
Natural Disease Course (Without Blood Screening):
Diagnosed by clinical symptoms at age 62 ────────────────── Death at age 65
[Calculated Survival: 3 Years]
Screen-Detected Course (With Blood Screening):
Diagnosed by blood signal at age 57 ───────────────────── Death at age 65
[Calculated Survival: 8 Years]
The measured survival interval increases from 3 to 8 years, while the age at death remains unchanged.
Cancer-specific mortality in well-designed randomized screening trials is a key way to determine whether earlier detection actually extends life, rather than merely lengthening the time a person lives with a diagnosis.
5. The Regulatory and Commercial Race
While randomized trials continue to evaluate long-term outcomes, commercial access to multi-cancer screening has expanded through distinct regulatory frameworks.
In the United States, Galleri has been marketed since 2021 as a laboratory-developed test (LDT). Under the Clinical Laboratory Improvement Amendments (CLIA) of 1988, administered by the Centers for Medicare & Medicaid Services (CMS), federal regulations set quality standards for laboratories performing clinical testing, including requirements for laboratories to establish and verify the performance specifications of their testing methods, as applicable under CLIA. CLIA establishes quality requirements for laboratories performing clinical testing, including performance specifications for methods that have not been cleared or approved by the FDA. CLIA compliance does not, by itself, establish a test’s clinical validity or demonstrate that its use improves patient outcomes. FDA premarket review is a separate regulatory process, and as of early October 2026 Galleri had not received FDA approval. The FDA’s 2024 rule asserting oversight of LDTs was vacated by a federal court in March 2025, and the agency later issued a rule reverting the regulation in September 2025 [17].
GRAIL lists the Galleri test at $949 [10], although the price paid may vary by provider and payment arrangement. The test is prescription-only. Coverage varies by plan and program, so patients should confirm any potential reimbursement or discount directly with their insurer, employer benefit program, or other applicable payer. Insurance coverage remains distinct from FDA premarket authorization.
On September 23, 2026, the FDA’s Molecular and Clinical Genetics Panel reviewed GRAIL’s formal Premarket Approval (PMA) application, evaluating whether the test should receive official FDA approval as an adjunct to guideline-recommended screening in adults aged 50 and older. According to the FDA 24-Hour Meeting Summary, the committee’s advisory votes were split across three specific regulatory determinations [7]:
- Safety: Voted 10–0 that reasonable assurance of safety was demonstrated.
- Effectiveness: Split 6–4 on whether reasonable assurance of effectiveness was demonstrated.
- Benefit-Risk Balance: Voted 7–2 (with 1 abstention) that the benefits outweighed the risks for the proposed indication.
The panel’s deliberations reflected differing viewpoints on the clinical trade-offs. Panel members voting in favor noted that the test identified cancers lacking routine screening options (such as pancreatic and head and neck cancers) with high tissue-of-origin accuracy. Dissenting panel members raised concerns regarding the claim of “early” detection, the absence of long-term mortality data, and the clinical burden of false-positive evaluations. Several supportive votes were conditioned on post-approval study requirements, labeling safeguards, and physician education [7]. An FDA advisory committee vote is a formal recommendation; it does not constitute final regulatory approval, and FDA authorization should not be assumed until the agency issues a formal decision. As of early October 2026, reporting indicated that the agency had not announced one.
The test has also entered international markets:
- Brazil: In October 2026, Galleri launched commercially in Brazil through Genesis Genomics, a joint initiative between Hospital Israelita Albert Einstein and Grupo Fleury. As reported in Folha de S.Paulo, the test was introduced at an initial launch price of R$6,990 (approximately US$1,250) [11], offered on an out-of-pocket, self-pay basis requiring a physician’s prescription, with reporting confirming it is not covered by private health insurance plans.
- India: In India, multi-cancer blood tests are not included in national cancer screening programs. Under the Ministry of Health and Family Welfare’s Operational Guidelines on Prevention, Screening and Control of Common NCDs, public health screening initiatives remain focused on oral, breast, and cervical cancers [15]. The policy does not establish routine population screening or public funding for commercial multi-cancer early detection blood tests.
6. Who Benefits, and What Happens Next?
The economic debate over multi-cancer blood testing centers on whether the clinical benefits of screening justify its purchase price and the costs of follow-up care.
For commercial developers and sequencing providers, population screening represents a substantial market opportunity. Reuters, citing Gabelli Funds asset managers, reported that analysts saw the potential for a US market approaching $60 billion annually by 2030 if insurance coverage for multi-cancer screening became widespread [12]. This is a conditional market projection, not a measured current market size or a guaranteed forecast.
For healthcare systems and insurance payers, however, the direct cost of the screening assay represents only the initial expenditure. Positive results can require multiple laboratory tests, imaging studies, and, in some cases, invasive procedures. If MCED screening is introduced at scale, health systems will need to assess whether existing diagnostic services can accommodate the additional demand without delaying care for symptomatic patients. In PATHFINDER 2, participants with a positive result underwent diagnostic workups that could involve multiple imaging modalities and clinical visits to evaluate a signal [9]. Patients and healthcare systems may bear the financial and procedural costs of investigations that do not identify cancer.
To establish the clinical evidence required for broad population adoption, public health agencies are pursuing larger randomized trial designs. The National Cancer Institute is conducting its Vanguard Study, a pilot trial enrolling up to 24,000 participants [8] to evaluate feasibility, participant adherence, diagnostic pathways, and clinical logistics of multi-cancer screening. The Vanguard Study is designed to inform a potential large-scale randomized controlled trial powered to evaluate whether multi-cancer screening reduces cancer mortality. Notably, the Vanguard pilot selected two other MCED assays (Avantect and Shield) rather than Galleri to evaluate screening performance in its cohort [8].
Simultaneously, extended follow-up of the NHS-Galleri trial cohort in England is expected to provide further evidence on cancer-specific mortality and longer-term outcomes. Researchers and public health authorities, including the UK National Screening Committee, can consider those findings alongside other emerging evidence when assessing future population-screening policy.
7. Conclusion
The ability to detect cancer-associated epigenetic signals in peripheral blood represents a significant technological achievement in molecular diagnostics. Yet the transition from analytical detection to demonstrated public health benefit requires overcoming substantial clinical and epidemiological hurdles.
As clinical trials and regulatory reviews have made clear, identifying a molecular signal is not equivalent to establishing a clinical diagnosis, and diagnosing a tumor earlier is not synonymous with extending a patient’s life. The inconclusive late-stage cancer reduction in the NHS-Galleri trial underscores the complexity of transforming biological signals into clinical outcomes. Until randomized trials demonstrate definitive reductions in advanced-stage disease and cancer-specific mortality, multi-cancer early detection tests remain an adjunct to established screening programs rather than a replacement for them. The central task for clinical oncology and public health authorities over the coming decade will be determining whether detecting cancer sooner in a tube of blood ultimately saves more lives.
Sources and Further Reading
The following sources provide the research, clinical evidence and policy context discussed in this article. Bracketed numbers in the text refer to this list.
- Estimated proportion of cancer deaths not addressed by current cancer screening efforts in the United States. Research on the proportion of cancer deaths not covered by existing screening recommendations, including the estimates for adults aged 50 and older. The study includes authors affiliated with GRAIL and the American Cancer Society. https://pmc.ncbi.nlm.nih.gov/articles/PMC12288388/
- National Cancer Institute: Esophageal Cancer Statistics. SEER data on the distribution of esophageal cancer cases by stage at diagnosis. https://seer.cancer.gov/statfacts/html/esoph.html
- National Cancer Institute: Pancreatic Cancer Statistics. SEER data on pancreatic cancer stage at diagnosis. https://seer.cancer.gov/statfacts/html/pancreas.html
- Effect of Screening with Multicancer Early-Detection Test on Late-Stage Cancer Diagnosis. The primary randomized NHS-Galleri trial results, published in The New England Journal of Medicine. https://www.nejm.org/doi/full/10.1056/NEJMoa2505723
- NHS England: NHS to pilot potentially revolutionary blood test that detects more than 50 cancers. The original November 2020 NHS England announcement of the 165,000-person trial and its conditional expansion plan. https://www.england.nhs.uk/2020/11/nhs-to-pilot-potentially-revolutionary-blood-test/
- The BMJ: Galleri promises to detect multiple cancers, but new evidence casts doubt on this much-hyped blood test. Reporting on the evidence available in 2024 and the decision not to proceed directly to the proposed large-scale pilot at that stage. https://www.bmj.com/content/386/bmj.q1706
- US Food and Drug Administration: Advisory Committee Meeting Materials. FDA materials relevant to the advisory committee’s review of multicancer early-detection testing. https://www.fda.gov/media/195032/download
- National Cancer Institute: Vanguard Study. Information about the NCI’s pilot study evaluating multicancer detection tests to inform future research. https://www.cancer.gov/research/areas/screening/vanguard-study
- Performance and safety of a multi-cancer early detection test: the PATHFINDER 2 study. Nature Medicine, September 22, 2026. https://www.nature.com/articles/s41591-026-04618-w
- Galleri Test Information for Patients. Information from the test manufacturer about the Galleri test. https://www.galleri.com/patient/the-galleri-test
- Reporting on Galleri’s Launch in Brazil. Coverage of the test’s reported Brazilian launch and pricing. https://www1.folha.uol.com.br/equilibrioesaude/2026/10/exame-de-sangue-que-busca-detectar-mais-de-50-tipos-de-cancer-chega-ao-brasil-entenda.shtml
- Reuters: Grail’s cancer blood test nears landmark FDA approval, but insurance coverage looms. Reporting on the potential US market for multicancer screening and the role of insurance coverage in that projection. https://www.reuters.com/legal/litigation/grails-cancer-blood-test-nears-landmark-fda-approval-insurance-coverage-looms-2026-10-02/
- Liu et al., Annals of Oncology: Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA. https://pmc.ncbi.nlm.nih.gov/articles/PMC8274402/
- Welch and Black, Journal of the National Cancer Institute: Overdiagnosis in cancer. https://pubmed.ncbi.nlm.nih.gov/20413742/
- Ministry of Health and Family Welfare, Government of India: Operational Guidelines on Prevention, Screening and Control of Common Non-Communicable Diseases. https://www.mohfw.gov.in/sites/default/files/Operational%20Guidelines%20on%20Prevention%2C%20Screening%20and%20Control%20of%20Common%20NCDs_1.pdf
- Life and death of circulating cell-free DNA. Review of cfDNA release, clearance pathways and half-life estimates. https://pmc.ncbi.nlm.nih.gov/articles/PMC6606043/
- US Food and Drug Administration: Laboratory Developed Tests. Current FDA information on the regulatory framework for laboratory-developed tests, including the 2025 vacatur and regulatory reversion. https://www.fda.gov/medical-devices/in-vitro-diagnostics/laboratory-developed-tests
- Clonal Hematopoiesis in Liquid Biopsy: From Biological Noise to Valuable Clinical Implications. Review of age-related somatic mutations and their potential to confound cell-free DNA analysis. https://pmc.ncbi.nlm.nih.gov/articles/PMC7463455/
