nabu.science
MethodologyAboutContact

 

  • Snapshot
  • Key Claims
  • Strengths & Limitations
  • Commentary
  • From the Authors
  • Author Note / Field Feedback
  • Dimension breakdown
  • Impact Profile
Free · About 5 minutes
Report #NBU-37C1B048·Framework v5.5·Pipeline v0.4·Evaluated 21 Jul 2026·OECD: 3.2 Clinical medicine·Subfield: Cancer Research·Basis: Full text·Type: Diagnostic accuracy study

Toward the simultaneous detection of multiple diseases with a highly cost-effective cell-free DNA methylome test

Weihua Zeng et al.Weihua Zeng, S. Guo-Dong Li, Shuo Li et al.·2026·10.1073/pnas.2518347123 ↗

Primary claim

MethylScan achieved an area under the receiver operating characteristic curve (AUROC) of 0.938 (95% CI: 0.920 to 0.954), with a sensitivity of 63.3%

Supported · internally cross-validated diagnostic performance is consistently reported, without independent external validation
In brief

MethylScan advances low-cost cfDNA methylome diagnostics across multiple tasks, but internal validation limits clinical confidence.

ReliabilityCaveats4 of 4 checks completeRead the 2 entries ↓
Fragile — the claim's weakest requirement is that the effect exists at all. Experimental. Every point below is traceable to a quoted passage, but this forecast has not yet been validated against known replication outcomes.
Abstract

Plasma cell-free DNA (cfDNA), originating from multiple organs, holds significant potential for noninvasive diagnostics and prognostics. Current cfDNA methylation assays primarily focus on single clinical indications by targeting specific genomic loci. In contrast, comprehensive profiling of cfDNA methylome can enable simultaneous detection of multiple diseases by capturing organ-specific methylation signatures, thereby offering a holistic view of health, when disease etiology is unclear or when conventional biochemical diagnostics are unavailable. However, deep sequencing required for sensitive detection of methylation abnormalities remains prohibitively expensive, limiting widespread clinical use. To overcome this barrier, we developed MethylScan , a highly cost-effective approach for cfDNA methylome sequencing. We demonstrated its broad clinical utility in a cohort of 1,061 individuals across diverse applications, including multicancer detection in general population, liver cancer surveillance in high-risk individuals, liver disease classification, identification of organ abnormalities, and race prediction from cfDNA. In multicancer detection (liver, lung, ovarian, and stomach cancers), MethylScan achieved an area under the receiver operating characteristic curve (AUROC) of 0.938 (95% CI: 0.920 to 0.954), with a sensitivity of 63.3% (95% CI: 58.9 to 67.9%) at 98.0% specificity for all cancer stages. For early-stage cancers, the AUROC was 0.916 (95% CI: 0.890 to 0.940), with 55.3% sensitivity (95% CI: 49.1 to 62.1%) at the same specificity. In liver cancer surveillance, MethylScan achieved an AUROC of 0.927 (95% CI: 0.889 to 0.959), with 79.6% sensitivity (95% CI: 70.6 to 87.8%) at 90.4% specificity. The assay also demonstrated strong performance in additional diagnostic tasks, supporting its potential as a versatile platform for comprehensive cfDNA-based health monitoring.

Key Claims

Primary finding · Supported

MethylScan achieved an area under the receiver operating characteristic curve (AUROC) of 0.938 (95% CI: 0.920 to 0.954), with a sensitivity of 63.3%

Fragile — the claim's weakest requirement is that the effect exists at all. Experimental. Every point below is traceable to a quoted passage, but this forecast has not yet been validated against known replication outcomes.

internally cross-validated diagnostic performance is consistently reported, without independent external validation

Abstract; Results, Fig. 6C
Secondary · Supported

For early-stage cancers, the AUROC was 0.916 (95% CI: 0.890 to 0.940), with 55.3% sensitivity

stage-specific cross-validation metrics are reported, though sensitivity remains moderate

Abstract; Results, Fig. 6D
Secondary · Supported

In liver cancer surveillance, MethylScan achieved an AUROC of 0.927 (95% CI: 0.889 to 0.959), with 79.6% sensitivity

high-risk surveillance analysis reports appropriate metrics, but remains internally validated

Abstract; Results, Liver Cancer Detection, p. 7
Secondary · Tentative

we developed MethylScan, a highly cost-effective approach for cfDNA methylome sequencing

assay enrichment and depth reductions support direction, but direct cost benchmarking is limited

Abstract; Introduction pp. 1–2; Results, Fig. 2–3
Exploratory · Tentative

The assay also demonstrated strong performance in additional diagnostic tasks

additional tasks are heterogeneous, with partial liver-disease classification and uneven class performance

Abstract; Results pp. 7–8

Derived from the full evaluation — not a separate score.

Strengths · Limitations

Strengths

#contribution

Coherent lower-cost assay strategy

The assay targets regions where blood-cell cfDNA background is consistently hypomethylated and uses MSRE digestion plus targeted capture to enrich disease-informative hypermethylated fragments. This directly addresses the sequencing-depth cost barrier identified in the introduction.

↳ Introduction pp. 1–2; Results, Fig. 1 and Fig. 2

#methodological rigour

Analytical spike-in validation included

The paper evaluates tumor signal detection using a liver tumor gDNA dilution series spanning 0% to 5% tumor fraction, with replicates and comparison to the assay without digestion. This provides an analytical bridge between assay chemistry and downstream diagnostic modeling.

↳ Results, Fig. 3; Methods, Panel Validation by the Spike-In Dilution Series Experiment

#reporting

Main cancer metrics transparently reported

For multicancer detection and liver cancer surveillance, the article reports AUROC, sensitivity, specificity, confidence intervals, and repeated cross-validation details. Stage-specific breakdowns are also provided for cancer detection, including early-stage performance.

↳ Results, Fig. 6C–D; Results, Liver Cancer Detection, p. 7

Limitations

#methodological rigour

No independent external validation cohort

The headline diagnostic results are based on stratified fivefold cross-validation repeated 50 times within the study cohort. That is useful for development-stage evidence but does not establish performance in independent intended-use populations.

↳ Results, Fig. 6C; Methods, Marker Selection and Predictive Model for Cancer Detection

#reporting

Broad simultaneous-detection framing overreaches

The title, abstract, and significance statement describe simultaneous detection and comprehensive health monitoring, but the reported evidence consists of separate models for distinct applications. This creates a reliability-relevant framing concern rather than a contradiction in the numerical results.

↳ Title; Abstract; Significance statement; Discussion pp. 8–9

#methodological rigour

Uneven non-cancer task maturity

The liver disease classifier assigns reliable labels to only 62.4% of patients, and ALD and MASLD performance is weaker than HBV and HCV. These caveats matter because the abstract summarizes additional tasks as strong platform evidence.

↳ Results, “Using MethylScan For Noninvasively Classifying Multiple Liver Diseases,” p. 7

Does this match your read? Add your assessment ↓

Nabu Commentary

The adjudicated quality score is driven by a strong assay-development contribution and competent modeling workflow, anchored by the spike-in dilution validation and repeated cross-validation results. The largest methodological constraint is the absence of an independent external or prospective validation cohort for the main diagnostic claims. The broad platform framing is directionally plausible but stronger than the evidence for several secondary tasks, particularly liver disease classification and race prediction. Contextual engagement with cfDNA methylation literature is substantive, but direct comparator benchmarking and ethical discussion of race prediction remain limited.

Evaluate another paper

Free · About 5 minutes · Research preview

Author Note / Field Feedback

Nabu’s assessment, alongside the field’s view.

Are you an author of this paper?

Dimension breakdown

Evidence Score

Out of 5
012345
3.7±0.2

Contribution
3.9

The score reflects a meaningful platform contribution: MSRE-based depletion of blood-cell background cfDNA combined with broad demonstration across cancer and liver-related applications. The contribution is constrained by internal validation and by cost-effectiveness claims that are supported indirectly through enrichment and sequencing-depth evidence rather than direct economic benchmarking.

Abstract; Introduction pp. 1–2; Results, Fig. 3 and Fig. 6

“a highly cost-effective approach for cfDNA methylome sequencing”

Methodological Rigour
3.6

The methods include analytical spike-in validation, tissue-based marker discovery for cancer tasks, stratified fivefold cross-validation repeated 50 times, SVD dimensionality reduction, and LSVM classifiers. The score is limited by lack of independent external validation, heterogeneous sample sources, and plasma-based marker discovery and validation within cross-validation for liver disease classification.

Methods, pp. 8–10; Results, Fig. 3, Fig. 6, Fig. 7

“stratified fivefold cross-validation… randomly repeated 50 times”

Reporting
3.6

The article’s main structure moves clearly from assay rationale to analytical validation and clinical applications, with headline performance metrics reported in conventional diagnostic terms. Reporting is limited by broad language around simultaneous multi-disease detection and by several supplied results sections where narrative detail is partially replaced by figure or supplement references.

Abstract; Significance statement; Results pp. 5–8

“supporting its potential as a versatile platform”

Positioning
3.5

The paper situates MethylScan in relation to cfMethyl-Seq, MCED methylation studies, CancerDetector, CancerLocator, and tissue deconvolution work, and it acknowledges MSRE digestion and methylation-signal limitations. The score is held below the high range because direct performance comparison with leading clinical/commercial alternatives, STARD-style framing, and ethical context for race prediction are limited.

Introduction pp. 1–2; Discussion pp. 8–9; References 1–4, 13–19

“potential confounding factor in methylation-based diagnostics”

Reliability

Caveats4 of 4 checks


Internal CoherenceNoted

The headline numerical performance metrics are internally consistent, but the article’s broad simultaneous-detection framing runs ahead of the separately evaluated retrospective models. Liver disease classification is disclosed with a substantial non-classification rate, but that caveat is less prominent than the overall multi-application framing.

Research ConductNoted

Human genomic data availability and code availability are specifically stated, and commercial conflicts are explicitly disclosed. The disclosed concentration of author employment, stock, consulting, executive roles, and patent interests in a company tied to the assay is a conduct watch-out for interpretation, especially given internal-only validation.

Flags: 3 declared / 5 total

Reference IntegrityClean

35 references verified via PubMed, Crossref, or OpenAlex.

References not verified automatically are resolved manually within 24 hours.

Post-Publication RecordClean

No retraction or expression of concern found in Crossref (includes Retraction Watch).

Last checked: 17 Oct 2026

Sources: Crossref (incl. Retraction Watch) ✓

Impact Profile

Impact areaHealth · Cancer and neoplasms · Liver cancer
Type of advanceMethodological
Route to useNamed
AdvanceExtends
BeneficiariesLarge
CitationsToo early
Policy · patentNot yet traced
ReachedNot reachedNot assessable or unverified
ReadinessPiloted
Foundational
Controlled
Piloted
Validated
Implementable
FoundationalImplementable

Performance was demonstrated in a 1,061-person cohort but only via internal repeated cross-validation on the same data, without independent external validation or prospective clinical deployment.

From the paper
“with the potential to transform early disease detection and health monitoring across diverse clinical settings”

The paper names clinical settings and populations (general-population cancer screening, high-risk liver cancer surveillance) where the assay would be used as a blood-based diagnostic test.

Pathway

Diagnostic developers and clinical laboratories would implement MethylScan as a low-cost blood test for multicancer screening in general populations and liver cancer surveillance in high-risk clinics.

Barrier

The assay still requires independent prospective validation in large, diverse real-world cohorts and regulatory approval before it can enter routine clinical screening practice.

Evaluated 13 Nov 2026 · GBD 2023 · glm_v1_3_3 · impact_profile_v1_3_3

Report an issue →

AI-generated, human-governed. Something look off? Contact us to request a review.

Keep reading

Related evaluations

Other papers Nabu has evaluated blind, starting with the same field.

  1. Technical feasibility of a long read, fourth generation sequencing platform in diagnostic profiling of clinical routine samples: a proof-of-concept study

    Vitale, Colarossi, Russo, et al. · 2026 · Clinical medicine

    Primary claim

    The adapted ONT workflow demonstrated high concordance rate in detecting clinically relevant molecular alterations on short-read fragments, achieving comparable accuracy with standardized second generation NGS platforms on tissue and liquid biopsy samples.

    Tentative
    Evidence Score
    2.7±0.2
    Reliability
    Caveats
  2. Neoadjuvant immunotherapy and chemotherapy regimens for the treatment of high-risk, early-stage triple-negative breast cancer: a systematic review and network meta-analysis

    Cortes, Haiderali, Huang, et al. · 2023 · Basic medicine

    Primary claim

    Neoadjuvant pembrolizumab + chemotherapy followed by adjuvant pembrolizumab confers benefits in response and survival outcomes versus alternative neoadjuvant treatments for early-stage TNBC.

    Tentative
    Evidence Score
    3.3±0.2
    Reliability
    Caveats
  3. Platelet-derived factors dysregulate placental sphingosine-1-phosphate receptor 2 in human trophoblasts

    Freya Lyssy, Jacqueline Guettler, Beatrice A Brugger, et al. · 2023 · Basic medicine

    Primary claim

    S1PR2 was found to be significantly down-regulated in BeWo cells when co-incubated with platelet-derived factors (P = 0.0055).

    Tentative
    Evidence Score
    3.1±0.2
    Reliability
    Caveats

How the rubric works →