---

Article 1 — Guo X et al. — ccf-mtDNA fragmentomics for ovarian cancer detection

PMID 42908017 | Genome Medicine | 🔴 EARLY_CANCER_DETECTION

Dimension	Score	Rationale
Scientific Novelty	8	ccf-mtDNA fragmentomics is a genuinely novel biomarker class for OC detection; most liquid biopsy work focuses on nuclear ctDNA. Applying fragmentomics specifically to mitochondrial cfDNA and achieving this performance across multi-ethnic cohorts is an advance on prior art.
Clinical Relevance	9	Ovarian cancer is routinely detected at late stage with catastrophic survival consequences. 91.67% sensitivity at Stage I (95% specificity) using a blood test would directly address the most clinically critical gap. AUC 0.982 for malignant vs. benign discrimination outperforms CA-125.
Population Reach	8	Ovarian cancer affects ~300,000 women annually worldwide; detecting it at Stage I could prevent the majority of ~200,000 annual deaths. Multi-ethnic validation extends applicability.
Implementation Speed	6	Liquid biopsy infrastructure is increasingly available; however, ccf-mtDNA sequencing and fragmentomics pipelines are not yet standard-of-care. Regulatory approval pathway, reimbursement, and prospective screening trial data required before adoption.
Evidence Strength	7	Multi-center design with 4 external validation cohorts including minority ethnic groups is methodologically strong for a biomarker study. ML models validated prospectively. Key limitation: abstract-only access; sample size across cohorts not precisely reported; no head-to-head comparison to existing multi-marker panels (e.g., OVA1, ROMA) stated.

Key quantitative result: AUC 0.987 (OD model, OC vs. healthy); AUC 0.982 (BMD model, malignant vs. benign); 91.67% sensitivity for Stage I OC at 95% specificity.

External validation: Yes — 4 external cohorts including Chinese minority ethnic group. Strongest validation design in this batch.

Main limitation: Abstract-only access prevents full assessment of cohort sizes, blinding procedures, and comparison to standard-of-care CA-125 + HE4 multimarker panels. Generalizability to non-Asian populations not yet established.

Equity implications: Multi-ethnic design is a notable strength; Chinese Han and minority populations included. Western/African-ancestry populations not described — a meaningful gap for a global screening tool. Access to sequencing infrastructure may disadvantage lower-income settings.

Evidence Maturity: ✅ Confirmed Validated — multi-cohort external validation warrants this classification, with a caveat that prospective screening trials have not yet been reported.

---

Article 2 — Huang L et al. — Patient-specific genomic MRD markers in AML

PMID 42902921 | British Journal of Haematology | 🟢 NEAR_TERM_IMPLEMENTABLE

Dimension	Score	Rationale
Scientific Novelty	7	Patient-specific PCR-MRD approaches are established in ALL; applying them systematically to myeloid malignancies using NGS/WGS is a meaningful extension. The 97.6% assay success rate and sensitivity advantage over ALL-MRD markers are genuinely novel contributions.
Clinical Relevance	8	AML MRD monitoring is a recognized clinical need — treatment intensification and transplant decisions hinge on it. Existing methods (flow cytometry, RNA-PCR) are not universally applicable. This approach would enable universal sensitive MRD tracking for the first time.
Population Reach	6	AML incidence ~20,000/year in the US; ~100,000 globally. Myeloid malignancies broadly affect hundreds of thousands, but this remains a specialized oncology application.
Implementation Speed	7	NGS/WGS infrastructure is increasingly available in academic hematology centers. Protocol uses EuroMRD-aligned standards, easing adoption. Multi-center prospective data suggest near-term clinical readiness in resource-adequate settings.
Evidence Strength	7	Prospective multi-center cohort with assay development + clinical outcome correlation in 40 patients. Limitation: n=40 in the prospective arm limits outcome data power; abstract only; no randomized comparison to flow cytometry standard-of-care.

Key quantitative result: 97.6% assay development success rate; 50% of myeloid patients had markers at ≤5×10⁻⁵ sensitivity vs. 29% of ALL patients; serial MRD correlated with outcome.

External validation: Multi-center Australian design provides internal validation; independent external replication not described.

Main limitation: Prospective cohort n=40 is underpowered for definitive clinical outcome validation. Real-world NGS/WGS turnaround time and cost in community oncology settings not addressed.

Equity implications: Performed in Australia with multi-center adult and pediatric coverage. Resource-intensive WGS/NGS requirement will disadvantage patients in lower-income countries and community settings without sequencing infrastructure. Pediatric inclusion is a strength.

Evidence Maturity: ✅ Confirmed Validated — prospective multi-center design with outcome correlation supports this, with a note that larger outcome studies are needed before practice change.

---

Article 3 — Brown BL et al. — CNS-targeted AAV gene therapies

PMID 42906551 | Cell Reports Medicine | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	5	Review synthesizes existing clinical and postmarketing data; does not generate new findings. Value is in authoritative synthesis from a leading center (CHOP).
Clinical Relevance	7	Four FDA-approved CNS AAV therapies are in clinical use; practical synthesis of toxicity, delivery routes, and barriers directly informs clinical decision-making for neurologists and gene therapy teams.
Population Reach	5	Rare neurological diseases individually have small patient counts, but collectively CNS AAV-eligible conditions (SMA, AADC deficiency, CLN disorders, MPSIII, etc.) affect tens of thousands globally. Unmet need is extreme.
Implementation Speed	5	Therapies already approved; review accelerates evidence-based practice for those already in use. Novel delivery approaches and capsids described are not yet clinical.
Evidence Strength	6	Review quality is constrained by source heterogeneity (trials + postmarketing reports). COI disclosure (BLD/Latus Bio) noted. Strong CHOP pedigree.

Main limitation: Review design; COI with industry. Cannot generate causal or quantitative effect estimates.

Equity implications: Gene therapies are extremely expensive (e.g., onasemnogene: ~$2.1M); access is severely limited in LMICs and by insurance coverage in high-income countries. Review does not address access equity.

Evidence Maturity: ✅ Confirmed Validated — for approved therapies; remainder is Exploratory for pipeline approaches.

---

Article 4 — Stocker LD et al. — Allo-HSCT in mature T-cell lymphomas (Cochrane)

PMID 42905121 | Cochrane Database | ⬜ STANDARD | Unsolicited find

Dimension	Score	Rationale
Scientific Novelty	5	Cochrane review synthesizes existing evidence; novelty is in rigor of aggregation rather than discovery. T-cell lymphoma allo-HSCT evidence base is known to be weak.
Clinical Relevance	7	Allo-HSCT is a potentially curative but high-morbidity intervention in T-cell lymphoma. Cochrane-level synthesis directly guides treatment selection and transplant referral decisions.
Population Reach	4	Mature T-cell lymphomas are rare (~15% of all NHLs); patient numbers are limited. High unmet need within this small population.
Implementation Speed	7	Allo-HSCT is already in use; this review informs immediate transplant decision-making protocols. Cochrane reviews are frequently incorporated into guidelines.
Evidence Strength	7	Cochrane systematic review is highest design quality for evidence synthesis; however, classification_confidence = medium and abstract was not retrieved — quantitative outcomes unknown. Downgrade applies for lack of abstract-level access.

Main limitation: Evidence base for allo-HSCT in T-cell lymphomas is largely observational; Cochrane methodology cannot compensate for sparse RCT data in this disease. Abstract not retrieved — effect sizes and included study quality unknown.

Equity implications: T-cell lymphomas are more common in certain ethnic groups (e.g., HTLV-1-associated ATL in Caribbean, Japanese, West African populations). Whether this review stratifies by subtype or ethnicity is unknown.

Evidence Maturity: Revised to Validated (for synthesis quality) with caveat that underlying evidence base is weak.

---

Article 5 — Franceschini K et al. — AI delirium risk prediction vs. 4AT-Score

PMID 42906075 | JMIR Formative Research | 🟢 NEAR_TERM_IMPLEMENTABLE

Dimension	Score	Rationale
Scientific Novelty	6	AI-based delirium prediction tools exist; this is a prospective head-to-head clinical comparison against a validated tool in a specific trauma ward setting. Novelty is in the prospective comparative design and magnitude of difference.
Clinical Relevance	7	Delirium affects 14–56% of hospitalized older adults and significantly worsens outcomes. Tripling sensitivity (87.3% vs. 28.0%) over the standard screening tool has direct workflow implications for preventive care protocols.
Population Reach	7	Delirium is one of the most common hospital complications; trauma patients are particularly vulnerable. Trauma wards exist globally; the target population is large.
Implementation Speed	7	EHR-integrated random forest model; infrastructure is increasingly common. JMIR Formative Research signals a developmental stage, but the tool has prior validation. Adoption barriers include EHR integration, workflow change, and staff training.
Evidence Strength	7	Prospective cohort, n=2002, single institution (Graz). Well-defined comparison. Limitation: single-center, single-country; generalizability uncertain; no patient outcome data reported (only delirium identification, not outcomes of early intervention).

Key quantitative result: AI sensitivity 87.3% vs. 28.0% (4AT); AUC 0.884.

Main limitation: Single-center; no downstream outcome data (does early identification translate to reduced delirium morbidity?). Random forest model developed externally — implementation fidelity in other EHR systems unknown.

Equity implications: Trauma populations include disproportionate numbers of older adults, men, and lower-SES individuals; AI tools built on single-center EHR data may generalize poorly to hospitals with different demographic mixes or documentation practices.

Evidence Maturity: Confirmed Validated — prospective design with head-to-head comparison.

---

Article 6 — Shuttleworth SL et al. — Proteomic atlas of T cell exhaustion

PMID 42906538 | Cell | ⚪ PROMISING_PRELIMINARY

Dimension	Score	Rationale
Scientific Novelty	9	First comprehensive multi-omic proteomic atlas of T cell exhaustion; identification of proteostatic collapse and differential E3 ligase activity as central mechanisms is a genuinely new conceptual framework for a phenomenon previously understood primarily at the transcriptomic level.
Clinical Relevance	4	Per cap: non-human primary study. Human correlation data (E3 ligase expression vs. immunotherapy outcomes) is intriguing but correlational. Clinical translation of RNF166 manipulation requires many development steps.
Population Reach	6	T cell exhaustion is relevant to all solid and hematologic tumors where immunotherapy is used — a very large potential beneficiary population. But translation is distant.
Implementation Speed	2	Preclinical; target identification stage. Clinical application is 7–10+ years away minimum without major acceleration.
Evidence Strength	5	Multi-omic rigor is high; mouse model data are well-supported; human correlation is cross-sectional. COI noted (TCura Bioscience, Oncorock Therapeutics). Non-human primary data cap (≤5 on Clinical Relevance).

Main limitation: Primarily preclinical; human data is correlative only. COI with commercial ventures targeting this pathway. Clinical relevance of RNF166 manipulation in human tumors is unproven.

Equity implications: Immunotherapy access is highly inequitable globally; fundamental discoveries here would primarily benefit populations in high-income countries in the near term.

Evidence Maturity: Confirmed Exploratory.

---

Article 7 — Kodali S et al. — Cholangiocarcinoma management review

PMID 42906872 | Clinical Gastroenterology and Hepatology | 🟢 NEAR_TERM_IMPLEMENTABLE

Dimension	Score	Rationale
Scientific Novelty	4	Review synthesizes existing breakthroughs (pemigatinib, ivosidenib, durvalumab + GemCis); no new data generated. Value is in clinical synthesis and practice guidance.
Clinical Relevance	6	CCA management has changed substantially; this review is timely for gastroenterologists/hepatologists now selecting molecular testing and frontline regimens. Directly actionable.
Population Reach	5	CCA is rare (~8,000 new cases/year US; ~200,000 globally); rising incidence. High unmet need.
Implementation Speed	6	NGS testing, FGFR2 inhibitors, and PD-L1-based regimens are already FDA-approved; this review accelerates appropriate utilization.
Evidence Strength	5	Review design; no new data; CGA level. Authoritative but limited by design.

Main limitation: Review only; no new primary data. Rapid pace of CCA drug approvals means this may require frequent updating.

Evidence Maturity: Confirmed Validated for described approved therapies.

---

Article 8 — Hofseth LJ et al. — GLP-1 RAs and cancer risk in obesity

PMID 42906947 | Obesity Reviews | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	6	The cancer-risk modifying potential of GLP-1 RAs is an emerging and genuinely uncertain area; narrative review synthesizes a growing but inconclusive literature. Cancer prevention angle is relatively novel vs. the established cardiovascular literature.
Clinical Relevance	6	With hundreds of millions of potential GLP-1 RA users globally, even modest cancer risk modification would have enormous population-level implications. For individual patients, the medullary thyroid carcinoma signal is directly actionable.
Population Reach	9	GLP-1 RAs are among the most widely prescribed drugs in history; obesity affects ~1 billion adults globally. Population reach is exceptional.
Implementation Speed	4	Narrative review; mechanisms are not sufficiently validated for cancer prevention practice changes. More trial data needed.
Evidence Strength	4	Narrative review — selection bias possible. Underlying evidence is largely observational (pharmacoepidemiology) with limited prospective cancer outcome data.

Main limitation: Narrative design with heterogeneous underlying evidence. Most GLP-1 RA trials were not powered for cancer endpoints.

Evidence Maturity: Revised to Exploratory — cancer risk modification by GLP-1 RAs is not validated; this is hypothesis-generating.

---

Article 9 — Mahuad C & Korin L — N-AVD vs. BrECADD in Hodgkin lymphoma

PMID 42907308 | Clinical Lymphoma, Myeloma & Leukemia | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	4	Synthesizes two landmark trials (SWOG S1826, GHSG HD21); no new data. Value is in the comparative clinical decision framework.
Clinical Relevance	7	Both regimens are now standard-of-care options; practical individualization guidance is highly relevant for oncologists selecting frontline therapy. Directly actionable.
Population Reach	5	cHL is relatively common among young adults; ~85,000 new cases/year globally. Highly curable — but optimization of frontline therapy affects quality of life and late effects.
Implementation Speed	8	Both regimens are already approved and in use. This review's guidance can be applied immediately.
Evidence Strength	5	Review of two Phase III trials; no meta-analysis. COI noted. Indirect evidence for decision framework.

Main limitation: COI (authors received pharma honoraria). No head-to-head trial data between the two regimens; individualization recommendations are expert opinion layered on separate trial results.

Evidence Maturity: Confirmed Validated for the underlying trial data; the decision framework itself is expert consensus.

---

Article 10 — Heinzmann A et al. — Somatic CAG expansions in Huntington disease

PMID 42906305 | Current Opinion in Genetics & Development | ⚪ PROMISING_PRELIMINARY

Dimension	Score	Rationale
Scientific Novelty	7	The somatic instability model for HD onset variability is gaining rapid traction; synthesizing its therapeutic implications (FAN1, MMR pathway targeting) is timely and represents a genuine conceptual shift from purely inherited-length determinism.
Clinical Relevance	5	No approved therapy yet targets somatic instability in HD. Promising mechanistic framework but distant from clinical application.
Population Reach	3	~30,000 HD patients in the US; ~hundreds of thousands globally. High unmet need within a small population.
Implementation Speed	2	Review of preclinical and genetic mechanisms; therapeutic agents targeting somatic instability not in clinical trials yet.
Evidence Strength	5	Review; medium classification confidence; abstract not fully retrieved. Underlying genetic evidence for somatic instability is compelling (GWAS-validated modifiers).

Main limitation: Review design; therapeutic applications are preclinical. Medium confidence due to partial abstract access.

Evidence Maturity: Confirmed Exploratory.

---

Article 11 — Penel N et al. — Head & neck desmoid-type fibromatosis

PMID 42905937 | Cancer Medicine | 🟡 UNDERSERVED_POPULATION

Dimension	Score	Rationale
Scientific Novelty	5	Largest series in these ultra-rare locations; descriptive natural history data fills an important evidence vacuum but is methodologically incremental.
Clinical Relevance	5	Directly informs management decisions for a patient group that currently lacks site-specific guidance; relevant within the narrow desmoid-treating specialist community.
Population Reach	2	Extremely rare; desmoid-type fibromatosis affects ~3–4 per million; head/neck/shoulder locations are a fraction of that. High relative unmet need.
Implementation Speed	5	Natural history data can immediately inform watch-and-wait vs. intervention decisions in specialist centers.
Evidence Strength	4	Retrospective cohort; sample size not reported in abstract; medium confidence.

Main limitation: Retrospective design; sample size unknown; abstract not retrieved. French network bias — may not reflect non-European populations.

Evidence Maturity: Confirmed Exploratory — descriptive natural history.

---

Article 12 — Stangl-Kremser J & Salami SS — Risk-stratified prostate cancer screening 2026

PMID 42907259 | European Urology Focus | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	3	Review of established/approved tools (PHI, 4Kscore, SelectMDx, etc.); no new data or novel biomarkers described.
Clinical Relevance	6	Practical for urologists navigating an increasingly complex screening toolkit; risk-stratification is becoming standard in guidelines.
Population Reach	7	Prostate cancer is the most common male cancer; screening affects millions of men annually.
Implementation Speed	6	Tools reviewed are already FDA-cleared; review facilitates existing adoption.
Evidence Strength	5	Narrative review; high confidence classification but limited design quality.

Evidence Maturity: Confirmed Validated for described tools.

---

Article 13 — Shu A et al. — Inferring tissue origin from cfDNA in metastatic cancer

PMID 42905206 | Cancer Metastasis Reviews | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	5	Active and evolving area; review covers multiple modalities. No new primary data.
Clinical Relevance	5	Cancer of unknown primary is a genuine clinical challenge (~3–5% of cancers); cfDNA-based tissue-of-origin inference could guide therapy. Not yet standard.
Population Reach	5	CUP patients are a significant minority; metastatic cancer patients broadly benefit from improved tissue-of-origin accuracy.
Implementation Speed	3	Methods reviewed are not yet validated for routine clinical use; regulatory pathway unclear.
Evidence Strength	4	Review; medium confidence; abstract not retrieved.

Evidence Maturity: Confirmed Exploratory.

---

Article 14 — Petranović M et al. — ACR Lung-RADS updates

PMID 42907936 | Cancer Imaging | 🟢 NEAR_TERM_IMPLEMENTABLE

Dimension	Score	Rationale
Scientific Novelty	3	Guideline update review; incremental refinement, not discovery.
Clinical Relevance	6	Lung-RADS is widely used; updates directly affect how ~250,000 annual US screening scans are reported.
Population Reach	7	Low-dose CT lung screening eligibility has expanded; millions of current/former smokers are candidates.
Implementation Speed	8	Guideline updates are immediately implementable by radiology departments.
Evidence Strength	5	Review; medium confidence; abstract not retrieved.

Evidence Maturity: Confirmed Validated for existing Lung-RADS framework; AI integration components remain Exploratory.

---

Article 15 — Muthukumaran M et al. — ClinNetDx: AI for ALL detection

PMID 42907972 | BMC Cancer | ⚪ PROMISING_PRELIMINARY

Dimension	Score	Rationale
Scientific Novelty	6	Hybrid CNN-DNN with tabular-to-spatial transformation and dual explainability (SHAP + Grad-CAM) is a methodologically interesting contribution; but the AI-for-ALL-detection space is crowded.
Clinical Relevance	4	No prospective clinical validation; public dataset only. High accuracy metrics on benchmarks do not translate directly to clinical performance.
Population Reach	5	ALL affects ~6,000 adults and children annually in the US; global burden higher in LMICs where it may be most impactful.
Implementation Speed	3	Requires prospective validation in real clinical environments before adoption.
Evidence Strength	4	Public dataset evaluation only; k-fold cross-validation within a single dataset. Significant risk of overfitting and optimistic accuracy estimates.

Main limitation: No prospective clinical validation; dataset not representative of real-world diagnostic variation. Ethics approval not applicable per COI statement — unusual framing.

Evidence Maturity: Confirmed Exploratory.

---

Article 16 — Almuhanna HNB et al. — USP7-FOXO1 axis in CLL

PMID 42904581 | Molecular Oncology | ⚪ PROMISING_PRELIMINARY

Dimension	Score	Rationale
Scientific Novelty	7	USP7's role in FOXO1 nuclear exclusion in CLL is a novel mechanistic finding. Combination of DUB inhibitor + ibrutinib is conceptually novel as a dual BCR-pathway strategy. Use of primary patient cells strengthens the finding.
Clinical Relevance	4	Preclinical; primary patient cell use is a plus but no in vivo or clinical data.
Population Reach	5	CLL affects ~20,000 new patients/year in the US; it is the most common adult leukemia. Ibrutinib resistance is a clinically pressing problem.
Implementation Speed	3	Early preclinical; USP7 inhibitors are in early development; combination trials are years away.
Evidence Strength	5	Primary CLL patient cells + cell line; mechanistic rigor is appropriate for the claims. Single lab; no in vivo mouse model.

Evidence Maturity: Confirmed Exploratory.

---

Article 17 — Tremblay D et al. — Cytoreduction in MPN-SVT

PMID 42906183 | Blood Advances | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	4	Evidence synthesis in a known evidence-poor area; no new data.
Clinical Relevance	6	SVT in MPN is a life-threatening complication; this synthesis directly addresses a clinical gap where no standard exists.
Population Reach	3	MPN-SVT is rare; small but clinically important patient group.
Implementation Speed	5	Synthesis can immediately guide clinical practice even if evidence is limited.
Evidence Strength	4	Review; medium confidence; abstract not retrieved.

Evidence Maturity: Confirmed Exploratory — the underlying evidence base is weak by authors' own characterization.

---

Article 18 — Batalov AI et al. — CT perfusion for PCNSL vs. GBM

PMID 42905194 | Neuroradiology | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	4	CT perfusion in PCNSL/GBM differentiation has been explored before; this adds to the evidence base but is not conceptually new.
Clinical Relevance	6	Distinguishing PCNSL from GBM before biopsy or corticosteroid treatment is clinically critical (steroids cause PCNSL regression, confounding pathology).
Population Reach	3	Brain mass differential diagnosis is relatively specialized; PCNSL is rare (~1,500 cases/year US).
Implementation Speed	5	CT perfusion is widely available; if validated, integration is feasible.
Evidence Strength	4	Diagnostic accuracy study; sample size unknown; abstract not retrieved. Medium confidence.

Evidence Maturity: Confirmed Exploratory.

---

Article 19 — Kounatidis DC et al. — GLP-1 RA safety profile review

PMID 42905300 | Current Obesity Reports | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	4	The safety literature for GLP-1 RAs is maturing; fact-vs-myth framing is a practical contribution but not conceptually novel.
Clinical Relevance	7	Directly addresses prescriber and patient uncertainty about a drug class used by tens of millions; actionable safety guidance on medullary thyroid risk, pancreatitis, and muscle mass is immediately relevant.
Population Reach	9	GLP-1 RAs are now among the most widely used drugs globally.
Implementation Speed	7	Safety guidance can be implemented immediately in clinical practice.
Evidence Strength	4	Narrative review; medium confidence; abstract not retrieved.

Evidence Maturity: Confirmed Validated for established safety signals; some controversies remain Exploratory.

---

Article 20 — Lebeda D et al. — RF3 degradation and MECP2 readthrough in Rett syndrome

PMID 42907672 | Journal of Molecular Medicine | ⚪ PROMISING_PRELIMINARY

Dimension	Score	Rationale
Scientific Novelty	7	RF3 degradation as a strategy to amplify aminoglycoside readthrough is a novel mechanistic approach. Combining translational readthrough enhancement with release factor manipulation represents a creative two-pronged strategy for Rett syndrome.
Clinical Relevance	3	In vitro only; Rett syndrome has extremely high unmet need but this is early-stage lab work. Capped at ≤5 for non-human/in vitro studies.
Population Reach	3	Rett syndrome affects ~1 in 10,000 female births globally; ~9,000 patients in the US. Relative to unmet need: high impact within this small population.
Implementation Speed	1	Lab stage; no in vivo data. Translation pathway is long.
Evidence Strength	4	In vitro cell model only; medium confidence; abstract not retrieved.

Evidence Maturity: Confirmed Exploratory.

---

Article 21 — Sheldrick S et al. — Monoamine dysfunction in Prader-Willi syndrome

PMID 42906282 | Journal of Psychiatric Research | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	5	Monoamine involvement in PWS has been discussed previously; a unifying framework synthesizing dopamine, serotonin, and NE is a useful conceptual contribution but not a discovery.
Clinical Relevance	4	Review only; therapeutic implications are speculative. No new treatments.
Population Reach	3	PWS affects ~1 in 15,000; ~400,000 patients globally. High unmet need relative to population size.
Implementation Speed	3	Mechanistic review; no clinical applications ready.
Evidence Strength	4	Review; medium confidence; abstract not retrieved.

Evidence Maturity: Confirmed Exploratory.

---

Article 22 — Lopedote P et al. — Glofitamab + polatuzumab vedotin in r/r CNS lymphoma

PMID 42906182 | Blood Advances | 🟠 NOVEL_TREATMENT

Dimension	Score	Rationale
Scientific Novelty	8	First reported use of glofitamab + polatuzumab vedotin in CNS lymphoma; demonstration of CNS penetration by both agents and rapid CSF clearance is clinically significant. The combination of bispecific antibody + ADC for CNS lymphoma is novel.
Clinical Relevance	7	r/r CNS lymphoma has virtually no effective salvage options; median survival measured in months. 100% ORR / 89% CR — even with n=9 — is clinically striking and likely to influence off-label use and trial design.
Population Reach	3	CNSL is rare (~1,500–2,000 primary cases/year US; secondary CNS involvement more common). Extreme unmet need within a small population.
Implementation Speed	4	Both agents are FDA-approved for systemic DLBCL; off-label CNS use is feasible at academic centers. However, n=9 limits guideline-level adoption — prospective trial needed.
Evidence Strength	4	Retrospective single-center case series, n=9. Highest risk of selection bias. Cannot be ranked #1 per scoring rules despite compelling results.

Key quantitative result: 100% ORR, 89% CR rate, n=9; rapid CSF lymphoma clearance observed.

Main limitation: n=9, single-center, retrospective; selection bias likely; no comparator arm; follow-up duration not specified.

Equity implications: Requires academic center with access to both agents; insurance coverage for off-label CNS use is highly variable.

Evidence Maturity: Confirmed Exploratory — hypothesis-generating case series.

---

Article 23 — Zhang Y et al. — Gas6 senokine and aortic valve calcification

PMID 42902865 | Aging Cell | ⬜ STANDARD

Dimension	Score	Rationale
Scientific Novelty	7	Identifying Gas6 as a senokine that drives aortic valve calcification via CaMKII-DRP1-mediated mitochondrial vesicle biology is a genuinely novel mechanistic contribution linking cellular senescence to valve disease.
Clinical Relevance	3	Mixed species model; primarily preclinical. Aortic valve calcification is a major clinical problem, but therapeutic application of this finding is distant. Capped per non-human study rule.
Population Reach	6	Aortic valve calcification affects ~2–3% of adults >75; tens of millions globally. A preventive therapy would have enormous reach.
Implementation Speed	2	Mechanistic preclinical study; no drug candidates at clinical stage.
Evidence Strength	4	Mixed species (likely cell + animal); medium confidence; abstract not retrieved.

Evidence Maturity: Confirmed Exploratory.

---

---

Conflict Notes

GLP-1 RA articles (Articles 8 & 19) are complementary rather than conflicting: Article 8 addresses cancer risk modification (net potentially beneficial for obesity-related cancers), while Article 19 addresses overall safety profile including thyroid risk. Both are narrative reviews; neither contradicts the other, though the evidence base they describe is itself contested. Users should read both in tandem.

Ovarian cancer detection (Article 1) vs. cfDNA tissue-of-origin (Article 13): These address different questions and are not conflicting.

---

Composite Impact Score Calculation

Weights: Clinical Relevance 30%, Population Reach 25%, Scientific Novelty 20%, Implementation Speed 15%, Evidence Strength 10%

Rank	PMID	Article	CR (30%)	PR (25%)	SN (20%)	IS (15%)	ES (10%)	Composite	Triage Score	Flag
1	42908017	ccf-mtDNA for OC detection	9	8	8	6	7	7.90	9	🔴
2	42902921	Genomic MRD in AML	8	6	7	7	7	7.20	9	🟢
3	42906182	Glofitamab + pola-V in r/r CNSL	7	3	8	4	4	5.65	4	🟠
4	42906075	AI delirium prediction vs. 4AT	7	7	6	7	7	6.95	6	🟢
5	42906538	T cell exhaustion proteomic atlas	4	6	9	2	5	5.05	6	⚪
6	42906551	CNS AAV gene therapies review	7	5	5	5	6	5.75	7	⬜
7	42905121	Allo-HSCT in T-cell lymphoma (Cochrane)	7	4	5	7	7	5.90	7	⬜
8	42905300	GLP-1 RA safety: facts & myths	7	9	4	7	4	6.55	5	⬜
9	42906947	GLP-1 RAs and cancer risk	6	9	6	4	4	6.00	6	⬜
10	42907308	N-AVD vs. BrECADD in cHL	7	5	4	8	5	5.90	6	⬜
11	42906872	Cholangiocarcinoma management review	6	5	4	6	5	5.30	6	🟢
12	42907936	ACR Lung-RADS updates	6	7	3	8	5	5.80	5	🟢
13	42907259	Risk-stratified PCa screening 2026	6	7	3	6	5	5.55	5	⬜
14	42904581	USP7-FOXO1 in CLL	4	5	7	3	5	4.80	5	⚪
15	42906305	Somatic CAG expansions in HD	5	3	7	2	5	4.55	6	⚪
16	42907672	RF3 degradation + MECP2 readthrough in Rett	3	3	7	1	4	3.65	5	⚪
17	42905206	cfDNA tissue-of-origin in metastatic cancer	5	5	5	3	4	4.55	5	⬜
18	42906183	Cytoreduction in MPN-SVT	6	3	4	5	4	4.60	5	⬜
19	42905194	CT perfusion for PCNSL vs. GBM	6	3	4	5	4	4.60	5	⬜
20	42905937	Head/neck desmoid-type fibromatosis	5	2	5	5	4	4.25	6	🟡
21	42907972	ClinNetDx AI for ALL detection	4	5	6	3	4	4.45	5	⚪
22	42906282	Monoamine dysfunction in PWS	4	3	5	3	4	3.85	5	⬜
23	42902865	Gas6 senokine and aortic valve calcification	3	6	7	2	4	4.35	4	⬜

---

Ranked Summary Table

Rank	Article	Impact Score	CR	PR	SN	IS	ES	Triage	Design	Flag
1	ccf-mtDNA ovarian cancer detection	7.90	9	8	8	6	7	9	Multi-center biomarker validation, ML	🔴
2	Genomic MRD in AML/myeloid malignancies	7.20	8	6	7	7	7	9	Prospective multi-center cohort	🟢
3	AI delirium prediction vs. 4AT	6.95	7	7	6	7	7	6	Prospective cohort, n=2002	🟢
4	GLP-1 RA safety: facts & myths	6.55	7	9	4	7	4	5	Narrative review	⬜
5	GLP-1 RAs and cancer risk	6.00	6	9	6	4	4	6	Narrative review	⬜
6	Glofitamab + pola-V in r/r CNSL	5.65	7	3	8	4	4	4	Retrospective case series, n=9	🟠
7	Allo-HSCT in T-cell lymphoma (Cochrane)	5.90	7	4	5	7	7	7	Cochrane SR/MA	⬜
8	ACR Lung-RADS updates	5.80	6	7	3	8	5	5	Guideline review	🟢
9	CNS AAV gene therapies review	5.75	7	5	5	5	6	7	Clinical/postmarketing review	⬜
10	N-AVD vs. BrECADD in cHL	5.90	7	5	4	8	5	6	Phase III trial synthesis	⬜
11	Cholangiocarcinoma management	5.30	6	5	4	6	5	6	Clinical practice review	🟢
12	Risk-stratified PCa screening 2026	5.55	6	7	3	6	5	5	Narrative review	⬜
13	T cell exhaustion proteomic atlas	5.05	4	6	9	2	5	6	Multi-omic lab + mouse + human correlation	⚪
14	USP7-FOXO1 in CLL	4.80	4	5	7	3	5	5	Primary patient cell lab study	⚪
15	cfDNA tissue-of-origin in metastatic cancer	4.55	5	5	5	3	4	5	Review	⬜
16	Somatic CAG expansions in HD	4.55	5	3	7	2	5	6	Review	⚪
17	Cytoreduction in MPN-SVT	4.60	6	3	4	5	4	5	Evidence synthesis	⬜
18	CT perfusion for PCNSL vs. GBM	4.60	6	3	4	5	4	5	Diagnostic accuracy study	⬜
19	Gas6 senokine/aortic valve calcification	4.35	3	6	7	2	4	4	Mechanistic lab study	⬜
20	ClinNetDx AI for ALL	4.45	4	5	6	3	4	5	ML model development, public datasets	⚪
21	Head/neck desmoid-type fibromatosis	4.25	5	2	5	5	4	6	Retrospective multi-center cohort	🟡
22	Monoamine dysfunction in PWS	3.85	4	3	5	3	4	5	Mechanistic review	⬜
23	RF3 degradation + MECP2 readthrough	3.65	3	3	7	1	4	5	In vitro cell model	⚪

---

Rank Justification Highlights

Rank 1 — ccf-mtDNA ovarian cancer: Earns the top position on near-unanimous strength across all five dimensions. Ovarian cancer is the gynecological cancer killer that kills quietly — because it almost always presents at Stage III or IV, when 5-year survival drops below 30%. A blood test that catches it at Stage I with 91.67% sensitivity is the kind of finding that could, if it holds up, fundamentally shift the survival curve for this disease. Four external validation cohorts, including ethnically diverse populations, is an unusually rigorous validation design for a biomarker study. Evidence Strength of 7 reflects the multi-cohort prospective design without being able to fully assess blinding, cohort sizes, or head-to-head comparisons against CA-125/HE4 panels from the abstract alone.

Rank 2 — Genomic MRD in AML: Sits just below Article 1 by composite score but is potentially more immediately implementable. AML MRD monitoring is already embedded in treatment guidelines, but current methods (flow cytometry, RNA PCR for NPM1/CBF fusions) fail in a substantial proportion of patients who lack trackable molecular targets. A 97.6% assay success rate using patient-specific WGS-derived PCR means this approach can, in principle, be universally applied. Prospective multi-center data with outcome correlation elevates this beyond a methods paper.

Rank 3 — AI delirium prediction: Ranks higher than its triage score (6) would suggest because the composite formula rewards its combination of broad population reach (delirium is ubiquitous in hospital medicine), immediate implementability, and solid prospective evidence. Tripling sensitivity over a validated clinical tool in a 2,002-patient prospective cohort is a meaningful result. The key gap — whether earlier identification translates to better patient outcomes — is the most important unanswered question.

Rank 6 — Glofitamab + polatuzumab in r/r CNSL (ranked below Article 7 and 8 by composite but warranting special note): The 🟠 NOVEL_TREATMENT flag is appropriate. A 100% ORR in CNS lymphoma — a uniformly fatal condition at relapse — would be practice-changing if confirmed. The evidence strength cap at 4 (retrospective, n=9) is not pessimism about the finding but appropriate calibration of confidence. This article is arguably the highest-priority watchlist item for prospective trial initiation.

---

PHASE 4 — Deep Dives

---

---

[HOOK]

Ovarian cancer is one of medicine's cruelest silences. By the time most women are diagnosed, the cancer has already spread — and the five-year survival rate falls from over 90% at Stage I to under 30% at Stage III. We don't have a reliable screening test. CA-125, the blood marker clinicians have relied on for decades, misses too many early cancers and raises alarms in too many healthy women. A new study may have found a better way — and it works by listening to signals that are already circulating in the blood.

[THE DISCOVERY]

Researchers from multiple centers across China developed and validated two machine-learning models that analyze fragments of mitochondrial DNA floating freely in the bloodstream — what scientists call circulating cell-free mitochondrial DNA, or ccf-mtDNA. They found that the pattern of how these mitochondrial DNA fragments are cut and sized differs meaningfully between women with ovarian cancer and healthy women, and between women whose adnexal masses are malignant versus benign. Their first model, designed to detect ovarian cancer outright, achieved an AUC of 0.987 — approaching the theoretical ceiling of diagnostic test performance. Critically, it caught more than 91% of Stage I cancers at 95% specificity. Their second model, designed to help triage the millions of women with adnexal masses who need a biopsy-or-watch decision, outperformed standard serum biomarkers across four independent validation cohorts — including populations from Chinese ethnic minority groups.

Think of it this way: conventional tests look at the amount of a single biological molecule in the blood. This approach instead reads a kind of molecular fingerprint — the specific pattern of how mitochondrial DNA is broken into pieces — and uses machine learning to interpret what that pattern says about what's happening in the body.

[THE SCIENCE BEHIND IT]

This was a multi-center diagnostic validation study. The team developed their ML models on a training cohort, then tested performance in four separate external validation cohorts — a study design that is considerably more rigorous than the single-center "train and test on the same dataset" approach that plagues many biomarker papers. The inclusion of a Chinese ethnic minority cohort is an important detail: most liquid biopsy studies are developed in homogeneous populations and then assumed to work universally, which often proves optimistic. Both the OD model (cancer vs. healthy) and BMD model (malignant vs. benign mass) maintained AUC above 0.979 across all validation cohorts. The key limitation is that we are working from the abstract only — full cohort sizes, blinding procedures, and direct comparison to CA-125 plus HE4 multimarker panels are not reported here. It is also important to note that these cohorts were drawn from women already being evaluated for adnexal masses or known cancer, not from a general screening population. Performance in an asymptomatic general population — where the cancer prevalence is far lower — would need separate evaluation.

[WHO THIS HELPS]

The most immediate beneficiaries are women with detected adnexal masses who need help deciding urgency of surgical referral — a daily clinical decision in gynecology and primary care. If validated in general screening populations, the benefit would extend to all women, but especially those at elevated genetic risk (BRCA1/2, Lynch syndrome) who currently rely on imperfect surveillance. The multi-ethnic validation is a meaningful step toward equity in detection, though generalizability to Black, Hispanic, and other non-Asian populations remains to be established.

[THE REAL-WORLD IMPACT]

If this enters clinical practice as a triage tool for adnexal masses, it could reduce unnecessary surgeries for benign cysts while accelerating referral for malignant ones. If it were eventually deployed as a population screening test, the arithmetic is stark: most of the women who die of ovarian cancer each year — roughly 13,000 in the US alone — are diagnosed too late. Shifting even a fraction to Stage I detection would dramatically improve survival without the treatment intensity that advanced-stage disease requires.

[WHAT WE STILL DON'T KNOW]

The most important unknown is how this performs in a true screening population — women who don't yet know they have cancer, and where cancer prevalence is much lower. In low-prevalence populations, even a highly specific test generates a meaningful number of false positives. The cost and infrastructure requirements for ccf-mtDNA sequencing and fragmentomics analysis are not yet established at a clinical scale. And we don't yet know how this compares head-to-head to the best current multimarker approaches, or whether it could be combined with them for even better performance.

[LIKELIHOOD OF MAKING A DIFFERENCE]

• Scientific Confidence: High — for the biomarker performance claims in the studied populations
• Translation Speed: 5–10 years — regulatory validation, prospective screening trials, and infrastructure development are all required
• Barrier Analysis:
  • Regulatory: FDA/CE validation in prospective screening populations required before approval as a screening tool
  • Reimbursement: Liquid biopsy sequencing is expensive; coverage pathways are not established for ovarian cancer screening
  • Infrastructure: Requires sequencing capability and bioinformatics pipeline at clinical scale
  • Equity: Access to sequencing in LMICs and community settings is a major barrier; Western/African-ancestry validation still needed
  • Awareness: Adoption among gynecologists will require guideline incorporation

[CALL TO ACTION / CLOSING]

Ovarian cancer has been waiting decades for a reliable early detection test — and this study is the most rigorous step toward one that we've seen from the liquid biopsy field. Watch this space: if prospective screening trials confirm what these validation cohorts suggest, the way we think about ovarian cancer could fundamentally change.

---

---

[HOOK]

In leukemia treatment, remission doesn't always mean gone. Thousands of malignant cells can persist invisibly after chemotherapy — too few to see on a bone marrow biopsy, but enough to spark a relapse that kills. Measuring this hidden disease burden, called minimal residual disease or MRD, has become one of the most important tools in leukemia management. The problem is that for acute myeloid leukemia — the most common and deadly form of adult leukemia — we've never been able to do it reliably for every patient. Until now.

[THE DISCOVERY]

A multi-center Australian team has developed a way to create a personalized genetic blood test for nearly every patient with AML or related myeloid malignancies. Using next-generation sequencing data from each patient's own tumor at diagnosis, they designed patient-specific PCR assays — ultra-sensitive tests that can detect one leukemia cell among 100,000 normal cells — and then used them to track disease over time. Critically, they succeeded in developing these assays for 97.6% of patients, a near-universal success rate that contrasts sharply with existing methods like flow cytometry or RNA-based PCR, which only work for subsets of patients with specific genetic markers. In the 40 patients monitored prospectively, serial MRD levels tracked with clinical outcomes: rising MRD preceded relapse, and clearance corresponded with durable remission.

[THE SCIENCE BEHIND IT]

The approach builds on well-established technology in childhood leukemia (ALL), where patient-specific DNA-based MRD assays have been standard of care for years. This study extends that framework to myeloid malignancies — a technically more complex challenge because AML is genomically more heterogeneous. The team used NGS or whole-genome sequencing to identify patient-specific somatic variants at diagnosis, then designed PCR assays targeting those variants for longitudinal blood and bone marrow monitoring. This follows EuroMRD-aligned standardization guidelines, which meaningfully increases the chance of cross-center comparability. The main limitation: the prospective monitoring cohort was 40 patients, which is enough to demonstrate feasibility and outcome correlation but not powered for definitive clinical decision-making. Abstract-only access means full outcome data, time-to-relapse correlations, and comparison to flow cytometry head-to-head are not available here.

[WHO THIS HELPS]

The immediate beneficiaries are AML patients across pediatric and adult settings whose disease doesn't carry the specific molecular markers (NPM1 mutations, CBF fusions) that current RNA-based MRD tests require. That represents a substantial majority of AML patients. Pediatric patients are particularly well-served by this approach because treatment intensification decisions — including whether to proceed to stem cell transplant — can now be guided by more sensitive disease monitoring. Multi-center adoption in Australia means this isn't a single-lab curiosity; it's a protocol approaching clinical readiness.

[THE REAL-WORLD IMPACT]

If adopted widely, patient-specific genomic MRD monitoring would change how AML is managed in three ways. First, it enables treatment intensification — including transplant decisions — to be based on molecular data rather than morphology alone, potentially preventing relapse before it becomes clinically apparent. Second, it could identify patients who achieve deep molecular remission and may not need the most intensive (and toxic) consolidation chemotherapy. Third, it enables earlier detection of relapse in post-treatment surveillance, when salvage therapy is most likely to work. The cost of WGS and assay development per patient is the central implementation barrier — but sequencing costs have fallen dramatically and continue to do so.

[WHAT WE STILL DON'T KNOW]

The most important clinical question is whether acting on MRD information actually improves survival. MRD can tell us that leukemia cells are present; it cannot yet tell us definitively what to do about it in every clinical situation. The optimal MRD thresholds for treatment decisions in AML, and whether preemptive intervention based on MRD positivity improves outcomes in randomized comparisons, remain active research questions. Access to WGS at diagnosis in community oncology settings — where most AML patients are initially treated — is not yet routine.

[LIKELIHOOD OF MAKING A DIFFERENCE]

• Scientific Confidence: High — for the assay development and feasibility claims; Moderate — for clinical outcome impact pending larger prospective data
• Translation Speed: 2–5 years — infrastructure is available at major centers; guideline incorporation and cost-coverage will determine speed
• Barrier Analysis:
  • Regulatory: Laboratory-developed tests require validation but not necessarily PMA-level approval in many jurisdictions; CE/FDA pathway for companion diagnostic would take longer
  • Reimbursement: WGS + bespoke assay development is expensive; payer coverage for personalized MRD is inconsistent
  • Infrastructure: Academic and major regional cancer centers can implement now; community oncology will lag
  • Equity: Resource-intensive approach disadvantages patients in LMICs and those treated in community settings; pediatric vs. adult center disparity may also apply
  • Awareness: Hematologists already receptive to MRD monitoring — the main barrier is cost and access, not clinical buy-in

[CALL TO ACTION / CLOSING]

For the first time, virtually every patient with AML may have access to a personalized molecular compass that can track their disease burden with extraordinary sensitivity — the kind of tool that has already changed the game in childhood leukemia. The question now is whether the healthcare system can build the infrastructure to make that compass available to every patient who needs it, not just those fortunate enough to be treated at a center with genomic sequencing capabilities.