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-E45AEAAD·Framework v5.9·Pipeline v0.4·Evaluated 08 Oct 2026·OECD: 3.2 Clinical medicine·Subfield: Psychiatry and Mental health·Basis: Full text·Type: Prediction or prognostic modeling

Accelerated long-term forgetting as a predictor of clinical onset in presymptomatic autosomal dominant Alzheimer's disease.

Magill et al.Magill, Tank, Ferguson et al.·2025·Journal of neurology, neurosurgery, and psychiatry·10.1136/jnnp-2025-336750 ↗

Primary claim

Severity of ALF is not only associated with the presence of AD pathology but also predictive of clinical onset, identifying those at the highest risk of imminent decline.

Tentative · small observational cohort without validated time-to-event prediction, compounded by inconsistent headline statistics
In brief

Longitudinal ALF testing identifies a promising presymptomatic ADAD prognostic signal, but internally inconsistent headline statistics prevent confident interpretation.

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

BACKGROUND: In Alzheimer's disease (AD), sensitive measures of cognitive decline prior to overt symptoms are urgently needed. Accelerated long-term forgetting (ALF), where new information is retained normally over conventional testing intervals but is then lost at an accelerated rate over the following days and weeks, has been identified cross-sectionally in presymptomatic autosomal dominant and sporadic AD cohorts. We aimed to assess whether ALF testing is predictive of proximity to future symptom onset.

METHODS: 20 asymptomatic autosomal dominant AD mutation carriers who performed normally on standard cognitive testing underwent ALF assessment with (1) a list, (2) a story and (3) a visual figure, with the testing of 30-min recall and 7-day recall. Participants were followed up annually for a median of 7 years and assessed each time with the Clinical Dementia Rating (CDR) scale.

RESULTS: 9/20 participants developed symptoms (CDR global>0) during follow-up. Those who became symptomatic had lower baseline ALF scores for both the list (progressors=30 (IQR, 30-36.4) and non-progressors=58.3 (IQR, 33-66.7), p=0.03) and story (progressors=58.8 (IQR, 44-66) and non-progressors=81.2 (IQR, 69.1-87.8), p<0.001). Story ALF (area under curve (AUC)=0.82) and list ALF (AUC=0.73) discriminated between those who did and did not develop symptoms.

CONCLUSIONS: Severity of ALF is not only associated with the presence of AD pathology but also predictive of clinical onset, identifying those at the highest risk of imminent decline. ALF testing offers promise in aiding presymptomatic trial recruitment, as a presymptomatic cognitive endpoint and potentially as a screening tool in the wider population.

Key Claims

Primary finding · Tentative

Severity of ALF is not only associated with the presence of AD pathology but also predictive of clinical onset, identifying those at the highest risk of imminent decline.

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

small observational cohort without validated time-to-event prediction, compounded by inconsistent headline statistics

Abstract, Conclusions
Quote verified
Secondary · Supported

Those who became symptomatic had lower baseline ALF scores for both the list

reported medians support the direction within this small cohort, although inferential uncertainty remains

Results, paragraph 1; Abstract
Quote verified
Secondary · Tentative

Story ALF (area under curve (AUC)=0.82) and list ALF (AUC=0.73) discriminated between those who did and did not develop symptoms.

in-sample AUCs from nine events lack uncertainty and conflict with the reported rank-test p-values

Abstract; Results, Figure 1C
Quote verified
Secondary · Tentative

No differences were found between those who did become symptomatic and those who did not become symptomatic for any of the other more conventional cognitive assessments.

the same 20-person cohort is underpowered to establish that conventional measures lack prognostic association

Results, paragraph 3; Table 1
Quote verified

Derived from the full evaluation — not a separate score.

Strengths · Limitations

Strengths

#contribution

Extended follow-up in genetically defined carriers

The cohort comprised asymptomatic ADAD mutation carriers with annual clinical assessment over a median of seven years, allowing baseline ALF to be related prospectively to later CDR-defined progression.

↳ Methods, Participants and Procedures; Results, paragraph 1

#methodological rigour

Relevant within-cohort cognitive comparators included

Three ALF modalities and eight conventional tests spanning memory, executive function, and global intellectual function were analysed in the same cohort. The negative figure result was reported alongside the verbal findings.

↳ Methods, Statistical Analysis; Results, paragraphs 1 and 3

#application pathway

Clear connection to presymptomatic trial needs

The Discussion identifies trial recruitment, cognitive endpoints, and treatment stratification as concrete potential uses while also noting implementation burden and the need for replication.

↳ Discussion, paragraphs 3–4 and 6

Limitations

#methodological rigour

Headline statistics are internally inconsistent

For the same two-group ALF predictors, the reported rank-test p-values imply substantially stronger separation than the AUCs in Figure 1C. This discrepancy affects the numerical basis of the central discrimination claim and requires resolution.

↳ Results, paragraph 1; Figure 1C; Discussion, paragraph 1

#methodological rigour

Discrimination estimates lack validation and uncertainty

AUCs are estimated and evaluated in the same 20-person cohort, with no confidence intervals, calibration assessment, cross-validation, or external validation. Nine progression events therefore support all predictive interpretation.

↳ Methods, Statistical Analysis; Results, Figure 1C

#reporting

Clinical language exceeds the studied evidence

Claims about identifying imminent decline and wider-population screening go beyond an ADAD-only cohort with no validated time-to-event prediction or demonstrated transfer to sporadic disease.

↳ Abstract, Conclusions; Discussion, final paragraph

Does this match your read? Add your assessment ↓

Nabu Commentary

The extended prospective follow-up and genetically defined cohort create a meaningful advance over predominantly cross-sectional ALF evidence. Methodological confidence is limited by nine progression events, in-sample AUCs without uncertainty or validation, and uncorrected testing across multiple cognitive measures. More importantly, the p-values reported for list and story ALF do not align with their corresponding AUCs, so the headline numerical result requires correction or explanation. The study therefore supports further prognostic investigation of verbal ALF, not current claims of individual-level prediction or wider screening.

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
3.2±0.2

Contribution
3.7

The study extends mostly cross-sectional ALF evidence by linking baseline performance to clinical progression over a median seven years in genetically confirmed carriers. The advance is meaningful but remains preliminary because only 20 participants and nine progressions underpin it.

Introduction, final paragraph; Results, paragraphs 1–3

We investigated the value of ALF assessment in predicting future onset of clinical decline

Methodological Rigour
2.8

The prospective cohort, annual CDR outcome ascertainment, three ALF modalities, and conventional-test comparators fit the research question. Methodological Rigour is constrained by in-sample AUCs without confidence intervals or validation, uncorrected multiple testing, descriptive median splits, and the p-value/AUC inconsistency affecting the headline results.

Methods, Statistical Analysis; Results, paragraphs 1–3 and Figure 1

Receiver operating characteristic curves were generated using logistic regression models

Reporting
3.2

The paper is short and structurally easy to follow from question through follow-up and results. Its language becomes materially stronger than the evidence when the Abstract claims prediction of imminent decline and wider-population screening from an unvalidated 20-person ADAD cohort.

Abstract, Conclusions; Discussion, final paragraph

identifying those at the highest risk of imminent decline

Positioning
3.6

The paper relates its findings to prior cross-sectional ALF studies, a 12-month predictive study, trial endpoints, biomarkers, and conventional cognitive testing. It acknowledges replication, generalisability, and implementation burden, although the small sample is partly reframed as evidence of a strong effect.

Introduction, paragraphs 1–4; Discussion, paragraphs 2–6

Replication in both familial and sporadic presymptomatic AD cohorts will be important.

Lower confidence on Methodological Rigour — domain match limited.

Reliability

Caveats4 of 4 checks


Results checkNoted
  • Story ALF (area under curve (AUC)=0.82) and list ALF (AUC=0.73) discriminated between those who did and did not develop symptoms: "story (progressors=58.8 (IQR, 44–66), non-progressors=81.2 (IQR, 69.1–87.8), p<0.001)" (Results, first paragraph) vs "story ALF (area under curve (AUC)=0.82)" (Results, first paragraph). Using 9 progressors and 11 non-progressors, treating the rounded AUC as the rank-order probability and ignoring ties gives z=(0.82−0.5)√(12×9×11/21)≈2.407 and a two-sided normal-approximation p≈0.016, rather than an exact reconstruction of the stated Kruskal–Wallis test. → Conclusion holds
Reporting and transparencyClean

No supplied concern establishes a reporting or research-conduct finding.

Conflict of interest: not stated · Ethics approval: declared · Preregistration: not stated · Data availability: not stated · Code availability: not stated

Reference IntegrityClean

18 of 18 checkable references verified

18 references in manuscript

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 · Neurological · Alzheimer's disease and other dementias
Type of advanceEmpirical
Route to useNamed
AdvanceExtends
BeneficiariesLarge
CitationsToo early
Policy · patentNot yet traced
ReachedNot reachedNot assessable or unverified
ReadinessPiloted
Foundational
Controlled
Piloted
Validated
Implementable
FoundationalImplementable

Predictive value was demonstrated once in a small single cohort of 20 carriers, with the authors themselves calling for replication in other cohorts.

From the paper
“ALF testing offers promise in aiding presymptomatic clinical trial recruitment as a presymptomatic cognitive endpoint”

The paper names trial teams and, elsewhere, screening settings as those who would use ALF testing for recruitment, endpoints and population screening.

Pathway

Clinical trial teams and memory clinics would adopt ALF testing to stratify presymptomatic individuals for trial recruitment and treatment decisions.

Barrier

The finding rests on only 20 mutation carriers in one cohort and the two-visit testing format is resource-intensive, so independent replication and streamlined delivery are needed.

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. Effect of intensive blood pressure control on cognitive outcomes and brain structures in adults with type 2 diabetes: prespecified analysis of secondary outcomes of BPROAD randomised controlled trial

    Wang, Xu, Xu, et al. · 2026 · Clinical medicine

    Primary claim

    no statistically significant difference in incidence of a composite of all cause dementia or mild cognitive impairment was observed among adults with type 2 diabetes receiving standard treatment of SBP and intensive treatment

    Well-supported
    Evidence Score
    4.1±0.1
    Reliability
    Caveats
  2. Smoking and dementia risk: an updated systematic review and meta-analysis of 61 prospective cohort studies and implications for tobacco control

    Mario Siervo, Jacob Brain, Sarah Beasley, et al. · 2026 · Health sciences

    Primary claim

    Current smoking was associated with increased risk of all-cause dementia (RR 1.31 [95% CI 1.23–1.39]; number of studies (k) = 39), Alzheimer's disease (1.40 [1.08–1.80]; k = 15), and vascular dementia (1.52 [1.20–1.93]; k = 9).

    Supported
    Evidence Score
    3.3±0.2
    Reliability
    Caveats
  3. Decreased awareness of cognitive decline is associated with multimodal Alzheimer's disease biomarkers in cognitively unimpaired individuals

    2026 · Clinical medicine

    Primary claim

    decreased ACD was associated with greater AD pathological burden

    Supported
    Evidence Score
    3.2±0.2
    Reliability
    Caveats

How the rubric works →