# Genetic risk of Alzheimer's disease is associated with loss of brain network segregation in midlife.

- **DOI:** [10.1038/s42003-026-10282-0](https://doi.org/10.1038/s42003-026-10282-0)
- **Rating:** Sound (3.0 out of 5)
- **Verdict:** A meaningful midlife longitudinal neuroimaging study identifies APOE-related DMN segregation change, but its mechanistic interpretation exceeds the observational evidence.
- **Framework version:** v5.9
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-10-08
- **Evaluation page:** https://www.nabu.science/paper/10.1038%2Fs42003-026-10282-0

## Key claims

1. Cognitively healthy midlife APOE ε4 carriers have higher segregation across the brain cross-sectionally, accompanied by greater longitudinal decline in the DMN over two years, relative to non-carriers.
   - Support: Supported
   - Type: Primary finding
   - Basis: adjusted observational models support the direction, but the longitudinal interaction is modest and based on a limited paired sample
   - Location: Abstract; Results, Figures 3–5 and Table 5
2. Higher DMN segregation is associated with better episodic and relational memory across the PREVENT cohort.
   - Support: Tentative
   - Type: Secondary
   - Basis: cross-sectional association only, with unclear correction across the three cognitive factors described in Methods
   - Location: Abstract; Results, Network Segregation and Cognitive Performance in PREVENT; Figure 4
3. Higher-order networks showing the most significant age-related decline are the default mode (DMN), frontal-parietal control (FPN) and salience (SN) networks.
   - Support: Tentative
   - Type: Secondary
   - Basis: network effects derive from cross-sectional age differences rather than observed longitudinal decline
   - Location: Abstract; Results, Network-Level Segregation in Cam-CAN; Figure 2
4. a role for segregation loss in the DMN as a mechanism for incipient impairment of episodic memory, from midlife.
   - Support: Tentative
   - Type: Secondary
   - Basis: observational associations and null longitudinal cognition coupling cannot identify a mechanism of impairment
   - Location: Discussion; Conclusion

## Strengths

- **Independent network selection and testing** Age-sensitive networks were identified in the 652-participant Cam-CAN cohort before APOE associations were examined in PREVENT, reducing reliance on wholly data-driven network selection within the target cohort.
  - Dimension: contribution
  - Location: Introduction; Results, Network-Level Segregation in Cam-CAN; Figure 2
- **Multiple technical robustness analyses** The principal findings were checked using an alternative atlas, mutual-information connectivity, normalised participation coefficients, and a harmonised preprocessing pipeline.
  - Dimension: methodological rigour
  - Location: Methods, Functional Brain Network Construction; Edge Definition; Validation Analyses
- **Unexpected direction reported transparently** The paper states that carriers had higher baseline segregation, contrary to its prediction, and discusses this result in relation to midlife hyperconnectivity and antagonistic pleiotropy.
  - Dimension: reporting
  - Location: Introduction, Predictions; Results, PREVENT Baseline; Discussion, paragraphs 2–3

## Limitations

- **Mechanistic conclusion exceeds observed association** The conclusion calls DMN segregation loss a mechanism for incipient memory impairment, although changes in segregation were not significantly associated with changes in memory and no mediation analysis was reported.
  - Dimension: methodological rigour
  - Location: Results, Longitudinal Effects; Methodological Considerations; Conclusion
- **Biomarker comparison lacks direct validation** The paper presents segregation as earlier and complementary to CSF and blood biomarkers without measuring AD pathology, comparing biomarkers, predicting dementia, or evaluating diagnostic performance.
  - Dimension: positioning
  - Location: Abstract; Discussion; Conclusion
- **Cognitive multiplicity reporting remains unclear** Methods refer to three cognitive factors and multiplicity correction, while the main Results report only episodic/relational memory with p values of 0.03–0.04. The tested family and corrected inference therefore require clarification.
  - Dimension: methodological rigour
  - Location: Methods, Cross-sectional Effects; Results, Network Segregation and Cognitive Performance in PREVENT

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted
  - Cognitively healthy midlife APOE ε4 carriers have higher segregation across the brain cross-sectionally, accompanied by greater longitudinal decline in the DMN over two years, relative to non-carriers: "Results suggest an early impact of the genetic risk for late-life sporadic AD on loss of functional segregation across the brain, and in the DMN, as well as role for segregation loss in the DMN as a mechanism for incipient impairment of episodic memory, from midlife." (Conclusion) vs "APOE ε4 carriers had significantly higher segregation than ε4 non-carriers (Fig. 3a)." (Results: Global segregation in the PREVENT cohort at baseline). The Conclusion's statement of segregation loss across the brain conflicts with the explicitly reported higher global segregation in carriers at baseline, while the longitudinal Results also report no significant global time × genotype interaction. → Not supported by the paper's own data
- **Reporting and transparency:** Noted. The cognitive-analysis description conflicts over whether one or three factors were assessed, and the reported ε2ε4 prevalence cannot represent a participant count in the stated sample; neither issue establishes a corrected key-result estimate.
  - Conflict of interest: not stated · Ethics approval: declared · Data availability: not stated · Code availability: not stated
- **Reference integrity:** Clean. 104 of 104 checkable references verified
  - 105 references in manuscript
  - 1 have no canonical index record — counted, but not index-checkable
  - 1 reference confirmed by manual review
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-10-17
