# Post-mortem evidence of pathogenic angiogenesis and abnormal vascular function in early Alzheimer's disease.

- **DOI:** [10.1093/brain/awaf394](https://doi.org/10.1093/brain/awaf394)
- **Rating:** Limited (2.9 out of 5)
- **Verdict:** A broad regional post-mortem study strengthens vascular associations in early Alzheimer’s disease, but its causal framing exceeds what cross-sectional biomarkers can establish.
- **Framework version:** v5.9
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-10-08
- **Evaluation page:** https://www.nabu.science/paper/10.1093%2Fbrain%2Fawaf394

## Key claims

1. The present data provide evidence of widespread cerebral hypoperfusion associated with pathogenic angiogenesis and vascular remodelling in AD.
   - Support: Tentative
   - Type: Primary finding
   - Basis: pathogenic framing outruns cross-sectional proxy-marker associations and is affected by directional reporting inconsistencies
   - Location: Abstract, final paragraph
2. MAG:PLP1 was lower in BSIII–IV than BS0–II controls in all cortical regions studied, as well as the putamen.
   - Support: Tentative
   - Type: Secondary
   - Basis: regional group differences support the scoped result, but the broader abstract wording incorrectly includes all regions
   - Location: Results, Fig. 1
3. EDN1 correlated strongly with Aβ1–42 in the BS0–II group in almost all cortical regions.
   - Support: Supported
   - Type: Secondary
   - Basis: reported regional correlations are consistent, though based on eight controls and extensive uncorrected testing
   - Location: Results, “EDN1 correlates with Aβ1–42 in low-pathology controls”; Fig. 4
4. PDGFRβ:CD31 was lower in BSIII–IV than BS0–II in all grey matter regions.
   - Support: Supported
   - Type: Secondary
   - Basis: the regional group comparison is reported consistently, although interpretation as pericyte loss remains indirect
   - Location: Results, “PDGFRβ levels are lower…”; Fig. 6A
5. BBB leakiness, evidenced by elevated fibrinogen in brain tissue homogenates, was highest in BSV-VI in most brain regions and correlated with VEGF-A, CD31 and ptau-231 concentrations.
   - Support: Supported
   - Type: Secondary
   - Basis: regional associations support the direction, but bulk-homogenate fibrinogen is an indirect BBB-leakiness proxy
   - Location: Abstract; Results, Fig. 7

## Strengths

- **Broad regional mapping across disease stages** The study measures vascular and pathological markers across six cortical regions, putamen and trigonal white matter in three Braak-stage groups, extending earlier single-region work.
  - Dimension: contribution
  - Location: Methods, Study Cohort; Results, Figs. 1–7
- **Detailed biochemical assay procedures** ELISA methods identify assay sources, dilution procedures, duplicate measurements and total-protein normalization, with carry-over samples used to monitor plate variation.
  - Dimension: methodological rigour
  - Location: Methods, Brain Tissue Homogenization through Statistical Analyses
- **Limitations explicitly tied to interpretation** The Discussion acknowledges the small control group, absence of cell-type resolution, unavailable vascular comorbidities and inability of post-mortem correlations to prove causality.
  - Dimension: positioning
  - Location: Discussion, limitations paragraph

## Limitations

- **Central correlation directions are contradictory** The text labels CD31–MAG:PLP1 as inverse despite positive coefficients and PDGFRβ:CD31–MAG:PLP1 as positive despite negative coefficients. These inconsistencies affect relationships used in the vascular-remodelling interpretation.
  - Dimension: reporting
  - Location: Results, Figs. 3B and 6B
- **Causal conclusions exceed study design** The Conclusion describes Aβ1-42–EDN1-induced hypoperfusion and VEGF-A-induced remodelling, although the study is cross-sectional and the limitations state that causality cannot be proven.
  - Dimension: methodological rigour
  - Location: Discussion, limitations paragraph; Conclusion
- **Multiplicity and small subsets limit inference** Numerous regional correlations lack a stated global correction, while the 55-protein profiler uses only n=3, n=3 and n=4. This makes nominally significant findings vulnerable to instability and multiplicity.
  - Dimension: methodological rigour
  - Location: Methods, Angiogenic Proteomic Profiler and Statistical Analyses; Results, Fig. 5

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted
  - The present data provide evidence of widespread cerebral hypoperfusion associated with pathogenic angiogenesis and vascular remodelling in AD: "Concentrations of CD31, a marker of endothelial cells, were raised in BSIII–IV in most brain regions and correlated inversely with MAG:PLP1 and positively with VEGF-A." (Discussion, CD31 and endoglin paragraph) vs "CD31 correlated inversely with MAG:PLP1 in the temporal cortex (r = 0.62, P < 0.001) and trigonal white matter (r = 0.53, P < 0.01) and weakly in the calcarine, frontal and parietal cortex (Fig. 3B)." (Results, widespread markers of cerebrovascular dysfunction, CD31 correlation paragraph). The printed temporal and trigonal correlation coefficients, 0.62 and 0.53, are positive and therefore contradict the stated inverse association. → Not supported by the paper's own data
  - The present data provide evidence of widespread cerebral hypoperfusion associated with pathogenic angiogenesis and vascular remodelling in AD: "The PDGFRβ:CD31 ratio, a marker of pericyte content adjusted for vessel density, was lower in BSIII-IV than BS0-II and correlated positively with MAG:PLP1 and inversely with Aβ1-42." (Abstract) vs "PDGFRβ:CD31 correlated positively with MAG:PLP1 in the frontal (r = −0.69, P < 0.01) and temporal cortex (r = −0.60, P < 0.001), and more weakly in the calcarine cortex and putamen (Fig. 6B)." (Results, PDGFRβ levels are lower in early-stage AD). The printed frontal and temporal correlation coefficients, −0.69 and −0.60, are negative and therefore contradict the stated positive association. → Not supported by the paper's own data
  - EDN1 correlated strongly with Aβ1–42 in the BS0–II group in almost all cortical regions: "Elevated EDN1 and reduced PDGFRβ were both strongly related to Aβ42 concentration in early AD." (Discussion, opening paragraph) vs "VEGF-A, EDN1 and CD31 did not correlate with pathological markers across the entire cohort but analysis within groups stratified according to BS stage revealed that EDN1 correlated strongly with Aβ1–42 in the BS0–II group in almost all cortical regions:" (Results, EDN1 correlates with Aβ1–42 in low-pathology controls). The Discussion assigns the strong EDN1–Aβ42 relationship to early AD, whereas the Results explicitly assign it to BS0–II low-pathology controls rather than the BSIII–IV early-stage/intermediate group. → Not supported by the paper's own data
- **Reporting and transparency:** Clean. No existing reporting or conduct concern supports a retained finding after excluding unavailable supplementary material and ungrounded reporting-error allegations.
  - Ethics approval: declared · Data availability: declared · Code availability: not stated
- **Reference integrity:** Noted
  - 50 references in manuscript
  - 50 of 50 checkable references found in an index
  - Citation details diverge for 3 references:
    - #7 Benzinger TL , Blazey T , Jack CR Jr , et al Regional variability of imaging biomarkers in autosomal dominant Alzheimer’s disease . Proc Natl Acad Sci U S A . 2013 ; 110 : E4502 – E4509 . 24194552 10.1073/pnas.1317918110 PMC3839740 (DOI resolves to: “Regional variability of imaging biomarkers in autosomal dominant Alzheimer’s disease” — Benzinger, 2013 (Proceedings of the National Academy of Sciences). Differs from the DOI’s record on: container, authors) https://doi.org/10.1073/pnas.1317918110
    - #45 Argaw AT , Gurfein BT , Zhang Y , Zameer A , John GR . VEGF-mediated disruption of endothelial CLN-5 promotes blood-brain barrier breakdown . Proc Natl Acad Sci U S A . 2009 ; 106 : 1977 – 1982 . 19174516 10.1073/pnas.0808698106 PMC2644149 (DOI resolves to: “VEGF-mediated disruption of endothelial CLN-5 promotes blood-brain barrier breakdown” — Argaw, 2009 (Proceedings of the National Academy of Sciences). Differs from the DOI’s record on: container, authors) https://doi.org/10.1073/pnas.0808698106
    - #49 Bennett RE , Robbins AB , Hu M , et al Tau induces blood vessel abnormalities and angiogenesis-related gene expression in P301L transgenic mice and human Alzheimer’s disease . Proc Natl Acad Sci U S A . 2018 ; 115 : E1289 – E1298 . 29358399 10.1073/pnas.1710329115 PMC5819390 (DOI resolves to: “Tau induces blood vessel abnormalities and angiogenesis-related gene expression in P301L transgenic mice and human Alzheimer’s disease” — Bennett, 2018 (Proceedings of the National Academy of Sciences). Differs from the DOI’s record on: container, authors) https://doi.org/10.1073/pnas.1710329115
  - 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
