# Abnormalities in core AD biomarkers precede inflammatory and glial markers in CSF in Autosomal Dominant Alzheimer's Disease.

- **DOI:** [10.64898/2026.03.31.26349851](https://doi.org/10.64898/2026.03.31.26349851)
- **Rating:** Sound (3.3 out of 5)
- **Verdict:** A meaningful ADAD proteomic resource maps biomarker chronology and benchmarks multiplex assays, but predictive superiority is not sustained in clinical converters.
- **Framework version:** v5.9
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-10-08
- **Evaluation page:** https://www.nabu.science/paper/10.64898%2F2026.03.31.26349851

## Key claims

1. Our results suggest that the main CSF proteomic landscape changes in ADAD are due to the presence of a pathogenic mutation and occur prior to symptom onset.
   - Support: Tentative
   - Type: Primary finding
   - Basis: observational carrier comparisons and threshold-based chronology do not fully identify mutation-driven presymptomatic change
   - Location: Abstract, Interpretation
2. known early alterations in CSF amyloid and tau were followed by inflammatory and neurodegenerative responses
   - Support: Supported
   - Type: Secondary
   - Basis: EYO-based ordering supports the direction, but selected abnormality thresholds could shift individual timings
   - Location: Abstract, Findings; Results, chronological changes
3. we found a multi-protein composite approach for predicting EYO that outperformed single biomarker values.
   - Support: Tentative
   - Type: Secondary
   - Basis: superiority appears in the main test set but not in the held-out clinical converters
   - Location: Abstract, Findings; Results, multiprotein models
4. These results suggest that mutation in these three genes drive changes that are similar regardless of the affected ADAD gene.
   - Support: Tentative
   - Type: Secondary
   - Basis: small gene subgroups and weak or negative effect-size correlations do not establish shared profiles
   - Location: Results, gene-stratified comparison; Figure 1D
5. Overall, we showed that NULISA measurements strongly correlate to the established immunoassay and IP-MS methods in the DIAN study
   - Support: Supported
   - Type: Secondary
   - Basis: most analytes show strong cross-platform correlations, although SNAP25 is only moderately correlated
   - Location: Results, benchmarking section; Figure 1A

## Strengths

- **Large rare-disease proteomic resource** The study analyses 972 CSF samples from 484 DIAN participants, combining cross-sectional and repeated observations across the ADAD continuum.
  - Dimension: contribution
  - Location: Methods, Study Participants; Table 1
- **Orthogonal assay benchmarking and drift handling** NULISA measurements are compared with Lumipulse, IP-MS, SMC, and other immunoassays, and suspected Aβ freeze/thaw drift is explicitly investigated and handled in the chronology analysis.
  - Dimension: methodological rigour
  - Location: Methods, Protein Measurements, Benchmarking, and Chronological Changes; Results, Figure 1A
- **Substantive engagement with prior findings** The Discussion distinguishes replications from non-replications and offers platform or sample-selection explanations for discordant SNAP25, NPTX, and YWHAZ findings.
  - Dimension: positioning
  - Location: Discussion, replication and non-replication paragraphs

## Limitations

- **Converter results undercut predictive superiority** In 24 clinical converters, the multi-protein model’s MAE was 7.88, numerically worse than p-tau217, p-tau181, PiB PET, and age alone. This qualification is not carried into the abstract or the Discussion’s robust-superiority language.
  - Dimension: reporting
  - Location: Results, converter paragraph; Abstract, Interpretation; Discussion, prediction paragraph
- **Chronology depends on selected thresholds** The temporal sequence depends on NC-derived 90th or 10th percentiles chosen for group separation and a 50% abnormality probability threshold, without reported sensitivity analyses around those choices.
  - Dimension: methodological rigour
  - Location: Methods, Chronological Changes of Protein Abundance
- **Cross-gene similarity lacks statistical support** The claim of shared dynamics across genes is difficult to reconcile with effect-size correlations of −0.09 for PSEN2 and −0.22 for APP, alongside an unexplained negative PSEN1 correlation.
  - Dimension: methodological rigour
  - Location: Results, gene-stratified comparison; Figure 1D

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted
  - we found a multi-protein composite approach for predicting EYO that outperformed single biomarker values: "The multi-protein model achieved an MAE of 7.88 (RMSE=9.11), comparable to single-biomarker models including p-tau217 (MAE=5.86, RMSE = 6.68), p-tau181 (MAE=6.86, RMSE=7.75), Aβ42/40 (MAE=7.88, RMSE=8.45), and PiB PET (MAE=6.81, RMSE=7.45)." (Results, Multiprotein models improve the prediction of EYO, clinical-converter paragraph) vs "The model containing only age, also exhibited inferior performance (MAE=7.57, RMSE=9.15)." (Results, Multiprotein models improve the prediction of EYO, clinical-converter paragraph). In converters, the multi-protein MAE exceeds age-only MAE by 7.88−7.57=0.31 years and p-tau217 MAE by 7.88−5.86=2.02 years, whereas its RMSE is 9.15−9.11=0.04 years lower than age-only RMSE. → Not supported by the paper's own data
  - These results suggest that mutation in these three genes drive changes that are similar regardless of the affected ADAD gene: "When PSEN1 carriers were compared to NC, the effect sizes of the two previously tested analyses were highly correlated (ρ=−0.86, p=3.20×10−39). Differences in p-values were most likely driven by differences in sample size. Similar results were observed when comparing PSEN2 (ρ=−0.09, p=0.34) or APP (ρ=−0.22, p=0.01) carriers to non-carriers (Figure 1D)." (Results, CSF proteome reveals widespread proteomic alterations, gene-stratified comparison) vs "These results suggest that mutation in these three genes drive changes that are similar regardless of the affected ADAD gene." (Results, CSF proteome reveals widespread proteomic alterations, gene-stratified comparison). The reported effect-size correlations are strongly inverse for PSEN1 (−0.86), near zero for PSEN2 (−0.09), and weakly inverse for APP (−0.22), rather than consistently strong correlations demonstrating similar changes across all three genes. → Not supported by the paper's own data
  - Minor reporting notes (1):
    - Descriptive: "total tau (MAPT; effect size=0.84, p=6.00×10−24)" (Results, CSF proteome reveals widespread proteomic alterations, strongest mutation-status associations) vs "SMOC1 (effect size=0.84, p=6.00×10−24)" (Results, CSF proteome reveals widespread proteomic alterations, strongest mutation-status associations). MAPT and SMOC1 have identical printed effect sizes and p-values, but the supplied text does not establish whether these are genuine matching results or a transcription error. → Effect on result unclear
- **Reporting and transparency:** Noted. The regression description does not distinguish linear from logistic modeling, the panel designation differs between the Abstract and Results, and the text does not quantify sample attrition during quality control.
  - Ethics approval: declared · Preregistration: declared · Data availability: declared · Code availability: not stated
- **Reference integrity:** Clean. 60 of 60 checkable references verified
  - 61 references in manuscript
  - 1 have no canonical index record — counted, but not index-checkable
  - 4 references 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
