Coherent genetics-to-application narrative
The article links Alzheimer’s genetic architecture to immune pathways, risk prediction, biomarkers, clinical trials, and hiPSC models in a readily traceable sequence.
↳ Full text, paragraphs 3–8
Abstract
Discovering relationships between clinical diseases and variation in DNA provides connections to putative disease pathways and novel targets for future therapies. This genetic strategy is and will continue to shape our understanding of disease mechanisms, epidemiology, advance clinical trials and drug discovery and revolutionise therapeutic development. Over the last decade genome-wide association studies (GWAS) have driven most of these advances for common diseases with genetic components (Sims et al. 2020). The effects of human genetic variability combined with the small individual effect sizes of disease associations commonly observed, have been overcome by powerful GWAS studies with stringent levels of proof, underpinned by large international collaborations with in-built replication across different laboratories and countries. These studies have been particularly informative for brain diseases, of which Alzheimer’s disease is a prime example. This commentary will provide a brief overview of the genetics of Alzheimer’s disease, illustrate how this can inform disease mechanisms and treatments and review whether we are exploiting the full potential of genetic discoveries, highlighting the failure of pharmaceutical companies to harness the wealth of genetic knowledge in clinical trials, for example.
Key Claims
Primary findingTentative
Notably, few if any, clinical trials of new therapies for AD, test individuals’ polygenic AD risk, even though this is easily achieved.
no trial inventory or citation substantiates the near-universal claim about current practice
Full text, paragraph 5
SecondaryTentative
Matching individual therapies to those showing biological risk to the disease pathway/s targeted, can only improve success rates for clinical trials.
the commentary provides a rationale but no comparative trial evidence establishing improved success
Full text, paragraph 5
SecondarySupported
In 2015 we developed a polygenic risk score which has an accuracy of around 80% to predict AD in neuropathologically confirmed AD case-control samples.
the estimate is sample-specific and accompanied by population-bias and ancestry caveats
Full text, paragraph 5
SecondaryTentative
We have already observed significant differences between high and low AD polygenic risk in several independent microglial functions.
no methods, effect estimates, or detailed results are reported for the stated functional differences
Full text, paragraph 7
Derived from the full evaluation — not a separate score.
Strengths · Limitations
Strengths
Specific mechanistic pathway synthesis
The TREM2–PLCγ2 discussion connects risk and protective variants with microglial activation, survival, plaque responses, and therapeutic tractability.
↳ Full text, paragraph 4
Concrete audiences and proposed uses
The commentary identifies pharmaceutical companies, trial designers, biomarker researchers, epidemiologists, and model developers as potential users of genetic stratification.
↳ Full text, paragraphs 5–8
Limitations
Central trial claim lacks evidence
The statement that few, if any, Alzheimer’s trials test polygenic risk is unsupported by a trial inventory, citation, or review method.
↳ Full text, paragraph 5
Prescription stated as guaranteed benefit
The claim that matching therapies to pathway-specific genetic risk “can only improve” trial success is not supported by comparative trial evidence.
↳ Full text, paragraph 5
Implementation context remains unexamined
The article does not engage practical, regulatory, or predictive-performance constraints that could explain limited adoption of polygenic trial enrichment.
↳ Abstract; full text, paragraphs 5–6
Nabu Commentary
The article provides a coherent synthesis of Alzheimer’s genetics, particularly the TREM2–PLCγ2, complement, polygenic-risk, biomarker, and hiPSC discussions. Its principal prescription is less secure: the statement that few, if any, trials assess polygenic risk is not supported by a trial inventory or citation. The further assertion that pathway matching can only improve trial success converts a plausible proposal into a conclusion without comparative evidence. Consequently, the article is most useful as an advocacy agenda rather than as a demonstrated case for changing trial practice.
Check another paper
Author Note / Field Feedback
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Dimension breakdown
Study Quality
Limited2.8
Confidence mediumContribution
2.8
The commentary identifies a concrete agenda for integrating genetics into Alzheimer’s trials, biomarkers, and disease models, but largely synthesises established findings and the authors’ prior framework. Its advance is therefore present but incremental.
Abstract; full text, paragraphs 5–9
“few if any, clinical trials of new therapies for AD, test individuals’ polygenic AD risk”
Methodological Rigour
2.7
The progression from genetic architecture through mechanisms, risk prediction, biomarkers, and models is coherent. The central claims about current trial practice and guaranteed improvement are uncited, and no source-selection method is reported.
Full text, paragraphs 4–8
“Matching individual therapies to those showing biological risk to the disease pathway/s targeted, can only improve success rates”
Reporting
3.1
The conceptual sequence is accessible and readily followed, but categorical wording in the abstract and trial discussion presents an untested prescription as established. The reversed early- and late-onset age definitions also create avoidable confusion.
Abstract; full text, paragraphs 2 and 5
“highlighting the failure of pharmaceutical companies to harness the wealth of genetic knowledge”
Positioning
2.8
The commentary covers major Alzheimer’s genetic pathways and acknowledges ancestry, control, and survival limitations of polygenic scores. It does not engage the practical, regulatory, or existing-genotyping context needed to substantiate its central criticism of clinical trials.
Full text, paragraphs 4–6
“more diversity has yet to be tested and population specific polygenic scores calculated”
Reliability
Caveats4 of 4 checks
Internal CoherenceNoted
A background definition reverses the age thresholds for early- and late-onset Alzheimer’s disease, while later text uses the conventional threshold for late onset. The commentary also relies repeatedly on the authors’ prior framework and resources, warranting caution about selection without reaching a reliability-concern threshold.
Research ConductClean
Ethics approval, registration, and data or code availability are not applicable to this commentary, and no conduct concern is established from the supplied text. No conflict-of-interest statement is visible, but this is treated as an assessability gap rather than a conduct concern.
Flags: 0 declared / 5 total
Reference IntegrityClean
19 references in manuscript 19 of 19 checkable references found in an index
Post-Publication RecordClean
No retraction notice found in Retraction Watch.
Last checked: 16 Oct 2026
Sources: Retraction Watch ✓
Readiness
Where this paper’s evidence sits on the path from initial observation to real-world use.
Mechanistic
The proposed trial-stratification pathway remains at the advocacy and hypothesis-generating stage. No comparative trial evidence or implementation protocol demonstrates improved recruitment, power, or outcomes.
Full text, paragraph 5
“can only improve success rates for clinical trials”
AI-generated, human-governed. Something look off? Contact us to request a review.