Published: June 02, 2026, 16:53 UTC
The Largest Alzheimer’s Genetics Study Ever: 48 New Genes Pinpoint Novel Drug Targets Beyond Amyloid
The genetics of Alzheimer’s disease has been one of the most intensively studied subjects in human genomics. Over the past two decades, successive genome-wide association studies (GWAS) have identified roughly 80 risk loci — genetic regions where variation influences the likelihood of developing the disease. These discoveries have shaped the field’s understanding of Alzheimer’s biology, pointing toward immune function, lipid metabolism, and endocytosis as key pathways.
But 80 loci is not enough. To transform genetic association into drug development, researchers need more signals — rarer variants, ancestry-diverse populations, and statistical power sufficient to detect the many small-effect variants that collectively determine disease risk.
The latest study from the Psychiatric Genomics Consortium’s Alzheimer’s Disease working group (PGC-ALZ) delivers precisely that. With 183,620 Alzheimer’s cases and 2.6 million controls — roughly 2.8 million individuals in total — it is the largest Alzheimer’s GWAS ever conducted. The results: 127 risk loci, of which 48 are entirely novel, more than doubling the known genetic architecture of the disease.
The study, led by a team at Vrije Universiteit Amsterdam (first author Emil Uffelmann, senior investigators Danielle Posthuma and Ole Andreassen) has been accepted at Nature Genetics.
How Big Is This, Really?
To appreciate the scale, consider the previous landmark study: Bellenguez et al. (2022) in Nature Genetics analyzed roughly 111,000 cases and 1.2 million controls and identified 75 loci. The new PGC-ALZ study nearly doubles the case count to 183,620 — combining clinically diagnosed Alzheimer’s with carefully defined proxy cases from biobanks — and extends the control pool to 2.6 million.
This increase in sample size is not incremental. It transformed what was detectable. The 48 novel loci represent the largest single tranche of new Alzheimer’s-associated genetic regions ever discovered in one study.
Importantly, the study is multi-ancestry, including individuals of European, African American, and other ancestries. Of the 127 total loci, 118 were found in the multi-ancestry analysis, with 9 additional ancestry-specific loci emerging from individual population analyses. Alzheimer’s genetics has been heavily Eurocentric; the inclusion of diverse populations means the findings are more likely to generalize globally.
Beyond Amyloid: What the New Genes Reveal
The expansion of the genetic map confirms some existing hypotheses and introduces new ones. The novel loci fall into several functional categories:
- Neuroinflammation and microglial function: Multiple new loci implicate genes expressed in microglia, the brain’s resident immune cells. This reinforces the growing consensus that neuroinflammation is not merely a consequence of Alzheimer’s pathology but a causal driver — and that microglial genes are among the most promising therapeutic targets.
- Neuronal signaling — a first: For the first time in an Alzheimer’s GWAS, the study found clear enrichment for genes expressed in specific neuronal subtypes, including excitatory and inhibitory neurons. Previous studies had primarily implicated microglia and other glial cells. The neuronal signals suggest that Alzheimer’s risk may involve cell-autonomous vulnerabilities in neurons themselves, not just the immune environment around them.
- Synaptic function and vesicle trafficking: Multiple novel loci converge on genes involved in synaptic transmission and endocytosis, consistent with the early synaptic loss observed in Alzheimer’s.
- Lipid metabolism: Several new signals implicate pathways involved in cholesterol and phospholipid metabolism, connecting Alzheimer’s risk to systemic metabolic processes.
- Protein catabolism: Genes involved in protein degradation and clearance pathways, relevant to the accumulation of amyloid-beta and tau aggregates.
From Loci to Drugs
A GWAS identifies where in the genome risk-associated variation occurs, but not which gene is responsible or how it works. The PGC-ALZ team used gene prioritization methods — integrating expression quantitative trait loci (eQTL) data, chromatin interaction maps, and protein interaction networks — to identify the most likely causal genes at each locus.
The prioritization is crucial for drug development. Many of the new loci map to genes that are druggable — encoding proteins that can be modulated by small molecules or biologics. Some of these genes already have compounds developed for other indications, raising the possibility of drug repurposing.
“The immediate value is in prioritizing targets that have human genetic validation,” says Posthuma. “A drug target backed by human genetics is far more likely to succeed in clinical trials than one discovered purely through animal models or cell assays.”
The shift beyond amyloid is significant. Anti-amyloid antibodies (lecanemab, donanemab) represent the first class of disease-modifying therapies to reach regulatory approval, but their clinical effects are modest, and they target only one aspect of the disease. The PGC-ALZ data suggests that Alzheimer’s is genetically heterogeneous — different patients may have disease driven by different biological pathways — and that effective treatment may require targeting multiple mechanisms.
Relevance to the GLP-1 Connection
The emphasis on neuroinflammation and microglial biology connects directly to one of the most exciting emerging areas in neurodegeneration research: the neuroprotective effects of GLP-1 receptor agonists like semaglutide (Ozempic). These drugs have well-characterized anti-inflammatory properties in the brain, and large observational studies and small clinical trials have suggested they may reduce the risk of Alzheimer’s and Parkinson’s. The genetic validation of inflammatory pathways as causal drivers of Alzheimer’s risk provides a mechanistic foundation for these observations — and suggests that anti-inflammatory strategies, including GLP-1 drugs, may be addressing a root cause rather than a downstream consequence.
The Limits
Even at this unprecedented scale, GWAS has limitations. The 127 loci explain an estimated 19% of Alzheimer’s heritability in European populations — up from ~12% in earlier studies, but still leaving the majority of genetic risk unaccounted for. The remaining “missing heritability” likely resides in rare variants, structural variants, and gene-gene interactions that require even larger sample sizes or different study designs to detect.
The polygenic risk score (PRS) derived from the study explains up to 17% of variance in case-control status — an improvement but not yet clinically useful for individual prediction. And the novel neuronal signals, while exciting, require functional validation to confirm that the prioritized genes actually play causal roles in neurodegeneration.
What the study does provide is a massively expanded target map — 48 new starting points for a field that urgently needs more. After two decades of amyloid dominance, the genetic architecture of Alzheimer’s is telling a more complex story. It is now the job of the drug development community to read it.
Reference: Uffelmann, E., Wightman, D. P., Bahrami, S., Shadrin, A. A., Fominykh, V., Posthuma, D., Andreassen, O. A., et al. (2026). Genomic analyses reveal new insights into Alzheimer’s disease. Nature Genetics. Preprint at medRxiv DOI: 10.1101/2025.10.10.25337470. Accepted for publication.