AI summary · DOI: 10.64898/2026.11.13.26367865 · medRxiv: Neurology

PSEN1 loss-of-function variants do not cause early-onset Alzheimer's disease

A. Farinas, A. Pena-Tauber, T. J. Hohman, A. G. Contreras, T. D. Jensen, Y. Le Guen, M. D. Greicius
Original paper Published on 13 Nov. 2026 medRxiv preprint License: CC BY-NC-ND 4.0

How to read this summary? Illustrated guide

The essentials

Across the cohorts examined, carriers of predicted PSEN1 null variants had no recorded early-onset dementia, while disease-associated variants could produce altered PSEN1 proteins.

Question

Do true gene-level loss-of-function variants in PSEN1 cause early-onset Alzheimer’s disease, or does disease require production of an altered PSEN1 protein? The study also re-examines the reported K109* variant in an affected father and son.

Method

The authors searched ADSP, UK Biobank, and BioVU for high-confidence PSEN1 variants, identifying 45 carriers of frameshift or stop-gain variants predicted to trigger nonsense-mediated decay; they assessed dementia diagnoses and age at last observation. They also examined 13 carriers of variants potentially producing truncated or alternatively spliced protein and used AlphaGenome, SpliceAI, and Pangolin to predict splicing effects of K109*.

Findings

  • Result None of the 45 carriers of predicted PSEN1 null variants had recorded early-onset dementia; 42 had no recorded dementia, including 17 followed beyond age 65.
    See the passage in the paper
    Of the 45 individuals carrying high-confidence PSEN1 LoF mutations, none had early-onset dementia. 42 individuals remained dementia-free through their last assessment, including 17 carriers who were observed past age 65 (Figure 1).
  • Result Six of the 13 carriers of potentially protein-producing splice or truncation variants had early-onset MCI or Alzheimer’s disease, and all six carried known exon 9 acceptor splice variants.
    See the passage in the paper
    Of these individuals, 6 had early-onset mild cognitive impairment (MCI) or EOAD, with disease onset before age 65. All early-onset cases were splice variant carriers at the exon 9 acceptor site (14:73206385:G:A and 14:73206385:G:T).
  • Result Three splice-prediction models indicated that K109* could create a cryptic donor, producing an in-frame deletion rather than only a null allele.
    See the passage in the paper
    Three sequence-based splicing predictors all supported increased use of this cryptic donor and decreased use of the canonical exon 4 donor for the alternate allele (Figure 3C).
  • Authors’ interpretation The authors infer that familial Alzheimer’s disease associated with PSEN1 requires an altered PSEN1 protein and that one normal allele appears sufficient for relevant PSEN1 functions.
    See the passage in the paper
    PSEN1-mediated FAD seems to require production of an altered PSEN1 protein product.

Limitations

  • Dementia diagnoses and observation periods came from clinical records and cohort assessments, so undetected or later-onset diagnoses may not be captured.
    See the passage in the paper
    We also identified fewer than 20 carriers of PSEN1 LoF mutations in the AoU database; while specific findings could not be disclosed under AoU data dissemination policies, the observed phenotypes were consistent with our conclusions.
  • The proposed cryptic splicing effect of K109* is computationally predicted in this study, rather than directly established by the reported cohort analysis.
    See the passage in the paper
    Three sequence-based splicing predictors all supported increased use of this cryptic donor and decreased use of the canonical exon 4 donor for the alternate allele (Figure 3C).
  • The paper is a preprint that has not been peer reviewed.
    See the passage in the paper
    This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

Why it matters

The authors argue that the findings distinguish gene-level PSEN1 loss of function from disease-causing variants that produce altered proteins, and provide a rationale for selectively suppressing mutant PSEN1 while preserving expression from the wild-type allele.

Written by AI from the paper. Passages are checked word for word against the text we analysed; the rest may contain errors.

AI-generated keywords: PSEN1 loss-of-function variants early-onset Alzheimer’s disease nonsense-mediated decay cryptic splicing

Read further

  • Comprehensive Summary
  • Layman's Summary
  • Blog article
  • Original abstract
This medRxiv preprint investigates whether genuine loss-of-function (LoF) variants in *PSEN1* can cause early-onset Alzheimer’s disease (EOAD). Pathogenic *PSEN1* variants are a major cause of familial Alzheimer’s disease, but researchers have debated whether they cause disease by reducing normal presenilin activity or through effects of an altered mutant protein. The question drew renewed attention after a father and son with EOAD were reported to carry the *PSEN1* stop-gain variant p.K109*. The authors tested whether variants expected to eliminate the protein produced from one *PSEN1* allele are associated with early-onset dementia, and re-examined p.K109*. They searched genetic and clinical data from the Alzheimer’s Disease Sequencing Project (ADSP), UK Biobank, and Vanderbilt’s BioVU. Using LOFTEE annotations, they identified high-confidence *PSEN1* LoF variants, then focused their main analysis on frameshift and stop-gain variants predicted to trigger nonsense-mediated decay (NMD), which degrades RNA containing premature stop signals. They assessed carriers’ dementia diagnoses and ages at last observation. They also examined variants that might produce truncated or alternatively spliced proteins. For p.K109*, they used three computational splice-prediction tools: AlphaGenome, SpliceAI, and Pangolin. The initial search identified 58 people carrying 42 unique variants annotated as high-confidence LoF. Some variants were splice changes or were predicted to produce truncated proteins rather than undergo NMD, so the main analysis focused on 34 frameshift or stop-gain variants expected to trigger NMD. Among 45 carriers in the three primary databases, none had a recorded diagnosis of early-onset dementia. Forty-two had no recorded dementia by their last assessment, including 17 who were observed beyond age 65. Three received dementia diagnoses after age 65, with onset considered more consistent with late-onset dementia than familial EOAD. The authors also identified fewer than 20 carriers in the All of Us program. Data policies prevented them from reporting detailed results, but the observed phenotypes were consistent with their overall findings. In contrast, six of the 13 carriers of variants that could produce altered or truncated PSEN1 proteins had early-onset mild cognitive impairment or Alzheimer’s disease. All six carried known exon 9 acceptor splice variants that cause in-frame exon skipping and produce a shortened protein. The authors interpret these cases as evidence that specific altered PSEN1 proteins can cause familial disease, rather than evidence that reducing PSEN1 dosage alone is sufficient. Reanalysis of p.K109* suggested that it may not be a straightforward null variant. All three splice-prediction methods indicated that the change could create a cryptic splice donor. Use of this site would remove five amino acids, including the premature stop, potentially allowing production of a near-full-length altered protein that escapes NMD. The authors note that earlier experimental models may not have captured this splicing effect. The authors conclude that heterozygous, gene-level *PSEN1* LoF is unlikely to be a Mendelian cause of EOAD and that *PSEN1*-related familial disease appears to require production of an altered protein. This supports investigating selective suppression of mutant *PSEN1* while preserving the normal allele, but the paper presents this as a therapeutic rationale, not evidence of treatment efficacy. Important limitations include the study’s observational design and preprint status; the findings have not been certified by peer review. Dementia status was based on available cohort assessments and health records, which may miss cases. The proposed splice effects—particularly those for p.K109*—are computational predictions that require experimental validation.

Was this synthesis useful?

Score: 0

Why do we need votes?

Votes are used to determine whether we need to re-run our summarizing tools. If the count reaches -10, our tools can be restarted.

Ask questions about this paper to our AI assistant

How to chat with this paper? Illustrated guide

You can also chat with multiple papers at once here.

Figures and plots

Ask about a figure by number or use its question button to request visual analysis. Up to two figures and four image panels per question. Values read from plots are approximate; check the original data.

AI assistant instructions?

Tools for this paper

Explain it differently 1 free rewrite left today
Save to a collection Group papers in collections to compare them, chat with them and export them. Continue with Google Create a free account Log in

Advanced tools Premium

For researchers who follow a field closely: citation alerts, journal-club slides, version comparison and Zotero export. Discover Premium

Zotero Zotero guide
Citation alerts Citation alerts guide
Weekly letter Presented in "Brain and behaviour this week" of 16 November 2026. Brain and behaviour this week: timing, repair and the origins of attraction · Receive one by email for free

Papers around this one

See them on a map

Ranked by how close their title and abstract are to this paper’s. Select some to read them together.

Closest papers

  1. Close Same question
    Presenilin 1 Mutations Converge on Aβ42/Aβ40 Elevation via Distinct Mechanisms
    Isabel Bestard-Lorigados, Keenan Sterling, Li-Wei Li et al. · 2026 · MedComm · cited 1 time · Not summarized yet

    Presenilin 1 Mutations Converge on Aβ42/Aβ40 Elevation tests how PSEN1 mutations alter protein processing, while this study asks whether PSEN1 loss-of-function variants cause EOAD.

    Abstract

    Mutations in Presenilin 1 (PS1) are the most common genetic cause of familial Alzheimer's disease (AD). However, the exact mechanisms by which these mutations alter amyloid precursor protein (APP) processing and drive amyloid β (Aβ) production remain unclear. In this study, five PS1 mutations (M146V, G384A, C410Y, Y115H, and the Notch‐sparing mutation ΔS169) were transfected into neuroblastoma cells lacking endogenous PS1/PS2 expression and human embryonic kidney cells stably overexpressing either wild‐type APP or the APP Swedish mutation. We investigated the effects of these mutations on APP processing, Aβ42/Aβ40 ratio, cytotoxicity, BACE1 expression, and PS1 endoproteolysis. Our results de…

  2. Close Broader context
    Motor manifestations in PSEN1-Related early-onset Alzheimer’s disease: a case series and systematic review
    M. Genz, Ingrid Lorena da Silva Gomes, F. Junqueira et al. · 2026 · ANPAbstract_CBN26 · Not summarized yet

    Motor manifestations in PSEN1-Related early-onset Alzheimer’s disease reviews clinical features among PSEN1-related EOAD cases, whereas this study tests whether PSEN1 null variants cause EOAD.

    Abstract

    Background: Autosomal dominant early-onset Alzheimer’s disease (EOAD) is a distinct subset of dementia, often affecting individuals during their working age. It is most frequently caused by pathogenic variants in PSEN1, accounting for the majority of genetically confirmed cases worldwide. Although typically characterized by amnestic cognitive decline, increasing evidence highlights substantial phenotypic variability among PSEN1 carriers, suggesting manifestations extend beyond the cognitive domain. Motor features, including pyramidal signs, dystonia and parkinsonism, have been reported in different stages of the disease, but are not routinely integrated into the recognized clinical spectrum.…

  3. Close Broader context
    Effect of high-risk SORL1 variants on clinical presentation in symptomatic Alzheimer’s disease
    M. D. de Waal, Janna I. R. Dijkstra, L. Vermunt et al. · 2026 · Alzheimer's Research & Therapy · Not summarized yet

    Effect of high-risk SORL1 variants on clinical presentation examines clinical outcomes in symptomatic AD carriers of another risk gene, rather than PSEN1 loss-of-function and EOAD.

    Abstract

    Rare, high-risk variants in SORL1 are established genetic contributors to Alzheimer’s disease (AD), but their impact on clinical presentation remains uncharacterized. This study investigates the effect of rare high-risk SORL1 variant ( SORL1 hi ) carriers on the clinical presentation, cognitive decline and time from diagnosis to death in symptomatic AD. Patients with a biomarker-confirmed diagnosis of symptomatic AD and available genetic information ( n = 1546, age at diagnosis 64.3 ± 7, 52% female, 69% APOE ε4 carriers) were included from the Amsterdam Dementia Cohort. SORL1 hi carriers were identified per established criteria, and wild-type (WT) carriers ( n = 1511) served as controls. Bas…

  4. Close Broader context
    Tau A152T risk variant drives distinct cell type transcriptional signatures in AD and PSP
    Meredith T. Lilley, Ying-Xue Ren, Disha M. Soni et al. · 2026 · Acta Neuropathologica · Not summarized yet

    Tau A152T risk variant drives distinct cell type transcriptional signatures studies another AD-associated genetic variant and its brain effects, not PSEN1 loss-of-function.

    Abstract

    The A152T point mutation is a rare genetic variant in the tau gene (MAPT) that is linked to risk of Alzheimer’s disease (AD) and frontotemporal dementia spectrum disorders, including progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD). To uncover mechanisms through which the A152T variant contributes to risk of tauopathy, we performed single-nucleus RNA sequencing (snRNA-seq) on postmortem brain tissue collected from A152T carriers and pathology-matched non-carriers with either AD or PSP, along with age-matched controls. These efforts revealed that despite similar cell type composition across groups, A152T carriers exhibited distinct cell type-specific transcriptional ch…

  5. Related* Broader context
    Pregnancy History and Loss of Normal PSEN2 Function Impact Female Late-Life Cognitive Processing, Inhibitory Signaling, and Gliosis
    Jen-Nie I. Fang, Leanne Lehmann, K. Prater et al. · 2026 · bioRxiv · preprint, not peer reviewed · Not summarized yet

    Pregnancy History and Loss of Normal PSEN2 Function examines PSEN2 function and cognition in female mice, while this study evaluates PSEN1 loss-of-function and EOAD.

  6. Related Same method, another problem
    Data from the ROPAD study corroborate an association between ITSN1 loss-of-function variants and Parkinson’s disease
    C. Ganoza, A. Westenberger, J. J. Paul et al. · 2026 · npj Parkinson's Disease · Not summarized yet

    Data from the ROPAD study corroborate an association between ITSN1 loss-of-function variants and Parkinson’s disease uses genetic-database analyses to assess disease association of loss-of-function variants in another gene.

    Abstract

    We read with interest the recent studies by Skuladottir and colleagues and Coganandcolleagues,whichreportanovelassociationbetweenmonoallelic loss-of-function(LoF)variantsinthe ITSN1 gene and risk for Parkinson ’ s disease (PD) 1,2 . The fi rst study observed that heterozygous LoF variants in ITSN1 confer a large risk for developing PD in analyses combining the deCODE, UK Biobank, and Accelerating Medicines Partnership Parkinson ’ s Disease (AMP-PD) datasets 1 . The second study identi fi ed fi ve PD patients from three families who harbored distinct heterozygous frameshift ITSN1 variants and reported their clinical phenotype 2 . Yet another recent study reported similar results in an analysis o…

  7. Related Broader context
    Genomic Landscape of Early-Onset and Familial Latin American Parkinson’s Patients
    E. Waldo, H. M. Chaparro-Solano, M. Isayan et al. · 2026 · medRxiv · preprint, not peer reviewed · Not summarized yet

    Genomic Landscape of Early-Onset and Familial Latin American Parkinson’s Patients screens disease-related genes for variants in early-onset and familial Parkinson’s disease.

    Abstract

    Background Parkinson's disease (PD), the most common neurodegenerative movement disorder, is commonly thought of as an aging and sporadic disease; however, 5-14% of individuals experience disease onset before the age of 50 years (early-onset PD; EOPD) and about 20% have a positive family history. In the case of both EOPD and people with a family history of PD, evidence suggests higher rates of a disease-causing genetic contribution. Objectives Our goal was to perform a variant screening of seven PD-related genes and 21 genes associated with related parkinsonian disorders to identify pathogenic/likely pathogenic variants and variants of uncertain significance within EOPD and family-history-po…

  8. Related Broader context
    Alzheimer's disease genetics converges on microglial and neuronal dysregulation.
    E. Uffelmann, D. Wightman, S. Bahrami et al. · 2026 · Nature · Not summarized yet

    Alzheimer's disease genetics converges on microglial and neuronal dysregulation analyzes genetic risk loci and biological pathways in AD, rather than PSEN1 null alleles specifically.

    Abstract

    Alzheimer's disease (AD) is the most common cause of dementia, with global case numbers projected to reach 153 million in 20501. AD is highly heritable, with twin-based heritability estimates of 60-80%2. Although 1,200 causal loci are predicted to exist for AD3, approximately 80 loci have been associated with AD in two recent studies4,5. Low single-nucleotide polymorphism (SNP)-based heritability estimates have prompted the suggestion that little further common-variant risk remains to be found6. Here we analysed data from 109,479 cases, 74,141 proxy cases, 2,131,799 controls and 499,708 proxy controls from diverse ancestries, identifying 118 loci in a multi-ancestry analysis and 9 additional…

How are these papers found?

We compare the title and abstract of this paper with those of every paper summarized on this site, and with the papers Semantic Scholar recommends, those it cites and those citing it (for a much-cited paper, its most recent citations and, from OpenAlex, the most cited ones). Among the prior work and the follow-ups, the papers that use each other in depth, a method it uses and the much-cited follow-ups come a little higher. The levels (very close, close, related) come from the similarity of the texts, not from a judgment of quality; * marks a paper without an abstract, compared by its title. The sentence under a paper is written by an AI from the two abstracts: check it in the papers. A paper marked not relevant by three readers leaves the list.

Updated on 16 November 2026.

The illustrated guide to the papers around a paper

Disclaimer: The AI-based summarization tool and virtual assistant provided on this website may not always provide accurate and complete summaries or responses. We encourage you to carefully review and evaluate the generated content to ensure its quality and relevance to your needs.