Early-Onset Autosomal Dominant Alzheimer Disease (ADAD/EOFAD): Comprehensive Research Report

1. Disease Information

Overview: Early-Onset Autosomal Dominant Alzheimer Disease (ADAD, also called Early-Onset Familial Alzheimer Disease, EOFAD) is a fully penetrant, Mendelian form of Alzheimer's disease caused by pathogenic variants in APP, PSEN1, or PSEN2. It produces the same core neuropathological and clinical phenotype as sporadic late-onset AD (amyloid-β plaques, neurofibrillary tangles, progressive dementia) but with a much earlier, highly predictable age of onset — typically before age 65, and often in the 30s–50s — and with a family history consistent with autosomal dominant transmission across generations. ADAD accounts for a small minority of all AD (~1% of total AD cases; ~5–10% of early-onset AD, EOAD), but it has been disproportionately important to AD research because carriers can be identified presymptomatically decades before expected symptom onset, enabling prospective study of the full preclinical-to-symptomatic biomarker cascade (the basis of the Dominantly Inherited Alzheimer Network, DIAN).

Key Identifiers: - MONDO: MONDO:0015140 (early-onset autosomal dominant Alzheimer disease) - Orphanet: ORPHA:1020 - OMIM (gene-specific subtypes): - AD1 — OMIM:104300 (APP, 21q21.3) - AD3 — OMIM:607822 (PSEN1, 14q24.2) - AD4 — OMIM:606889 (PSEN2, 1q42.13) - Additional related OMIM entries returned by Orphanet cross-reference include 104310, 602096, 604154, 605055 (familial early-onset AD with coexisting amyloid and prion pathology), 605526, 606187, 607116, 609636, 609790, 611073, 611152, 611154 (locus/phenotype variant entries) - ICD-11: 8A20.00 (Alzheimer disease, early onset) — ICD-10: G30.0 - MeSH: D000544 (Alzheimer Disease); no distinct MeSH term for the familial subtype specifically, indexed under "Alzheimer Disease" + genetic subheadings - HPO (disease-level phenotype set anchor): HP:0002511 (Alzheimer disease); HP:0031060 (early-onset Alzheimer disease, if present in current HPO builds)

Synonyms: Early-onset familial Alzheimer disease (EOFAD); Familial Alzheimer disease (FAD); Autosomal dominant Alzheimer disease (ADAD); Dominantly Inherited Alzheimer Disease (DIAD, used specifically by the DIAN consortium); Presenilin-related familial Alzheimer disease.

Data provenance: Nearly all mechanistic and biomarker knowledge for ADAD is aggregated disease-level knowledge synthesized from large observational cohorts (DIAN Observational Study, the Colombian PSEN1 E280A "Paisa" kindred cohort, the UK/European ADAD case series) rather than isolated EHR mining — this is a rare-disease field built on international, prospectively phenotyped, longitudinally followed multi-generational kindreds. (Orphanet: Early onset autosomal dominant Alzheimer disease; GARD: Early-onset autosomal dominant Alzheimer disease)

2. Etiology

Disease causal factors: ADAD is caused by heterozygous, fully penetrant (or near-fully penetrant) pathogenic variants in one of three genes, each converging on abnormal amyloid-β (Aβ) generation: - PSEN1 (presenilin-1, the catalytic subunit of γ-secretase) — the most common cause, accounting for ~69–70% of genetically-solved ADAD; over 300 pathogenic variants reported (missense predominant). - APP (amyloid precursor protein) — missense variants near the β- or γ-secretase cleavage sites, or whole-gene duplication (as in Down syndrome trisomy 21, which produces an ADAD-like phenotype via APP gene-dosage) — ~13–16% of cases (variants) plus ~7.5% (duplications). - PSEN2 (presenilin-2, homologous γ-secretase subunit) — rarest cause (~2%), notably enriched in the Volga German kindreds (N141I variant).

Genetic screening studies place combined APP/PSEN1/PSEN2 solve rates at roughly 60–70% of clinically defined familial (autosomal-dominant pedigree) early-onset AD (PLOS Medicine genetic screening study).

Risk factors: - Genetic: Inheriting one causal variant is both necessary and (with near-complete penetrance) sufficient — this is a monogenic disorder, not a polygenic-susceptibility one. Modifier loci shift age of onset but do not determine disease occurrence: APOE genotype (ε4 accelerates amyloid accumulation timing in some cohorts, though effects are more modest/variable than in sporadic AD — PSEN1 E280A ε4 carriers showed amyloid-PET-positivity ~3 years earlier than non-carriers); DAOA rs2391191 (A/A genotype associated with later onset in PSEN1 A431E carriers, PMC12387094); specific γ-secretase "Aβ profile" generated by a given PSEN1/PSEN2/APP variant (ratio and composition of Aβ37/38/40/42/43 species correlates linearly with age at onset across all three genes — PMC12032737, unifying "spectrum of γ-secretase dysfunction" model). - Environmental: Age is the dominant non-genetic modifier of symptom expression (variant-specific mean onset age, but individual variance of years-to-decades exists even within families carrying the identical variant). Educational attainment is protective/moderating for cognitive trajectory independent of neuropathological burden (cognitive reserve effect; Nature Communications, PMC10447560). Head trauma and cardiovascular risk factors are implicated as accelerants in sporadic AD but are less well-characterized as modifiers specifically in ADAD. - Reduced penetrance exceptions: Rare variants such as PSEN1 H163Y show documented reduced penetrance with symptom-free survival into advanced age in some carriers (22-year follow-up study, PMC5944151), indicating unmeasured genetic/epigenetic/environmental modifiers exist even in "fully penetrant" genes.

Protective factors: APOE ε2 has been associated with delayed cognitive decline/onset in some ADAD cohorts (contrasted with ε4 acceleration). No validated environmental protective factor is specific to ADAD; general cognitive-reserve/educational-attainment effects (moderating clinical expression of a fixed pathological burden) are the best-supported protective modifiers.

Gene-environment interactions: Formal GxE studies in ADAD are sparse given the small, geographically dispersed kindred sizes; the strongest documented interaction is the education × APOE interaction on cognitive trajectory (education moderates but does not eliminate the negative effect of APOE ε4 on cognition in mutation carriers).

3. Phenotypes

ADAD presents the core sporadic-AD phenotype (progressive episodic memory impairment leading to global dementia) plus a materially higher rate of "atypical"/non-amnestic and neurological accompaniments than late-onset sporadic AD, particularly with PSEN1 variants.

Core cognitive/behavioral phenotype: - Episodic memory impairment (initial and near-universal presenting symptom) — HP:0002354 (Memory impairment) / HP:0031830 (?) — best mapped to HP:0002354. - Progressive dementia — HP:0000726 (Dementia) - Behavioral/personality change (frequently an early presenting feature alongside or preceding memory loss) — HP:0000708 (Behavioral abnormality) / HP:0000722 (Irritability) / HP:0100716 (Self-injurious behavior, situational) - Language impairment / aphasia — HP:0002381 (Aphasia) - Executive dysfunction — HP:0002357 (?), best represented as HP:0002344 (Attention deficit) or a general HP:0031466 executive-function term depending on HPO build.

Neurological accompaniments enriched in ADAD (especially PSEN1): - Myoclonus — HP:0001336 — onset relatively early in disease course in ~70% of those affected; strong predictor of subsequent seizures. - Seizures — HP:0001250 — early-onset seizures in ~30% of affected individuals with myoclonus, late-onset in ~50%; individuals with myoclonus are 40% (PSEN1) to 50% (APP) more likely to develop seizures than those without (Lancet Neurology case series; ScienceDirect DIAN-OBS comparison, S1474442216302290). - Spastic paraparesis — HP:0007256 (Progressive spastic paraparesis) — variably associated with specific PSEN1 variants (e.g., the "spastic paraparesis variant" phenotype); pyramidal signs seen in ~25% of PSEN1 carriers in some series, restricted to PSEN1 (not seen with APP/PSEN2) — a genotype-phenotype correlation. - Extrapyramidal signs (parkinsonism) — HP:0002548 (Extrapyramidal sign) - Cerebellar ataxia — HP:0001251 — rare/exceptional. - Intracerebral hemorrhage — HP:0001342 — associated particularly with certain APP variants causing cerebral amyloid angiopathy (CAA), e.g., Dutch/Iowa-type variants. - Visual disturbance / posterior cortical features — HP:0000505 (Visual impairment)

Phenotype characteristics: - Age of onset: Highly variant-specific. PSEN1 mutations cause the earliest onset overall (mean often 30s–50s; documented range from as young as ~24 years to the 60s). PSEN2 and APP variants show delayed onset relative to PSEN1 — recent quantitative work estimates mean delays of ~27 years for PSEN2 and ~8 years for APP relative to PSEN1 (2025 study cited above). - Severity/progression: Uniformly progressive; disease course pattern is relentlessly progressive (not relapsing-remitting), typically over 8–10 years from symptom onset to death, though this varies by variant. - Frequency of specific neurological signs: Varies substantially by variant and gene (see above percentages); a UK case series and DIAN-OBS comparison found systematic differences between published literature phenotype summaries and directly-observed DIAN-OBS cohort data, cautioning against over-generalizing single-kindred phenotype reports (PMC12738114).

Quality of life impact: Progressive loss of independent function (activities of daily living), caregiver burden, and — because of early working-age onset — substantial socioeconomic and family-planning impact distinct from late-onset AD (loss of employment/income during peak career/parenting years; documented psychological burden and elevated suicidal ideation risk in at-risk relatives undergoing predictive testing, PMC10046467).

4. Genetic/Molecular Information

Causal genes: | Gene | HGNC | OMIM (gene) | Locus | Approx. % of solved ADAD | |---|---|---|---|---| | PSEN1 | HGNC:9508 | 104311 | 14q24.2 | ~69% | | APP | HGNC:620 | 104760 | 21q21.3 | ~13% (point variants) + ~7.5% (duplication) | | PSEN2 | HGNC:9509 | 600759 | 1q42.13 | ~2% |

Pathogenic variants: - PSEN1 — >300 reported pathogenic missense variants distributed across all 9 transmembrane domains; classified pathogenic/likely pathogenic per ACMG/AMP in ClinVar. Loss of normal γ-secretase substrate-processivity (rather than simple gain- or loss-of-function) is now understood as the shared mechanism: variants shift the ratio of Aβ species toward longer, more amyloidogenic forms (Aβ42, Aβ43) relative to Aβ40. Example founder variants: E280A (Colombian "Paisa" kindred, the largest single-family ADAD cohort in the world), A431E (Jalisco, Mexico, founder variant). - APP — Variants cluster near the β-secretase (Swedish, KM670/671NL), α-secretase (e.g., London V717I), and γ-secretase cleavage sites, shifting APP processing toward increased total Aβ production or toward the more aggregation-prone Aβ42 species; the APP locus duplication (as in trisomy 21 / Down syndrome) causes ADAD-like early-onset AD purely through increased gene dosage/protein level, without any coding variant — direct evidence that increased Aβ production alone is sufficient to cause the disease. - PSEN2 — Fewer variants known; N141I is the classic Volga German founder variant; generally later, more variable onset and comparatively reduced penetrance versus PSEN1. - Variant type spectrum: Predominantly missense for PSEN1/PSEN2; missense and structural (duplication) for APP. Frameshift/nonsense variants are rare/atypical since loss-of-function null alleles of PSEN1 generally do not cause the classical ADAD phenotype (informing a "partial loss of normal function / gain of abnormal function" model for missense variants rather than simple haploinsufficiency). - Allele frequency: Pathogenic ADAD-causing variants are essentially absent or present at extremely low frequency (rarely >1 allele) in population reference databases (gnomAD), consistent with strong negative selection against a childhood/reproductive-age-compatible but ultimately lethal, fully penetrant dominant disease. - Somatic vs. germline: Germline in essentially all classical ADAD kindreds; no established somatic-mosaicism mechanism is a primary ADAD cause (contrast with some other neurogenetic disorders), though germline mosaicism has been raised as a genetic-counseling consideration for apparent de novo cases. - Functional consequence: The unifying functional theme (PMC12032737, "spectrum of γ-secretase dysfunction") is altered γ-secretase processivity of APP substrate, producing a quantitatively predictable shift in the Aβ42(43)/Aβ40 ratio that correlates linearly with age at onset across PSEN1, PSEN2, and APP variants — a genotype→biochemistry→age-of-onset relationship that is one of the most quantitatively robust genotype-phenotype correlations in neurodegenerative disease.

Modifier genes: APOE (ε2/ε3/ε4), DAOA (rs2391191), with ongoing multiomic work (PMC11699654) identifying "common endotypes" across PSEN1/PSEN2/APP mutation carriers suggesting shared downstream modifier pathways (inflammatory/astrocytic) independent of the specific causal gene.

Epigenetic information: No established disease-defining epigenetic lesion in ADAD itself (unlike, e.g., imprinting disorders); epigenetic dysregulation (DNA methylation changes at AD-associated loci, histone modification changes) is an active downstream/consequence research area in both sporadic and familial AD but is not established as causal.

Chromosomal abnormalities: The APP locus duplication (21q21.3) is the clearest chromosomal/structural mechanism; complete trisomy 21 (Down syndrome) produces an obligate ADAD-like phenotype via 3-copy APP dosage, with virtually universal AD neuropathology by the 40s and clinical dementia in a majority of individuals with Down syndrome by their 50s–60s — this is a widely used "genetically determined AD" comparator model in the field.

5. Environmental Information

ADAD is, by definition, monogenically determined with near-complete penetrance, so environmental/lifestyle factors are best understood as modifiers of the timing and expression of an essentially inevitable disease rather than determinants of whether disease occurs. - Environmental toxin/occupational factors: No ADAD-specific toxin exposure has been established as a modifier; general AD-epidemiology environmental factors (air pollution, pesticide exposure) are studied in sporadic AD but not specifically validated in ADAD cohorts. - Lifestyle factors: Educational attainment/cognitive reserve is the best-evidenced modifier of clinical (not necessarily pathological) trajectory in ADAD mutation carriers (Nature Communications, PMC10447560). Cardiovascular risk-factor management (as in sporadic AD) is plausibly beneficial but not specifically validated as modifying ADAD's core amyloid-driven course. - Infectious agents: Not implicated as a cause of ADAD. (The broader "infectious hypothesis" of sporadic AD, e.g., HSV-1, periodontal pathogens, remains investigational and is not part of the ADAD causal model, which is fully explained by the monogenic amyloidogenic mechanism.)

6. Mechanism / Pathophysiology

Overview causal chain (the amyloid cascade hypothesis, as directly demonstrated by ADAD genetics):

Trigger (molecular scale): Pathogenic PSEN1/PSEN2 variant alters γ-secretase complex processivity, OR pathogenic APP variant/duplication alters substrate availability or cleavage-site accessibility → shift in the ratio and length distribution of Aβ peptides generated from sequential β- and γ-secretase cleavage of APP, favoring longer, more aggregation-prone species (Aβ42, Aβ43) over Aβ40.
Aβ aggregation and amyloidosis (molecular→cellular): Increased relative Aβ42/43 production → oligomerization → fibrillization → extracellular amyloid plaque deposition (diffuse and neuritic plaques) and, for certain APP variants, cerebral amyloid angiopathy (vascular Aβ deposition). This is the earliest detectable biomarker change, beginning ~20–25 years before expected symptom onset in DIAN cohort data (Bateman et al., NEJM 2012 — landmark longitudinal biomarker cascade paper; NEJM full text).
Downstream tauopathy: Amyloid pathology precipitates (via still partially defined mechanisms involving synaptic dysfunction, oxidative stress, and kinase dysregulation) hyperphosphorylation of microtubule-associated protein tau by kinases including GSK-3β; presenilins themselves (via γ-secretase-dependent and -independent roles) normally help regulate GSK-3β subcellular localization and restrain tau hyperphosphorylation, so presenilin dysfunction can directly promote tauopathy in addition to its Aβ-generating effect. Hyperphosphorylated tau detaches from microtubules, impairing axonal transport, and self-aggregates into paired helical filaments forming intraneuronal neurofibrillary tangles — occurring downstream of and temporally after amyloid changes in the DIAN cascade (tau PET/CSF changes emerge and progress after amyloid positivity).
Neuroinflammation / microglial dysfunction (cellular): Aβ plaques and early oligomers activate microglia; TREM2 (via DAP12-SYK signaling) mediates microglial phagocytic clearance of Aβ and, together with complement (C1q) and APOE, mediates aberrant microglial synaptic pruning/engulfment. Chronic activation produces a maladaptive, disease-associated microglial (DAM) state contributing to a self-perpetuating neuroinflammatory milieu rather than effective clearance.
Synaptic and neuronal injury (cellular→tissue): Soluble Aβ oligomers directly impair synaptic function and induce neuronal hyperactivity; combined amyloid/tau/neuroinflammatory burden drives synaptic loss (the pathological correlate best correlated with cognitive impairment), followed by neuronal loss.
Neurodegeneration and atrophy (tissue→organism): Progressive regional brain atrophy (starting in medial temporal lobe/hippocampus, later diffuse), declining cerebral glucose metabolism (FDG-PET hypometabolism), and clinical cognitive decline — occurring last in the DIAN cascade, ~10–20 years after the earliest amyloid biomarker changes.

Suggested GO terms (biological processes): - GO:0034205 amyloid-beta formation - GO:1902430 negative regulation of amyloid-beta formation (relevant to pathway perturbation) - GO:0007172 signal complex assembly (γ-secretase complex context) / GO:0070765 gamma-secretase complex - GO:0006338 chromatin remodeling (epigenetic downstream, if modeled) - GO:0050890 cognition (phenotypic endpoint) - GO:0001764 neuron migration / GO:0007399 nervous system development (not primary) - GO:0043523 regulation of neuron apoptotic process - GO:0045087 innate immune response (microglial activation) - GO:0006979 response to oxidative stress

Suggested CL terms (cell types): - CL:0000540 neuron (generic); CL:0000031 (neuroblast, not relevant) - CL:0000679 glutamatergic neuron (pyramidal/cortical neurons preferentially affected) - CL:0000129 microglial cell - CL:0000127 astrocyte (reactive astrogliosis) - CL:0002453 oligodendrocyte precursor cell (secondary white-matter involvement, less central)

Molecular functions/UniProt/PDB: APP (UniProt P05067); PSEN1 (UniProt P49768); PSEN2 (UniProt P49810); γ-secretase complex structures resolved by cryo-EM (PDB entries exist for human γ-secretase, e.g., 5A63, 5FN2).

Metabolic/biochemical: Downstream mitochondrial dysfunction and impaired glucose metabolism (FDG-PET hypometabolism is a core biomarker); altered lipid metabolism (cholesterol trafficking) links to APOE modifier effects.

Molecular profiling / advanced technologies: Multi-omic integration work across PSEN1/PSEN2/APP carriers has identified convergent ("common endotype") transcriptomic/proteomic signatures independent of the specific causal gene (PMC11699654), and iPSC-derived astrocyte models of PSEN1 variants show disrupted regulated intramembrane proteolysis predisposing to inflammatory phenotypes (PMC12181884) — evidence that presenilin dysfunction has γ-secretase-independent, immune-relevant cellular consequences beyond simple amyloidogenesis.

7. Anatomical Structures Affected

Organ level: Primary organ — brain (UBERON:0000955). Body system — nervous system (UBERON:0001016). Secondary/complication involvement: cerebral vasculature (in APP-variant-associated cerebral amyloid angiopathy, risk of intracerebral hemorrhage); spinal cord/corticospinal tract involvement clinically manifesting as spastic paraparesis in specific PSEN1 variants.

Tissue/cell level: - Cerebral cortex (UBERON:0000956), especially medial temporal lobe / hippocampus (UBERON:0002421) — earliest and most severe atrophy. - Entorhinal cortex (UBERON:0002728) — early tau pathology site (Braak staging origin). - Precuneus/posterior cingulate — early amyloid-PET/FDG-PET signal. - Cell populations: large glutamatergic pyramidal neurons (CL:0000679) selectively vulnerable; microglia (CL:0000129) and astrocytes (CL:0000127) reactively involved; cerebrovascular smooth muscle/endothelium involved in CAA.

Subcellular level: Extracellular amyloid plaques (GO:0097418 neurofibrillary tangle for the intracellular counterpart); GO:0005789 endoplasmic reticulum membrane and GO:0031090 organelle membrane (site of γ-secretase activity — PSEN1/2 are ER/Golgi/plasma-membrane transmembrane proteins); GO:0005739 mitochondrion (downstream dysfunction); GO:0045202 synapse (site of Aβ oligomer toxicity and microglial synaptic pruning).

Localization: Bilateral, generally symmetric involvement (contrast with focal neurodegenerative syndromes); progresses from medial temporal to diffuse neocortical involvement per Braak/Thal staging schemes, mirrored in amyloid-PET (Thal phases) and tau-PET (Braak-like patterns) in DIAN cohort imaging.

8. Temporal Development

Onset: Adult onset, defined as pre-senile (<65 years) and typically much earlier — commonly 30s–50s, occasionally younger depending on variant (as young as mid-20s reported for some aggressive PSEN1 variants). Onset pattern is insidious, not acute — a subtle prodromal decline over months to a few years before diagnosis.

Progression (the DIAN biomarker cascade — a hallmark, extensively validated natural-history finding): 1. Aβ42 CSF changes and amyloid-PET positivity — begin ~20–25 years before expected symptom onset (estimated from parental age at onset within a kindred). 2. Cerebral glucose hypometabolism (FDG-PET) — begins ~10–15 years before onset (~7–10 years after earliest amyloid change). 3. Hippocampal/structural atrophy and initial subtle cognitive decline — begin roughly 10–15 years, becoming clearly measurable ~5 years before clinical onset. 4. Clinical Dementia Rating (CDR) transition to symptomatic/impaired — the defined "onset." 5. Progressive dementia over subsequent years to death (variable, commonly 8–10 years post-diagnosis, similar to sporadic AD trajectory once symptomatic).

(Bateman et al. 2012 NEJM, "Clinical and Biomarker Changes in Dominantly Inherited Alzheimer's Disease"; subsequent DIAN longitudinal work confirming and refining timing, e.g., Neurology 2018 "Longitudinal cognitive and biomarker changes in dominantly inherited Alzheimer disease," and the 2025 "15 years of longitudinal…measures in DIAN" (npj Dementia) update.)

Disease stages: Preclinical (biomarker-positive, cognitively normal) → prodromal/MCI (mild cognitive impairment due to AD) → mild dementia → moderate dementia → severe/end-stage dementia — the CDR and DIAN "Estimated Years to Symptom Onset" (EYO) framework are the field-standard staging tools, distinct from generic AJCC-style staging (not applicable to a non-oncologic disease).

Progression rate/course: Once symptomatic, relentlessly progressive, non-remitting; among the more rapid dementia trajectories compared with typical sporadic late-onset AD, partly reflecting younger baseline health and higher pathological burden at a given clinical stage in some series, though this is debated.

Patterns: No spontaneous or treatment-induced remission is described; anti-amyloid immunotherapy trials (below) aim to slow but not yet reverse progression. The presymptomatic ~20-year biomarker-positive window is the field's key "critical period" for intervention — the rationale for the DIAN-TU prevention trials targeting mutation carriers before symptom onset.

9. Inheritance and Population

Epidemiology: - EOAD overall (<65y onset, all causes) prevalence ≈ 41.2 per 100,000 persons at risk in population-based estimates; incidence in the 45–64 age band ≈ 6.3/100,000/year, prevalence ≈ 24.2/100,000, rising steeply approaching age 65. - ADAD specifically (autosomal-dominant EOAD, ADEOAD): population prevalence ≈ 5.3 per 100,000 persons at risk (same population-based study cited above). - ADAD is estimated to account for ~1% of all AD cases overall, and ~5–10% of early-onset AD cases are attributable to dominantly inherited mutations, with the remainder of EOAD considered sporadic/complex/oligogenic. - Among familial EOAD (patients with ≥1 affected first-degree relative, 35–60% of EOAD), only 10–15% show a clear autosomal-dominant transmission pattern solvable by APP/PSEN1/PSEN2 testing.

Inheritance pattern: Autosomal dominant (HP:0000006). Penetrance is near-complete/"fully penetrant" for the great majority of PSEN1 variants and the APP variants/duplication; documented exceptions with reduced/age-dependent penetrance exist (e.g., PSEN1 H163Y). PSEN2 variants (e.g., N141I) show somewhat more variable, occasionally incomplete penetrance and later/more variable onset than PSEN1.

Expressivity: Variable — age of onset varies not only between genes and between different variants within a gene, but also between individuals carrying the identical variant within the same family, implicating modifier genes (APOE, DAOA) and other unmeasured factors.

Genetic anticipation: Not a defined feature of ADAD (unlike repeat-expansion disorders); no systematic earlier-onset-in-successive-generations pattern is established.

Germline mosaicism: A recognized genetic-counseling consideration for apparent de novo PSEN1/APP cases (parental germline mosaicism can produce unaffected parents with an affected child and residual recurrence risk for future pregnancies), though formally documented cases specific to ADAD genes are limited in the literature relative to other dominant disorders.

Founder effects: Well documented — PSEN1 E280A in the Colombian "Paisa" kindred (~6,000-member, ~1,200 carrier-lineage pedigree, the world's largest single ADAD kindred, central to the Colombia-API prevention trial); PSEN1 A431E in the Jalisco, Mexico population; PSEN2 N141I in Volga German-descended families in the United States.

Consanguinity: Not a relevant risk factor for this autosomal dominant disorder (in contrast to recessive conditions) — a single inherited or de novo copy is sufficient.

Carrier frequency: Population carrier frequency of any single pathogenic ADAD variant is extremely low (rare/private variants, or locally elevated only within founder-effect populations/kindreds); no meaningful general-population carrier screening frequency exists comparable to recessive-disease carrier screening.

Population demographics: No strong evidence for differential susceptibility by ethnicity beyond the founder-population enrichments noted above (Colombian, Mexican, Volga German lineages) — these reflect genealogical founder effects, not differential biological susceptibility. Sex ratio: approximately equal (autosomal, not sex-linked), consistent with Mendelian expectation, though APOE-related modifier effects (and sporadic-AD female-predominance patterns) may subtly influence age-of-onset/progression statistics by sex in some analyses. Age distribution of affected individuals is, by definition, the defining "early-onset" feature (predominantly 30s–50s at symptom onset, contrasted with the ≥65y peak of sporadic AD).

10. Diagnostics

Clinical tests: - Biomarkers (CSF): CSF Aβ42/Aβ40 ratio (decreased) and phosphorylated tau (p-tau181, increasingly p-tau217, p-tau231) are core, guideline-incorporated AD biomarkers; CSF p-tau217/Aβ42 ratio and p-tau217 phosphorylation occupancy show improved performance over p-tau181 for detecting amyloid and tau pathology (Nature Aging, PMC study cited above). - Blood-based biomarkers: Plasma p-tau217 and the p-tau217/Aβ1-42 ratio now show AUC 0.94–0.97 for detecting amyloid pathology against CSF/PET reference standards, a major recent (2024–2025) advance enabling non-invasive, scalable case detection and monitoring — highly relevant for at-risk ADAD family members considering testing and for trial screening/enrollment. - Imaging: Amyloid-PET (e.g., florbetapir, Pittsburgh compound B) — earliest positive biomarker in the DIAN cascade; tau-PET (flortaucipir) — shows variant-specific signatures, e.g., a characterized flortaucipir signature specific to PSEN1 A431E carriers (PMC12740027); structural MRI for hippocampal/cortical atrophy; FDG-PET for regional hypometabolism. - Neurophysiology: EEG may show abnormalities correlating with myoclonus/seizure activity, which are enriched in ADAD relative to sporadic AD. - Neuropathology (postmortem/rare biopsy): Amyloid plaques (neuritic and diffuse), neurofibrillary tangles (Braak staging), cerebral amyloid angiopathy (especially with certain APP variants); the OMIM:605055 entry specifically flags documented coexisting amyloid and prion pathology in some familial early-onset AD cases, an important differential/comorbidity note.

Genetic testing: - Recommended approach: targeted sequencing (single-gene or small panel) of APP, PSEN1, PSEN2 in individuals with early-onset dementia (<65y) and/or a family history consistent with autosomal dominant inheritance; panel/gene-panel testing is standard of care (per GeneReviews/ClinGen-informed protocols) rather than routine WGS/WES as first-line, though WES/WGS is used when panel testing is uninformative or the phenotype is atypical. - Predictive (presymptomatic) testing in at-risk, asymptomatic relatives follows Huntington-disease-style genetic counseling protocols (given the comparably severe, currently incurable, fully penetrant nature of the result) — pre- and post-test counseling, psychological support, and consideration of implications for insurance/employment/family planning are essential; documented elevated suicidal-ideation risk in this population underscores the need for structured counseling (PMC10046467). - Chromosomal microarray/karyotyping is relevant specifically to detect APP locus duplication or trisomy 21 as a cause. - Not relevant: mitochondrial DNA testing, repeat-expansion testing (ADAD is not a repeat-expansion disorder).

Omics-based diagnostics: Not yet standard clinical practice for ADAD; used in research contexts (multi-omic endotyping, PMC11699654) but blood p-tau217 is the closest omics-adjacent tool nearing clinical/trial-screening utility.

Clinical criteria: Standard AD clinical diagnostic frameworks (NIA-AA criteria, now biomarker-integrated) apply; DSM-5 criteria for Major/Mild Neurocognitive Disorder due to Alzheimer's Disease apply clinically. Differential diagnosis must consider: frontotemporal dementia (especially when behavioral/language-predominant presentation), other early-onset dementia genes (e.g., MAPT, GRN, C9orf72 for FTD spectrum; prion disease given the amyloid/prion overlap noted in OMIM:605055), and reversible causes of cognitive decline in a young patient (metabolic, autoimmune, structural).

Screening: Cascade genetic testing in at-risk relatives once a family proband variant is identified is the standard screening approach (not population newborn screening, given adult onset). Preimplantation genetic diagnosis (PGD/PGT-M) has been successfully used to select unaffected embryos for at-risk couples (documented case for an APP V717L family resulting in birth of an unaffected child, PMID:11866650).

11. Outcome/Prognosis

Survival/mortality: ADAD is ultimately fatal; life expectancy after symptom onset is broadly comparable to or somewhat shorter than sporadic AD, commonly on the order of 8–10 years from diagnosis to death, though this varies by causal variant and access to supportive care. No disease-modifying therapy has yet been shown to alter mortality.

Morbidity/function: Progressive functional decline through the classic AD trajectory — loss of instrumental then basic activities of daily living, eventual total care dependency. Given the working-age onset, morbidity burden (loss of employment, caregiving strain on a typically younger family unit with dependent children) is proportionally more disruptive than in late-onset sporadic AD.

Quality of life: Progressive decline across cognitive, functional, and psychological domains for both patients and caregivers; documented elevated psychological distress (including suicidal ideation) in unaffected at-risk relatives navigating predictive testing decisions is itself a distinct QoL/mental-health outcome domain specific to this genetically-predictable disease.

Complications: Seizures and myoclonus (notably more frequent than sporadic AD, particularly with PSEN1); intracerebral hemorrhage in CAA-associated APP variants; aspiration pneumonia and other end-stage-dementia complications as the ultimate proximate causes of death, as in sporadic AD.

Prognostic factors: Causal gene and specific variant (via its Aβ-species "biochemical signature") is the single strongest predictor of age at onset and, to a lesser extent, of rate of progression and accompanying neurological features (myoclonus/seizures/spasticity cluster with PSEN1). APOE genotype and, more speculatively, DAOA genotype and educational attainment act as secondary modifiers of onset timing and/or cognitive trajectory. Biomarker trajectory (rate of amyloid accumulation, tau-PET spread) is an active area of prognostic-biomarker research within DIAN and the newer anti-tau trials (e.g., etalanetug/E2814 targeting tau spread).

12. Treatment

Pharmacotherapy (symptomatic, shared with sporadic AD): - Cholinesterase inhibitors (donepezil, rivastigmine, galantamine) — MAXO term: pharmacotherapy generically (NCIT:C15986); symptomatic cognitive treatment, not disease-modifying. - NMDA receptor antagonist (memantine) — symptomatic, moderate-to-severe stages. - Symptomatic management of myoclonus/seizures (anti-seizure medications, e.g., levetiracetam) given their elevated frequency in ADAD.

Advanced/disease-modifying therapeutics (the major current research/clinical frontier, largely trialed specifically in ADAD via DIAN-TU because of presymptomatic identifiability): - Anti-amyloid monoclonal antibodies: - Gantenerumab and solanezumab — tested in the DIAN-TU-001 platform trial (2012–2019 double-blind phase): did not meet the primary endpoint of slowing cognitive decline across combined symptomatic/asymptomatic cohorts, but gantenerumab produced significant, dose-dependent amyloid plaque reduction and reduced CSF tau/synaptic-degeneration/neuroinflammation markers, especially in asymptomatic carriers (Nature Medicine 2021, JAMA Neurology downstream-biomarker analysis). A subsequent open-label extension (2020–2023, up to 10 years cumulative treatment in some asymptomatic carriers) examined whether sustained, high-dose treatment yields clinical benefit (Lancet Neurology 2025); amyloid-related imaging abnormalities-edema (ARIA-E) occurred in 19.2% of gantenerumab-treated participants vs. 2.5% placebo, an important safety signal shared across the anti-amyloid antibody class. - Lecanemab — now the backbone anti-amyloid therapy in the ongoing DIAN-TU Tau NexGen study (initiated January 2022), combined with the investigational anti-tau antibody etalanetug (E2814) to test whether reducing pathological tau spread adds benefit atop amyloid removal — trial evaluates outcomes over 208 weeks; interim baseline-characteristics/6-month safety data were presented in 2025. - Donanemab — approved and studied extensively in sporadic AD (TRAILBLAZER-ALZ program) but not yet reported with ADAD-specific dominantly-inherited-cohort trial data in the current literature search. - Anti-tau therapy: Etalanetug (E2814), an anti-tau monoclonal antibody targeting extracellular tau spread, is being trialed specifically in DIAD mutation carriers as an adjunct to lecanemab (DIAN-TU NexGen) — this represents the field's leading edge in testing combination amyloid+tau disease-modification in a genetically-defined, presymptomatic population. - Gene-directed/other advanced modalities: No approved gene therapy, RNA-based therapy (ASO/siRNA), or cell therapy for ADAD specifically exists as of this report; given the well-defined single-gene cause, ADAD is a plausible future candidate for allele-specific ASO knockdown approaches (as pursued in other dominant neurodegenerative diseases), but no ADAD-specific program has reached the clinical literature reviewed here.

Surgical/interventional: Not applicable as a primary treatment modality for ADAD itself.

Supportive/rehabilitative: Multidisciplinary dementia supportive care (physical/occupational/speech therapy as needed for functional decline), caregiver support programs, and psychiatric/behavioral symptom management (MAXO:0000950 supportive care; MAXO:0000011 physical therapy where applicable).

Experimental/clinical trials (selected, with identifiers where available): - DIAN-TU Platform Trial extensions — NCT06424236 ("Dominantly Inherited Alzheimer Network Trial: An Opportunity to Prevent Dementia") - DIAN Observational Study — NCT00869817 - Colombia-API (Alzheimer's Prevention Initiative) autosomal-dominant AD trial in the PSEN1 E280A kindred (crenezumab, subsequently discontinued as primary endpoint not met, though biomarker signals were reported)

Treatment outcomes/adverse events: ARIA-E/ARIA-H (amyloid-related imaging abnormalities — edema/hemorrhage) is the dominant class-wide safety concern for anti-amyloid antibodies, with meaningfully elevated incidence versus placebo across trials; monitoring via serial MRI is now standard trial (and increasingly clinical) protocol. To date, no anti-amyloid or anti-tau agent has demonstrated a clearly established clinical cognitive benefit specific to the ADAD/DIAD population, despite robust biomarker (amyloid/tau/neurodegeneration marker) modification — a key ongoing translational gap actively being addressed by the tau-directed combination trials.

Treatment strategy/personalized medicine: Presymptomatic/prevention-oriented dosing strategy (treating carriers years before expected onset, leveraging the known ~20-year presymptomatic biomarker window) is the field's central strategic hypothesis, distinguishing ADAD trial design fundamentally from sporadic late-onset AD trials, which necessarily enroll after some degree of established pathology/symptoms.

13. Prevention

Primary prevention: Not currently achievable pharmacologically (no proven method to prevent disease onset in a confirmed mutation carrier), but is the explicit long-term goal of the presymptomatic DIAN-TU/NexGen and historical API trials — administering anti-amyloid (and now anti-tau) therapy during the decades-long presymptomatic biomarker-positive window in the hope of preventing or substantially delaying clinical onset.

Secondary prevention: Early biomarker detection (CSF/plasma p-tau217, amyloid-PET) in known mutation carriers enables early trial enrollment and monitoring, functioning as the operational "secondary prevention" framework in this genetically-predictable disease, though it does not yet translate into a proven clinical intervention.

Reproductive/genetic prevention: - Preimplantation genetic diagnosis (PGD/PGT-M) — successfully used for ADAD-causing APP variants to select unaffected embryos, resulting in births of unaffected children (PMID:11866650). - Prenatal testing is possible for known-carrier pregnancies (analogous framework to other dominant, adult-onset, fully penetrant disorders such as Huntington disease), though ethically and psychologically complex given adult onset. - Cascade/predictive genetic testing in at-risk relatives, paired with formal genetic counseling, is the standard risk-stratification and family-planning-guidance approach (NSGC/ACMG-informed protocols, modeled closely on Huntington disease predictive-testing guidelines given comparable ethical weight).

Behavioral/lifestyle interventions: General brain-health/cardiovascular-risk-reduction measures (as recommended for sporadic AD risk reduction) are reasonable adjuncts but are not established to meaningfully alter the essentially deterministic course in confirmed mutation carriers.

Public health/environmental interventions: Not a primary prevention lever for this monogenic disorder; population-level public-health prevention strategies relevant to sporadic AD (cardiovascular risk factor control, education, physical activity) do not have established ADAD-specific efficacy data.

Prophylaxis: The presymptomatic anti-amyloid/anti-tau trial paradigm described above is, in effect, the field's prophylaxis research program, though no agent is yet validated/approved for this indication.

14. Other Species / Natural Disease

Taxonomy: No naturally occurring, spontaneous orthologous disease with an equivalent PSEN1/PSEN2/APP-driven fully penetrant dominant Alzheimer-like dementia has been well documented in non-human species under natural conditions. Aged dogs (Canis lupus familiaris, NCBITaxon:9615) and aged non-human primates can develop age-related, sporadic-AD-like cognitive dysfunction syndrome with some amyloid pathology, but this is an aging-associated phenomenon, not a monogenic dominant disease analogous to human ADAD.

Orthologous genes: App, Psen1, Psen2 are conserved across mammals (mouse orthologs: App MGI:88059, Psen1 MGI:104779, Psen2 MGI:108086); rodents do not naturally develop human-like amyloid pathology even when carrying these orthologs unmodified, which is precisely why transgenic/knock-in humanized models were developed (see Section 15).

Veterinary relevance: Canine cognitive dysfunction syndrome (CCDS) is studied as a naturally occurring, sporadic aging model with some amyloid deposition, of comparative interest but not a direct ADAD model (OMIA does not list a canonical monogenic canine ADAD ortholog disease).

Comparative biology: The core γ-secretase/APP processing pathway is deeply evolutionarily conserved, which is precisely why humanized knock-in and transgenic rodent models (below) can be constructed to express human pathogenic variants and partially recapitulate amyloid pathology, despite the absence of natural disease in these species.

Zoonotic potential: Not applicable — ADAD is a purely genetic, non-transmissible disease.

15. Model Organisms

Rodent transgenic overexpression models: Classic APP/PS1 double-transgenic mice (overexpressing mutant human APP and PSEN1 from non-physiological promoters) robustly and reproducibly develop amyloid plaque pathology and are the most widely used preclinical AD model, but suffer from significant limitations: APP overexpression from an early age (unlike the gradual human increase), non-physiological accumulation of APP fragments (CTFs, AICD) not seen in human AD brain that may independently cause artifactual endosomal/transcriptional abnormalities, and — most importantly for translational validity — frequent failure to fully recapitulate human-comparable tau pathology, neurodegeneration, and behavioral/cognitive decline patterns at physiologically relevant timescales.

Knock-in models (designed to address overexpression artifacts): App^NL-G-F^ and related humanized App knock-in lines (Saito/Saido-style knock-ins) express mutant human Aβ sequence at physiological levels under the endogenous promoter, better modeling the gradual human amyloid accumulation trajectory and revealing profound microglial metabolic dysregulation (Molecular Neurodegeneration, PMC9188195); however, recent work (bioRxiv 2024) reports that App-KI mice do not display the hallmark age-dependent cognitive decline seen in overexpression models or in human disease, and PSEN1 knock-in alone (without an accompanying APP mutation) is generally insufficient to induce Aβ pathology at all, indicating that a combined, humanized APP+PSEN1(or PSEN2) knock-in approach is needed to approach physiological relevance, and even then, cognitive-behavioral phenotype fidelity to human ADAD remains incompletely validated (an appropriate candidate for a HUMAN_MODEL_MISMATCH framing in mechanistic curation, per this project's conventions).

Immunodeficient humanized models: Combined APP/PSEN1 knock-in immunodeficient mice have been reported to exhibit intraneuronal Aβ pathology, microgliosis, and extensive neuronal loss (PMC11975631) — an attempt to better model the human innate-immune/microglial contribution by permitting engraftment of human microglia or other human cellular elements.

Other model systems: - Drosophila melanogaster presenilin models (used to dissect γ-secretase-independent presenilin toxicity mechanisms, e.g., ALZFORUM-reported "protease or not" eye-phenotype screens). - Human iPSC-derived neuronal and astrocyte models carrying patient-specific PSEN1/PSEN2/APP variants — increasingly central for mechanistic study (e.g., iPSC-astrocyte models showing PSEN1-variant-driven inflammatory predisposition via disrupted regulated intramembrane proteolysis, PMC12181884) and for modeling human-specific cellular contexts not well captured by rodents. - Organoid models — cerebral/brain organoids from ADAD patient iPSCs are an emerging platform for studying 3D tissue-context amyloid/tau pathology, though standardization and long-term maturation remain limitations.

Applications: Mouse and iPSC models are used to dissect the amyloid-generation mechanism, test candidate anti-amyloid/anti-tau therapeutics preclinically (informing the DIAN-TU trial pipeline), and study microglial/neuroinflammatory contributions; they are less reliable for modeling the full human cognitive-behavioral phenotype or the neurological accompaniments (myoclonus, seizures, spastic paraparesis) seen in human PSEN1 carriers.

Model limitations (summary): The recurring, well-documented gap across essentially all current models is a mismatch between robust amyloid (and sometimes neuroinflammatory) pathology on one hand, and inconsistent or absent age-dependent cognitive/behavioral decline and incomplete tau pathology/neurodegeneration on the other — meaning translational claims from rodent efficacy data to expected human clinical benefit should be treated cautiously, consistent with the repeated failure of preclinically-promising anti-amyloid agents to show unambiguous cognitive benefit in DIAN-TU human trials despite clear target engagement.

Resources: MGI (Mouse Genome Informatics) for App/Psen1/Psen2 alleles; Alzforum "Research Models" database (the field-standard curated registry of AD mouse/model lines, including full genotype/phenotype-recapitulation summaries); JAX (Jackson Laboratory) repository for physical strain distribution.

Summary Table: Key Ontology Term Suggestions

Table (click to expand)

Category	Term	ID
Disease	Early-onset autosomal dominant Alzheimer disease	MONDO:0015140 / ORPHA:1020
Gene	PSEN1	HGNC:9508
Gene	PSEN2	HGNC:9509
Gene	APP	HGNC:620
Phenotype	Dementia	HP:0000726
Phenotype	Memory impairment	HP:0002354
Phenotype	Myoclonus	HP:0001336
Phenotype	Seizures	HP:0001250
Phenotype	Progressive spastic paraparesis	HP:0007256
Phenotype	Extrapyramidal sign	HP:0002548
Phenotype	Intracerebral hemorrhage	HP:0001342
Inheritance	Autosomal dominant inheritance	HP:0000006
GO (BP)	Amyloid-beta formation	GO:0034205
GO (BP)	Regulation of amyloid-beta formation	GO:1902430
GO (CC)	Gamma-secretase complex	GO:0070765
CL	Glutamatergic neuron	CL:0000679
CL	Microglial cell	CL:0000129
CL	Astrocyte	CL:0000127
UBERON	Hippocampus	UBERON:0002421
UBERON	Cerebral cortex	UBERON:0000956
MAXO	Pharmacotherapy	NCIT:C15986
Drug class	Monoclonal antibody	NCIT:C20401