¶ AD/PD 2026: Precision Medicine and Genetic Stratification
Dates: March 17-21, 2026
Location: Copenhagen, Denmark
Organizer: Kenes Group
Precision medicine — tailoring diagnosis, prevention, and treatment to individual genetic profiles — has transitioned from theoretical aspiration to clinical reality at AD/PD 2026. The conference showcased how genetic stratification is reshaping clinical trial design, enabling targeted therapies, and guiding diagnostic pathways for both Alzheimer's disease and Parkinson's disease[@bellenguez2022][@correia2023].
The field has matured beyond single-gene associations toward polygenic risk scores that integrate dozens to hundreds of genetic variants, and beyond pharmacogenetics toward mechanistic stratification that links genetic risk to specific biological pathways. For AD, APOE4 remains the central risk factor with new understanding of its mechanistic role in tau pathology. For PD, LRRK2 and GBA represent the most actionable genetic targets, with multiple disease-modifying therapies in development specifically for these populations.
One of the most significant conceptual advances presented at AD/PD 2026 was the reframing of APOE4 homozygosity not merely as a risk factor but as a distinct genetic form of Alzheimer's disease[@chen2023]. Key evidence supporting this reclassification:
- Penetrance: Nearly 100% of APOE4/4 homozygotes develop AD pathology by age 85, approaching monogenic disease penetrance
- Pathological mechanisms: Distinct from amyloid-driven AD, with APOE4/4 showing accelerated tau pathology independent of amyloid burden
- Treatment response: Differential response to anti-amyloid therapies (lower lecanemab efficacy, higher ARIA risk in APOE4/4)
- Biomarker trajectory: Accelerated amyloid accumulation, earlier tau PET positivity, more rapid neurodegeneration
Mechanistic understanding:
APOE4 drives AD risk through multiple pathways beyond amyloid[@lester2024]:
- Impaired amyloid clearance: APOE4 binds amyloid with reduced efficiency compared to APOE3, impairing microglial clearance and perivascular drainage
- Tau vulnerability: APOE4 directly enhances neuronal susceptibility to tau-mediated toxicity through dysregulated lipid metabolism
- Synaptic dysfunction: APOE4 fragment accumulation disrupts mitochondrial function and synaptic homeostasis
- Vascular contributions: APOE4 impairs blood-brain barrier integrity and cerebral blood flow
- TREM2 interaction: APOE4 carriers show reduced TREM2 expression, compounding microglial dysfunction
¶ APOE Allele Distribution and AD Risk
| Genotype |
Relative Risk vs. E3/E3 |
Lifetime Risk (Women) |
Lifetime Risk (Men) |
| E2/E2 |
0.4x (protective) |
~5% |
~3% |
| E2/E3 |
0.6x (protective) |
~10% |
~7% |
| E3/E3 |
Reference (1x) |
~15% |
~10% |
| E3/E4 |
2-3x elevated |
~25-30% |
~15-20% |
| E4/E4 |
12-15x elevated |
~55-60% |
~35-40% |
The dose-dependent effect of APOE4 reflects its semi-dominant inheritance pattern, with heterozygotes showing intermediate risk[@genin2011].
APOE genotyping in clinical practice:
- Diagnostic utility: Supports AD probability assessment in symptomatic patients
- Risk communication: Enables discussion of lifetime risk, though requires careful counseling
- Treatment selection: Critical for anti-amyloid therapy decision-making (ARIA risk stratification)
- Clinical trial stratification: Essential for trial design, particularly for prevention studies
APOE-guided anti-amyloid therapy decisions:
- Lecanemab: APOE4/4 carriers show 30-40% less amyloid reduction vs. non-carriers
- ARIA risk: APOE4/4 increases ARIA-E incidence 3-4 fold, ARIA-H 2-fold compared to E3/E3
- Dosing considerations: Some protocols recommend reduced dosing for E4 homozygotes
- Monitoring: More frequent MRI monitoring warranted in E4/4 patients
¶ Development and Validation
Polygenic risk scores (PRS) aggregate the effects of hundreds to thousands of genetic variants associated with AD risk, providing continuous risk stratification that complements single-gene testing[@truchardet2023].
AD PRS construction:
- GWAS variants: 100-500+ variants from large-scale AD GWAS (predominantly European-ancestry)
- Effect size weighting: Each variant weighted by its association strength (beta coefficient)
- Ancestral portability: PRS developed in one population requires recalibration for others
- Performance: PRS explains 7-15% of phenotypic variance, with top quintile showing 3-4x increased risk vs. bottom quintile
AD/PD 2026 PRS findings:
-
Multi-ancestry PRS validation:
- Trans-ancestry PRS developed using African, Asian, and Hispanic ancestry cohorts
- Performance maintained across populations after recalibration
- Critical need for continued expansion of non-European GWAS samples
-
PRS combined with biomarkers:
- PRS in combination with blood biomarkers (p-tau217) achieved AUC >0.95 for AD
- PRS provides independent information from fluid biomarkers
- Combined model enables risk prediction 10-15 years before symptoms
-
PRS for clinical trial enrichment:
- PRS-enriched prevention trials demonstrate feasibility
- Top PRS quintile allows smaller sample sizes for equivalent power
- Ongoing debate about ethical implications of genetic enrichment
- Ancestry bias: Most PRS derived from European-ancestry cohorts, limiting utility in non-European populations
- Risk communication: Translating probabilistic risk scores into actionable clinical recommendations remains challenging
- Regulatory framework: No established regulatory pathway for PRS-based clinical decision support
- Clinical utility evidence: Demonstration of improved outcomes from PRS-guided care is still accumulating[@jensen2024]
¶ Monogenic and Early-Onset AD
¶ APP, PSEN1, and PSEN2 Mutations
Early-onset autosomal dominant Alzheimer's disease (ADAD) is caused by mutations in APP, PSEN1, and PSEN2. These rare variants provide mechanistic insight applicable to sporadic AD.
PSEN1 mutations (most common):
- Over 300 pathogenic variants identified
- Age of onset typically 30-60 years
- Mechanisms: increased Aβ42 production, altered Aβ42/Aβ40 ratio
- Genotype-phenotype correlations being refined
APP duplications:
- Phenocopy of Down syndrome-associated AD (APP triplication)
- Typical onset 40-55 years
- Associated with cerebral amyloid angiopathy
PSEN2 mutations:
- Rarer than PSEN1
- Later onset than PSEN1 (typically 50-70 years)
- Some variants show reduced penetrance
AD/PD 2026 findings:
- PSEN1 mutation carriers show highly predictable amyloid accumulation rates
- Aβ42/40 ratio in CSF may help differentiate PSEN1 from PSEN2
- Treatment responses in ADAD cohorts inform sporadic AD treatment development
- DIAN-TU and other prevention trials continue to refine intervention windows[@simons2024]
¶ ABCA7 and Other Moderate-Risk Genes
Beyond APOE, several genes confer moderate AD risk (odds ratios 1.2-2.0):
| Gene |
Function |
Risk Effect |
Mechanism |
| ABCA7 |
Lipid transport |
OR ~1.2-1.3 |
Amyloid processing, phagocytosis |
| CLU (Apolipoprotein J) |
Complement regulation |
OR ~1.1-1.2 |
Amyloid clearance, inflammation |
| PICALM |
Synaptic vesicle trafficking |
OR ~1.1 |
Endocytosis, synaptic function |
| BIN1 |
Membrane trafficking |
OR ~1.1-1.2 |
Tau pathophysiology |
| SORL1 |
Retromer function |
OR ~1.2-1.5 |
APP trafficking, amyloid production |
| TREM2 |
Microglial activation |
OR ~2-3 (loss-of-function) |
Amyloid clearance, inflammation[@zhan2023] |
TREM2 variants:
Loss-of-function TREM2 variants (including R47H) confer approximately 3-fold increased AD risk[@zhan2023]. The variant impairs microglial transition to disease-associated states, reducing amyloid clearance capacity. This has direct therapeutic implications — TREM2-activating antibodies (e.g., AL002) are in clinical development specifically for TREM2 variant carriers.
¶ G2019S and Other Pathogenic Variants
LRRK2 mutations are the most common genetic cause of Parkinson's disease, accounting for 1-5% of sporadic PD and up to 40% of familial PD in certain populations (particularly those of North African Arab and Ashkenazi Jewish ancestry)[@correia2023].
G2019S (Gly2019Ser):
- Most common LRRK2 variant (5-6% of familial PD)
- Located in the kinase domain — results in increased kinase activity
- Autosomal dominant inheritance with age-dependent penetrance (~30% by age 80)
- Phenotype generally resembles idiopathic PD with typical motor features
- Good response to dopaminergic therapy
- Reduced non-motor symptoms compared to idiopathic PD in some cohorts
Other LRRK2 variants:
- N1437H, R1441C/G/H, Y1699C — all in ROC domain
- Variable penetrance and phenotypic expression
- Similar kinase hyperactivity mechanism
LRRK2 has emerged as one of the most tractable therapeutic targets in PD, with multiple disease-modifying therapies in development:
Kinase inhibitors:
- DNL151 / BIIB094 (Denali): LRRK2 inhibitor, Phase 2 completed — demonstrated target engagement and good safety profile
- DNL343 (Denali): CNS-penetrant LRRK2 inhibitor, earlier-stage development
- JM10 (Neuroscience): Brain-penetrant LRRK2 inhibitor in early clinical testing
Mechanism of action:
LRRK2 kinase inhibitors reduce the pathological hyperactivity of mutant LRRK2, normalizing:
- Lysosomal function (LAMP2A, GBA interaction)
- Primary cilia function (implicated in dopamine neuron survival)
- Synaptic vesicle trafficking
- Neuroinflammation via microglial effects
AD/PD 2026 LRRK2 therapeutic updates:
- DNL151 showed sustained target engagement at doses used in Phase 2
- Biomarker evidence of reduced neurodegeneration markers in LRRK2 inhibitor-treated patients
- Next-generation inhibitors with improved brain penetration advancing
- Patient selection strategies: LRRK2 variant carriers show greatest treatment response
flowchart TD
A["LRRK2 G2019S<br/>Mutation"] --> B["Kinase Domain<br/>Hyperactivity"]
B --> C1["Lysosomal<br/>Dysfunction"]
B --> C2["Primary Cilia<br/>Disruption"]
B --> C3["Synaptic<br/>Vesicle Defects"]
B --> C4["Microglial<br/>Activation"]
C1 --> D["Alpha-Synuclein<br/>Aggregation"]
C2 --> D
C3 --> D
C4 --> D
D --> E["Dopaminergic<br/>Neuron Loss"]
E --> F["Parkinson's<br/>Disease Motor<br/>Symptoms"]
style A fill:#e1f5fe,stroke:#333
style E fill:#ffcdd2,stroke:#333
style F fill:#f99,stroke:#333
¶ GBA1 Variants and PD Risk
Heterozygous GBA variants represent the most common genetic risk factor for Parkinson's disease, present in 5-15% of PD patients depending on ancestry[@sidransny2020].
Pathogenic GBA variants:
- N370S, L444P, E326K, N408S (rs356182)
- Complex alleles (recombinations, recombinant haplotypes)
- Variant-specific penetrance and phenotype
Risk quantification:
- N370S: OR ~5-7 for PD development
- L444P: OR ~8-10 (associated with more severe phenotype)
- E326K: OR ~2-3 (lower penetrance, more prevalent)
- Population frequency varies by ancestry (highest in Ashkenazi Jewish: 15-20% carrier rate)
PD phenotype in GBA carriers:
- Earlier onset (average 5-7 years earlier than non-carriers)
- More rapid progression
- Higher prevalence of cognitive impairment and dementia (up to 30%)
- More prominent non-motor features (autonomic dysfunction, REM sleep behavior disorder)
- Greater tau pathology burden on PET imaging
Mechanism-based approach:
GBA variants reduce glucocerebrosidase enzyme activity, leading to:
- Accumulation of glucocerebroside (GluCer) in lysosomes
- Impaired lysosomal protein degradation
- Alpha-synuclein accumulation (feedback loop)
- Mitochondrial dysfunction
- Endoplasmic reticulum stress
Therapeutic strategies:
-
Enzyme enhancement:
- Ambroxol: Acidic beta-glucosidase (GCase) chaperone, increases GBA activity
- Venglustat (GZ/SAR402671): GBA substrate reduction therapy
- AD/PD 2026: Ambroxol trials showed increased GCase activity and reduced alpha-synuclein in CSF
-
Substrate reduction:
- Prevents accumulation of toxic glucocerebroside metabolites
- Multiple programs in clinical development
-
Gene therapy:
- AAV-GBA delivery for sustained expression
- Preclinical promise, early clinical planning
AD/PD 2026 GBA therapeutic updates:
- Phase 2 ambroxol trials: 20-30% increase in GCase activity in CSF, trend toward clinical stabilization
- Venglustat: Phase 2 data showed biomarker effects but limited clinical benefit in initial readout
- Next-generation chaperones with improved CNS penetration in development
Point mutations and duplications:
- SNCA A53T (most common pathogenic point mutation)
- SNCA triplication — gene dose correlates with severity and earlier onset
- A30P, E46K, H50Q, G51D — rare variants with variable penetrance
- PD phenotype with prominent autonomic and cognitive features
SNCA and prodromal markers:
- RBD prevalence higher in SNCA mutation carriers
- Hyposmia and constipation may precede motor symptoms by years
- SAA positivity can be detected in prodromal SNCA carriers
PD PRS development:
Recent GWAS have identified >90 independent risk loci for Parkinson's disease, enabling construction of polygenic risk scores[@singh2024].
PRS applications:
- Risk stratification in prodromal populations
- Clinical trial enrichment
- Understanding genetic architecture of PD subtypes
- Integration with environmental risk factors
AD/PD 2026 PD PRS updates:
- PRS composed of 90+ variants explains ~7-8% of phenotypic variance
- Top PRS decile shows 2.5-3x increased PD risk vs. bottom decile
- PRS combined with environmental risk factors (pesticides, head trauma) improves prediction
- Trans-ancestry PRS development ongoing to address European bias
Indications for genetic testing in PD:
- Early-onset PD (age <50): Higher yield for LRRK2, GBA, and other genetic causes
- Family history: Affected first-degree relatives, especially with autosomal dominant pattern
- Ancestry: Ashkenazi Jewish, North African Arab — elevated LRRK2 and GBA prevalence
- Atypical features: Rapid progression, early dementia, prominent autonomic dysfunction (suggests GBA)
- Clinical trial participation: Required for genotype-stratified studies
- Family planning: Relevant for reproductive decisions
Genetic testing panels:
- PD panel: LRRK2, GBA, SNCA, PARK2 (parkin), PINK1, DJ1, VPS35, ATP13A2
- AD panel: APOE, PSEN1, PSEN2, APP, ABCA7, TREM2, SORL1
- Next-generation sequencing: Whole exome or targeted panels are standard approach
Ethical considerations:
- Penetrance variability complicates risk communication
- Limited therapeutic options for positive results in some cases
- Insurance and employment discrimination concerns
- Family implications of cascade testing
AD and PD share some genetic risk factors, reflecting common cellular vulnerability pathways:
| Gene |
AD Effect |
PD Effect |
Shared Mechanism |
| SNCA |
Weak |
Strong |
Alpha-synuclein aggregation |
| MAPT |
Weak |
Moderate |
Tau pathology, microtubule dynamics |
| HLA region |
Moderate |
Moderate |
Immune regulation, microglial function |
| CLU |
Moderate |
Weak |
Complement, inflammation |
| TREM2 |
Moderate |
Weak |
Microglial activation |
| GBA |
Weak |
Strong (via lysosomal) |
Lysosomal function, alpha-synuclein |
¶ Multi-Morbidity and Mixed Pathology
Genetic risk profiling increasingly recognizes that many patients have overlapping pathology:
- APOE4 carriers with PD show more rapid cognitive decline
- GBA carriers show more tau PET burden than non-carriers
- Combined amyloid + synuclein pathology detectable via SAA panels
The integration of genetic risk with fluid biomarkers creates a powerful stratification framework:
flowchart TD
A["Patient Assessment"] --> B["Genetic Panel:<br/>APOE, LRRK2, GBA,<br/>SNCA, TREM2"]
B --> C{"High-Risk<br/>Variant?"}
C -->|"APOE4/4"| D["APOE4-Homozygous<br/>AD Protocol"]
C -->|"LRRK2 G2019S"| E["LRRK2 Kinase<br/>Inhibitor Trial"]
C -->|"GBA Variant"| F["GBA Chaperone<br/>Therapy Trial"]
C -->|"None"| G["Sporadic Protocol:<br/>Fluid Biomarkers"]
G --> H["p-tau217 / NfL /<br/>Alpha-syn SAA"]
H --> I{"Pathology<br/>Profile"}
I -->|"Amyloid+ Tau+"| J["AD-Specific<br/>Therapies"]
I -->|"Synuclein+"| K["PD-Specific<br/>Therapies"]
I -->|"Both+"| L["Multi-Target<br/>Approach"]
style D fill:#e1f5fe,stroke:#333
style E fill:#c8e6c9,stroke:#333
style F fill:#c8e6c9,stroke:#333
style J fill:#f3e5f5,stroke:#333
style K fill:#f3e5f5,stroke:#333
style L fill:#fff3e0,stroke:#333
Genetic stratification is transforming clinical trial design:
Genotype-specific trials:
- LRRK2 inhibitor trials: Require LRRK2 G2019S or other pathogenic variants
- GBA chaperone trials: Require heterozygous GBA variant
- TREM2 agonist trials: Require TREM2 loss-of-function variant (R47H)
Stratified populations:
- APOE4/4: Separate trials for anti-amyloid therapy response
- Early-onset AD: Different endpoints, shorter timelines
- Prodromal carriers: Prevention trial designs
Basket trials:
- Genotype-matched baskets across diseases (e.g., LRRK2 inhibitors for PD and Crohn's disease)
- Platform trials with molecular eligibility criteria
- Genetic diversity: Underrepresentation of non-European ancestry in GWAS limits precision for diverse populations
- Variant interpretation: Many variants of uncertain significance (VUS) complicate clinical testing
- Therapeutic access: Genetic stratification only有价值 if matched therapies are available
- Economic feasibility: Comprehensive genetic + biomarker testing faces reimbursement challenges
- Clinical infrastructure: Genetic counseling capacity insufficient for population-scale testing
- LRRK2 inhibitor approval: First disease-modifying therapy for genetically defined PD likely
- GBA chaperone pivotal trials: Ambroxol and other GCase enhancers advancing to Phase 3
- APOE-guided anti-amyloid therapy protocols: Standardized approaches for APOE4 carrier management
- Multi-ancestry PRS: Validated across diverse populations with clinical utility
- SAA-guided genetic testing: Using seed amplification to guide genetic testing decisions
- Combination genetic + biomarker risk scores: Integrated models incorporating genetics, fluid biomarkers, imaging, and lifestyle factors
- Gene editing for monogenic cases: CRISPR-based approaches for ADAD and monogenic PD
- Preventive intervention based on genetic risk: Nudge-based or pharmacological prevention for high-PRS individuals
- Mechanism-matched clinical trials: Every clinical trial stratified by molecular mechanism
- Population screening: At-risk population genetic screening with cascade counseling