Alzheimer Disease Genetic Test: APOE, APP, PSEN1, PSEN2, and Results

Alzheimer disease genetic testing can answer very different questions depending on which genes are examined. Rare pathogenic variants in APP, PSEN1, or PSEN2 can cause autosomal dominant Alzheimer disease, usually with symptoms beginning before age 65. APOE testing is different: the ε4 allele can raise the chance of late-onset Alzheimer disease and influence the risk of treatment-related brain swelling or bleeding with some anti-amyloid medicines, but it does not prove that a person has or will develop Alzheimer disease. Testing is most informative when it follows a careful neurologic evaluation, a three-generation family history, and genetic counseling. A positive result may clarify a diagnosis, guide testing for relatives, or affect treatment discussions. A negative or uncertain result often leaves meaningful residual risk because many cases are not caused by a single identifiable variant. The laboratory report must therefore be interpreted in the context of symptoms, age at onset, family pattern, ancestry, and the exact type of test ordered.

• APP, PSEN1, and PSEN2 pathogenic variants can cause inherited Alzheimer disease, often with onset before age 65 and a 50% chance of transmission to each child.
• APOE ε4 increases susceptibility but is not diagnostic; many carriers never develop dementia, and many people with Alzheimer disease do not carry ε4.
• APOE testing may be considered before certain anti-amyloid treatments because two ε4 copies are associated with the highest risk of amyloid-related imaging abnormalities.
• Testing usually requires only blood or saliva, and fasting is not needed unless another test is ordered at the same visit.
• A variant of uncertain significance should not be used to predict disease or test healthy relatives until stronger evidence supports reclassification.
• Genetic counseling is especially important before predictive testing in an unaffected adult, because results can affect emotional health, relatives, insurance planning, and reproductive choices.

Table of Contents

• What the Test Examines
• Who May Benefit From Testing
• How Testing Is Performed
• How to Read the Results
• APOE and Treatment Safety
• Family and Personal Implications
• Next Steps After Testing

What the Test Examines

An Alzheimer disease genetic test may evaluate a single known family variant, sequence several dementia-related genes, or report an APOE genotype. These approaches are not interchangeable. The most useful test depends on whether the aim is to diagnose a symptomatic person, identify a familial cause, estimate susceptibility, or support a medication-safety discussion.

APP, PSEN1, and PSEN2 are high-impact genes

APP, PSEN1, and PSEN2 help regulate the production or processing of amyloid-beta, a protein involved in Alzheimer disease biology. Certain pathogenic variants alter this pathway strongly enough to cause autosomal dominant Alzheimer disease. In a family with such a variant, each child of an affected person has a 1-in-2 chance of inheriting it.

These variants are rare in the population and account for only a small fraction of all Alzheimer disease. They are most likely when several relatives in successive generations developed similar progressive cognitive symptoms, especially when onset occurred in the 30s, 40s, 50s, or early 60s. PSEN1 is the most common of the three in clearly inherited early-onset families. APP variants and duplications are less common, while PSEN2 variants are rarer and may show wider variation in age at onset.

A pathogenic variant in one of these genes is often called deterministic or causal, but those words still require nuance. Penetrance is very high for many established variants, yet the exact age at onset and symptom pattern can differ even within one family. Some people first develop memory loss, while others have language problems, movement findings, seizures, or behavioral changes.

APOE is primarily a susceptibility gene

The APOE gene has three common alleles: ε2, ε3, and ε4. Everyone inherits one allele from each biological parent, producing genotypes such as ε3/ε3, ε3/ε4, or ε4/ε4.

APOE ε4 is the strongest common genetic risk factor for late-onset Alzheimer disease. One ε4 copy generally raises risk compared with ε3/ε3, and two copies raise it further. However, APOE is not a simple yes-or-no test. Risk varies by age, sex, ancestry, cardiovascular health, family history, and other genes. An ε4 carrier may remain cognitively healthy throughout life. Conversely, Alzheimer disease can occur in a person with no ε4 allele.

APOE ε2 is often associated with lower Alzheimer disease risk than ε3, but it is not complete protection. It also has other medical associations, including a rare lipid disorder in some people with ε2/ε2. A genotype should therefore be interpreted for the specific clinical question rather than labeled universally “good” or “bad.”

People who want a broader explanation of risk versus diagnostic results may find genetic diagnostic test result categories helpful. APOE susceptibility testing is also distinct from a presymptomatic test for a known familial pathogenic variant.

Who May Benefit From Testing

Testing is most likely to provide a clear answer when the personal and family history suggests a single-gene form of dementia. It is less likely to settle the diagnosis in a person with typical late-onset symptoms and no strong family pattern.

A clinician may consider testing APP, PSEN1, PSEN2, and other dementia genes when one or more of the following apply:

• Cognitive decline began before age 65, particularly before age 55.
• Several relatives in more than one generation had early or unusually similar dementia.
• The family history suggests autosomal dominant inheritance, with affected parents and children across generations.
• The clinical presentation is atypical, overlaps with frontotemporal dementia, or includes seizures, movement symptoms, or prominent language decline.
• A relative already has a pathogenic variant and targeted family testing is available.
• Establishing a molecular diagnosis would change counseling, reproductive planning, or eligibility for a study.

The preferred first person to test is usually a living relative who has symptoms. Testing an affected person gives the laboratory and clinical team the best chance of linking a variant to the family’s condition. If a pathogenic variant is found, unaffected adult relatives can then choose a focused familial variant test. Testing an unaffected relative first with a broad panel can produce a negative or uncertain result that is much harder to interpret.

Routine APOE testing is generally not recommended solely to predict whether a healthy person will develop Alzheimer disease. Its predictive precision is limited, and no genotype can provide a personal age of onset. Direct-to-consumer results may also lack counseling and can be misunderstood as a diagnosis. Clinical confirmation may be appropriate before making health decisions.

APOE testing has a more defined role when a person with mild cognitive impairment or mild Alzheimer dementia is considering an anti-amyloid therapy whose safety profile differs by APOE genotype. Even then, the result supports a treatment discussion; it does not replace confirmation of Alzheimer pathology through the clinician’s recommended biomarker pathway.

Testing is not a substitute for a complete evaluation of memory or thinking problems. Depression, sleep disorders, medication effects, thyroid disease, vitamin deficiencies, vascular injury, seizures, infection, and other neurologic disorders can mimic or worsen cognitive decline. Genetic testing should sit within—not ahead of—that broader assessment.

How Testing Is Performed

Most Alzheimer disease genetic tests use DNA from a blood sample. Some laboratories accept saliva or a cheek swab. No fasting or medication pause is usually required. Blood often provides the most consistent DNA quality, especially for large panels or follow-up methods.

The laboratory method should match the suspected variant type:

Test approach	Best use	Important limitation
Targeted variant analysis	Tests for a pathogenic variant already identified in the family	Does not evaluate other genes or variants
Sequence analysis	Detects many single-letter changes and small insertions or deletions in APP, PSEN1, PSEN2, or a panel	May miss large deletions, duplications, repeat changes, or difficult regions
Deletion/duplication analysis	Detects copy-number changes such as an APP duplication	May require a separate laboratory method
Dementia multigene panel	Evaluates Alzheimer, frontotemporal dementia, prion, and other overlapping genes	Raises the chance of uncertain or unexpected findings
APOE genotyping	Reports common ε2, ε3, and ε4 alleles for risk or treatment-safety context	Does not diagnose Alzheimer disease

A broad dementia panel may be more useful than testing only APP, PSEN1, and PSEN2 when symptoms overlap several disorders. Genes such as MAPT, GRN, C9orf72, PRNP, SORL1, and others may be relevant depending on the history. Some repeat expansions, especially in C9orf72, require a dedicated assay and may not be reliably detected by ordinary sequencing. The ordering clinician should confirm that the laboratory’s methods include the variant types of concern.

Results often take two to eight weeks, although timing varies by laboratory, insurance review, and whether relatives’ samples are needed. Before testing, the consent discussion should cover possible positive, negative, uncertain, and secondary findings. A multigene panel test can improve diagnostic reach, but it also increases interpretive complexity.

How to Read the Results

A laboratory report classifies each relevant variant and explains whether it is known to affect gene function. The clinical meaning depends on the gene, variant classification, symptoms, and family evidence. The same word—“positive,” for example—can carry very different implications for APOE and PSEN1.

Pathogenic or likely pathogenic result

A pathogenic or likely pathogenic variant in APP, PSEN1, or PSEN2 can establish a molecular diagnosis of autosomal dominant Alzheimer disease when it fits the clinical picture. “Likely pathogenic” generally means the evidence strongly supports disease causation, although laboratories reserve “pathogenic” for variants meeting a higher evidence threshold.

For an unaffected adult from a family with a known variant, a positive targeted result means the person inherited that familial variant. It does not predict the exact year symptoms will begin or how quickly they will progress. Predictive testing should use a structured process that includes neurologic or genetic assessment, informed consent, psychological support, and a plan for receiving the result.

An APOE result is not classified in the same deterministic way. A report showing ε3/ε4 or ε4/ε4 indicates increased susceptibility and may affect anti-amyloid treatment counseling, but it does not establish that current symptoms are due to Alzheimer disease. Amyloid or tau biomarkers, clinical findings, and imaging remain separate parts of diagnosis.

Negative result

A negative result means the test did not find a reportable pathogenic variant in the regions and genes analyzed. Its reassurance depends on the starting situation.

• If a person tests negative for a known familial APP, PSEN1, or PSEN2 variant, that person usually did not inherit the family’s specific autosomal dominant risk.
• If an affected person with a strong family history has a negative panel, an inherited cause may still exist in a gene or variant type the test did not detect.
• If an unaffected person is tested without first identifying a family variant, a negative result is often uninformative and does not return risk to zero.
• An APOE genotype without ε4 does not rule out future or current Alzheimer disease.

Laboratory coverage, mosaicism, structural variants, repeat expansions, and evolving gene knowledge can all affect sensitivity. Reanalysis may be useful after several years if the clinical suspicion remains high.

Variant of uncertain significance

A variant of uncertain significance, or VUS, is a DNA change with insufficient or conflicting evidence. It is not a diagnosis and should not be used by itself to predict disease, select treatment, or test healthy relatives. Family studies may sometimes help the laboratory determine whether the variant tracks with disease, but many VUS results remain unresolved for years.

Clinical care should continue to follow the person’s symptoms and established diagnostic findings. The laboratory or genetics clinic may issue an amended report if the variant is later reclassified. The meaning of a VUS result is especially important in large dementia panels, where uncommon benign variation is frequently found.

APOE and Treatment Safety

APOE genotyping now has a practical role in discussions about some anti-amyloid monoclonal antibodies used for early symptomatic Alzheimer disease. These medicines can cause amyloid-related imaging abnormalities, commonly shortened to ARIA. ARIA may appear as temporary brain swelling or fluid leakage, called ARIA-E, or as small areas of bleeding and iron deposition, called ARIA-H.

People with APOE ε4—especially those with two copies—have a higher frequency of ARIA during treatment than noncarriers. Most imaging abnormalities are asymptomatic, but some cause headache, confusion, visual changes, dizziness, nausea, seizures, focal weakness, or serious bleeding. Because risk differs by genotype, the prescribing clinician may recommend APOE testing before treatment and use the result during informed consent.

The genotype does not decide treatment by itself. The clinician also considers:

• Confirmation that the person has amyloid pathology and is in the treatment-eligible disease stage
• Baseline brain MRI findings, including microbleeds or superficial siderosis
• Use of anticoagulants, antiplatelet drugs, or thrombolytic medicines
• History of stroke, seizures, bleeding disorders, or severe small-vessel disease
• Ability to complete scheduled MRI monitoring and report new neurologic symptoms promptly
• The person’s values regarding possible slowing of decline, infusion burden, and adverse effects

A person with ε4/ε4 may still be considered for treatment, but the risk discussion is more complex. An ε4-negative result does not eliminate ARIA. Treatment plans, MRI schedules, and emergency instructions should follow the specific medicine’s current prescribing information and the treating center’s protocol.

APOE testing for treatment safety should not be confused with using APOE to diagnose the cause of memory loss. A person can carry ε4 and have another dementia, and a noncarrier can have biomarker-confirmed Alzheimer disease. Genetics and disease biomarkers answer related but separate questions.

Family and Personal Implications

A causal result belongs biologically to a family even though the medical record belongs to the person tested. Finding a pathogenic APP, PSEN1, or PSEN2 variant may clarify risk for siblings, adult children, and more distant relatives. It may also reveal that an apparently sporadic case arose from a new variant or from a family history obscured by small family size, early deaths, adoption, or misdiagnosis.

Genetic counseling helps the family translate the result into a testing strategy. The usual sequence is to confirm the variant in an affected person, offer targeted testing to competent adults who want it, and avoid predictive testing in children for an adult-onset condition with no childhood medical benefit. Each adult relative should make an independent choice about whether and when to learn the result.

Before predictive testing, people may want to consider:

• Who will accompany them when results are disclosed
• How they have handled uncertain or serious health information before
• Whether they want life, disability, or long-term-care insurance in place first
• What protections and exclusions apply in their country or state
• Whether the result may affect employment, military service, or immigration processes
• How and when they would tell relatives, partners, or adult children
• Whether they are considering pregnancy, donor gametes, prenatal diagnosis, or preimplantation genetic testing

In the United States, federal protections generally limit health-insurance and employment discrimination based on genetic information, but they do not cover every form of insurance. Laws differ elsewhere and can change. A genetics professional or qualified legal adviser can explain current local rules before testing.

Reproductive options may include natural conception without testing, prenatal diagnosis, in vitro fertilization with preimplantation genetic testing, use of donor eggs or sperm, adoption, or deciding not to have children. Some laboratories can perform exclusion testing designed to reduce disclosure of an at-risk parent’s own status, but this requires specialized planning.

APOE results also carry family information, but they are usually not handled like a deterministic familial variant. Because ε4 is common and incompletely predictive, testing relatives solely because one family member carries ε4 rarely gives a clear clinical answer.

Next Steps After Testing

The most useful follow-up begins with the exact report, not a verbal summary or consumer dashboard. Ask for the gene, transcript, DNA and protein change, zygosity, classification, test method, coverage limitations, and laboratory interpretation. Keep the original report because relatives and future clinicians may need the precise variant notation.

For a pathogenic APP, PSEN1, or PSEN2 result, the next steps may include neurologic follow-up, baseline cognitive assessment, counseling about expected but variable disease course, referral to a specialty memory or inherited-dementia clinic, and discussion of research studies. At-risk relatives can be referred for independent counseling rather than tested automatically.

For an APOE result, clarify why the test was ordered. If it was obtained for treatment planning, review ARIA risk, MRI requirements, medication interactions, and warning symptoms with the prescribing team. If it was obtained for general prediction, a counselor can help place the result within age, ancestry, family history, and modifiable health factors without treating it as destiny.

For a negative or uncertain result, ask whether the test included deletion/duplication analysis, repeat expansions, and other dementia genes suggested by the phenotype. A genetics team may recommend testing another affected relative, storing DNA, broadening the panel, or revisiting the result as knowledge advances. Whole-exome or whole-genome sequencing may be considered in selected families, but neither method detects every clinically important variant and both can generate uncertain findings.

Regardless of genotype, seek medical evaluation for progressive memory loss, impaired judgment, language change, getting lost, personality change, or loss of daily function. Sudden confusion, new weakness, severe headache, seizure, or abrupt speech difficulty needs urgent assessment because it may represent stroke, infection, a medication problem, or another acute condition rather than gradual Alzheimer disease.

Genetic information can refine risk and sometimes identify a cause, but it works best alongside clinical examination, cognitive testing, imaging, and validated Alzheimer biomarkers. A well-chosen test can end a long diagnostic search; a poorly chosen one can create more uncertainty. The ordering question, family structure, and plan for using the result should be clear before the sample reaches the laboratory.

References

• The Alzheimer’s Association clinical practice guideline for the Diagnostic Evaluation, Testing, Counseling, and Disclosure of Suspected Alzheimer’s Disease and Related Disorders (DETeCD-ADRD): Executive summary of recommendations for primary care 2024 (Guideline)
• Lecanemab Therapy and APOE Genotype 2024 (Review)
• Implications of Emerging Uses of Genetic Testing for Alzheimer’s Disease 2023 (Review)
• Alzheimer’s Disease: An Updated Overview of Its Genetics 2023 (Review)
• Genetic testing of common and rare variants in dementia: prediction and correlation with clinical diagnosis 2025 (Review)
• APP, PSEN1, and PSEN2 Variants in Alzheimer’s Disease 2021 (Review)

Disclaimer

This article provides general education about Alzheimer disease genetic testing and does not diagnose disease or replace individualized medical or genetic counseling. Genetic results should be interpreted by a qualified clinician using the person’s symptoms, family history, ancestry, laboratory methods, and current treatment guidance. Seek urgent medical care for sudden neurologic symptoms or severe symptoms during anti-amyloid treatment.