Disease-causing variants and risk variants

A disease-causing variant can be sufficient to produce disease

Some variants disrupt a gene in a way that can cause a particular disorder. If the evidence is strong, a clinical laboratory may classify the variant as pathogenic or likely pathogenic.

Examples include pathogenic variants in APP, PSEN1, or PSEN2 in rare families with autosomal dominant Alzheimer's disease, and pathogenic variants in GRN, C9orf72, or MAPT in inherited frontotemporal degeneration.

Even within this category, a result is not a complete forecast. Penetrance describes how often people who carry a variant develop the associated condition. Some pathogenic variants have very high penetrance. Others have reduced or age-related penetrance, which means some carriers remain well for many years or may never develop symptoms.

Expression describes how a condition appears. Relatives who carry the same pathogenic variant can have different first symptoms, ages of onset, severity, and rates of change.

A risk variant shifts probability

A risk variant is associated with a higher or lower likelihood of disease in a group of people. It is neither necessary nor sufficient for the disease to occur.

APOE e4 is associated with a higher probability of Alzheimer's disease. Many people with APOE e4 never develop Alzheimer's, and many people with Alzheimer's do not carry APOE e4. APOE e2 is associated with lower risk in many studies, but it does not make a person immune.

Risk estimates are not fixed properties of the gene alone. They can differ with age, ancestry, sex, the population studied, other genetic variants, and how the outcome was defined. A population-level association should not be converted into a precise personal percentage without appropriate data and context.

The distinction is useful, but biology is not always binary

The terms disease-causing and risk are a practical starting point, not a perfect division of nature.

Some pathogenic variants have reduced penetrance. Some risk variants have a larger effect than others. A gene can also be associated with more than one condition or act differently depending on the particular variant. For example, some GBA1 variants cause Gaucher disease when present in both gene copies and also increase Parkinson disease risk in carriers. The exact result matters.

This is why the name of a gene alone is not enough. Interpretation requires the specific variant, its classification, the inheritance pattern, the person's symptoms, and the family history.

What each kind of result can tell a family

A pathogenic variant in a disease-causing gene

It may:

• explain the condition in a person who has symptoms;
• identify a specific familial variant that adult relatives can choose to test for;
• clarify an inheritance pattern;
• support eligibility for a gene-specific study or clinical trial;
• inform reproductive and long-term planning.

It usually cannot specify the exact age of onset, first symptom, severity, or pace for an individual.

A risk variant

It may:

• move estimated probability higher or lower;
• contribute to a treatment-safety discussion in a defined clinical setting;
• help answer a research question.

It does not diagnose the disease, establish that symptoms are caused by that disease, or predict one person's future with certainty.

A negative result means different things in different settings

If a family has a confirmed pathogenic variant and a relative tests negative for that exact variant, the result can be a true negative for that familial cause.

If no familial cause has been identified and a broad panel is negative, the result is usually less complete. The test may not include every relevant gene or method. Current science may not recognize the cause yet. A negative panel does not erase the family history.

Similarly, not carrying APOE e4 does not remove the possibility of Alzheimer's disease. That result only addresses one risk factor.

What an uncertain variant means

A variant of uncertain significance is not a weaker positive. It means the evidence is insufficient to decide whether the variant contributes to disease.

Uncertain variants may be reclassified as more evidence accumulates. Until then, medical and family decisions should rest on the clinical picture and established findings rather than treating the uncertain result as causal.

The first question to ask about a report

When a report names a gene, ask which category the result belongs to:

• Is this a pathogenic or likely pathogenic variant?
• Is it a risk allele?
• Is it a variant of uncertain significance?
• Is it a negative result, and what did the test actually examine?

That category determines what the result can reasonably mean.