Causation versus Susceptibility: Alzheimer’s Disease, Genetics, & APOE

By Chase Beeler, Rachel Westman, and Devin Shuman

New Developments

A recent study published in Nature Medicine, and a follow up article in the New York Times, has added to and challenged our understanding of the genetics of Alzheimer’s Disease. Based on the data analyzed, this study is claiming that a specific variant in the gene APOE almost always causes Alzheimer’s Disease, rather than just increasing the lifetime risk of developing Alzheimer’s Disease, as has previously been understood. 

The conclusions of this article have been framed by the authors and news outlets as definitive and relevant to patient care for hundreds of thousands of people. However, we recommend using caution in redefining Alzheimer’s Disease from based on a combination of clinical symptoms (dementia), imaging studies, and lab test results (genetic or biochemical) to only lab test results. This caution is particularly important when considering claims from one study with limited data and ascertainment bias. What most individuals want to know is “will I develop symptoms of Alzheimer’s Disease” and this study (and APOE4 testing) do not answer that question. Now let us walk you through why we say this. 

Background on Alzheimer’s Disease & APOE

Dementia is a generic term used to describe a disease in which an individual experiences a decline in cognitive functioning, often including changes in memory, judgment, or reasoning. Far and away the most common cause of dementia is Alzheimer’s Disease, accounting for about 60-80% of cases. About 1 in 10, or about 10-12%, of those in the general population who are 65 and older will have dementia due to Alzheimer’s, with more than 30% being affected by age 85.  

Because of how common Alzheimer’s Disease is in the general population, it is common for individuals to have at least one close family member who has been affected. Those family members may wonder whether their relatives’ diagnosis increases their risk to develop Alzheimer’s Disease over the course of their lifetime. 

As with many common disorders, our understanding of the causes and genetics of Alzheimer’s Disease continues to evolve as more information and research becomes available. For most individuals and families, Alzheimer’s Disease is considered a ‘multifactorial condition’ meaning that there are both genetic and environmental factors at play. In rare cases, there is a single gene that causes an individual/family’s Alzheimer’s Disease, especially when Alzheimer’s Disease is affecting multiple people at younger ages (less than 65 years old) in the same family (More information Here: Alzheimer’s & Genetics).

But in most cases, the genetic factors at play for Alzheimer’s Disease are what we call ‘susceptibility genes’. Susceptibility genes refer to variations within genes that increase the chance to develop Alzheimer’s Disease but do NOT guarantee that an individual will. Many different susceptibility factors (both genetic and environmental) must, together, reach a threshold before symptoms of Alzheimer’s Disease develop. The most significant and well researched susceptibility gene for Alzheimer’s Disease is called APOE.

APOE comes in three different versions sometimes referred to as “alleles”. These are termed APOE2, APOE3, and APOE4. We each have two copies of the APOE gene and an individual may have any combination of the three possible alleles (e.g., you may have one APOE2 allele inherited from one parent and an APOE3 allele inherited from the other). The APOE3 allele is the most common allele and has no effect on an individual’s risk to develop Alzheimer’s Disease. The APOE2 allele is reported to reduce the risk of developing Alzheimer’s Disease. Finally, the APOE4 allele has been shown to increase an individual’s risk to develop Alzheimer’s Disease. 

The degree to which APOE4 increases the risk is still being researched, but historical data has shown that those of White (European) background with one copy of the APOE4 allele are estimated to have a 3-fold increased risk for late-onset Alzheimer’s Disease (the most common form) and those with two copies of the APOE4 allele are estimated to have a 15-fold increased risk of late-onset Alzheimer’s Disease. For other populations, these estimates may be higher or lower, but overall the risk for individuals with two copies of the APOE4 allele has historically been said to be between 25-50%, with some reports saying the risk may be even higher than that. 

The Studies Findings

The recent study published in Nature Medicine was interested in reevaluating what we know about the true Alzheimer’s Disease risks for individuals with two copies of the APOE4 allele. 

In order to do this, the researchers analyzed brain findings (examination of brain tissue from affected individuals) from the National Alzheimer’s Coordinating Center (NACC) and biomarker data from several large multicenter patient groups including Alfa plus (Alzheimer’s and Families cohort), A4 Clinical Trial, OASIS (Open Access Series of Imaging Studies), ADNI (Alzheimer’s Disease Neuroimaging Initiative), and WRAP (Wisconsin Registry for Alzheimer’s Prevention). NACC is one of the largest databases of data from individuals with Alzheimer’s Disease. Each of the databases and patient registries used by this new study include both individuals with and without dementia. For this study they analyzed data from over 3,000 brain donors and biomarkers from over 10,000 individuals.

First, the researchers analyzed data from the brains of donors who were both affected and unaffected with Alzheimer’s Disease and showed:

  • Brain changes were present in about 95% of all individuals with two copies of APOE4 and 85% of those with only one copy of APOE4, compared to 57% of those with two copies of APOE3 had these same changes and 40% of those with at least one copy of APOE2. 

Second, they analyzed biomarkers associated with Alzheimer’s Disease in individuals with two copies of APOE4 compared to one copy of APOE4 and one copy of APOE3 or two copies of APOE3. In particular, they analyzed for Tau and Amyloid (proteins linked to the development of  Alzheimer’s Disease). Both age and clinical status was considered in this analysis and it showed: 

  • Individuals over the age of 55 with two copies of APOE4 had higher levels of biomarkers that are linked with Alzheimer’s Disease compared to individuals with two copies of APOE3.
  • Nearly all individuals 65 years old and older with two copies of APOE4 had abnormal levels of amyloid in their cerebrospinal fluid and 75% had abnormal amyloid levels on brain scans.

Finally, they compared the age of symptom onset for those with Alzheimer’s Disease both with and without two copies of APOE4. They found:

  • Alzheimer’s Disease symptom onset was approximately 7-10 years earlier for those with two copies of APOE4 compared to individuals with two copies of APOE3.

The Studies Claims

Based on the data presented, this study draws the conclusion that the penetrance (the chance that someone with a given genetic status will manifest a disease) for individuals with two copies of APOE4 is almost 100%.

In particular, they claim two copies of APOE4 should be considered causative of Alzheimer’s Disease rather than a risk factor (susceptibility gene).

The Studies Limitations

Biomarkers vs. Dementia: 

The value of the claims being made in this study depend on how we choose to define Alzheimer’s Disease. 

The main data points that indicated that two copies of APOE4 causes Alzheimer’s Disease were biomarkers not clinical symptoms. The study showed that the chance an individual with two copies of APOE4 would have increased biomarker levels was close to 100% by age 65 years old. However, this does not necessarily mean that person has (or will have) dementia or cognitive impairment. 

In recent years, data has been published that redefines Alzheimer’s Disease as a biological entity that can theoretically be diagnosed by biomarkers that are believed to eventually lead to clinical symptoms. The presence of biomarkers in the absence of clinical symptoms is often referred to as “preclinical Alzheimer’s Disease”. This redefinition is helpful in research to further categorize and understand Alzheimer’s Disease and aid in identifying new therapeutic options, but can be misleading clinically. In the clinical setting, at this time, we are unable to fully interpret and predict how a biomarker may impact an individual’s risk for associated symptoms.

Ascertainment Bias:

As mentioned above, the data that was used for this study came from NACC and several large multicenter trials or registries. NACC is a national database of individuals with personal and family histories of Alzheimer’s Disease, as well as genetic variants associated with Alzheimer’s Disease. The large multicenter patient groups include a combination of those who do and do not have personal and/or family histories of clinical symptoms or biomarker evidence for Alzheimer’s Disease. Data from NACC and these other patient groups are not necessarily representative of the general population, with possible bias towards those who have higher risks related to Alzheimer Disease as they are the individuals who are most likely to learn about these trials and want to participate. The sampling of these patient groups done for this study (pulling individual data of those with two copies of APOE4) may have also added further bias. Often when there is a possibility of bias in the data used (called sampling/ascertainment bias) studies make statistical adjustments to their data. While the researchers in their supplemental data acknowledge the convenience of their sampling, further adjustments based on this possible bias was not done. 

Further, the demographics of the patients included in this analysis are not provided. Information regarding the sex (female or male) and ages are included in this analysis. However, as stated by the researchers “…all participants came from the USA or Europe and were predominantly white.” While they go on to acknowledge the differences in APOE frequencies and risks related to geographical location and ethnicities and the need for further studies to include more diverse populations, they do not explicitly state the limitations on their conclusions. The data presented by the researchers, at most, can be applied to the population included (i.e. individuals of European ancestry) in this analysis and caution is needed in the use of this data for the broader population.

The Way Forward

Historically, genetic testing of the APOE gene has not been recommended. The American College of Medical Genetics and National Society of Genetic Counselors currently recommend against APOE gene testing for asymptomatic individuals. This is because the predictive value – our ability to use the results of this testing to predict risks for an individual – is limited.  

In July of 2023, the drug Leqembi (lecanamab) was approved by the FDA and may be considered for those with early onset Alzheimer’s Disease. This treatment has not been approved for asymptomatic individuals or those with moderate or advanced disease. There are known side effects, including ARIA (amyloid-related imaging abnormalities) related to brain bleeding and/or swelling. Those who carry one or two APOE4 alleles are at an increased risk for ARIA. Testing for APOE status of symptomatic individuals is encouraged when considering initiating treatment so patients can be informed of their risks to develop ARIA (For more information see our previous article).

The study discussed above highlights that defining the effect of genetic variants is a constantly moving target. Given the data that was presented, it is certainly possible that having two copies of APOE4 represents a higher risk for Alzheimer’s Disease than was previously understood and that the biomarkers linked with Alzheimer’s Disease are consistently present in these individuals once they reach a certain age. However, more data is needed to define two copies of APOE4 as clinically causative of Alzheimer’s Disease. 

Individuals who are interested in discussing their family history of Alzheimer’s Disease or who have previously undergone genetic testing related to APOE, or other Alzheimer’s/Dementia related genes, may want to meet with a genetic counselor who can evaluate personal and family histories on an individual basis, interpret your previous testing, and discuss any further options for you and your family. 

For further reading on Alzheimer’s Disease Check out these previous posts: