Major Finding: Particular variants of the TOMM40 gene are associated with declines in regional gray matter volume and in verbal learning and memory that may presage Alzheimer's disease.
Data Source: Two prospective studies of healthy volunteers with TOMM40 genotyping, one involving MR imaging of 117 participants and another involving verbal learning and memory testing in 337 participants.
Disclosures: Both studies were funded by the National Institute on Aging. Neither investigator had any potential financial conflict.
People who have a newly discovered gene variant that increases the risk of Alzheimer's disease appear to experience subtle changes in both cognition and brain structure years before even the earliest signs of Alzheimer's appear.
The gene, TOMM40, was associated with small but measurable declines in verbal learning and memory, and with decreases in gray matter volume, particularly in the posterior cingulate and precuneus–both areas that are highly involved with memory retrieval.
“Brain changes in these regions are particularly interesting because they are also associated with a high amyloid burden in people with Alzheimer's,” Sterling Johnson, Ph.D., said in an interview. “Many studies have shown that these areas are active when you are recalling the recent past. So seeing changes there fits very nicely with a major symptom of Alzheimer's: recent memory loss.”
Dr. Johnson and his colleague, Dr. Mark Sager, both of the University of Wisconsin, Madison, both said the work may open an important door on the ability to risk-stratify patients for early treatment, potentially identifying them before they experience significant brain damage from the disease.
“Right now we're spending billions of dollars on trying to find a disease-modifying therapy, but we don't really know who we would give that to,” Dr. Johnson said.
In the same way that those with high cholesterol benefit most from statins, patients with a demonstrated risk of Alzheimer's would be the best candidates for any drug therapy, he noted. “Our work may eventually be able to identify in a much more targeted way groups of people who could benefit the most from early intervention, hopefully early enough to prevent them from having serious cognitive loss from Alzheimer's.”
TOMM40 has variable lengths that are defined by how many thymidine bases are contained in a specific section of the gene. The very long form (30 or more thymidine bases) is associated with an earlier onset of Alzheimer's disease, while the very short length (20 or fewer) is associated with a later onset. Some people have a long form (between 20 and 30 bases) that confers intermediate risk for Alzheimer's disease.
In 2009, Dr. Allen D. Roses of Duke University, Durham, N.C., first showed that among patients who developed Alzheimer's after 60 years of age and were homozygous or heterozygous for the apolipoprotein E e3 (APOE-3) allele, those with two copies of the long TOMM40 developed the disease an average of 7 years earlier than did those with shorter TOMM40 lengths (Pharmacogenomics J. 2009 Dec. 22 [doi:10.1038/tpj.2009.69]).
Dr. Johnson's study consisted of 117 healthy middle-aged volunteers (mean age, 57 years) who were known to be homozygous for the APOE-3 allele. APOE-3 homozygotes are thought to have a neutral risk for Alzheimer's disease. He then tested the subjects to determine what variant of TOMM40 they had.
In Dr. Johnson's study group, 38 patients were homozygous for the very short-length TOMM40, 44 had one short and one long or very long allele, and 35 were homozygous for the very long form of TOMM40.
All of the volunteers underwent brain scanning with voxel-based morphometry to assess gray matter volume in different brain regions.
The subjects with two copies of the very long TOMM40 allele had significantly less gray matter in both the ventral posterior cingulate and the precuneus than did those with two short alleles. These are both regions that show early deterioration in late-onset Alzheimer's disease.
“This is very important, because we are finding brain changes in people who are quite young,” Dr. Johnson said.
“The brain differences between the groups were very similar to, but less severe than, what is observed in full-blown Alzheimer's. It may be that the TOMM40 gene will be a useful measure of Alzheimer's risk in middle age, but additional research with longitudinal follow-up is necessary.”
Dr. Sager examined the gene's potential effect on memory in a cohort of 337 adults (mean age, 54), who were genotyped for APOE and TOMM40. Of these, 128 were homozygous for the short form of TOMM40, indicating a lower risk. The long or very long forms were seen in 219 subjects. In the low-risk group, 57% had a family history of Alzheimer's, compared with 77% of the high-risk group–a significant difference.