Skip to main content
Premium Trial:

Request an Annual Quote

Alzheimer’s Genome Project Discovers New Risk Genes. Will New Drugs Follow?

Premium
The identification of four new candidate risk genes for Alzheimer’s disease by researchers from the Alzheimer’s Genome Project increases the chances that scientists may soon develop genetically targeted therapies and predictive tests.
 
But holding up such innovations is the lack of a reliable algorithm that can analyze protective and risk genes to determine patients’ chances of developing the disease, according to a senior study author.
 
In a paper published last week in the American Journal of Human Genetics, researchers from Harvard Medical School and elsewhere used microarrays to assess 500,000 SNPs in about 1,400 individuals from more than 400 families. The study identified four candidate genes, two of which have already been validated in large replication studies.
 
“Because of the way we did the analysis, we are the first [group] to find genome-wide significant hits beyond APOE ε4,” Rudolph Tanzi, a neurologist at Harvard Medical School and director of the Genetics and Aging Research Unit at Massachusetts General Hospital and the lead researcher in the project, told Pharmacogenomics Reporter this week. He added that the family-based genome-wide association studies contributed to the success of the researchers.
 
The family-based approach “allows you to make these comparisons against a more or less normalized genetic background, because you are comparing at any given moment just siblings, or first-degree relatives, within a family,” Tanzi said. “So, that removes a lot of the type 1 and type 2 error that plagues case control studies, where you are comparing cases and controls with mixed genders, mixed ethnicities, and mixed ideologies of the diseases, because Alzheimer’s is so heterogeneous.”
 
Following the publication of the study results, Tanzi hopes that other research groups will test these genes in case-control studies to see if they hold up in a population-based approach. The Alzheimer’s Genome Project will then conduct a meta-analysis of these case-control studies, to determine which “gene associations look real.”
 
“We want to just put the data out there and we hope that many different groups will test these SNPs in their case-control series,” Tanzi noted. “What we expect, depending on the size and composition of the case control samples that [are] being used, each group may or may not see an association.”
 
After the genetic hits are validated, which will be published in Cure Alzheimer’s Alzgene.org database, the possibility of developing a targeted drug or predictive chip from this data may be closer, according to Tanzi.
 
In fact, the Japanese pharmaceutical company Eisai is already looking into developing therapeutics based on the genetic targets discovered by the initial screen, Tanzi said. However, he noted that developing a predictive chip for lifetime Alzheimer’s risk is more complex, particularly since there is currently no algorithm that can incorporate known genes associated with the disease into a risk score.
 
“I think we’re probably three to five years from [developing a predictive chip] because we have to wait for all the genome-wide screens to be done; we have to wait for the whole genome to be covered; we have to wait for the papers to come out; and we have to wait for enough groups to do case control studies in at least four independent samples to reliably do the meta-analyses,” Tanzi said.
 
But even then, “there is still a problem,” he continued. “Even when you have the genes, what we don’t have yet is the algorithm that allows you to look for the gene-to-gene interaction with that many genes.”
 
The work of the Alzheimer’s Genome Project is funded by the Cure Alzheimer’s Fund, which is composed of four founding families with the aim of funding research that improves treatments for the disease. 
 
The Alzheimer’s Genome Project “wants to find these genes, find a reliable way to do prediction and diagnosis. Most importantly, the goal is to learn what’s going wrong, so we can incorporate novel therapies and we can empower people,” Tanzi said.
 
The Study
 
For the study published in the American Journal of Human Genetics, Tanzi and his colleagues looked specifically at late-onset Alzheimer’s disease using the Affymetrix GeneChip Human Mapping 500K Array Set to evaluate 404,604 SNPs in 1,376 individuals from 410 families. This study is believed to be the largest family-based genome-wide association study of Alzheimer’s thus far.
 
Following the initial screen, Tanzi’s team tested genetic associations in replication samples of nearly 2,700 individuals from almost 900 families, drawn from three different studies.
 
Previous studies have linked three genes – amyloid beta (A4) precursor protein (APP), presenilin-1 (PSEN1), and presenilin-2 (PSEN2) – with early-onset Alzheimer’s. These gene variants, which Tanzi’s lab previously discovered, are fully penetrant but only explain between 1 and 2 percent of Alzheimer’s cases.
 

“Even when you have the genes, what we don’t have yet is the algorithm that allows you to look for the gene-to-gene interaction with that many genes.”

There is only one risk gene, APOE ε4, which is conclusively linked to late-onset of the Alzheimer’s, accounting for 95 percent of the disease. While the risk gene increases a person’s chances of getting Alzheimer’s, it is not fully penetrant, meaning there is no guarantee that a person with the APOE ε4 variant will get the disease. 
 
According to Tanzi, the three early onset genes for Alzheimer’s account for 30 percent of the genetic variants in the disease, and the vast bulk of research in the field focuses on these and APOE ε4. “So, the argument is if we’ve gone this far on four genes, and 70 percent of the puzzle remains unsolved, let’s find the remaining genes and let’s see what we can do with that,” Tanzi said.
 
The initial screen by Tanzi and his colleagues turned up five loci associated with Alzheimer’s. One of these, rs4420638, was about 11,000 bases away from APOE-ε4. Two of the four other variants — rs11159647 and rs3826656 — were also significantly associated with Alzheimer’s in the replication study. Meanwhile, a third SNP, rs179943, showed a trend toward association and a fourth was marginally associated with Alzheimer’s.
 
The researchers found the strongest association between the rs11159647 SNP on chromosome 14 and Alzheimer’s age-of-onset. According to Tanzi, this SNP doesn’t fall within any characterized genes, but it is located in the same part of the genome as PSEN1, one of the characterized early-onset Alzheimer’s genes.
 
The rs179943 SNP falls within the intron of a gene called ataxin-1, or ATXN1, on chromosome six. Mutations in ATXN1 cause another neurodegenerative disease called spinocerebellar ataxia, in which the cerebellum, brain stem, and spinal cord progressively degenerate. “Just like in Alzheimer’s, you lose neurons,” Tanzi explained. “But you lose them in the cerebellum and spinal cord.”
 
Additionally, having also identified a SNP in the CD33 gene, which codes for a cell-surface receptor with a role in fighting bacterial infections in the innate immune system, the research team is now analyzing how that infection may be linked to Alzheimer’s.
 
However, the four SNPs identified by Tanzi’s group did not appear to be associated with Alzheimer’s in the small group of African-American families tested. There are other loci that appear to influence Alzheimer’s risk in African American families, and Tanzi said he wants to work with other groups to test these and other associations in a larger set of African-American families.
 
“I think all four of them are worthy of follow-up,” Tanzi said.
 
Translating Research
 
Tanzi, who in 1983 was part of the team led by James Gusella at Massachusetts General Hospital to discover the first gene associated with a disease (Huntington’s), believes that the current study in Alzheimer’s was possible due to advanced screening tools and statistical methods.
 
“We didn’t really have the means to [this kind of study] until the Illumina and the Affymetrix types chips came out, and the HapMap databases from the Human Genome Project, and new statistical software that allows us to do these advanced family based studies,” Tanzi said.
 
In 1983, the search for a gene associated with Huntington’s disease “took two years to run a handful of variants in a handful of families,” he recalled. “It took us six months to run a million variants through 400 families. It’s amazing what 25 years will do.”
 
In the first screen, researchers used the Affy 500K chip, which covers 65 percent of the genome. However, the researchers are planning to run the same samples again using the Affy 6.0 chip, which covers 1 million SNPS and covers 95 percent of the genome.
 
With the 6.0 chip, “we expect to see the same hits but we expect to see new hits as well because we are covering more of the genome,” Tanzi said.
 
After these family-based gene associations are tested in case controlled studies, and Tanzi’s group conducts meta-analysis to further validate the associations, then the work on translating this findings into new treatments and tests can begin.
 
Tanzi noted that the gene associations his group looked at, along with previously validated genes, can be incorporated into a predictive chip within the next half-decade.
 
“As the genome-wide hits get tested in independent samples by case control and those go on to alzgene.org, and we see which ones hold up, then I think you’re going to start seeing enough hits to start going on some sort of a risk assessment chip,” he said.
 
By garnering patents on some of their gene associations, the Alzheimer’s Genome Project plans to spur this kind of development along. Although currently, none of these gene associations are patented, there are several opportunities for industry partnerships to develop targeted treatments.
 
Eisai Pharmaceuticals is looking into developing targeted therapeutics based on the findings of the original screen.
 
Meanwhile, TorreyPines Therapeutics was originally slated to partner with the Alzheimer’s Genome Project on developing novel therapeutics based on the findings of the original screen. However, these plans fell through when the La Jolla, Calif.-based firm decided to move away from the Alzheimer’s space.
 
However, Tanzi, who helped found TorreyPines, said he plans to spin out a new company, called Neurogenetic Pharmaceuticals, with TorreyPines’ Alzheimer’s assets. Neurogenetic Pharmaceuticals, while it is currently awaiting VC funding, could potentially be an industry partner translating some of these findings.
 
“The hope is to start this new company, Neurogenetic Pharmaceuticals, and between that and Eisai, find a home for a lot of these genetic hits in terms of drug development,” Tanzi noted
 
On the diagnostic side, however, Tanzi also noted that what might delay the innovation of such a chip is the absence of an algorithm that helps calculate people’s chances of getting the disease by factoring in not just risk genes, but also protective genes.
 
“We just don’t have a reliable algorithm yet that can … come up with a program to look at lifetime genetic risk based on several dozen both protective and risk conferring variants,” he said.

Filed under

The Scan

Not Yet a Permanent One

NPR says the lack of a permanent Food and Drug Administration commissioner has "flummoxed" public health officials.

Unfair Targeting

Technology Review writes that a new report says the US has been unfairly targeting Chinese and Chinese-American individuals in economic espionage cases.

Limited Rapid Testing

The New York Times wonders why rapid tests for COVID-19 are not widely available in the US.

Genome Research Papers on IPAFinder, Structural Variant Expression Effects, Single-Cell RNA-Seq Markers

In Genome Research this week: IPAFinder method to detect intronic polyadenylation, influence of structural variants on gene expression, and more.