There was nothing genetic in the first pass that the researchers could say unequivocally was associated with Alzheimer's disease.
But a second pass, this one on even more patients and controls, uncovered a strong association between a mutation on the PCDH11X gene and Alzheimer's disease.
"It was significant enough to convince us we had something real," Younkin said.
The findings were given added weight because the researchers examined autopsies on participants who had died to confirm the diagnosis of Alzheimer's. Clinical diagnoses are about 90 percent accurate anyway, Porsteinsson said.
Women with two copies of the gene had a 75 percent increased risk for Alzheimer's. Women with only one copy of the gene had a 26 percent increased risk, while men with one copy had an 18 percent increased risk.
The gene is part of a family of genes linked with the nervous system and are involved in cell adhesion.
"We're basically looking at a finding that seems to encode a gene that has something to do with central nervous system development," Porsteinsson said. "And Alzheimer's is ultimately a central nervous system disease."
It's unclear how the gene may actually work to increase the risk of Alzheimer's. Once that is known, however, the search for new treatments could be accelerated, the researchers said.
"It's trite but true -- the reason we all search for genes is that every one opens a door to identify people at risk, and every one opens a door to identify potential therapies if we can understand how it works," Younkin said.
But first, these findings need to be replicated. "This just means a lot of work ahead," Porsteinsson said.
To learn more, visit the Alzheimer's Association.
All rights reserved