Five years ago, Icelandic researchers uncovered a mutation in the amyloid precursor protein (APP) that protected carriers against developing AD, ostensibly by decreasing production of amyloid-beta (Aβ) peptides. Now, researchers led by Mikko Hiltunen at the University of Eastern Finland in Kuopio report that Finnish men with the protective mutation A673T have on average 28 percent less plasma Aβ than those without the gene. The results appeared in the Annals of Neurology online on May 26.

“This is a very important finding,” said Christian Haass of the German Center of Neurodegenerative Diseases in Munich. “It really shows what these protective mutations are doing—they reduce amyloid production over the lifetime and prevent the disease.” Haas said the work provides clear support for the amyloid-cascade hypothesis of AD, which puts amyloid production at the head of a pathological sequence of events leading to disease. 

The work also suggests a path to prevention, indicating that a less-than-total blockade of Aβ production, if maintained through life, yields benefits later on.

The Finnish group is the first to publish evidence for decreased plasma Aβ in A673T carriers, but Kári Stefánsson of deCODE Genetics, Reykjavik, Iceland, told Alzforum his group has similar, unpublished results. Their data on about 200 mutation carriers shows a 15 percent reduction in plasma Aβ, Stefánsson said. “This result is real and it is significant,” he told Alzforum. 

Stefánsson’s deCODE team was the first to identify the salubrious effects of the APP A673T mutation, which protects carriers from AD and age-related cognitive decline (see Jul 2012 news). The mutation renders APP resistant to β-secretase (BACE) cleavage, and lowers Aβ output from neurons in vitro. The mutation appears in a lucky few in Iceland and Scandinavia, but is practically missing from other parts of the world (Feb 2015 news). 

For his study, Hiltunen drew on a Finnish cohort of 10,000 men between 45 and 70 years old, which had been assembled to assess how type II diabetes affected comorbid diseases, including Alzheimer’s. The large number of participants gave researchers a unique opportunity to study A673T. The mutation was present in only 47 of 8,629 men who were genotyped. Hiltunen and colleagues measured Aβ42 and Aβ40 in plasma samples of these carriers and 141 noncarriers matched for age, BMI, and ApoE status. On average, APP A673T carriers had 28 percent lower concentrations of both peptides compared with their matched controls. The exact values reported were 109.5 versus 152.7 pg/ml for Aβ40, and 24.9 versus 34.4 pg/ml for Aβ42 in carriers compared to noncarriers, respectively. The 42/40 ratio was not different, and other Aβ species were not measured.

In the same study, the researchers looked at a second variant related to AD risk, also discovered by the Iceland group, that has the opposite effect of increasing Aβ production in vitro. They expected that carriers of the ABCA7 transporter rs200538373-C variant might show increased plasma Aβ; however, they saw no difference at all.

The A673T effect was surprising because it was so strong, said Hiltunen. “It clearly supports the idea that this gene is protective,” he told Alzforum.  He said his group has additional, unpublished genetic data bolstering the case that A673T protects against AD. More recent sequencing data of the APP gene in 1,600 Finnish AD cases and 14,000 controls found no instances of the mutation in any of the cases, but did find the expected rate of ~0.5 percent in the controls, he said.

Beyond reducing the risk of AD, lifelong Aβ suppression did not appear to affect health otherwise. The researchers found no significant differences between carriers and non-carriers in any of 248 cardiovascular or metabolic parameters, or in markers of general health status.

“We looked at blood pressure, metabolic values, type II diabetes-related factors, and we did not see adverse effects on anything. That tells us that 30 percent less Aβ is not harmful,” said Hiltunen.

The results should be reassuring for pharma as companies pursue BACE inhibitors for prevention of AD, Hiltunen said, “Now they know they can start as early as possible to try to reduce Aβ.” 

If 15 or 20 percent reduction of plasma Aβ is enough to stave off AD, that also increases the chances of success in preventive trials using BACE inhibitors, said Haass. “When people go in with a hammer to try to block an enzyme entirely, that causes side effects by hitting physiological substrates. If we want to start early with people at risk, like ApoE carriers, or people reaching a certain age, a low dose gives a much better chance of avoiding side effects and treating for a long time,” he said.

The results also raise the possibility of using plasma Aβ reduction as a readout for effective BACE inhibition. “You could dose the treatment until you have achieved an Icelandic mutation-like reduction” said Henrik Zetterberg, University of Gothenburg, Sweden.

But Stefánsson cautions that it is hard to extrapolate from the mutation data to a therapeutic Aβ target concentration. “What is the coefficient that converts lifelong reduction of 15 percent to reduction starting at age 50? We just don’t know,” he said.

Hiltunen is planning to extend his study to look at Aβ where it counts, that is, in the cerebrospinal fluid and brains of carriers and noncarriers, by recruiting participants for lumbar puncture and PiB-PET scans.

Zetterberg wants to see more study of plasma Aβ, too. “This data is clear and convincing, and to me it’s hopeful. It's a wonderful paper that should simulate a lot of discussion,” he told Alzforum. “I hope this will encourage other research groups to publish additional data sets on the topic of plasma Aβ levels in A673T mutants. This would help the field to assess the robustness of the findings.”—Pat McCaffrey.


  1. Replications are important in medical science, and this study essentially replicates the paper from deCode on this Alzheimer-protective mutation published five years ago. This replication is particularly important because its result points at a therapeutic strategy for Alzheimer’s disease.

    The fact that, previously, the Aβ-lowering effect of this APP mutation had only been seen in an Icelandic population was worrying. This new data in a Finnish population is therefore a welcome result.

  2. This study is an important contribution to the validation of the amyloid hypothesis in vivo. Since the identification of this Icelandic mutation, it has been debated whether the apparently protective effect of the mutation is primarily due to its lowered Aβ production or due to its reduced aggregation propensity or due to both. Evidence for both mechanisms had been described in vitro. With this new study, there is finally evidence in vivo in plasma, at least for the reduced Aβ production.

    It would be very informative to additionally know in the future whether the same reduction is seen in the CSF of these mutation carriers.

Make a Comment

To make a comment you must login or register.


Mutations Citations

  1. APP A673T (Icelandic)

News Citations

  1. Protective APP Mutation Found—Supports Amyloid Hypothesis
  2. Let’s All Move to Iceland: Anti-Dementia Allele Rare in U.S.

Further Reading


  1. . BACE1 Cleavage Site Selection Critical for Amyloidogenesis and Alzheimer's Pathogenesis. J Neurosci. 2017 Jul 19;37(29):6915-6925. Epub 2017 Jun 16 PubMed.

Primary Papers

  1. . Decreased plasma β-amyloid in the Alzheimer's disease APP A673T variant carriers. Ann Neurol. 2017 Jul;82(1):128-132. PubMed.