A previous epidemiological study in Northern Europe showed that the A673T mutation (Icelandic mutation) in the amyloid precursor protein gene (APP) can protect against Alzheimer's disease (AD). While the effect of the A673T mutation on APP processing has been investigated primarily in vitro, its in vivo impact has not been evaluated. This is mainly because most existing AD mouse models carry the Swedish mutation. The Swedish and Icelandic mutations are both located near the β-cleavage site, and each mutation is presumed to have the opposite effect on β-cleavage. Therefore, in the AD mouse models with the Swedish mutation, its effects could compete with the effects of the Icelandic mutation. Here, we introduced the A673T mutation into App knock-in mice devoid of the Swedish mutation (AppG-F mice) to avoid potential deleterious effects of the Swedish mutation and generated AppG-F-A673T mice. APP-A673T significantly downregulated β-cleavage and attenuated the production of Aβ and amyloid pathology in the brains of these animals. The Icelandic mutation also reduced neuroinflammation and neuritic alterations. Both sexes were studied. This is the first successful demonstration of the protective effect of the Icelandic mutation on amyloid pathology in vivo. Our findings indicate that specific inhibition of the APP-BACE1 interaction could be a promising therapeutic approach. Alternatively, introduction of the disease-protective mutation such as APP-A673T using in vivo genome editing technology could be a novel treatment for individuals at high risk for AD, such as familial AD gene mutation carriers and APOE ε4 carriers.Significance statement The A673T mutation (Icelandic mutation) in the APP gene can protect against AD. The effect of the A673T mutation on amyloid pathology has not been evaluated in vivo. Utilizing a new AD mouse model that we have recently developed, we show that the APP-A673T attenuates amyloid pathology in vivo. We demonstrate that its protective effects are exerted by inhibiting β-cleavage and reducing the production of Aβ in the brain. Furthermore, we reveal that the Icelandic mutation also reduced neuroinflammation and neuritic alterations. Our findings indicate that specific inhibition of the APP-BACE1 interaction or introduction of protective variants via in vivo genome editing could be a promising therapeutic approach.