Alzheimer's Brain Markers Vary with Age

by John Gever for MedPage Today:Relationships between inflammation markers and dementia risk in the "oldest old" appear to differ from those in individuals a few years younger, suggesting that pathological processes may change with age, a researcher said here.Previous studies have shown strong positive correlations between inflammation markers and the risk of new-onset cognitive impairment in people in their late 60s and early 70s, explained Andrea Metti, a doctoral student at the University of Pittsburgh.But data from a prospective study of older women showed no such relationship, she told attendees at the Alzheimer's Association International Conference. In fact, among participants whose interleukin (IL)-6 receptor levels increased during the 10-year study or that started high and remained high, the risk of developing dementia was actually significantly lower.She said the findings suggested that the underlying mechanisms of dementia in the oldest old may differ from those producing cognitive impairments at younger ages.Many studies have shown that neurodegeneration is at least partly an inflammatory process, marked by cytokine release and/or immune cell activation. But the specifics of how these processes contribute to neuron death and brain atrophy in different disorders are complex and still poorly understood.In the current study, Metti and colleagues used data from the 20-year Study of Osteoporotic Fractures. Among the many types of data collected were serum levels of inflammatory markers -- including IL-6, soluble receptors for the cytokine (IL-6 sR), and tumor necrosis factor soluble receptors 1 (TNF sR1) -- in about 10% of the study's 10,000 participants.The blood samples were collected at year 10 of the study, when participants were 78 on average (SD 2.8). A 380-participant subset of this group had a second analysis of serum inflammatory markers at year 16. Cognitive outcomes were determined at years 10 and again at year 20, when participants had a mean age of 88.After adjusting for age, education level, cystatin-C levels, and use of statins and nonsteroidal anti-inflammatory drugs, no relationships were seen at year 10 between cognitive status (mild impairment or overt dementia) and levels of IL-6, IL-6 sR, or TNF sR1.At year 16, however, individuals with high levels of IL-6 sR, relative to those with low levels, were at only half the risk for a diagnosis of dementia (adjusted odds ratio 0.47, 95% CI 0.25 to 0.87), although no such effect was seen with the IL-6 cytokine or TNF sR1.Metti and colleagues also looked at whether the trajectories of inflammation marker levels from year 10 to year 16 were related to dementia risk.High levels of IL-6 sR at year 16 were associated with dramatically lower rates of dementia at year 20 whether these levels were already high at year 10 or not. In the unadjusted data, the odds ratios for dementia were both 0.38 for participants with consistently high levels and for those whose levels went from low to high, relative to women with low IL-6 sR levels at both time points.After adjustments, the odds ratios in the low-high and high-high IL-6 sR groups were both below 0.30 (all P<0.05). Levels of the other inflammation markers were not associated with dementia risk.Metti suggested that transport of cytokines across the blood-brain barrier may provide a mechanism to connect inflammatory markers and cognitive outcomes. IL-6 sR are known to have a strong regulatory role, she said, and their binding to the IL-6 cytokine may, in the oldest old, counteract damaging effects in the brain.She cautioned that the findings were based on relatively few participants and that at least one important potential confounder, APOE genotype, was not measured.Session moderator Maria Glymour, ScD, of Harvard School of Public Health, told MedPage Todaythat the findings were interesting but that they should be replicated in other samples before firm conclusions could be drawn.Source: http://www.medpagetoday.com/MeetingCoverage/AAIC/33840?