Brain atrophy plagues more than just people with neurodegenerative disorders. Even seemingly healthy individuals lose brain matter over time. Can doctors and researchers separate brain changes due to Alzheimer's disease and other dementias from those caused by normal aging, and if so, which brain regions offer the clearest and earliest signs of disease?
Two recent papers illustrate the complexities of using brain atrophy as a marker for neurodegeneration. Researchers reported in the May 8 Journal of Neuroscience that over just one year, brain volume wanes in healthy people judged unlikely to develop Alzheimer's, although not in the same way as those with the disease. Another study, published in the May JAMA Neurology, found that dementia and AD pathology correlate with expansion of the brain ventricles, although not with hippocampal atrophy, as other research has indicated.
Because brain atrophy in Alzheimer's patients precedes clinical symptoms, researchers have proposed using it as a surrogate marker for pathology in clinical trials and longitudinal studies. For example, in October 2011, the European Medicines Agency deemed low hippocampal volume an acceptable selection marker for clinical trials of people in the early, predementia, stages of Alzheimer’s disease (see ARF related news story).
However, not all brain volume changes portend dementia. In 2009, researchers led by Anders Fjell at the University of Oslo, Norway, found that the cortex, known to shrink in patients with mild cognitive impairment and Alzheimer's disease, also diminished in older healthy people within as little as one year. Were those losses signs of disease to come? Fjell and colleagues now believe that is highly unlikely. Their follow-up data show that those volunteers remained free of mild cognitive impairment and Alzheimer's disease for at least three years.
As reported in the Journal of Neuroscience, Fjell and colleagues studied 132 elderly adults who took part in the Alzheimer's Disease Neuroimaging Initiative (ADNI) (see ARF related news story). The investigators used a variety of measures, including memory tests, cerebrospinal fluid markers of amyloid pathology, and ApoE4 status to identify participants at very low risk of developing AD. Magnetic resonance imaging (MRI) scans showed that over a year, these seemingly healthy elders lost, on average, 0.44 percent of whole brain volume. Most areas of the brain were affected, with the hippocampus and amygdala losing about 1 percent and cortical regions about 0.5 percent.
"There are a lot of changes in the brain that cannot be ascribed to incipient dementia," said Fjell. "These changes could be due to either normal aging or other pathologies that we are not able to detect."
"This paper makes the very important point that shrinkage and other changes happening in the aging brain are not just caused by Alzheimer's," said William Jagust at the University of California, Berkeley. Others were cautious to point out that AD may still be partly to blame for the shrinkage. Nick Fox of University College London, U.K., noted there is no guarantee that the changes seen are due to aging or non-Alzheimer's pathology. "At a theoretical level, a component of the net loss may be due to very early Alzheimer's disease in individuals who remain healthy for another three to four years," he said. "Even so, changes due to aging or other diseases are very important [to understand]," said Fox. "Just having some change doesn’t mean that you are destined to deteriorate and go down the Alzheimer's path."
What separates losses due to normal aging from those caused by disease? The affected part of the brain may be key. In the low-risk participants, Fjell and colleagues saw the greatest atrophy in the default-mode network (DMN), a series of interconnected brain regions that become more active when the mind is at rest and unfocused. This network is susceptible to Alzheimer's pathology (see ARF related news story). While DMN losses also occurred in patients with MCI or AD, a distinct pattern emerged in the low-risk population. They exhibited greater atrophy in the prefrontal cortex, especially the orbitofrontal region, compared to patients with mild cognitive impairment or Alzheimer's disease.
The speed of tissue loss marks another important distinction between the normal and the Alzheimer's brain. "Even if there are changes in people who age normally, the critical issue is the rate of loss," said Fox. "Once people pass from mild cognitive impairment to Alzheimer's, then the rates of whole-brain atrophy are three times those seen in the healthy brain, and for the medial temporal lobe they are even higher," he said. Fjell agreed. Although he found that hippocampal volume waned in the low-risk population, atrophy may still indicate early Alzheimer's disease as long as rate of change is considered, he said. "In Alzheimer's disease the hippocampus may lose 3 to 4 percent a year, whereas loss in a normal brain may be less than 1 percent," he noted.
The most useful region in which to track volume changes may not be in brain tissue per se, but rather in the culvert-like structures within the brain that contain cerebrospinal fluid. In JAMA Neurology, researchers led by Deniz Erten-Lyons of Oregon Health and Science University, Portland, report that swelling in the ventricles associates strongly with Alzheimer's pathology and with dementia diagnosis confirmed on autopsy.
The investigators measured brain volumes in MRI scans from 71 healthy adults who had taken part in the Oregon Brain Aging Study. All participants were 65 or older, with an average age of 94 when they died. All had an MRI scan within three years of death. On autopsy, the researchers measured neuritic plaques, neurofibrillary tangles, and blood vessel damage.
The researchers found that age, dementia diagnosis, and AD pathologies closely correlated with enlargement of the brain ventricles but not with total brain volume loss (white and gray matter). They note that ventricles can enlarge in response to various conditions. "Ventricular enlargement reflects changes in both the white matter and gray matter of the brain and can be caused by both cerebrovascular disease and Alzheimer's disease," said Erten-Lyons.
The researchers also found no correlation between AD pathology and hippocampal atrophy, which is widely reported in AD (see, e.g., Barnes et al., 2009). This was surprising, said Erten-Lyons. She suggested that hippocampal loss may have occurred earlier in the course of the disease. "Potentially, we didn't start imaging our subjects early enough to see changes in the hippocampus," she said. However, the authors did find that hippocampal shrinkage associated with cortical amyloid angiopathy (CAA), or deposits of amyloid in blood vessels. The investigators argued that the hippocampus is more susceptible to reduced blood flow that occurs in CAA.
Further complicating the story, Erten-Lyons and colleagues found that when they controlled for AD pathology, volume changes still associated with dementia and mild cognitive impairment. "This suggests that there may be other factors not measured in this study that could be contributing to brain atrophy in those with cognitive impairment," wrote the authors. Erten-Lyons suggested that genetics, lifestyle, and nutrition are worth considering. “The relationship between pathology and brain atrophy is not simple and linear," said Erten-Lyons. Neither is the distinction between aging and disease, said Jagust. "It is complicated because we don’t know where disease begins and where it ends," he said.—Susan Young
- The EMA Deems Brain Atrophy Valid Trial Selection Measure
- Wrapping Up Enrollment, ADNI 2, PPMI Focus on Earlier Disease
- Network Diagnostics: "Default-Mode" Brain Areas Identify Early AD
- Barnes J, Bartlett JW, van de Pol LA, Loy CT, Scahill RI, Frost C, Thompson P, Fox NC. A meta-analysis of hippocampal atrophy rates in Alzheimer's disease. Neurobiol Aging. 2009 Nov;30(11):1711-23. PubMed.
- Erten-Lyons D, Dodge HH, Woltjer R, Silbert LC, Howieson DB, Kramer P, Kaye JA. Neuropathologic basis of age-associated brain atrophy. JAMA Neurol. 2013 May 1;70(5):616-22. PubMed.
- Fjell AM, McEvoy L, Holland D, Dale AM, Walhovd KB, . Brain changes in older adults at very low risk for Alzheimer's disease. J Neurosci. 2013 May 8;33(19):8237-42. PubMed.