Tau imaging of patients with different kinds of dementia and a peptide microarray that detects antibody patterns in Alzheimer's disease blood samples were among the highlights of the American Academy of Neurology annual meeting, held April 26-May 3 in Philadelphia. Alzforum coverage includes the latest on the newly approved amyloid tracer florbetaben, and a snapshot of this year’s Potamkin Prize winner.
Mesulam Awarded Potamkin Prize
Marek-Marsel Mesulam of Northwestern University in Chicago wins this year’s Potamkin Prize for Research in Pick’s, Alzheimer’s, and Related Diseases. Supported by philanthropic donations from the Potamkin family of Colorado, Philadelphia, and Miami, the $100,000 prize goes to researchers who make critical advances in neurodegenerative disease research. Mesulam received the award for his contributions to the discovery and understanding of primary progressive aphasia (PPA). His work showed that this language disorder can manifest differently in people with Alzheimer’s disease and frontotemporal dementia. Mesulam received the award April 30 at the 2014 American Academy of Neurology meeting held April 26 through May 3 in Philadelphia.
Mesulam first described PPA in 1982, in a paper describing six patients who had language difficulties that progressed slowly in the absence of dementia (see Mesulam, 1982). Even that small study hinted that the disorder was quite heterogeneous, as some of the patients only had trouble recalling words, whereas one could not understand language at all. Mesulam and others built upon these initial findings over the next three decades, and now PPA is classified into three clinical subcategories: agrammatic aphasia (difficulty constructing sentences that are grammatically correct), semantic aphasia (difficulty recalling words), and logopenic aphasia (delayed or uneven speech due to slow word retrieval).
While PPA symptoms may appear in the absence of dementia, Mesulam’s work ultimately helped show that the different forms of PPA are caused by underlying neurodegenerative disease. Mesulam and others have observed profound degeneration in the left hemisphere of the brain in people with PPA, and found that around 30 percent of PPA cases had frontotemporal lobular degeneration (FTLD) with tauopathy, another 30 percent had FLTD with TDP-43 aggregate pathology, and the remaining 40 percent displayed the characteristic Aβ deposits and neurofibrillary tangles of AD. Interestingly, his studies also revealed an odd pattern of neurofibrillary tangles in AD patients with PPA as compared with others with the disease. Those with PPA had tangles that skewed toward the language centers of the neocortex, rather than the entorhinal cortex (see Gefen et al., 2012, and Mesulam et al., 2014). While the reasons behind the variety of PPA manifestations aren’t known, Mesulam reported that people with learning disabilities have a heightened risk of developing PPA (see Feb 2008 news story). He hypothesized that damage to neuronal networks, rather than brain atrophy alone, may account for PPA symptoms (see Nov 2010 conference coverage).—Jessica Shugart
- Primary Progressive Aphasia—Learning Disabilities Point to Early Susceptibility
- Indianapolis: Neuroimaging Opens Window to Disease, Better Diagnosis
- Mesulam MM. Slowly progressive aphasia without generalized dementia. Ann Neurol. 1982 Jun;11(6):592-8. PubMed.
- Gefen T, Gasho K, Rademaker A, Lalehzari M, Weintraub S, Rogalski E, Wieneke C, Bigio E, Geula C, Mesulam MM. Clinically concordant variations of Alzheimer pathology in aphasic versus amnestic dementia. Brain. 2012 May;135(Pt 5):1554-65. PubMed.
- Mesulam MM, Weintraub S, Rogalski EJ, Wieneke C, Geula C, Bigio EH. Asymmetry and heterogeneity of Alzheimer's and frontotemporal pathology in primary progressive aphasia. Brain. 2014 Apr;137(Pt 4):1176-92. Epub 2014 Feb 25 PubMed.
- Mesulam MM. Primary progressive aphasia and the language network: the 2013 H. Houston Merritt Lecture. Neurology. 2013 Jul 30;81(5):456-62. PubMed.
Three’s Company: Florbetaben Approved, Excludes AD Diagnosis
A third amyloid imaging agent has joined the ranks of those approved for clinical use. Piramal Imaging’s PET tracer florbetaben, now rechristened Neuraceq, got the nod from the European Medicines Agency on February 20, and from the U.S. Food and Drug Administration on March 20. At the 66th annual meeting of the American Academy of Neurology, held 26 April to 3 May in Philadelphia, Marwan Sabbagh of Banner Sun Health Research Institute in Sun City, Arizona, presented the latest data from the Phase 3 trial that helped clinch florbetaben’s approval. Previously, Sabbagh detailed early data from this trial at the 2012 AAN meeting (see May 2012 news story). The study suggests that a negative Neuraceq scan excludes a diagnosis of Alzheimer's disease.
The Phase 3 florbetaben study enrolled 205 elderly patients with Alzheimer’s or other dementias from Australia, France, Germany, Japan, and the United States. In 2012, Sabbagh reported autopsy data from 31 participants; that number has now grown to 74. Sabbagh noted that this is the largest autopsy study to date for any amyloid imaging agent and the only autopsy study to correlate tracer binding with amyloid load in specific brain regions. Neuraceq demonstrated a sensitivity of 98 percent and specificity of 89 percent for detecting amyloid deposits, similar to the efficacy of the other approved tracers. Because one of the most useful clinical applications for amyloid imaging may be to rule out a diagnosis of Alzheimer’s disease, the researchers also analyzed the negative predictive value, which was 96 percent. This makes florbetaben a valuable tool for differential diagnosis, claimed Sabbagh.
The first approved amyloid tracer was Eli Lilly and Company’s Amyvid (florbetapir) (see Apr 2012 news story), followed last fall by GE Healthcare’s Vizamyl (flutemetamol) (see Nov 2013 news story). All three agents have the same label; they can determine whether amyloid is present in order to support or rule out a diagnosis of AD, but cannot diagnose Alzheimer’s on their own. The FDA requires radiologists and clinicians to complete reader training programs before using any of the tracers. Only Vizamyl is approved to analyze plaque density using false-color brain scans; the other two must be read in black and white.—Madolyn Bowman Rogers
- Me, Too: Florbetaben, Flutemetamol Look Good in Trial
- FDA Approves Amyvid for Clinical Use
- FDA Approves a Second Amyloid Imaging Agent
PET Tracers Enlighten at Neurology Conference
Imaging technologies have transformed the study of neurodegenerative diseases by giving scientists a glimpse at what is happening inside the brains of living patients. At the 66th annual meeting of the American Academy of Neurology, held 26 April to 3 May in Philadelphia, researchers touted the potential of several positron emission tomography (PET) tracers to improve diagnoses. A tau tracer under development looks promising for characterizing several types of frontotemporal lobar degeneration (FTLD), while amyloid imaging distinguishes effectively between FTLD and Alzheimer’s disease in young dementia patients, speakers said. Meanwhile, fluorodeoxyglucose (FDG) PET was confirmed once again to reveal dampened brain metabolism in people who carry one or more ApoE4 alleles, the main genetic risk factor for sporadic AD, though surprisingly, this effect occurred in people young and old, with or without amyloid deposits.
“We are in a revolutionary era of neurodegenerative disease research,” Brad Dickerson Massachusetts General Hospital, Boston, told Alzforum. “[New imaging technologies] will completely change the way we approach these diseases.”
These days the biggest buzz surrounds the potential of tau tracers, since tau pathology marks numerous neurodegenerative diseases. Several companies are developing such tracers (see Jan 2013 conference story; Aug 2013 conference story; Sep 2013 news story). At AAN, Dickerson reported preliminary results from imaging with Eli Lilly and Company’s T807 in about 15 patients with various tauopathies. These included behavioral-variant FTLD, frontotemporal aphasias, and the movement disorder progressive supranuclear palsy (PSP).
In all cases, the brain regions that T807 lit up most strongly matched the regions with the greatest atrophy and hypometabolism in that particular disease, Dickerson said. That means these three major brain imaging modalities used in living patients are aligned. In behavioral-variant FTLD patients, the frontal, insular, and anterior temporal cortices gave the strongest T807 signal, whereas Broca’s aphasia patients bound the most tracer in the inferior frontal and middle temporal gyri. These patterns correspond to the areas of greatest degeneration in each disease. Moreover, they are distinct from the profile seen in Alzheimer’s, where T807 lights up temporal regions. The findings suggest that T807 binding reflects tau deposits, but final confirmation will have to come from postmortem studies, Dickerson said.
How early in disease might a tau signal appear? Dickerson described one cognitively healthy participant who carried the tau mutation P301L, which causes frontotemporal dementia. Although the man was at least 10 years away from his expected age of symptom onset, his T807 signal was mildly elevated in the same regions of the brain that light up in symptomatic patients. The results hint that tau tracers could help identify people at preclinical disease stages.
One concern emerged from the data. Only about 40 percent of clinically diagnosed FTLD cases are tauopathies; many others are characterized instead by TDP-43 deposits. Ideally, a tau tracer would allow scientists to distinguish between these proteins. However, in this study, a participant who carried a progranulin mutation, which causes TDP-43 pathology, had an elevated T807 signal. “That was a major surprise,” Dickerson told Alzforum. This raises the question of whether T807 might react non-specifically with TDP-43; alternatively, this patient might have concurrent tau pathology. In ongoing work, Dickerson is scanning more people with different predicted pathologies to try to nail down how specific the tracer is. Eventually, he will validate in-vivo findings against autopsy results.
Despite the questionable specificity, other researchers found these results exciting. “In patients with PSP and tau mutations, the data looks very promising,” said Gil Rabinovici at the University of California, San Francisco.
A patient clinically diagnosed with Alzheimer’s was negative for amyloid imaging (left scans), but positive by FDG PET (right scans). Postmortem pathology identified the disease as corticobasal degeneration. [Image courtesy of Gil Rabinovici.]
FTLD cases can be difficult to distinguish from early onset AD. Amyloid imaging may help clinicians make this distinction, Rabinovici said in his AAN talk. Currently, many clinicians use FDG PET for differential diagnosis, because insurers such as the Centers for Medicare and Medicaid Services (CMS) reimburse for this scan. By contrast, CMS does not yet cover clinical use of amyloid imaging (see Jul 2013 news story). Many researchers believe amyloid imaging would do a better job than FDG PET of distinguishing these diseases. Rather than compare patterns of glucose use, as FDG does, amyloid tracers detect a specific, defining pathology present in AD that is typically absent in FTD. Researchers are currently gathering evidence to try to persuade the CMS to change its stance (see Oct 2013 news story).
For his part, Rabinovici compares the diagnoses made after FDG and PIB amyloid scans in patients being treated at memory clinics. Previously, he reported that amyloid imaging was more reliable between different raters than FDG PET, as well as more sensitive and equally specific, in a cohort of about 100 patients (see Rabinovici et al., 2011). In response, CMS noted that this data was promising, but lacked autopsy confirmation. At AAN, Rabinovici reported autopsy results from the first 44 patients in this cohort. Twenty-eight turned out to have had FTLD, 12 AD, 3 mixed FTLD and AD, and one had had prion disease.
Amyloid imaging correctly identified all the amyloid and mixed-pathology patients as having AD, and ruled out the prion patient. In the FTLD group, however, four patients had neuritic plaques that showed up on PIB PET scans, resulting in an AD diagnosis. Pathologists judged that these plaques were unlikely to have contributed to dementia, and called the cases FTLD. This gave amyloid imaging a sensitivity of 100 percent and a specificity of about 90 percent for distinguishing AD from FTLD. By contrast, FDG reads misclassified several AD and FTLD cases, and had an overall sensitivity of about 86 percent and specificity of 79 percent. All numbers favored amyloid imaging, but the cohort was too small to show statistical significance. Nonetheless, Rabinovici said the data indicate that amyloid imaging is at least as good as FDG PET for distinguishing FTLD and AD.
“The data definitely support the reimbursement of amyloid imaging in the FTLD versus AD scenario,” Rabinovici told Alzforum. Misdiagnosis can be harmful, because AD medications such as cholinesterase inhibitors and memantine have been found to worsen outcomes for FTLD patients in clinical trials, Rabinovici noted. However, he stressed that amyloid scans should be interpreted by experts, and only after a comprehensive clinical evaluation has been conducted. “Scans can be misinterpreted. A positive amyloid scan doesn’t rule out FTD, for example. I don’t think this technique is ready for widespread use in primary care,” he said.
While FDG PET may become passé for differential diagnosis of FTD versus AD, this technique has many other applications. At AAN, David Knopman at the Mayo Clinic in Rochester, Minnesota, reported using it to study metabolic changes in more than 800 cognitively normal adults in the Mayo Clinic Study of Aging. His data were published March 11 in Neurobiology of Aging. He found a modest decline with age in most cortical and sub-cortical brain regions. Previous studies had conflicted on this point, with some showing no age-related changes. The current study is much larger than most prior studies and used more rigorous methods, making the results more definitive, Knopman claimed.
Knopman also wanted to know how an ApoE4 allele would affect metabolism. The 209 study participants who carried at least one ApoE4 allele had lower glucose uptake than non-carriers did in the posterior cingulate, precuneus, lateral parietal, and inferior temporal regions. Intriguingly, these regions typically atrophy in AD. Knopman’s data confirm previous FDG PET findings in ApoE4 homozygotes by Eric Reiman and colleagues at Banner Alzheimer’s Institute, Phoenix, and others (see, e.g., Reiman et al., 1996; Reiman et al., 2004; Lehmann et al., 2014). The new data also extends earlier FDG PET observations on ApoE. In particular, it shows that age has no bearing. The ApoE4 effect looms as large in 30-something adults as in people in their 70s. Knopman said this surprised him. The magnitude of the drop in metabolism seen in carriers is about one-fourth of that seen in AD patients versus age-matched controls.
Notably, glucose metabolism in ApoE4 carriers was lower regardless of whether people had amyloid in their brain, agreeing with some previous data (see Dec 2012 news story). This suggests that ApoE4 may contribute to AD through some other mechanism, Knopman said. In future work, Knopman will compare FDG scans with both amyloid and tau imaging at multiple age ranges to try to gain more insight into the pathological mechanisms.
Other researchers found the results intriguing. Rabinovici wrote to Alzforum, “While ApoE4 is clearly a risk factor for amyloid aggregation, there is increasing evidence that ApoE4 also increases the risk for AD via a multitude of Aβ-independent pathways, some of which traverse the entire lifespan (see excellent review by Huang and Mucke, 2012, and interesting model proposed by Jagust and Mormino, 2011). This has major implications as the field develops strategies for AD treatment and prevention in E4 positive individuals.” (See full comment below.)—Madolyn Bowman Rogers
- HAI—Spotlight on Tau Tracers at Human Amyloid Imaging Meeting
- Tau Tracers Shine at Boston Conference
- Tau Tracer May Light Up All Tauopathies
- Coverage Denial For Amyloid Scans Riles Alzheimer’s Community
- Alzheimer’s Community Mobilizes to Show Benefits of Amyloid Scans
- Rabinovici GD, Rosen HJ, Alkalay A, Kornak J, Furst AJ, Agarwal N, Mormino EC, O'Neil JP, Janabi M, Karydas A, Growdon ME, Jang JY, Huang EJ, Dearmond SJ, Trojanowski JQ, Grinberg LT, Gorno-Tempini ML, Seeley WW, Miller BL, Jagust WJ. Amyloid vs FDG-PET in the differential diagnosis of AD and FTLD. Neurology. 2011 Dec 6;77(23):2034-42. PubMed.
- Reiman EM, Caselli RJ, Yun LS, Chen K, Bandy D, Minoshima S, Thibodeau SN, Osborne D. Preclinical evidence of Alzheimer's disease in persons homozygous for the epsilon 4 allele for apolipoprotein E. N Engl J Med. 1996 Mar 21;334(12):752-8. PubMed.
- Reiman EM, Chen K, Alexander GE, Caselli RJ, Bandy D, Osborne D, Saunders AM, Hardy J. Functional brain abnormalities in young adults at genetic risk for late-onset Alzheimer's dementia. Proc Natl Acad Sci U S A. 2004 Jan 6;101(1):284-9. PubMed.
- Lehmann M, Ghosh PM, Madison C, Karydas A, Coppola G, O'Neil JP, Huang Y, Miller BL, Jagust WJ, Rabinovici GD. Greater medial temporal hypometabolism and lower cortical amyloid burden in ApoE4-positive AD patients. J Neurol Neurosurg Psychiatry. 2013 Aug 21; PubMed.
- Huang Y, Mucke L. Alzheimer mechanisms and therapeutic strategies. Cell. 2012 Mar 16;148(6):1204-22. PubMed.
- Jagust WJ, Mormino EC. Lifespan brain activity, β-amyloid, and Alzheimer's disease. Trends Cogn Sci. 2011 Nov;15(11):520-6. PubMed.
Blood Patterns: Prowling for Alzheimer’s Clues in Plasma
Undeterred by past setbacks, researchers continue to pursue a blood test for Alzheimer’s disease. At the 66th annual meeting of the American Academy of Neurology, held 26 April to 3 May in Philadelphia, one presenter described a microarray technique that picks up a particular antibody signature present in the blood of people with AD, but not in most healthy controls. Researchers will test the method to see if it can predict disease progression. Other groups focused on plasma Aβ, which in past research has correlated poorly with amyloid in the brain. At AAN, however, researchers reported that blood Aβ does associate with vascular disease, a risk factor for AD. Moreover, a highly toxic form of Aβ, pyroglutamate Aβ, abounds in the blood of middle-aged Down’s syndrome patients. Together, the results suggest that the plasma may harbor signs of AD, after all.
Most attempts to develop a blood test for AD have focused on plasma proteins or lipids (see Oct 2007 news story; Mar 2014 news story; Jun 2013 Webinar). At AAN, Lucas Restrepo of the University of California, Los Angeles, described a different approach. He applied an antibody-based test developed by Stephen Johnston and Neal Woodbury at Arizona State University in Phoenix. In this method, researchers add a drop of blood to a microarray dotted with 10,000 random peptides, then look for antibodies that bind to the chip. Antibodies are more abundant than other blood proteins, making this test more sensitive than typical plasma assays, Johnston told Alzforum.
The researchers used blood samples from 44 AD patients and 53 age-matched controls from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). As previously reported, all AD samples displayed a similar pattern, with antibodies binding to a shared subset of peptides. A small number of these peptides mimic segments of Aβ, but for the others, it is unknown what native antigens they resemble. Blood from AD patients also failed to bind some peptides typically bound by antibodies from healthy controls. Altogether, about 200 peptides are differentially bound by AD and control blood, Johnston said (see image below). Three control samples exhibited the AD antibody binding pattern and the test misclassified them as AD; the rest were negative. This gave the test a sensitivity of 100 percent and specificity of 94 percent for AD diagnosis (see Restrepo et al., 2013). The results agree with previous research on distinctive autoantibodies in AD blood (see Aug 2011 news story).
The authors have now completed a follow-up study in 60 autopsy-confirmed AD patients and 60 age-matched controls from the Rush Memory and Aging Project. The researchers used a slightly different set of peptides in the array, but again saw a distinctive AD signature, with similar sensitivity and specificity to the earlier version. They plan to test the diagnostic in the Alzheimer’s Prevention Initiative’s Colombian study, which enrolls people from families with inherited AD. They will also test blood from people with mild cognitive impairment in other populations to see if their diagnostic can flag people with preclinical Alzheimer’s, Johnston said.
Johnston uses the same microarray to test for various types of cancer, as well as infectious diseases. In 2010 he co-founded a company, Health Tell, to commercialize the technology. The company has made a new chip that contains 350,000 random peptides and is more sensitive and accurate at detecting disease, Johnston claimed.
Other groups are still trying to base diagnostic tests on Aβ in the blood, though no consistent relationship between it and AD risk has been shown (see Jan 2011 news story; Apr 2012 news story). Instead, researchers led by Sara Kaffashian at INSERM, Paris, looked at the relationship between plasma Aβ and markers of cerebrovascular disease. They followed more than 1,000 non-demented French participants over the age of 65 for four years.
At AAN, Kaffashian reported that people who had the lowest Aβ40 at baseline and lowest Aβ42 at follow-up accumulated the greatest number of lesions in the white matter of their brain. Considered a marker of cerebrovascular disease, damaged white matter associates with AD risk, cognitive impairment, and brain amyloidosis, although what is cause and what is effect is unclear (see, e.g., Brickman, 2013; Noh et al., 2014). Moreover, the nearly 100 participants who had enlarged perivascular spaces at baseline also had the lowest plasma Aβ42. Perivascular spaces are the small clefts around blood vessels entering the brain that act as drainage pathways for interstitial fluid. Enlarged perivascular spaces correlate with lesions and white-matter damage in the brain and an elevated risk of stroke (see Rouhl et al., 2008; Doubal et al., 2010).
“Our findings support a relationship between vascular [damage] and neurodegenerative mechanisms in cerebral aging,” Kaffashian wrote to Alzforum. When the brain’s small vessels are diseased, less blood reaches brain tissue, which may reduce the clearance of Aβ along perivascular drainage pathways, she noted. That results in less Aβ in the blood. In future studies, she will look at whether those participants who have markers of cerebrovascular disease along with low-plasma Aβ also deposit more amyloid in brain or blood vessels. If so, it would strengthen the idea that low-plasma Aβ could serve as a biomarker, Kaffashian believes. The vascular contribution to AD has become a hot topic of research (see May 2014 conference story part 1 and part 2).
By contrast, researchers led by Pankaj Mehta and David Miller at the New York State Institute for Basic Research in Developmental Disabilities, Staten Island, chose to examine pyroglutamate Aβ (pyroGluAβ), a truncated form of the protein that aggregates readily. Previous studies have not found it in blood, perhaps because the peptide clumps together so quickly (see DeMattos et al., 2012).
Mehta and Miller turned to Down’s syndrome patients, who carry an extra copy of the amyloid precursor protein (APP) gene and have high levels of Aβ products in their blood throughout life. The researchers generated an antibody to pyroGluAβ in rabbits. They reported that it detects as little as 100 pg/ml pyroGluAβ and does not cross-react with Aβ40 or Aβ42. With this tool, the authors screened plasma of 35 Down’s patients whose average age was 47. They found the concentration of pyroGluAβ to be eightfold higher than in 32 age-matched controls (about 2.5 ng/ml versus 0.29 ng/ml). The authors do not know if this plasma pyroGluAβ originates from brain or periphery. At their age, all the Down’s patients in the study would have accumulated Aβ plaques in the brain (see Jun 2011 news story).
In future studies, the authors will gather longitudinal data to see whether plasma pyroGluAβ levels rise or fall over time as amyloidosis increases, and whether these changes relate to cognitive decline. Mehta noted that some participants with Down’s syndrome in this study had relatively low levels of pyroGluAβ in the blood compared to their peers. If that correlates with dementia severity, this marker could help identify people at the greatest risk for cognitive decline. Though dementia can be difficult to diagnose in people with Down’s syndrome, there are cognitive tests for this purpose (see Jun 2011 news story). Mehta also wants to adapt the method for AD patients. Like other researchers who presented at this annual meeting, Mehta has not given up on finding a blood test for Alzheimer’s.—Madolyn Bowman Rogers
- A Blood Test for AD?
- Do Lipids Hold the Key to Blood-Based Alzheimer’s Test?
- "Autoantibody-omics" Yields Potential Blood Biomarkers for AD
- In the Blood: What Can Plasma Aβ Tell Us About Alzheimer’s?
- Plasma Aβ as Biomarker: Not a Lost Cause, After All?
- It’s Not All About You, Neurons. Glia, Blood, Arteries Shine at Symposium
- Fluid Markers and Imaging Back Idea of Breached Blood-Brain Barrier
- Research Brief: Imaging Shows AD Pathology in Down’s Syndrome Brains
- Restrepo L, Stafford P, Johnston SA. Feasibility of an early Alzheimer's disease immunosignature diagnostic test. J Neuroimmunol. 2013 Jan 15;254(1-2):154-60. PubMed.
- Brickman AM. Contemplating Alzheimer's disease and the contribution of white matter hyperintensities. Curr Neurol Neurosci Rep. 2013 Dec;13(12):415. PubMed.
- Noh Y, Seo SW, Jeon S, Lee JM, Kim JH, Kim GH, Cho H, Yoon CW, Kim HJ, Ye BS, Kim ST, Choe YS, Lee KH, Kim JS, Ewers M, Weiner MW, Lee JH, Werring DJ, Kang DR, Kim CS, Na DL. White matter hyperintensities are associated with amyloid burden in APOE4 non-carriers. J Alzheimers Dis. 2014;40(4):877-86. PubMed.
- Rouhl RP, van Oostenbrugge RJ, Knottnerus IL, Staals JE, Lodder J. Virchow-Robin spaces relate to cerebral small vessel disease severity. J Neurol. 2008 May;255(5):692-6. Epub 2008 Feb 21 PubMed.
- Doubal FN, MacLullich AM, Ferguson KJ, Dennis MS, Wardlaw JM. Enlarged perivascular spaces on MRI are a feature of cerebral small vessel disease. Stroke. 2010 Mar;41(3):450-4. Epub 2010 Jan 7 PubMed.
- Demattos RB, Lu J, Tang Y, Racke MM, DeLong CA, Tzaferis JA, Hole JT, Forster BM, McDonnell PC, Liu F, Kinley RD, Jordan WH, Hutton ML. A plaque-specific antibody clears existing β-amyloid plaques in Alzheimer's disease mice. Neuron. 2012 Dec 6;76(5):908-20. PubMed.