In the March 27 JAMA Neurology online, scientists led by Matti Sillanpää, University of Turku, Finland, report that some middle-aged adults who had developed epilepsy as children had more amyloid plaques in their brains than controls of the same age. This was just as true when the disease was in remission as when patients were still experiencing seizures. ApoE4 allele carriers were especially vulnerable. The results yield new insights about the long-term consequences of childhood epilepsy, and help explain why it may lead to cognitive disorders such as Alzheimer’s disease (AD).

“There’s been a lot of research into the mechanisms by which β-amyloid deposition and tau aggregation might promote seizures,” said Zachary Miller, University of California, San Francisco, who was not involved in the research. This study suggests the relationship goes two ways, he said. “Seizure activity itself may make a person vulnerable to the disease.” Miller found the paper exciting because it supports the notion that early life differences shape the susceptibility to neurodegenerative disease later on.

Amyloid in Epilepsy: PiB uptake is higher in people who had epilepsy as children than in controls (orange scale). In ApoE4 carriers, the increase was more widely distributed (blue scale). [© 2017 American Medical Association. All rights reserved.]

Previous studies have reported a greater incidence of dementia and AD in people with epilepsy (see Breteler et al., 1991; Breteler et al., 1995). Vice versa, AD patients have an increased risk of seizures (Amatniek et al., 2006). How might the two disorders be related? In the 1990s, researchers analyzed brain tissue surgically removed from people with otherwise unmanageable temporal lobe epilepsy. They found unusually high levels of amyloid precursor protein and of amyloid plaques (see Sheng et al., 1994; Mackenzie and Miller, 1994). It was unclear if the amyloid deposits were limited to epileptic loci, or if they occurred more generally throughout the brain, since there was no way to measure brain amyloid in living people at that time.

To address this, first author Juho Joutsa, now at Massachusetts General Hospital in Charlestown, took advantage of amyloid PET scanning. He measured uptake of the ligand Pittsburgh compound B (PiB) in 41 middle-aged patients who had developed epilepsy beginning around 5 years old and 46 age-matched controls from the Turku Adult Childhood Onset Epilepsy (TACOE) study. Patients had struggled with epilepsy for 17 years on average and had been followed clinically for about 50 years. Most were in remission, but some still had active seizures.

On visual inspection of the scans, nine epilepsy patients (22 percent) were PiB-positive, compared to just three (7 percent) of the controls. The areas of PiB uptake did not correspond to epileptic foci but were distributed across the brain. PiB standard uptake value ratios did not associate with age at onset, duration of active epilepsy, or how long the participant had been taking epilepsy drugs. People in remission, who had stopped treatment decades ago, were among those with more amyloid.

At a group level, semi-quantitative analysis indicated that epilepsy patients who carried an ApoE4 allele had higher PiB SUVRs in the prefrontal cortex (1.66) and whole cortex (1.53) than did ApoE4-positive controls (1.40 and 1.36 respectively). Regions where PiB bound included those involved in the default mode network, an area susceptible to Aβ deposition. “That might tell us there are similar things going on in epilepsy and in AD,” Joutsa told Alzforum. He does not know why some epilepsy patients accumulate amyloid so young while others don’t, but assumes there are contributing factors not captured in this study. A larger, possibly more homogenous cohort might help figure that out, he said.

That the epilepsy-related plaques turn up in the default mode network, an area of the brain known to be active when the mind wanders, is intriguing, said Marcus Raichle, Washington University in St. Louis. “When I saw this picture, I was stunned,” he told Alzforum (see image above). “That’s an area that we know to be vulnerable to amyloid in AD.” He wondered how epilepsy enhances susceptibility in that region.

Raichle was also struck that amyloid sticks around even in people whose epilepsy is controlled or in remission. Previous studies in AD patients and in animal models of the disease report that increased neuronal activity leads to more Aβ production and seizures (see Dec 2005 newsSep 2007 news). That could mean a brain that has experienced epilepsy at some point is made marginally more active even once the seizures stop, he said.

Among people with active epilepsy, ApoE4 allele carriers deposited more amyloid than noncarriers (see image above). The ApoE4 allele mostly affected people with idiopathic disease, which scientists believe has a strong genetic basis, rather than “cryptogenic” epilepsy, which has less clear origins. Idiopathic patients could have additional genetic risk factors that make them more vulnerable to neuronal stress or faulty neuron repair, or that impair Aβ clearance, Joutsa said. ApoE itself has been linked to all these phenomena. Most recently, researchers led by Thomas Südhof, Stanford University, reported that ApoE has a differential effect on APP expression in human neurons, with the E4 allele driving APP protein levels up the most (see Jan 2017 news). 

Together, the results suggest that people with epilepsy deposit Aβ at a younger age than the general population, carrying amyloid loads typical of people 10 years older, the authors wrote. Joutsa is unsure whether this predisposes patients to AD, but plans to follow this cohort to find out, ideally with longitudinal PiB-PET scanning.

Another recent paper from Sillanpää and the TACOE study group reported that older epilepsy patients with ongoing seizures perform worse on tests of language, semantics, and visuomotor function (Karrasch et al., 2017). These scientists also found more brain atrophy and network abnormalities in patients with active epilepsy (see Garcia-Ramos et al., 2017).—Gwyneth Dickey Zakaib


  1. This study focuses on the blurry borderland of epilepsy and brain amyloid deposition that emerged even before Alzheimer’s description of Auguste D. (Alzheimer 1907), since the plaques he described with great clarity in 1907 had been first noted by Blocq and Marinesco in so-called “essential epilepsy” patients over a decade earlier (Buda et al., 2009). Over the years, little significance was attached to the coincidence of seizures and amyloid plaques in human cases or mouse models of AD until the arrival of compelling epidemiological (Scarmeas et al., 2009) and experimental evidence (Palop et al., 2007). Now, a century later, Joutsa and colleagues logically ask whether a 50-year history of “uncomplicated” epilepsy actually favors amyloid plaque formation. The authors assembled a cohort of individuals with early onset idiopathic and cryptogenic epilepsy to image the distribution of PiB-positive amyloid plaques and compare the patterns to age-matched, neurologically unaffected controls. Their hypothesis was that individuals with long-standing epilepsy and positive APOE4 status would be more likely to show an elevated plaque burden. The hypothesis is important, but unfortunately, the study design and data set are insufficient for a clinically meaningful answer, and indeed, there may be no single answer.

    The authors themselves point out four critical limitations of their pilot study. The first is the small size of the cohort, and its clinical heterogeneity. The lack of a group representing a single etiology of epilepsy is a major confound, since within the category of idiopathic and cryptogenic epilepsy there are many monogenic pathways leading to vastly different intrinsic seizure networks and patterns of Aβ protein release and metabolism in different brain regions, as well as distinct molecular reasons for amyloid plaques to locally accumulate, and the authors found this heterogeneity in their imaging results. The second issue is epilepsy severity. The seizure history is undocumented in each of these individuals. They are described as uncomplicated cases, many in remission, and the types of seizures they presented with, their location in the brain, and their severity are not included in the analysis. A third issue is timing. Imaging was performed at only a single time point, thus the chronological age and plaque accumulation rate is not known. Both seizures and rate of plaque deposition are determined by complex cellular and synaptic interactions during aging, and both vary in the young and older brain. Finally, the time point chosen for determination of amyloid plaque load was made at a mean age that precedes most late-onset dementia, and there was a selection bias against cases with substantial cognitive impairment.

    Whether a history of abnormal synaptic activity drives the accumulation of amyloid plaques, when and where it begins, and how this process contributes to the progression of dementia remain fundamental relationships to explore. A few insights from the laboratory illustrate the complicated road ahead. Single experimentally induced seizures in mice release soluble Aβ from synapses (Cirrito et al., 2005), but this may not be sustained. In a monogenic mouse channelopathy model (Kv1.1 knockout) of early onset epilepsy, despite a history of intense early seizure activity, there is no elevation of Aβ aggregates in the first month of life (Holth et al., 2013). In a transgenic model of Aβ overexpression on a seizure-prone versus seizure-resistant genetic background, mice with the more severe seizure phenotype actually showed less Aβ accumulation (Jackson et al., 2015). For its part, overexpression of Aβ, which triggers epilepsy in most mouse models (Noebels et al., 2011), appears to do so more readily when expressed early rather than late in adulthood (Born et al., 2014). Finally, cortical hyperexcitability occurs at levels of Aβ below, and earlier than, those required for plaque deposition (Cummings et al., 2015), suggesting an earlier biomarker for cognitive impairment than simple plaque load. 


    . Über eine eigenartige Erkrankung der Hirnrinde. Allgemeine Zeitschrift fur Psychiatrie und Psychisch-gerichtliche Medizin. 1907 Jan;64:146-8.

    . Georges Marinesco and the early research in neuropathology. Neurology. 2009 Jan 6;72(1):88-91. PubMed.

    . Seizures in Alzheimer disease: who, when, and how common?. Arch Neurol. 2009 Aug;66(8):992-7. PubMed.

    . Aberrant excitatory neuronal activity and compensatory remodeling of inhibitory hippocampal circuits in mouse models of Alzheimer's disease. Neuron. 2007 Sep 6;55(5):697-711. PubMed.

    . Synaptic activity regulates interstitial fluid amyloid-beta levels in vivo. Neuron. 2005 Dec 22;48(6):913-22. PubMed.

    . Tau loss attenuates neuronal network hyperexcitability in mouse and Drosophila genetic models of epilepsy. J Neurosci. 2013 Jan 23;33(4):1651-9. PubMed.

    . DBA/2J genetic background exacerbates spontaneous lethal seizures but lessens amyloid deposition in a mouse model of Alzheimer's disease. PLoS One. 2015;10(5):e0125897. Epub 2015 May 1 PubMed.

    . A perfect storm: Converging paths of epilepsy and Alzheimer's dementia intersect in the hippocampal formation. Epilepsia. 2011 Jan;52 Suppl 1:39-46. PubMed.

    . Genetic suppression of transgenic APP rescues Hypersynchronous network activity in a mouse model of Alzeimer's disease. J Neurosci. 2014 Mar 12;34(11):3826-40. PubMed.

    . First effects of rising amyloid-β in transgenic mouse brain: synaptic transmission and gene expression. Brain. 2015 Jul;138(Pt 7):1992-2004. Epub 2015 May 16 PubMed.

  2. As recently reviewed in some depth (Palop and Mucke, 2016) and prospectively demonstrated in the clinic (Vossel et al., 2016), there clearly is an important link between Alzheimer’s disease and epilepsy, which may involve a vicious cycle between aberrant neural network activity and the accumulation of pathogenic proteins in brain. The interesting findings reported by Joutsa et al. provide additional support for this hypothesis. At least in my opinion, the results of this study also indirectly support the idea that the effective suppression of network hypersynchrony could have symptomatic as well as disease-modifying therapeutic benefits in people with AD or at high risk of developing the disease (Bakker et al., 2012; Sanchez et al., 2012; Palop and Mucke, 2016). Although many epilepsy patients included in the study by Joutsa et al. had not received any antiepileptic drugs for a long time before they received the PiB PET scan, it is worth noting that most of them appear to have never been treated with levetiracetam, an antiepileptic that had beneficial effects in AD-related mouse models (Sanchez et al., 2012) and in patients with amnestic mild cognitive impairment (Bakker et al., 2012; Bakker et al., 2015). Instead, many patients in this cohort received antiepileptic drugs that strongly block voltage-gated sodium channels and, thus, might be expected to promote rather than suppress AD-related network dysfunction (Verret et al., 2012; Palop and Mucke, 2016). 


    . Network abnormalities and interneuron dysfunction in Alzheimer disease. Nat Rev Neurosci. 2016 Dec;17(12):777-792. Epub 2016 Nov 10 PubMed.

    . Incidence and impact of subclinical epileptiform activity in Alzheimer's disease. Ann Neurol. 2016 Dec;80(6):858-870. Epub 2016 Nov 7 PubMed.

    . Reduction of hippocampal hyperactivity improves cognition in amnestic mild cognitive impairment. Neuron. 2012 May 10;74(3):467-74. PubMed.

    . Levetiracetam suppresses neuronal network dysfunction and reverses synaptic and cognitive deficits in an Alzheimer's disease model. Proc Natl Acad Sci U S A. 2012 Oct 16;109(42):E2895-903. PubMed.

    . Network abnormalities and interneuron dysfunction in Alzheimer disease. Nat Rev Neurosci. 2016 Dec;17(12):777-792. Epub 2016 Nov 10 PubMed.

    . Response of the medial temporal lobe network in amnestic mild cognitive impairment to therapeutic intervention assessed by fMRI and memory task performance. Neuroimage Clin. 2015;7:688-98. Epub 2015 Feb 21 PubMed.

    . Inhibitory interneuron deficit links altered network activity and cognitive dysfunction in Alzheimer model. Cell. 2012 Apr 27;149(3):708-21. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Paper Alert: Synaptic Activity Increases Aβ Release
  2. Do "Silent" Seizures Cause Network Dysfunction in AD?
  3. ApoE Risk Explained? Isoform-Dependent Boost in APP Expression Uncovered

Paper Citations

  1. . Medical history and the risk of Alzheimer's disease: a collaborative re-analysis of case-control studies. EURODEM Risk Factors Research Group. Int J Epidemiol. 1991;20 Suppl 2:S36-42. PubMed.
  2. . Risk of dementia in patients with Parkinson's disease, epilepsy, and severe head trauma: a register-based follow-up study. Am J Epidemiol. 1995 Dec 15;142(12):1300-5. PubMed.
  3. . Incidence and predictors of seizures in patients with Alzheimer's disease. Epilepsia. 2006 May;47(5):867-72. PubMed.
  4. . Increased neuronal beta-amyloid precursor protein expression in human temporal lobe epilepsy: association with interleukin-1 alpha immunoreactivity. J Neurochem. 1994 Nov;63(5):1872-9. PubMed.
  5. . Senile plaques in temporal lobe epilepsy. Acta Neuropathol. 1994;87(5):504-10. PubMed.
  6. . Cognitive Outcome in Childhood-Onset Epilepsy: A Five-Decade Prospective Cohort Study. J Int Neuropsychol Soc. 2017 Apr;23(4):332-340. Epub 2017 Jan 10 PubMed.
  7. . Brain structure and organization five decades after childhood onset epilepsy. Hum Brain Mapp. 2017 Apr 3; PubMed.

Further Reading


  1. . Growing old with epilepsy: the neglected issue of cognitive and brain health in aging and elder persons with chronic epilepsy. Epilepsia. 2008 May;49(5):731-40. Epub 2007 Nov 21 PubMed.
  2. . Shared cognitive and behavioral impairments in epilepsy and Alzheimer's disease and potential underlying mechanisms. Epilepsy Behav. 2013 Mar;26(3):343-51. PubMed.

Primary Papers

  1. . Association Between Childhood-Onset Epilepsy and Amyloid Burden 5 Decades Later. JAMA Neurol. 2017 May 1;74(5):583-590. PubMed.