Alzheimer's disease and its severity reduce MEG responses to unexpected auditory stimuli

Alexandra Krugliak; Mats WJ van Es; Marlou N Perquin; Chun Shen; Haddy Fye; Cara Alcock; Juliette H Lanskey; Melek Karadag; Ece Kocagoncu; Rebecca Williams; Andrew J Quinn; Jemma Pitt; Tony Thayanandan; Stephen L Lowe; Vanessa Raymont; Krish D Singh; Mark W Woolrich; Anna C Nobre; James B Rowe

doi:10.1002/alz70856_102668

Abstract

Background: Alzheimer's disease (AD) impairs cognition with synaptic and neuronal loss related to amyloid and tau pathology. Drug development needs sensitive and reliable biomarkers for early phase clinical trials. The effects of AD on brain function can be measured by magneto/electroencephalography (M/EEG). The New Therapeutics in Alzheimer's Disease study (NTAD) aims to test the sensitivity and reliability of M/EEG‐based biomarkers to AD.

The entorhinal cortex and hippocampi are affected early by AD, underlying deficits in associative learning. However, it is difficult to directly measure physiological responses from medial temporal lobe by MEG due to its depth and orientation. In contrast, lateral auditory cortical responses to unexpected stimuli (e.g. mismatch negativity response) are readily detected by MEG. We therefore developed a hybrid task, based on associative learning, with auditory novelty and associative deviants to rapidly presented standard trials: the audio‐visual oddball task (VAB).

Method: We used concurrent M/EEG to measure neural responses to unexpected novel sounds and mismatches of associated image‐sound pairings. Cognitively healthy older controls (N = 31) and people with amyloid‐positive mild cognitive impairment (MCI) or early AD completed baseline assessment (N = 55). A subset of patients (N = 17) were re‐tested after two‐weeks to assess the reliability of the VAB.

Result: A sensor‐level analysis on gradiometers revealed that controls showed a stronger neural response to novel compared to mismatching sounds. In AD/MCI patients the response to mismatching sounds was increased compared to controls, indicating reduced differentiation between novel and learned sounds consistent with deficits in associative learning. This increase in neural response to mismatched sounds in patients related to cognitive deficits (Addenbrooke's Cognitive Examination, ACE‐R). Comparing the VAB measures of the baseline assessment and two‐week scans shows moderate to good reliability. An analysis in source space is in progress.

Conclusion: Our results demonstrate that the MEG responses in the VAB task are sensitive to presence of AD/MCI. With sufficient reliability we propose that MEG is suitable to support experimental medicine studies of AD/MCI.

Original language	English
Article number	e102668
Number of pages	2
Journal	Alzheimer's & Dementia
Volume	21
Issue number	S2
DOIs	https://doi.org/10.1002/alz70856_102668
Publication status	Published - 26 Dec 2025
Event	Alzheimer’s Association International Conference 2025 - Metro Toronto Convention Centre, Toronto, Canada Duration: 27 Jul 2025 → 31 Jul 2025

Access to Document

10.1002/alz70856_102668Licence: Creative Commons: Attribution (CC BY)

KrugliakA2025AlzheimersFinal published version, 88.8 KBLicence: Creative Commons: Attribution (CC BY)

Cite this

Krugliak, A., van Es, M. WJ., Perquin, M. N., Shen, C., Fye, H., Alcock, C., Lanskey, J. H., Karadag, M., Kocagoncu, E., Williams, R., Quinn, A. J., Pitt, J., Thayanandan, T., Lowe, S. L., Raymont, V., Singh, K. D., Woolrich, M. W., Nobre, A. C., & Rowe, J. B. (2025). Alzheimer's disease and its severity reduce MEG responses to unexpected auditory stimuli. Alzheimer's & Dementia, 21(S2), Article e102668. https://doi.org/10.1002/alz70856_102668

@article{13eb52a30e6b4ad8925ff2cd295b09a9,

title = "Alzheimer's disease and its severity reduce MEG responses to unexpected auditory stimuli",

abstract = "Background: Alzheimer's disease (AD) impairs cognition with synaptic and neuronal loss related to amyloid and tau pathology. Drug development needs sensitive and reliable biomarkers for early phase clinical trials. The effects of AD on brain function can be measured by magneto/electroencephalography (M/EEG). The New Therapeutics in Alzheimer's Disease study (NTAD) aims to test the sensitivity and reliability of M/EEG‐based biomarkers to AD. The entorhinal cortex and hippocampi are affected early by AD, underlying deficits in associative learning. However, it is difficult to directly measure physiological responses from medial temporal lobe by MEG due to its depth and orientation. In contrast, lateral auditory cortical responses to unexpected stimuli (e.g. mismatch negativity response) are readily detected by MEG. We therefore developed a hybrid task, based on associative learning, with auditory novelty and associative deviants to rapidly presented standard trials: the audio‐visual oddball task (VAB). Method: We used concurrent M/EEG to measure neural responses to unexpected novel sounds and mismatches of associated image‐sound pairings. Cognitively healthy older controls (N = 31) and people with amyloid‐positive mild cognitive impairment (MCI) or early AD completed baseline assessment (N = 55). A subset of patients (N = 17) were re‐tested after two‐weeks to assess the reliability of the VAB. Result: A sensor‐level analysis on gradiometers revealed that controls showed a stronger neural response to novel compared to mismatching sounds. In AD/MCI patients the response to mismatching sounds was increased compared to controls, indicating reduced differentiation between novel and learned sounds consistent with deficits in associative learning. This increase in neural response to mismatched sounds in patients related to cognitive deficits (Addenbrooke's Cognitive Examination, ACE‐R). Comparing the VAB measures of the baseline assessment and two‐week scans shows moderate to good reliability. An analysis in source space is in progress. Conclusion: Our results demonstrate that the MEG responses in the VAB task are sensitive to presence of AD/MCI. With sufficient reliability we propose that MEG is suitable to support experimental medicine studies of AD/MCI.",

author = "Alexandra Krugliak and \{van Es\}, \{Mats WJ\} and Perquin, \{Marlou N\} and Chun Shen and Haddy Fye and Cara Alcock and Lanskey, \{Juliette H\} and Melek Karadag and Ece Kocagoncu and Rebecca Williams and Quinn, \{Andrew J\} and Jemma Pitt and Tony Thayanandan and Lowe, \{Stephen L\} and Vanessa Raymont and Singh, \{Krish D\} and Woolrich, \{Mark W\} and Nobre, \{Anna C\} and Rowe, \{James B\}",

year = "2025",

month = dec,

day = "26",

doi = "10.1002/alz70856\_102668",

language = "English",

volume = "21",

journal = "Alzheimer's \& Dementia",

issn = "1552-5260",

publisher = "Elsevier",

number = "S2",

note = "Alzheimer{\textquoteright}s Association International Conference 2025, AAIC 2025 ; Conference date: 27-07-2025 Through 31-07-2025",

}

Krugliak, A, van Es, MWJ, Perquin, MN, Shen, C, Fye, H, Alcock, C, Lanskey, JH, Karadag, M, Kocagoncu, E, Williams, R, Quinn, AJ, Pitt, J, Thayanandan, T, Lowe, SL, Raymont, V, Singh, KD, Woolrich, MW, Nobre, AC & Rowe, JB 2025, 'Alzheimer's disease and its severity reduce MEG responses to unexpected auditory stimuli', Alzheimer's & Dementia, vol. 21, no. S2, e102668. https://doi.org/10.1002/alz70856_102668

TY - JOUR

T1 - Alzheimer's disease and its severity reduce MEG responses to unexpected auditory stimuli

AU - Krugliak, Alexandra

AU - van Es, Mats WJ

AU - Perquin, Marlou N

AU - Shen, Chun

AU - Fye, Haddy

AU - Alcock, Cara

AU - Lanskey, Juliette H

AU - Karadag, Melek

AU - Kocagoncu, Ece

AU - Williams, Rebecca

AU - Quinn, Andrew J

AU - Pitt, Jemma

AU - Thayanandan, Tony

AU - Lowe, Stephen L

AU - Raymont, Vanessa

AU - Singh, Krish D

AU - Woolrich, Mark W

AU - Nobre, Anna C

AU - Rowe, James B

PY - 2025/12/26

Y1 - 2025/12/26

N2 - Background: Alzheimer's disease (AD) impairs cognition with synaptic and neuronal loss related to amyloid and tau pathology. Drug development needs sensitive and reliable biomarkers for early phase clinical trials. The effects of AD on brain function can be measured by magneto/electroencephalography (M/EEG). The New Therapeutics in Alzheimer's Disease study (NTAD) aims to test the sensitivity and reliability of M/EEG‐based biomarkers to AD. The entorhinal cortex and hippocampi are affected early by AD, underlying deficits in associative learning. However, it is difficult to directly measure physiological responses from medial temporal lobe by MEG due to its depth and orientation. In contrast, lateral auditory cortical responses to unexpected stimuli (e.g. mismatch negativity response) are readily detected by MEG. We therefore developed a hybrid task, based on associative learning, with auditory novelty and associative deviants to rapidly presented standard trials: the audio‐visual oddball task (VAB). Method: We used concurrent M/EEG to measure neural responses to unexpected novel sounds and mismatches of associated image‐sound pairings. Cognitively healthy older controls (N = 31) and people with amyloid‐positive mild cognitive impairment (MCI) or early AD completed baseline assessment (N = 55). A subset of patients (N = 17) were re‐tested after two‐weeks to assess the reliability of the VAB. Result: A sensor‐level analysis on gradiometers revealed that controls showed a stronger neural response to novel compared to mismatching sounds. In AD/MCI patients the response to mismatching sounds was increased compared to controls, indicating reduced differentiation between novel and learned sounds consistent with deficits in associative learning. This increase in neural response to mismatched sounds in patients related to cognitive deficits (Addenbrooke's Cognitive Examination, ACE‐R). Comparing the VAB measures of the baseline assessment and two‐week scans shows moderate to good reliability. An analysis in source space is in progress. Conclusion: Our results demonstrate that the MEG responses in the VAB task are sensitive to presence of AD/MCI. With sufficient reliability we propose that MEG is suitable to support experimental medicine studies of AD/MCI.

AB - Background: Alzheimer's disease (AD) impairs cognition with synaptic and neuronal loss related to amyloid and tau pathology. Drug development needs sensitive and reliable biomarkers for early phase clinical trials. The effects of AD on brain function can be measured by magneto/electroencephalography (M/EEG). The New Therapeutics in Alzheimer's Disease study (NTAD) aims to test the sensitivity and reliability of M/EEG‐based biomarkers to AD. The entorhinal cortex and hippocampi are affected early by AD, underlying deficits in associative learning. However, it is difficult to directly measure physiological responses from medial temporal lobe by MEG due to its depth and orientation. In contrast, lateral auditory cortical responses to unexpected stimuli (e.g. mismatch negativity response) are readily detected by MEG. We therefore developed a hybrid task, based on associative learning, with auditory novelty and associative deviants to rapidly presented standard trials: the audio‐visual oddball task (VAB). Method: We used concurrent M/EEG to measure neural responses to unexpected novel sounds and mismatches of associated image‐sound pairings. Cognitively healthy older controls (N = 31) and people with amyloid‐positive mild cognitive impairment (MCI) or early AD completed baseline assessment (N = 55). A subset of patients (N = 17) were re‐tested after two‐weeks to assess the reliability of the VAB. Result: A sensor‐level analysis on gradiometers revealed that controls showed a stronger neural response to novel compared to mismatching sounds. In AD/MCI patients the response to mismatching sounds was increased compared to controls, indicating reduced differentiation between novel and learned sounds consistent with deficits in associative learning. This increase in neural response to mismatched sounds in patients related to cognitive deficits (Addenbrooke's Cognitive Examination, ACE‐R). Comparing the VAB measures of the baseline assessment and two‐week scans shows moderate to good reliability. An analysis in source space is in progress. Conclusion: Our results demonstrate that the MEG responses in the VAB task are sensitive to presence of AD/MCI. With sufficient reliability we propose that MEG is suitable to support experimental medicine studies of AD/MCI.

UR - https://alz.confex.com/alz/2025/meetingapp.cgi/Paper/102668

U2 - 10.1002/alz70856_102668

DO - 10.1002/alz70856_102668

M3 - Abstract

SN - 1552-5260

VL - 21

JO - Alzheimer's & Dementia

JF - Alzheimer's & Dementia

IS - S2

M1 - e102668

T2 - Alzheimer’s Association International Conference 2025

Y2 - 27 July 2025 through 31 July 2025

ER -

Alzheimer's disease and its severity reduce MEG responses to unexpected auditory stimuli

Abstract

Access to Document

Fingerprint

Cite this