Diagnosis of Mild Cognitive Impairment Due to Alzheimer’s Disease with Transcranial Magnetic Stimulation

Abstract

Background:

Considering the increasing evidence that disease-modifying treatments for Alzheimer’s disease (AD) must be administered early in the disease course, the development of diagnostic tools capable of accurately identifying AD at early disease stages has become a crucial target. In this view, transcranial magnetic stimulation (TMS) has become an effective tool to discriminate between different forms of neurodegenerative dementia.

Objective:

To determine whether a TMS multi-paradigm approach can be used to correctly identify mild cognitive impairment (MCI) due to AD (AD MCI).

Methods:

A sample of 69 subjects with MCI were included and classified as AD MCI or MCI unlikely due to AD (non-AD MCI) based on 1) extensive neurological and neuropsychological evaluation, 2) MRI imaging, and 3) cerebrospinal fluid analysis or/and amyloid PET imaging. A paired-pulse TMS multi-paradigm approach assessing short interval intracortical inhibition-facilitation (SICI-ICF), dependent on GABAergic and glutamatergic intracortical circuits, respectively, and short latency afferent inhibition (SAI), dependent on cholinergic circuits, was performed.

Results:

We observed a significant impairment of SAI and unimpaired SICI and ICF in AD MCI as compared to non-AD MCI. According to ROC curve analysis, the SICI-ICF / SAI index differentiated AD MCI from non-AD MCI with a specificity of 87.9% and a sensitivity of 94.4%.

Conclusions:

The assessment of intracortical connectivity with TMS could aid in the characterization of MCI subtypes, correctly identifying AD pathophysiology. TMS can be proposed as an adjunctive, non-invasive, inexpensive, and time-saving screening tool in MCI differential diagnosis.

Keywords

Alzheimer’s disease dementia mild cognitive impairment short interval intracortical inhibition short latency afferent inhibition transcranial magnetic stimulation

INTRODUCTION

Important innovations in ongoing clinical trials in Alzheimer’s disease (AD) include the use of preclinical/prodromal biomarkers, considering the increasing evidence that disease-modifying treatments must be administered early in the disease course [1, 2]. As a result, the development of diagnostic tools capable of accurately identifying AD pathophysiology at early disease stages has become a crucial target [3].

Indeed, AD can be represented by a continuum from cognitively normal individuals with evidence of amyloid accumulation in the brain to those with severe dementia, and the stage of mild cognitive impairment (MCI) can be considered as an intermediate phase between normal cognitive decline with aging and dementia, in which patients have a greater cognitive decline than expected for their age and educational level [4 –6].

However, only up to two-thirds of patients with amnestic MCI have an underlying AD pathology, while 15–25% have neurodegenerative diseases other than AD, such as frontotemporal dementia (FTD), and the remainder have normal age-related changes [7 –9]. These findings highlight how it has become critical to develop reliable biomarkers that reflect the underlying disease pathophysiology, even in the early phases of disease [10]. In this context, the diagnosis of MCI due to AD, which can be considered as part of the prodromal stage of AD, has deeply focused on biomarkers of cerebral amyloidosis, such as a decrease in cerebrospinal fluid (CSF) Aβ_{1 - 42} concentrations or an increase retention of amyloid tracers on positron emission tomography (PET) imaging.

In the context of biomarkers for AD pathology, our group has recently developed an index using transcranial magnetic stimulation (TMS) intracortical connectivity measures, yielding a diagnostic accuracy of 90% in identifying AD, with high accuracy even in the early phases of disease [11]. Short-latency afferent inhibition (SAI), assessing indirectly the function of cholinergic circuits, has been reported to be impaired in AD patients; conversely, short-interval intracortical inhibition (SICI) and intracortical facilitation (ICF), markers of GABA_Aergic and glutamatergic neurotransmission, respectively, have been found impaired in FTD patients [11]. In support of these findings, previous studies have observed a significant association between CSF biomarkers and neurophysiological measures of central cholinergic activity [12, 13], further highlighting the impact of TMS in detecting ongoing mechanisms of neurodegeneration in vivo. Indeed, if it is true that amyloid deposits are the neuropathological hallmark of AD, it is true that AD is characterized by a well-established cholinergic impairment [14, 15], even in the prodromal phases of disease [16, 17].

All the above observations defined the objective of this work, aimed at assessing and comparing the accuracy of TMS measures in a well-studied population of MCI subjects due to AD as compared to MCI unlikely due to AD to evaluate if this inexpensive, non-invasive and easy to perform evaluation may be suggested as a screening or additional marker in the diagnostic work-up of MCI.

METHODS

Standard protocol approvals, registrations, and patient consents

Full written informed consent was obtained from all participants according to the Declaration of Helsinki. The study protocol was approved by the local ethics committee (Brescia Hospital), #NP1965 approved 05.19.2015.

Participants

In the present study, 71 subjects with MCI were recruited from the Neurology Unit, Department of Clinical and Experimental Sciences, University of Brescia, from the Neurology Unit, Valle Camonica Hospital, Brescia, and from the Laboratory of Alzheimer’s Neuroimaging and Epidemiology, IRCCS San Giovanni di Dio Fatebenefratelli, Brescia, Italy.

MCI was defined as a new-onset cognitive decline reported by the patient, relatives, or physician for at least the previous 6 months, with a maintained independence in completing daily activities, and a consistent abnormal performance compared with mean age- and education-specific values on memory tests [5].

All participants underwent a neuropsychological assessment and a structural MRI brain scan. The neuropsychological evaluation consisted of the following tests: Mini-Mental State Examination, semantic and phonemic fluencies, Rey Complex Figure Copy and Recall, Short-Story Recall Test, Digit Span, Trail Making Test A and B, and Token Test. Instrumental and basic activities of daily living were assessed as well. Behavioral and psychiatric disturbances were evaluated with the Neuropsychiatry Inventory and the Geriatric Depression Scale. All patients were classified as multiple-domain MCI according to current criteria [6, 18].

Furthermore, in all participants, CSF analysis (48.3%), amyloid PET imaging (41.7%), or both (10.0%), was performed. A CSF AD-like profile was defined as CSF t-Tau≥400 ng/L and CSF Aβ_{1 - 42}≤650 ng/L using an ELISA assay (INNOTEST, Innogenetics, Ghent, Belgium) [19]. Amyloid PET imaging was acquired using 370 MBq (10 mCi) of ¹⁸F-florbetapir and visual readings were performed independently by two nuclear medicine physicians who were blinded to the patients’ diagnosis, following the procedures provided by the ligand manufacturer, as previously reported [20].

According to clinical and CSF/PET amyloid results, MCI subjects were classified in MCI due to AD (AD MCI) and MCI unlikely due to AD (non-AD MCI) in accordance with NIA-AA criteria [5]. Subjects with contrasting biomarkers were excluded from analysis.

None of the participants were treated with central nervous system active drugs that could influence cerebral cortex excitability in the previous months.

Transcranial magnetic stimulation

TMS was performed with a figure-of-eight coil (each loop diameter 70 mm) connected to a Magstim Bistim² system (Magstim Company, Oxford, UK), as previously reported [21]. The magnetic stimuli had a monophasic current waveform (rise time of 100μs, decaying back to zero over 800μs). Motor evoked potentials (MEPs) were recorded from the right first dorsal interosseous muscle (FDI) through surface Ag/AgCl electrodes placed in a belly-tendon montage and acquired using a Biopac MP-150 electromyograph (BIOPAC Systems Inc., Santa Barbara, CA, USA).

The TMS coil was held tangentially over the scalp region corresponding to the primary hand motor area contralateral to the target muscle, with the coil handle pointed 45° posteriorly and laterally to the sagittal plane.

The motor hot spot was defined as the location where TMS consistently produced the largest MEP size at 120% of the resting motor threshold (rMT) in the target muscle.

rMT was defined as the minimal stimulus intensity needed to produce MEPs with an amplitude of at least 50μV in 5 out of 10 consecutive trails during complete muscle relaxation, which was controlled by visually checking the absence of electromyography (EMG) activity at high-gain amplification [22].

SICI and ICF, which predominantly reflect GABA_Aergic and glutamatergic neurotransmission respectively, were studied at rest via a paired-pulse paradigm, delivered in a conditioning-test design with the conditioning stimulus (CS) set at an intensity of 70% of the rMT, while the test stimulus (TS) was adjusted to evoke a MEP approximately 1 mV peak-to-peak in the relaxed FDI. Different interstimulus intervals (ISIs) between the CS and TS were employed to investigate preferentially both SICI (1, 2, 3, 5 ms) and ICF (7, 10, 15 ms) [23, 24].

SAI, which primarily reflects cholinergic transmission, was studied using a previously described technique [25]. CS were single pulses (200μs) of electrical stimulation applied through bipolar electrodes to the right median nerve at the wrist (cathode proximal). The intensity of the CS was set at just over motor threshold for evoking a visible twitch of the thenar muscles while the TS was adjusted to evoke a MEP of approximately 1 mV peak-to-peak. The CS to the peripheral nerve preceded the TS by different ISI (^-4, 0, ⁺4, ⁺8 ms, determined relative to the latency of the N20 component of the somatosensory evoked potential).

Ten stimuli were delivered for each ISI for all stimulation paradigms and fourteen control MEPs in response to the TS alone were recorded, for each paradigm, in all participants in a pseudo-randomized sequence. The amplitude of the conditioning MEPs was expressed as a ratio of the mean unconditioned response. The inter-trial interval was set at 5 s (±10%).

SICI-ICF and SAI protocols were performed in a randomized order. Throughout the experiment, complete muscle relaxation was monitored by audio-visual feedback where appropriate. If quality of study data was degraded by patient movement, the protocol was recommenced, and the initial data discarded. Trials were discarded if EMG activity exceeded 100μV in the 250 ms prior to TMS stimulus delivery. All patients were able to understand instructions and obtain full muscle relaxation.

The operators who performed TMS were blinded to the subjects’ biomarker status and neuropsychological evaluation.

Statistical analysis

Clinical and demographic characteristics were compared using independent samples t-test or Fisher’s exact test. TMS measures were compared using a two-way mixed ANOVA with ISI as within-subjects factor and GROUP as between-subjects factor. If a significant main effect was reached, group differences were examined with post hoc tests (Bonferroni correction for multiple comparisons). To check and correct for sphericity violation, Mauchly’s test and Greenhouse-Geisser epsilon determination were used.

To determine the area under the curve (AUC), receiver operating characteristics (ROC) curves were used, including 95% confidence interval (CI) values, for each index (SICI-ICF, SAI and SICI-ICF / SAI ratio). The SICI-ICF/SAI ratio was defined as (average SICI at 1, 2, 3 ms)/(average ICF at 7, 10, 15 ms)/(average SAI at 0, +4 ms). Cut-off points were set to minimize the difference between sensitivity and specificity (Youden’s index).

Pearson’s r correlation coefficients were used to explore any influence of CSF Aβ_{1 - 42}, t-Tau and p-Tau₁₈₁ on SAI and SICI-ICF.

Statistical significance was assumed at p < 0.05. Data analyses were carried out using SPSS 21.0 software.

RESULTS

Participants

Seventy-one MCI subjects were included in the study. Among them, 31 MCI subjects performed only amyloid PET imaging and 31 subjects underwent only CSF analysis, while 9 subjects performed both amyloid PET and CSF analysis (see Table 1). Two participants with contrasting biomarkers (1 subject with decreased CSF Aβ_{1 - 42} and negative amyloid PET imaging, 1 subject with normal CSF Aβ_{1 - 42} and positive amyloid PET imaging) were excluded from analysis. Thus, the present analysis was conducted on 69 MCI subjects, namely 36 classified as AD MCI and 33 non-AD MCI subjects (see Fig. 1). No significant differences in demographic or clinical characteristics were observed between groups (see Table 2).

Table 1

Biomarker assessment in each group of patients

AD MCI (n = 37)	non-AD MCI (n = 34)	Total (n = 71)
Amyloid PET imaging only (%)	21 (56.8)	10 (29.4)	31
CSF analysis only (%)	9 (24.3)	22 (64.7)	31
Both (amyloid PET+CSF analysis) (%)	7 (18.9)	2 (5.9)	9

Fig.1

Flow diagram of the study. After the index test, results were sorted on the basis of the reference standard. AD MCI, mild cognitive impairment likely due to Alzheimer’s disease; non-AD MCI, MCI unlikely due to AD; SICI-ICF, short-interval intracortical inhibition - intracortical facilitation; SAI, short-latency afferent inhibition; SICI-ICF / SAI, ratio between SICI-ICF and SAI parameters.

Table 2

Demographic, clinical, and neurophysiological characteristics of included patients

	AD MCI	non-AD MCI	p*
Patients (n)	36	33	n.s.
Age, y	69.8±7.3	67.3±7.3	n.s.
Age at onset (y)	67.0±7.1	64.2±7.1	n.s.
Gender (% female)	55.6	60.6	n.s.
Education (y)	10.4±4.4	8.8±4.5	n.s.
MMSE	26.3±2.0	26.4±2.0	n.s.
NPI	8.0±8.8	8.5±8.9	n.s.
GDS	3.2±3.3	2.6±3.0	n.s.
BADL	0.1±0.3	0.1±0.2	n.s.
IADL	0.4±0.7	0.3±0.5	n.s.
Amyloid PET, positive	27/27 (100%)	0/11 (0%)	p < 0.001
Cerebrospinal fluid
t-Tau	858.9±493.3	300.4±144.7	p < 0.001
p-Tau₁₈₁	102.9±40.2	52.4±32.0	p < 0.001
Aβ_{1 - 42}	526.1±141.8	950.9±207.0	p < 0.001
TMS measures
rMT (% MSO)	43.0±8.3	45.6±10.0	n.s.
SICI-ICF	0.25±0.14	0.69±0.35	p < 0.001
SAI	0.82±0.11	0.58±0.18	p < 0.001
SICI-ICF / SAI	0.31±0.15	1.27±0.71	p < 0.001

Demographic and clinical characteristics, and neurophysiological parameters are expressed as mean±SD (otherwise specified); resting motor threshold is expressed as ratio of the MSO; SICI-ICF, and SAI are represented as ratio of mean motor evoked potential (MEP) amplitude related to the control MEP. AD MCI, mild cognitive impairment likely due to Alzheimer’s disease; non-AD MCI, MCI unlikely due to AD; MMSE, Mini-Mental State Examination; NPI, Neuropsychiatric Inventory; GDS, Geriatric Depression Scale; BADL, basic activities of daily living; IADL, instrumental activities of daily living; t-Tau, total Tau; p-Tau₁₈₁, phosphorylated Tau₁₈₁; Aβ_{1 - 42}, amyloid beta_{1 - 42}; PET, positron emission tomography; TMS, transcranial magnetic stimulation; rMT, resting motor threshold; MSO, percentage of maximal stimulator output; SICI-ICF, mean short interval intracortical inhibition (1, 2, 3 ms) / intracortical facilitation (7, 10, 15 ms); SAI, mean short latency afferent inhibition (0, ⁺4 ms); SICI-ICF / SAI, ratio between SICI-ICF and SAI parameters; n.s., not significant. *p-values for independent samples t-test or Fisher’s exact test, as appropriate.

Neurophysiological measures

For SICI-ICF, repeated measures ANOVA highlighted an ISI×GROUP interaction, F(2.73, 182.82) = 37.61, p < 0.001, partial η² = 0.36, ɛ= 0.46, with post hoc comparisons showing a significant difference between AD MCI and non-AD MCI at ISI 1, 2, 7, 10, 15 ms (all p < 0.001), but not at ISI 5 (p = 0.525) (see Fig. 2A).

Fig.2

Neurophysiological parameters in AD MCI and non-AD MCI. A) Short-interval intracortical inhibition (SICI) at ISI 1, 2, 3, 5 and intracortical facilitation (ICF) at ISI 7, 10, 15 ms, (B) short-latency afferent inhibition (SAI) at ISI –4, 0, +4, +8 ms, in AD MCI, non-AD MCI. Data are represented as a ratio to the unconditioned motor evoked potential amplitude; error bars represent standard errors. AD MCI, mild cognitive impairment likely due to Alzheimer’s disease; non-AD MCI, MCI unlikely due to AD; MEP, motor evoked potential; ISI, inter stimulus interval. *p < 0.05 using one-way ANOVA (post hoc tests with Bonferroni correction for multiple comparisons).

Repeated measures ANOVA performed on SAI revealed a significant ISI within-subjects effect, F(1.63, 108.85) = 24.00, p < 0.001, partial η² = 0.26, ɛ= 0.54, and a significant GROUP between-subjects effect, F(1.0, 67.0) = 23.78, p < 0.001, partial η² = 0.26, but not an ISI×GROUP interaction, p = 0.270. A significant difference between AD MCI and non-AD MCI was observed at ISI ^-4 (p = 0.026), at ISI 0, ⁺4 ms (all p < 0.001), but not at ISI ⁺8 (p = 0.210) after post hoc comparisons (see Fig. 2B).

Diagnostic accuracy of neurophysiological parameters

A ROC curve with AUCs was used to compare how effectively neurophysiological measures differentiated patients with AD MCI from those with non-AD MCI.

Mean SICI-ICF (1, 2, 3 ms / 7, 10, 15 ms) showed an AUC of 0.88 (p < 0.001, 95% CI 0.78–0.97) and mean SAI (0, ⁺4 ms) of 0.87 (p < 0.001, 95% CI 0.77–0.97). Using the SICI-ICF / SAI index, AUC was of 0.94 (p < 0.001, 95% CI 0.87–1.00) (see Fig. 3). At the best cutoff score of 0.544 for SICI-ICF/ SAI ratio, sensitivity was 94.4%, specificity was 87.9%, positive predictive value was 89.5%, negative predictive value was 93.5%, and accuracy was 91.3% (Table 3).

Fig.3

Receiver operating characteristic curve for neurophysiological parameters in differentiating AD MCI from non-AD MCI. AD MCI, mild cognitive impairment likely due to Alzheimer’s disease; non-AD MCI, MCI unlikely due to AD; SICI-ICF, mean short-interval intracortical inhibition (1, 2, 3 ms) / intracortical facilitation (7, 10, 15 ms); SAI, mean short-latency afferent inhibition (0, ⁺4 ms); SICI-ICF / SAI, ratio between SICI-ICF and SAI parameters.

Table 3

Area under the curve, sensitivity, specificity, accuracy, positive and negative predictive values for receiver operating characteristics curves in differentiating AD MCI from non-AD MCI

Neurophysiological test	SICI-ICF	SAI	SICI-ICF / SAI
AUC (95% CI)	0.88 (0.78–0.97)	0.87 (0.77–0.97)	0.94 (0.87–1.00)
Sensitivity (%)	97.2	97.2	94.4
Specificity (%)	81.8	75.8	87.9
Positive Predictive Value (%)	85.4	81.4	89.5
Negative Predictive Value (%)	96.4	96.2	93.5
Accuracy (%)	89.9	87.0	91.3

AD MCI, mild cognitive impairment likely due to Alzheimer’s disease; non-AD MCI, MCI unlikely due to AD; AUC, area under the curve; SICI-ICF, mean short interval intracortical inhibition (1, 2, 3 ms) / intracortical facilitation (7, 10, 15 ms); SAI, mean short latency afferent inhibition (0, ⁺4 ms); SICI-ICF / SAI, ratio between SICI-ICF and SAI parameters.

With regard to the two participants with contrasting biomarkers, 1 subject was a 56-year-old male with a positive PET amyloid imaging and a CSF non-AD like profile (t-Tau: 201 pg/mL, p-Tau₁₈₁: 41 pg/mL, Aβ_{1 - 42}: 1019), and a SICI-ICF / SAI ratio of 1.340 (classified as non-AD MCI by TMS). The other participant was a 72-year-old female with a negative PET amyloid imaging and a CSF AD-like profile (t-Tau: 494 pg/mL, p-Tau₁₈₁: 81 pg/mL, Aβ_{1 - 42}: 517), and a SICI-ICF / SAI ratio of 0.193 (classified as AD MCI by TMS).

Correlations between neurophysiological parameters and CSF measures

In the AD MCI group, correlation analyses revealed a strong positive correlation between mean SAI (0, ⁺4 ms) and t-Tau (r = 0.754, p = 0.001), p-Tau₁₈₁ (r = 0.585, p = 0.022) but not Aβ_{1 - 42} (r = 0.037, p = 0.895). Mean SICI-ICF (1, 2, 3 ms / 7, 10, 15 ms) correlated with t-Tau (r = 0.554, p = 0.032) but not p-Tau₁₈₁ (r = 0.456, p = 0.088) or Aβ_{1 - 42} (r = 0.040, p = 0.887).

In the non-AD MCI group, correlation analyses did not reveal any significant correlation between mean SAI (0, ⁺4 ms) and t-Tau (r = 0.000, p = 0.999), p-Tau₁₈₁ (r = –0.052, p = 0.815) or Aβ_{1 - 42} (r = –0.097, p = 0.658). Mean SICI-ICF (1, 2, 3 ms / 7, 10, 15 ms) showed a strong correlation with t-Tau (r = 0.563, p = 0.005) but not p-Tau₁₈₁ (r = 0.186, p = 0.395) or Aβ_{1 - 42} (r = –0.124, p = 0.572).

DISCUSSION

The recent development of new disease-modifying drugs will increase the relevance and urgency for an accurate diagnosis, to properly stage disease, and to monitor response to treatment, thus increasing the demand for accurate and early biomarkers.

In the present work, we observed a significant impairment in specific neurophysiological parameters in MCI. AD MCI subjects showed a remarkable disruption of cholinergic circuits, as highlighted by an impairment of SAI circuits. On the other hand, non-AD MCI subjects showed a significant disruption of GABA_Aergic and glutamatergic transmission, as highlighted by the impairment of SICI and ICF circuits, while cholinergic transmission resulted unaffected. These observations confirm previous findings in which an impairment of cholinergic circuits has been shown in amnestic MCI [26] and in AD patients [11, 15]. We have also observed an association between CSF biomarkers and neurophysiological parameters, with a significant correlation between SAI and t-Tau, p-Tau₁₈₁ but not Aβ_{1 - 42}, and a significant correlation between SICI-ICF and t-Tau but not p-Tau₁₈₁ or Aβ_{1 - 42} in AD MCI patients. These results partially confirm previous findings in AD patients, where SAI correlated with CSF p-Tau₁₈₁ but also with Aβ_{1 - 42} [12], which, however, was not confirmed in our study, while another study did not find any correlations between SAI and CSF biomarkers [13]. This could possibly be due to the nature of CSF Aβ_{1 - 42} levels, which have been shown to precede t-Tau accumulation and reach a plateau as early as the MCI phase, as supported by several pathophysiological models of dynamic biomarkers in AD [27, 28], or by different thresholds applied on CSF measurements.

Previous studies on MCI patients with a biomarker supported diagnosis, and thus classifiable as prodromal AD by current diagnostic criteria, are currently lacking.

The impairment of SICI-ICF in our sample of non-AD MCI, which were predominantly classified as FTD MCI, is in line with previous findings in symptomatic FTLD patients [11 , 30] and in the prodromal phases of monogenic FTD [31].

Taking into consideration these findings, we obtained high levels of diagnostic accuracy using the SICI-ICF / SAI ratio, with sensitivity of 94.4% and specificity of 87.9% in differentiating AD MCI from non-AD MCI, comparable to other validated biomarkers.

Indeed, CSF and amyloid PET imaging biomarkers, have shown levels of diagnostic accuracy of 85–95% in clinico-pathological studies [32, 33]. Thus, on the basis of the above evidence, we argue that the association between neurophysiological measures and other biomarkers, as CSF analysis or amyloid PET imaging, could represent a useful strategy to improve AD diagnostic accuracy, even in the prodromal phases of disease. Indeed, the most widely accepted diagnostic criteria assume that the greatest accuracy can be achieved with a combination of amyloidosis biomarkers and neurodegeneration markers. The findings with these combinations, however, are inconsistent [34 –36], and simply suggest that the use of more biomarkers might improve accuracy [37]. In this view, the role of additional biomarkers, as TMS, could become useful particularly in cases with contrasting biomarkers of neurodegeneration or amyloidosis obtained from different techniques, in cases where these biomarkers are unavailable, or contraindicated in the single patient. Furthermore, considering the high sensitivity of the technique (95%), the test could be particularly suitable to be used as a screening tool in the initial diagnostic assessment, selecting candidate patients to undergo further testing, as CSF analysis or amyloid PET imaging.

This concept is further emphasized by the possible drawbacks of currently used biomarkers. Amyloid PET imaging, besides being expensive and not always available in all centers, has been shown to have high sensitivity and specificity for brain amyloidosis, and not necessarily for AD, particularly in the elderly population, thus resulting more useful as an exclusion criteria for AD in cases without brain amyloidosis [38]. Moreover, CSF analysis is still an invasive procedure with a low, albeit possible risk of complications, with significant contraindications and pre-analytical limitations [39, 40].

Herein we suggest the SAI and the SICI-ICF/SAI ratio as candidate biomarkers with several advantages, as this is: 1) non-invasive, reliable, easy to apply and inexpensive; 2) not time-consuming; and 3) suitable as a screening tool to differentiate AD MCI from non-AD MCI with accuracy levels comparable to validated biomarkers. Considering that exclusion rates are low, patients with electronic implants or epilepsy are not suitable candidates for TMS procedures.

In conclusion, the assessment of intracortical connectivity with TMS might be suggested in the diagnostic work-up of subjects with suspected neurodegenerative disorder, to be added to currently used biomarkers. Comparative studies on larger samples, evaluation of TMS parameters in well-characterized subjects with different MCI subtypes and longitudinal follow-up might be helpful to support the use of TMS in the diagnostic algorithm of neurodegenerative dementias.

Footnotes

ACKNOWLEDGMENTS

This study was supported by grants from “AIRAlzh Onlus” and “ANCC-COOP” issued to VC. Amyloid PET imaging was carried out in the context of the “The Incremental Diagnostic Value of Amyloid PET with ¹⁸F-Florbetapir” (INDIA-FBP) study.

The authors are in debt with Dr. Silvana Archetti for laboratory analyses, with Drs. Barbara Paghera and Paolo Guerra for imaging scan acquisition, and with Dr. Roberto Grasso for patient data retrieval.

Authors’ disclosures available online ().

References

Blennow

, Hampel

, Zetterberg

(2014) Biomarkers in amyloid-β immunotherapy trials in Alzheimer’s disease. Neuropsychopharmacology 39, 189–201.

Andrieu

, Coley

, Lovestone

, Aisen

, Vellas

(2015) Prevention of sporadic Alzheimer’s disease: Lessons learned from clinical trials and future directions. Lancet Neurol 14, 926–944.

Dubois

, Feldman

, Jacova

, Hampel

, Molinuevo

, Blennow

, Dekosky

, Gauthier

, Selkoe

, Bateman

, Cappa

, Crutch

, Engelborghs

, Frisoni

, Fox

, Galasko

, Habert

M-O

, Jicha

, Nordberg

, Pasquier

, Rabinovici

, Robert

, Rowe

, Salloway

, Sarazin

, Epelbaum

, De Souza

, Vellas

, Visser

, Schneider

, Stern

, Scheltens

, Cummings

(2014) Advancing research diagnostic criteria for Alzheimer’s disease: The IWG-2 criteria. Lancet Neurol 13, 614–629.

Petersen

, Smith

, Waring

, Ivnik

, Tangalos

, Kokmen

(1999) Mild cognitive impairment: Clinical characterization and outcome. Arch Neurol 56, 303–308.

Albert

, Dekosky

, Dickson

, Dubois

, Feldman

, Fox

, Gamst

, Holtzman

, Jagust

, Petersen

, Snyder

, Carrillo

, Thies

, Phelps

(2011) The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer‘s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 270–279.

Winblad

, Palmer

, Kivipelto

, Jelic

, Fratiglioni

, Wahlund

, Nordberg

, Bäckman

, Albert

, Almkvist

, Arai

, Basun

, Blennow

, de Leon

, DeCarli

, Erkinjuntti

, Giacobini

, Graff

, Hardy

, Jack

, Jorm

, Ritchie

, van Duijn

, Visser

, Petersen

(2004) Mild cognitive impairment - Beyond controversies, towards a consensus: Report of the International Working Group on Mild Cognitive Impairment. J Intern Med 256, 240–246.

Bennett

, Wilson

, Boyle

, Buchman

, Schneider

(2012) Relation of neuropathology to cognition in persons without cognitive impairment. Ann Neurol 72, 599–609.

Jack

, Lowe

, Senjem

, Weigand

, Kemp

, Shiung

, Knopman

, Boeve

, Klunk

, Mathis

, Petersen

(2008) 11C PiB and structural MRI provide complementary information in imaging of Alzheimer’s disease and amnestic mild cognitive impairment. Brain 131, 665–680.

Abner

, Kryscio

, Schmitt

, Fardo

, Moga

, Ighodaro

, Jicha

, Yu

, Dodge

, Xiong

, Woltjer

, Schneider

, Cairns

, Bennett

, Nelson

(2017) Outcomes after diagnosis of mild cognitive impairment in a large autopsy series. Ann Neurol 81, 549–559.

10.

Frisoni

, Bocchetta

, Chételat

, Rabinovici

, de Leon

, Kaye

, Reiman

, Scheltens

, Barkhof

, Black

, Brooks

, Carrillo

, Fox

, Herholz

, Nordberg

, Jack

, Jagust

, Johnson

, Rowe

, Sperling

, Thies

, Wahlund

L-O

, Weiner

, Pasqualetti

, DeCarli

(2013) Imaging markers for Alzheimer disease: Which vs how. Neurology 81, 1–14.

11.

Benussi

, Di

Lorenzo F

, Dell’Era

, Cosseddu

, Alberici

, Caratozzolo

, Cotelli

, Micheli

, Rozzini

, Depari

, Flammini

, Ponzo

, Martorana

, Caltagirone

, Padovani

, Koch

, Borroni

(2017) Transcranial magnetic stimulation distinguishes Alzheimer disease from frontotemporal dementia. Neurology 89, 665–672.

12.

Martorana

, Esposito

, Di Lorenzo

, Giacobbe

, Sancesario

, Bucchi

, Bonní

, Bernardini

, Sorge

, Sancesario

, Bernardi

, Caltagirone

, Koch

(2012) Cerebrospinal fluid levels of Aβ42 relationship with cholinergic cortical activity in Alzheimer’s disease patients. J Neural Transm 119, 771–778.

13.

Koch

, Di Lorenzo

, Del Olmo

, Bonní

, Ponzo

, Caltagirone

, Bozzali

, Martorana

(2016) Reversal of LTP-like cortical plasticity in Alzheimer’s disease patients with tau-related faster clinical progression. J Alzheimers Dis 50, 605–616.

14.

Francis

, Palmer

, Snape

, Wilcock

(1999) The cholinergic hypothesis of Alzheimer’s disease: A review of progress. J Neurol Neurosurg Psychiatry 66, 137–147.

15.

Freitas

, Mondragón-Llorca

, Pascual-Leone

(2011) Noninvasive brain stimulation in Alzheimer’s disease: Systematic review and perspectives for the future. Exp Gerontol 46, 611–627.

16.

Mufson

, Binder

, Counts

, Dekosky

, de Toledo-Morrell

, Ginsberg

, Ikonomovic

, Perez

, Scheff

(2012) Mild cognitive impairment: Pathology and mechanisms. Acta Neuropathol 123, 13–30.

17.

Teipel

, Meindl

, Grinberg

, Grothe

, Cantero

, Reiser

, Möller

H-J

, Heinsen

, Hampel

(2011) The cholinergic system in mild cognitive impairment and Alzheimer’s disease: An in vivo MRI and DTI study. Hum Brain Mapp 32, 1349–1362.

18.

Petersen

(2004) Mild cognitive impairment as a diagnostic entity. J Intern Med 256, 183–194.

19.

Borroni

, Benussi

, Archetti

, Galimberti

, Parnetti

, Nacmias

, Sorbi

, Scarpini

, Padovani

(2015) CSF p-tau181/tau ratio as biomarker for TDP pathology in frontotemporal dementia. Amyotroph Lateral Scler Frontotemporal Degener 16, 86–91.

20.

Boccardi

, Altomare

, Ferrari

, Festari

, Guerra

, Paghera

, Pizzocaro

, Lussignoli

, Geroldi

, Zanetti

, Cotelli

, Turla

, Borroni

, Rozzini

, Mirabile

, Defanti

, Gennuso

, Prelle

, Gentile

, Morandi

, Vollaro

, Volta

, Bianchetti

, Conti

, Cappuccio

, Carbone

, Bellandi

, Abruzzi

, Bettoni

, Villani

, Raimondi

, Lanari

, Ciccone

, Facchi

, Di Fazio

, Rozzini

, Boffelli

, Manzoni

, Salvi

, Cavaliere

, Belotti

, Avanzi

, Pasqualetti

, Muscio

, Padovani

, Frisoni

; Incremental Diagnostic Value of Amyloid PET With [18F]-Florbetapir (INDIA-FBP) Working Group (2016) Assessment of the Incremental diagnostic value of Florbetapir F 18 imaging in patients with cognitive impairment: The Incremental Diagnostic Value of Amyloid PET with [18F]-Florbetapir (INDIA-FBP) Study. JAMA Neurol 73, 1417–1424.

21.

Benussi

, Cotelli

, Cosseddu

, Bertasi

, Turla

, Salsano

, Dardis

, Padovani

, Borroni

(2017) Preliminary results on long-term potentiation-like cortical plasticity and cholinergic dysfunction after miglustat treatment in Niemann-Pick Disease Type C. JIMD Rep 36, 19–27.

22.

Rossini

, Burke

, Chen

, Cohen

, Daskalakis

, Di Iorio

, Di Lazzaro

, Ferreri

, Fitzgerald

, George

, Hallett

, Lefaucheur

, Langguth

, Matsumoto

, Miniussi

, Nitsche

, Pascual-Leone

, Paulus

, Rossi

, Rothwell

, Siebner

, Ugawa

, Walsh

, Ziemann

(2015) Non-invasive electrical and magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: Basic principles and procedures for routine clinical and research application. An updated report from an I.F.C.N. Committee. Clin Neurophysiol 126, 1071–1107.

23.

Kujirai

, Caramia

, Rothwell

, Day

, Thompson

, Ferbert

, Wroe

, Asselman

, Marsden

(1993) Corticocortical inhibition in human motor cortex. J Physiol 471, 501–519.

24.

Ziemann

, Rothwell

, Ridding

(1996) Interaction between intracortical inhibition and facilitation in human motor cortex. J Physiol 496, 873–881.

25.

Tokimura

, Di

Lazzaro V

, Tokimura

, Oliviero

, Profice

, Insola

, Mazzone

, Tonali

, Rothwell

(2000) Short latency inhibition of human hand motor cortex by somatosensory input from the hand. J Physiol 523 Pt 2, 503–513.

26.

Nardone

, Bergmann

, Kronbichler

, Kunz

, Klein

, Caleri

, Tezzon

, Ladurner

, Golaszewski

(2008) Abnormal short latency afferent inhibition in early Alzheimer’s disease: A transcranial magnetic demonstration. J Neural Transm 115, 1557–1562.

27.

Jack

Jr , Knopman

, Jagust

, Shaw

, Aisen

, Weiner

, Petersen

, Trojanowski

(2010) Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol 9, 119–128.

28.

Jack

Jr , Knopman

, Jagust

, Petersen

, Weiner

, Aisen

, Shaw

, Vemuri

, Wiste

, Weigand

, Lesnick

, Pankratz

, Donohue

, Trojanowski

(2013) Tracking pathophysiological processes in Alzheimer’s disease: An updated hypothetical model of dynamic biomarkers. Lancet Neurol 12, 207–216.

29.

Burrell

, Kiernan

, Vucic

, Hodges

(2011) Motor neuron dysfunction in frontotemporal dementia. Brain 134, 2582–2594.

30.

Benussi

, Dell’Era

, Cantoni

, Ferrari

, Caratozzolo

, Rozzini

, Alberici

, Padovani

, Borroni

(2018) Discrimination of atypical parkinsonisms with transcranial magnetic stimulation. Brain Stimul 11, 366–373.

31.

Benussi

, Cosseddu

, Filareto

, Dell’Era

, Archetti

, Sofia Cotelli

, Micheli

, Padovani

, Borroni

(2016) Impaired long-term potentiation-like cortical plasticity in presymptomatic genetic frontotemporal dementia. Ann Neurol 80, 472–476.

32.

Shaw

, Vanderstichele

, Knapik-Czajka

, Clark

, Aisen

, Petersen

, Blennow

, Soares

, Simon

, Lewczuk

, Dean

, Siemers

, Potter

, Lee

VMY

, Trojanowski

, Alzheimer’s Disease Neuroimaging Initiative (2009) Cerebrospinal fluid biomarker signature in Alzheimer’s disease neuroimaging initiative subjects. Ann Neurol 65, 403–413.

33.

Clark

, Pontecorvo

, Beach

, Bedell

, Coleman

, Doraiswamy

, Fleisher

, Reiman

, Sabbagh

, Sadowsky

, Schneider

, Arora

, Carpenter

, Flitter

, Joshi

, Krautkramer

, Lu

, Mintun

, Skovronsky

(2012) Cerebral PET with florbetapir compared with neuropathology at autopsy for detection of neuritic amyloid-β plaques: A prospective cohort study. Lancet Neurol 11, 669–678.

34.

Garibotto

, Herholz

, Boccardi

, Picco

, Varrone

, Nordberg

, Nobili

, Ratib

(2017) Clinical validity of brain fluorodeoxyglucose positron emission tomography as a biomarker for Alzheimer’s disease in the context of a structured 5-phase development framework. Neurobiol Aging 52, 183–195.

35.

Sonni

, Ratib

, Boccardi

, Picco

, Herholz

, Nobili

, Varrone

(2017) Clinical validity of presynaptic dopaminergic imaging with 123 I-ioflupane and noradrenergic imaging with 123 I-MIBG in the differential diagnosis between Alzheimer’s disease and dementia with Lewy bodies in the context of a structured 5-phase development framework. Neurobiol Aging 52, 228–242.

36.

Mattsson

, Lönneborg

, Boccardi

, Blennow

, Hansson

(2017) Clinical validity of cerebrospinal fluid Aβ42, tau, and phospho-tau as biomarkers for Alzheimer’s disease in the context of a structured 5-phase development framework. Neurobiol Aging 52, 196–213.

37.

Frisoni

, Boccardi

, Barkhof

, Blennow

, Cappa

, Chiotis

, Démonet

J-F

, Garibotto

, Giannakopoulos

, Gietl

, Hansson

, Herholz

, Jack

Jr , Nobili

, Nordberg

, Snyder

, Kate Ten

, Varrone

, Albanese

, Becker

, Bossuyt

, Carrillo

, Cerami

, Dubois

, Gallo

, Giacobini

, Gold

, Hurst

, Lönneborg

, Lovblad

K-O

, Mattsson

, Molinuevo

, Monsch

, Mosimann

, Padovani

, Picco

, Porteri

, Ratib

, Saint-Aubert

, Scerri

, Scheltens

, Schott

, Sonni

, Teipel

, Vineis

, Visser

, Yasui

, Winblad

(2017) Strategic roadmap for an early diagnosis of Alzheimer’s disease based on biomarkers. Lancet Neurol 16, 661–676.

38.

Ossenkoppele

, Jansen

, Rabinovici

, Knol

, van der Flier

, van Berckel

BNM

, Scheltens

, Visser

, Amyloid PET Study Group, Verfaillie

SCJ

, Zwan

, Adriaanse

, Lammertsma

, Barkhof

, Jagust

, Miller

, Rosen

, Landau

, Villemagne

, Rowe

, Lee

, Na

, Seo

, Sarazin

, Roe

, Sabri

, Barthel

, Koglin

, Hodges

, Leyton

, Vandenberghe

, Van Laere

, Drzezga

, Forster

, Grimmer

, Sánchez-Juan

, Carril

, Mok

, Camus

, Klunk

, Cohen

, Meyer

, Hellwig

, Newberg

, Frederiksen

, Fleisher

, Mintun

, Wolk

, Nordberg

, Rinne

, Chételat

, Lleo

, Blesa

, Fortea

, Madsen

, Rodrigue

, Brooks

(2015) Prevalence of amyloid PET positivity in dementia syndromes: A meta-analysis. JAMA 313, 1939–1949.

39.

Engelborghs

, Niemantsverdriet

, Struyfs

, Blennow

, Brouns

, Comabella

, Dujmovic

, van der Flier

, Frölich

, Galimberti

, Gnanapavan

, Hemmer

, Hoff

, Hort

, Iacobaeus

, Ingelsson

, Jan de Jong

, Jonsson

, Khalil

, Kuhle

, Lleo

, de Mendonça

, Molinuevo

, Nagels

, Paquet

, Parnetti

, Roks

, Rosa-Neto

, Scheltens

, Skårsgard

, Stomrud

, Tumani

, Visser

, Wallin

, Winblad

, Zetterberg

, Duits

, Teunissen

(2017) Consensus guidelines for lumbar puncture in patients with neurological diseases. Alzheimers Dement (Amst) 8, 111–126.

40.

Ritchie

, Smailagic

, Noel-Storr

, Ukoumunne

, Ladds

, Martin

(2017) CSF tau and the CSF tau/ABeta ratio for the diagnosis of Alzheimer’s disease dementia and other dementias in people with mild cognitive impairment (MCI). Cochrane Database Syst Rev 2017, CD010803.