Related Experiment Video
Updated: Apr 18, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Mild cognitive impairment with suspected nonamyloid pathology (SNAP): Prediction of progression
Anna Caroli1, Annapaola Prestia1, Samantha Galluzzi1
1From the Medical Imaging Unit (A.C.), Biomedical Engineering Department, IRCCS Istituto di Ricerche Farmacologiche Mario Negri, Bergamo; LENITEM-Laboratory of Epidemiology Neuroimaging and Telemedicine (A.P., S.G., C.F., A.R., G.B.F.), IRCCS Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy; Alzheimer Center and Department of Neurology (W.M.v.d.F., R.O., P.S.), Department of Epidemiology & Biostatistics (W.M.v.d.F.), Department of Radiology & Nuclear Medicine (B.V.B., F.B.), and Neurochemistry Laboratorium and Biobank, Department of Clinical Chemistry (C.T.), VU University Medical Center, Amsterdam, the Netherlands; PET-Center (A.E.W.), Section of Nuclear Medicine & PET, Department of Radiology, Oncology and Radiation Sciences, Uppsala University; Alzheimer Neurobiology Center (S.F.C., M.S., I.H.C., A.N.), Karolinska Institutet, Stockholm, Sweden; Wolfson Molecular Imaging Centre (S.F.C.), University of Manchester, UK; MedTech West (M.S.), Sahlgrenska University Hospital, University of Gothenburg, Sweden; Department of Neuropsychiatry (I.H.C.), School of Medicine, Chosun University, Gwangju, Republic of Korea; Department of Psychiatry and Psychotherapy (T.G.), Klinikum rechts der Isar, Technische Universitat Muenchen, Germany; Department of Geriatric Medicine (A.N.), Karolinska University Hospital Huddinge, Stockholm, Sweden; Department of Nuclear Medicine (A.D.), University of Cologne, Germany; and Departments of Internal Medicine and Psychiatry (G.B.F.), University Hospitals and University of Geneva, Switzerland.
Objectives:
The aim of this study was to investigate predictors of progressive cognitive deterioration in patients with suspected non-Alzheimer disease pathology (SNAP) and mild cognitive impairment (MCI).
Methods:
We measured markers of amyloid pathology (CSF β-amyloid 42) and neurodegeneration (hippocampal volume on MRI and cortical metabolism on [(18)F]-fluorodeoxyglucose-PET) in 201 patients with MCI clinically followed for up to 6 years to detect progressive cognitive deterioration. We categorized patients with MCI as A+/A- and N+/N- based on presence/absence of amyloid pathology and neurodegeneration. SNAPs were A-N+ cases.
Results:
The proportion of progressors was 11% (8/41), 34% (14/41), 56% (19/34), and 71% (60/85) in A-N-, A+N-, SNAP, and A+N+, respectively; the proportion of APOE ε4 carriers was 29%, 70%, 31%, and 71%, respectively, with the SNAP group featuring a significantly different proportion than both A+N- and A+N+ groups (p ≤ 0.005). Hypometabolism in SNAP patients was comparable to A+N+ patients (p = 0.154), while hippocampal atrophy was more severe in SNAP patients (p = 0.002). Compared with A-N-, SNAP and A+N+ patients had significant risk of progressive cognitive deterioration (hazard ratio = 2.7 and 3.8, p = 0.016 and p < 0.001), while A+N- patients did not (hazard ratio = 1.13, p = 0.771). In A+N- and A+N+ groups, none of the biomarkers predicted time to progression. In the SNAP group, lower time to progression was correlated with greater hypometabolism (r = 0.42, p = 0.073).
Conclusions:
Our findings support the notion that patients with SNAP MCI feature a specific risk progression profile.
Insights
Patients with suspected non-Alzheimer disease pathology (SNAP) and mild cognitive impairment (MCI) show a distinct progression risk. Neurodegeneration markers, not amyloid, predict cognitive decline in SNAP MCI.
Area of Science:
- Neurology
- Neuroscience
- Gerontology
Background:
- Mild cognitive impairment (MCI) affects a significant portion of the aging population.
- Identifying predictors of cognitive deterioration is crucial for early intervention.
- Suspected non-Alzheimer disease pathology (SNAP) represents a distinct subtype of MCI.
Purpose of the Study:
- To investigate predictors of progressive cognitive deterioration in patients with MCI.
- To differentiate the progression profiles of patients with SNAP and other MCI subtypes.
- To identify specific pathological markers associated with cognitive decline in MCI.
Main Methods:
- 201 patients with MCI were assessed for amyloid pathology (CSF β-amyloid 42) and neurodegeneration (hippocampal volume, [18F]-fluorodeoxyglucose-PET).
- Patients were categorized into four groups: A-N-, A+N-, SNAP (A-N+), and A+N+.
- Clinical follow-up for up to 6 years identified progressive cognitive deterioration.
Main Results:
- SNAP and A+N+ patients had a significantly higher risk of cognitive progression compared to A-N- patients.
- SNAP patients exhibited more severe hippocampal atrophy and comparable hypometabolism to A+N+ patients.
- In SNAP patients, greater hypometabolism correlated with a shorter time to progression.
Conclusions:
- Patients with SNAP MCI possess a unique risk progression profile.
- Neurodegenerative markers are key predictors of cognitive decline in SNAP MCI.
- This study highlights the importance of differentiating MCI subtypes for understanding disease trajectory.
Related Concept Videos
Alzheimer's Disease: Overview
The clinical diagnosis of AD hinges on the presence of memory and other cognitive impairments. Biomarkers, such as changes in Aβ...
Alzheimer Disease l: Introduction
Dementia l: Introduction
Alzheimer Disease ll: Pathophysiology
Dementia
The progression of dementia is generally gradual....
Alzheimer's Disease: Treatment

