Clinical Summary
Clinical Summary: Neuropsychiatric Symptom Clusters and Their Association With Brain Structure in Alzheimer Disease
Neuropsychiatric symptoms are nearly ubiquitous in Alzheimer disease and often drive disability, distress, and faster clinical decline more than cognitive symptoms alone. This study helps clinicians think beyond single symptoms by showing that reproducible symptom clusters track with specific brain atrophy patterns and with worsening daily function over time.
Design
ONDRI is a multicenter, longitudinal cohort characterizing degenerative cognitive impairment across memory clinics in Ontario, Canada.
N
Of 126 participants, 111 (88%) had sufficient NPI-Q data for PCA
Population
participants with AD (mild cognitive impairment [MCI] or dementia due to AD) from the ONDRI Cohort
Duration
at baseline and at 1-and 2-year follow-ups
Key Findings
- Preliminary tests supported factorability (KMO = 0.74; Bartlett P < .001). PCA initially yielded 5 components, explaining 70.7% of the variance. The final model included 4 components, explaining 62.4% of the variance.
- The neurovegetative cluster showed the strongest longitudinal brain-volume associations, including left middle temporal (β=−0.50, 95% CI =−0.72 to −0.28, P<.001), right nucleus accumbens (β=−0.55, 95% CI =−0.75 to −0.35, P<.001), and left fusiform (β=−0.36, 95% CI =−0.58 to −0.14, P=.002) volumes.
- The affective cluster was longitudinally associated with lower volumes in left rostral anterior cingulate (β=−0.42, 95% CI =−0.67 to −0.16, P=.002), right entorhinal (β=−0.51, 95% CI =−0.84 to −0.17, P=.004), right medial orbitofrontal (β=−0.47, 95% CI =−0.75 to −0.20, P=.001), right pars opercularis (β=−0.44, 95% CI =−0.69 to −0.18, P= .001), and left putamen (β=−0.44, 95% CI =−0.72 to −0.17, P= .002).
- The hyperactivity cluster was longitudinally associated with left middle temporal (β=−0.24, 95% CI =−0.45 to −0.03, P=.025), right nucleus accumbens (β=−0.28, 95% CI =−0.48 to −0.08, P=.007), and left ventral diencephalon (β=−0.25, 95% CI =−0.44 to −0.05, P=.013) volumes.
- Psychosis symptoms were too infrequent for imaging follow-up analysis: 13.9%, 18.2%, and 14.3% of participants achieved scores of >0 at baseline and 1-and 2-year follow-up visits, respectively, so this cluster was excluded from further analyses.
Clinical Bottom Line
Together, these findings support a cluster-based symptom paradigm that reflects co-occurring behavioral dimensions, aligns with underlying circuitry, and relates to functional decline, informing biologically grounded stratification and potential targets for precision-oriented care.
Practice Implications
- Clinicians may use symptom clusters, rather than individual symptoms alone, to anticipate functional decline and guide monitoring and intervention priorities.
- Grouping co-occurring neuropsychiatric symptoms into clusters provides a biologically grounded framework that aligns specific behavioral profiles with regional brain atrophy.
- All clusters showed longitudinal associations with iADL performance; higher NPS burden, male sex, later visits, older age, and lower MoCA were associated with worse iADLs.
- Clusters 1 (hyperactivity), 3 (neurovegetative), and 4 (affective) were also associated with ADL performance, with time and age predicting worse ADLs across clusters.
