Heterogeneity in subjective cognitive decline in the Sino Longitudinal Study on Cognitive Decline(SILCODE): Empirically derived subtypes, structural and functional verification

AIMS: We evaluated whether Subjective Cognitive Decline (SCD) subtypes could be empirically derived within the Sino Longitudinal Study on Cognitive Decline (SILCODE) SCD cohort and examined associated neuroimaging markers, biomarkers, and clinical outcomes. METHODS: A cluster analysis was performed on eight neuropsychological test scores from 124 SCD SILCODE participants and 57 normal control (NC) subjects. Structural and functional neuroimaging indices were used to evaluate the SCD subgroups. RESULTS: Four subtypes emerged: (1) dysexecutive/mixed SCD (n = 23), (2) neuropsychiatric SCD (n = 24), (3) amnestic SCD (n = 22), and (4) cluster‐derived normal (n = 55) who exhibited normal performance in neuropsychological tests. Compared with the NC group, each subgroup showed distinct patterns in gray matter (GM) volume and the amplitude of low‐frequency fluctuations (ALFF). Lower fractional anisotropy (FA) values were only found in the neuropsychiatric SCD group relative to NC. CONCLUSION: The identification of empirically derived SCD subtypes demonstrates the presence of heterogeneity in SCD neuropsychological profiles. The cluster‐derived normal group may represent the majority of SCD individuals who do not show progressive cognitive decline; the dysexecutive/mixed SCD and amnestic SCD might represent high‐risk groups with progressing cognitive decline; and finally, the neuropsychiatric SCD may represent a new topic in SCD research.