Phenotyping Superagers Using Resting-State fMRI

Abstract

Background and purpose: Superagers are defined as older adults with episodic memory performance similar or superior to that in middle-aged adults. This study aimed to investigate the key differences in discriminative networks and their main nodes between superagers and cognitively average elderly controls. In addition, we sought to explore differences in sensitivity in detecting these functional activities across the networks at 3T and 7T MR imaging fields.

Materials and methods: Fifty-five subjects 80 years of age or older were screened using a detailed neuropsychological protocol, and 31 participants, comprising 14 superagers and 17 cognitively average elderly controls, were included for analysis. Participants underwent resting-state-fMRI at 3T and 7T MR imaging. A prediction classification algorithm using a penalized regression model on the measurements of the network was used to calculate the probabilities of a healthy older adult being a superager. Additionally, ORs quantified the influence of each node across preselected networks.

Results: The key networks that differentiated superagers and elderly controls were the default mode, salience, and language networks. The most discriminative nodes (ORs > 1) in superagers encompassed areas in the precuneus posterior cingulate cortex, prefrontal cortex, temporoparietal junction, temporal pole, extrastriate superior cortex, and insula. The prediction classification model for being a superager showed better performance using the 7T compared with 3T resting-state-fMRI data set.

Conclusions: Our findings suggest that the functional connectivity in the default mode, salience, and language networks can provide potential imaging biomarkers for predicting superagers. The 7T field holds promise for the most appropriate study setting to accurately detect the functional connectivity patterns in superagers.

FIG 1.

Flow chart of participant selection. RAVLT indicates the Rey Auditory Verbal Learning Test; y/o, years of age.

FIG 2.

Plots showing the classification results for superagers across several networks examined on 3T and 7T fields. These plots show the observed superager status for each participant (blue and red dots) plotted against the probability of being a superager predicted from the fitted model. The diagonal lines represent the mean difference between predicted probabilities for superagers and elderly controls. The steeper the gradient of the lines, the higher the superagers’ prediction.

FIG 3.

The lollipop plots in the 3T data set (A) and the 7T data set (B) indicate the nodes within networks that can differentiate superagers from elderly controls. Within the plots, we show the magnitude (dot) and the range (line) of the difference between superagers and elderly controls. ORs of >1 (OR > 1) suggest a larger influence on the predicted probability of being a superager (lollipops in green). ORs of < 1 indicate regions negatively discriminated as characteristic of a superager (lollipops in red). Cingp indicates posterior cingulate cortex; ContA, control A; ContB:, control B; ContC, control C; DorsAttnA, dorsal attention A; DorsAttnB, dorsal attention B; ExStrSup, extrastriate superior cortex; FrMed, frontal medial cortex; Ins, insula; IPL, inferior parietal lobule; IPS, intraparietal sulcus; LH, left hemisphere; OFC, orbital frontal cortex; ParOper, parietal operculum; PCC, precuneus posterior cingulate cortex; pCun, precuneus; PHC, parahippocampal cortex; PFCd, dorsal prefrontal cortex; PFCl, lateral prefrontal cortex; PFClv, lateral ventral prefrontal cortex; PFCm, medial prefrontal cortex; PFCmp, medial posterior prefrontal cortex; PFCv, ventral prefrontal cortex; PostC, postcentral cortex; RH, right hemisphere; Rsp, retrosplenial cortex; SalVentAttnA, salience/ventral attention A; SalVentAttnB, salience/ventral attention B; SPL, superior parietal lobule; Temp, temporal cortex; TempPar, temporoparietal junction; TempPole, medial temporal pole; TempOcc, temporo-occipital junction; VisPeri, peripheral visual.

FIG 4.

The most discriminative nodes among the DMN and SN in superagers compared with elderly controls. The heatmap varies from dark blue to dark red (denoting a higher prediction rate for classification as a superager using ORs). RH indicates right hemisphere; LH, left hemisphere; L, left; R, right.

FIG 5.

The most discriminative nodes among the ECN-L and ECN-R in superagers compared with elderly controls. The heatmap varies from dark blue to dark red (denoting a higher prediction rate for classification as a superager using ORs). RH indicates right hemisphere; LH, left hemisphere; L, left; R, right.

FIG 6.

The most discriminative nodes among the hippocampal and language networks in superagers compared with elderly controls. The heatmap varies from dark blue to dark red (denoting a higher prediction rate for classification as a superager using ORs). RH indicates right hemisphere; LH, left hemisphere; L, left; R, right.