The presymptomatic brain changes of granulin (GRN) disease, preceding by years frontotemporal dementia, has not been fully characterized. New approaches focus on the spatial chronnectome can capture both spatial network configurations and their dynamic changes over time. To investigate the spatial dynamics in 141 presymptomatic GRN mutation carriers and 282 noncarriers from the Genetic Frontotemporal dementia research Initiative cohort. We considered time-varying patterns of the default mode network, the language network, and the salience network, each summarized into 4 distinct recurring spatial configurations. Dwell time (DT) (the time each individual spends in each spatial state of each network), fractional occupacy (FO) (the total percentage of time spent by each individual in a state of a specific network) and total transition number (the total number of transitions performed by each individual in a specifict state) were considered. Correlations between DT, FO, and transition number and estimated years from expected symptom onset (EYO) and clinical performances were assessed. Presymptomatic GRN mutation carriers spent significantly more time in those spatial states characterised by greater activation of the insula and the parietal cortices, as compared to noncarriers (p < 0.05, FDR-corrected). A significant correlation between DT and FO of these spatial states and EYO was found, the longer the time spent in the spatial states, the closer the EYO. DT and FO significantly correlated with performances at tests tapping processing speed, with worse scores associated with increased spatial states’ DT. Our results demonstrated that presymptomatic GRN disease presents a complex dynamic reorganization of brain connectivity. Change in both the spatial and temporal aspects of brain network connectivity could provide a unique glimpse into brain function and potentially allowing a more sophisticated evaluation of the earliest disease changes and the understanding of possible mechanisms in GRN disease.

颗粒蛋白 (GRN)疾病的症状前脑变化，在数年前额颞叶痴呆之前，尚未完全表征。专注于空间计时组的新方法可以捕获空间网络配置及其随时间的动态变化。调查 141 个症状前GRN的空间动力学来自遗传性额颞叶痴呆研究计划队列的突变携带者和 282 名非携带者。我们考虑了默认模式网络、语言网络和显着网络的时变模式，每个模式都总结为 4 个不同的重复空间配置。停留时间（DT）（每个人在每个网络的每个空间状态中花费的时间），分数占用（FO）（每个人在特定网络状态中花费的时间的总百分比）和总转换数（总考虑了每个人在特定状态下执行的转换次数）。评估了 DT、FO 和过渡数之间的相关性以及从预期症状发作 (EYO) 和临床表现的估计年数。症状前GRN与非携带者相比，突变携带者在以更大程度激活岛叶和顶叶皮质为特征的空间状态中花费的时间明显更多（p < 0.05，FDR 校正）。发现这些空间状态的DT和FO与EYO之间存在显着相关性，在空间状态中花费的时间越长，EYO越接近。DT 和 FO 与测试攻丝处理速度的表现显着相关，更差的分数与空间状态的 DT 增加相关。我们的结果表明，症状前GRN疾病呈现出大脑连接的复杂动态重组。大脑网络连接的空间和时间方面的变化可以提供对大脑功能的独特了解，并可能允许对最早的疾病变化进行更复杂的评估，并了解GRN疾病的可能机制。