VanGilder, Jennapher Lingo; Hooyman, Andrew; Hakhu, Sasha; Schilling, Kurt G.; Hu, Leland S.; Zhou, Yuxiang; Caselli, Richard J.; Baxter, Leslie C.; Beeman, Scott C. (2026).听.听Experimental Gerontology, 218, 113112.听
This study explores how subtle changes in the brain may help identify people at risk for聽Alzheimer鈥檚 disease (AD)聽before symptoms appear. The researchers focused on the聽hippocampus, a brain region important for memory, and compared older adults who carry the聽APOE 蔚4 gene variant聽(a known genetic risk factor for AD) with those who do not. Using advanced brain imaging techniques, including聽diffusion MRI聽methods that examine the brain鈥檚聽microstructure聽(the fine, internal organization of brain tissue), they looked at how these features relate to memory performance.
The results showed that overall hippocampal size did not differ in a meaningful way. However, more detailed microstructural measures鈥攅specially a metric called聽orientation dispersion (ODI), which reflects how nerve fibers are organized鈥攚ere linked to better verbal memory performance in people with the APOE 蔚4 variant. In particular, higher ODI in a specific hippocampal subregion (the left subiculum) was associated with better recall of spoken information.
These findings suggest that looking at the brain鈥檚 microstructure, rather than just its size, may provide earlier and more sensitive clues about cognitive changes in people at genetic risk for Alzheimer鈥檚 disease.

Fig. 1.听Shown are the absolute values of log-transformed raw聽p-values for the APOE 蔚4 interaction across 10 hippocampal regions of interest (i.e., left and right CA1, CA2鈥�3, CA4, subiculum, and whole hippocampus), assessed for ODI, NDI, FA, MD, and volumetric metrics in relation to CFT recall and AVLT scores. Higher the magnitudes on the graph correspond to smaller p-values. The dashed line represents the threshold for statistical significance after Bonferroni correction for 10 comparisons (p聽=聽0.005). Notably, only the left subiculum was associated with AVLT, indicating significant interaction effects that persist beyond multiple comparison correction.