Back

Differential associations of insulin resistance with anterior hippocampal volume in mild cognitive impairment and Alzheimer's disease

Watanabe, N.; Ogawa, A.; Osada, T.; Adachi, Y.; Shirokoshi, T.; Kodama, H.; Oshima, Y.; Tanaka, S.; Kaga, H.; Tamura, Y.; Watada, H.; Kawamori, R.; Konishi, S.

2026-08-10 neuroscience
10.64898/2026.08.04.742662 bioRxiv
Show abstract

Insulin resistance is increasingly recognized as a metabolic factor associated with Alzheimers disease (AD); however, its relevance to hippocampal structural changes--a key pathological feature of AD--across disease stages is not fully understood. To address this issue, we investigated the relationship between insulin resistance, hippocampal gray matter volume, and cognitive performance using data from the Alzheimers Disease Neuroimaging Initiative (ADNI), a large-scale neuroimaging dataset. Insulin resistance was assessed using the Homeostatic Model Assessment of Insulin Resistance (HOMA-IR), and its relationship with brain structure and cognitive performance was evaluated across diagnostic groups. In the mild cognitive impairment (MCI) group, higher insulin resistance was associated with larger anterior hippocampal gray matter volume, whereas in the AD group this association was reversed in direction. Furthermore, in the MCI group, anterior hippocampal gray matter volume was also positively associated with higher Mini-Mental State Examination (MMSE) scores, and an exploratory mediation analysis suggested a significant indirect association linking HOMA-IR, anterior hippocampal volume, and cognitive performance through anterior hippocampal volume. These findings suggest that the relationship between insulin resistance and AD-related brain changes differs across diagnostic groups, highlighting the importance of considering metabolic alterations in relation to disease status.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.