Raw data of all participants in this study.

Objectives
To determine whether 6 months of combined androgen blockade (CAB) treatment would result in metabolic changes of hippocampus and whether metabolic changes correlate with changes in cognition in patients with advanced prostate cancer (PCa).
Materials and methods
This is an observational study. Proton magnetic resonance spectroscopy (1H-MRS) was used to observe the changes in the ratios of N-acetylaspartate/creatine (NAA/Cr) and choline-containing compounds/creatine (Cho/Cr) on the bilateral hippocampus for the patients before and 6 months after CAB treatment. Cognitive function was also assessed by the Beijing version of the Montreal Cognitive Assessment (MoCA-BJ) at the above two time points. Additionally, a certain number of matched individuals undergoing physical examination were selected as the control group.
Results
CAB group comprised 25 patients with advanced PCa completing follow-up, while control group had 22 healthy controls. Prior to CAB, no significant differences existed in MoCA-BJ scores (including sub-scores) or bilateral hippocampal NAA/Cr and Cho/Cr ratios between groups. Six months after CAB, CAB group exhibited marked declines in MoCA-BJ total score, delayed recall, visuospatial/executive, and attention functions, alongside reduced bilateral hippocampal NAA/Cr and elevated left hippocampal Cho/Cr (P < 0.05).The results of multiple linear regression indicate a positive correlation between NAA/Cr in the left hippocampus and MoCA-BJ total score (β = 4.66, P < 0.001), as well as delayed recall function (β = 2.76, P < 0.001). Mediation analysis confirms that testosterone influences the MoCA-BJ total score and delayed recall function by affecting NAA/Cr in the left hippocampus.
Conclusions
The impact of advanced PCa on cognitive performance could be negligible. However, patients experienced secondary hippocampal injury after CAB, which further led to cognitive dysfunction.