Abstract Purpose In children with brain tumors, radiation-induced hippocampal injury is associated with increased mean diffusivity (MD) on diffusion tensor imaging (DTI) and short-term memory decline. However, whether hippocampal microstructural injury and memory impairment are present prior to radiation therapy (RT) is unknown. We quantified hippocampal MD and its association with memory performance pre-RT using data from the CogRT study (NCT05658731). Methods Patients ≤26 years of age with newly diagnosed brain tumors enrolled on CogRT underwent paired DTI and neurocognitive testing prior to RT. Fractional anisotropy (FA) and MD were measured across 23 brain substructures, including the hippocampus, after excluding tumor and surgical cavities, and compared with age- and sex-matched healthy controls. The ratio of lateral ventricle volume to brain volume (LV/BV) was calculated as a marker of hydrocephalus. Memory was assessed using the California Verbal Learning Test (CVLT), with impairment defined as 1 SD below population mean. False discovery rate (FDR) correction was applied for multiple comparisons. Results Patients (N = 53; median age 15.1 years, range 2.5–25.6) demonstrated widespread microstructural abnormalities compared with controls (N = 55), with significantly lower FA and higher MD (pFDR0.05). LV/BV was significantly higher in patients vs. controls (mean 2.60 vs. 1.25, p 0.001). Impairment in short-term memory was observed in 35.6% of patients. Short-term memory impairment was associated with elevated MD in the left hippocampus (pFDR=0.019) and right superior frontal gyrus including underlying white matter (pFDR=0.047), as well as reduced FA in the left inferior cerebellar peduncle (pFDR=0.038). Conclusion Prior to RT, children with brain tumors exhibit widespread microstructural compromise, including hippocampal abnormalities, with one-third demonstrating pre-RT short-term memory impairment. These findings suggest that pre-existing structural injury may contribute to early cognitive deficits and should be considered when evaluating longitudinal neurocognitive outcomes and planning radiation treatment.
Chen et al. (Tue,) studied this question.