C. Pathiravasan1, R. T. Oglesby2, L. Chang3, E. O. Olatunji4, J. Chotiyanonta5, Y. Uchida5, J. Lee2, K. Oishi5, R. Peterson6,7, and S. Acharya2; 1Biostatistics, Johns Hopkins University, Baltimore, MD, 2Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins Medicine, Baltimore, MD, 3Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 4Johns Hopkins Medicine, Baltimore, MD, 5Radiology and Radiological Science, Johns Hopkins Medicine, Baltimore, MD, 6Psychiatry and Behavioral Sciences, Johns Hopkins Medicine, Baltimore, MD, 7Neuropsychology, Kennedy Krieger Institute, Baltimore, MD
Purpose/Objective(s):The relationship between the brain’s neurodevelopment and neurocognitive outcomes is not well characterized in childhood brain tumor survivors. Both mean diffusivity (MD) and fractional anisotropy (FA) are used to characterize microstructural integrity of white matter tracts. We investigated the association between longitudinal neurocognitive outcomes and change in MD and FA in childhood brain tumor survivors. Materials/
Methods: Patients diagnosed with a brain tumor at < 20 years of age with = 2 neurocognitive assessments and paired diffusion tensor imaging at each neurocognitive timepoint were retrospectively identified and eligible for inclusion. Multi-trajectory modeling was applied to cluster patients into distinct neurocognitive trajectories based on intelligent quotient, processing speed and working memory. Linear mixed models were used to determine whether the clusters were associated with change in MD and FA over time. Results: A total of 61 patients were eligible with 151 neurocognitive assessments paired with DTI. Mean follow-up was 4.8 years. Multi-trajectory modeling identified two clusters of patients with distinct neurocognitive trajectories. Compared to the low-performance group, the high-performance group (45.5%) had higher baseline neurocognitive scores with less decline in all neurocognitive outcomes. The high-performance group was associated with a greater decrease in MD and greater increase in FA of the hippocampus and several white matter tracts such as the corpus callosum, anterior limb of the internal capsule and inferior cerebellar peduncle, after adjusting for age and sex (all p<0.05). Additionally, the high-performance group was associated with a greater increase in FA in the superior, middle and inferior frontal gyri, after adjusting for age and sex (all p<0.05). When stratifying by treatment exposure into three subgroups: chemotherapy alone (n=11), radiation with/without chemotherapy (n=35) and no chemotherapy/radiation (n=15), the high-performance group consistently demonstrated greater decline in MD and greater increase in FA across all subgroups compared to the low-performance group, after adjusting for age and sex. Conclusion: The high-performance neurocognitive group was associated with better neurodevelopment over time, as measured by change in FA and MD in multiple brain substructures, compared to the low-performance group, irrespective of age, sex and treatment exposure.
