2197 - Radiation Effects in Corpus Callosum and Processing Speed Specificity: Implications for Cognitive Sparing
Presenter(s)
A. Saha1, H. Nguyen2, A. C. Puett3, A. B. Hopper4, R. Karunamuni5, J. S. Kohli6, K. R. Tringale4, C. McDonald6, and J. A. Hattangadi-Gluth4; 1UCSD School of Medicine, La Jolla, CA, 2University of California San Diego School of Medicine, La Jolla, CA, 3UC San Diego, La Jolla, CA, 4Department of Radiation Medicine and Applied Sciences, University of California San Diego, La Jolla, CA, 5Research Service, VA San Diego Healthcare System, San Diego, CA, 6Department of Psychiatry, University of California San Diego, La Jolla, CA
Purpose/Objective(s): Radiotherapy (RT) can injure deep association white matter (WM) tracts causing functional decline. Microstructural WM injury can be measured in vivo with diffusion MRI. We analyzed how radiation dose differentially impacts WM integrity and microstructural volume within the corpus callosum (CC) and identified dosimetric predictors of functional decline.
Materials/Methods: High-resolution volumetric MRI and diffusion tensor imaging were obtained pre-RT and at 3, 6, and 12 months post-RT for patients with primary brain tumors on a prospective clinical trial. Processing speed and executive function were assessed concurrently using well-validated cognitive tests frequently employed in brain tumor clinical trials (WAIS-IV Coding, D-KEFS). Robust automated processing (Dmean). Primary outcomes of WM integrity, microstructural volume, and neurocognitive function were evaluated using linear mixed-effects models adjusted for age, sex, education, tumor type (glioma versus non-glioma), systemic therapy, and surgery type, controlling for time. Analyses were adjusted for multiple comparisons using false discovery rate.
Results: Among the 81 patients longitudinally evaluated, most (60.5%) had gliomas, median age was 46.5 years, and median prescription dose was 59.4 Gy. Higher anterior CC Dmean predicted reduced FA (ß = -0.21, ) and increased MD (ß=0.003, p<0.001) with no association of dose with WM biomarkers in the posterior CC. Dichotomized by median anterior CC Dmean (23.5 Gy), higher doses were associated with earlier FA decline by 6 months (p=0.01) and 12 months (p=0.003). In contrast, Dmean predicted significant volume loss only in the posterior CC (p<0.001). Functionally, lower FA in the anterior CC, reflecting more WM injury, was the sole independent predictor of poorer processing speed performance (WAIS-IV Coding; whereas posterior CC volume was not associated (p=0.26). This relationship was specific to processing speed, as anterior FA did not predict executive function (p>0.1).
Conclusion: RT is associated with regional and biomarker-specific CC injury. Higher doses within the anterior CC predicted greater WM injury, and these WM changes were associated with downstream processing speed performance. These data support prospective evaluation of anterior CC dose sparing and DTI-informed planning strategies to preserve cognitive function.