The goal of this project is to investigate cerebral microbleed (CMB) measures of radiation-induced vascular injury from ultra-high field MRI in patients who previously received uniform, supratentorial cranial radiation therapy (CRT) for childhood brain tumors and to relate these findings to long-term neurocognitive outcome and associated imaging metrics. Cranial radiation therapy plays an integral role in the treatment of children with brain tumors despite its known long-term adverse effects on cognitive function. Although the underlying mechanism for neurocognitive decline is not well understood, there is evidence from other diseases, that it is related to vascular injury in the form of CMBs. Identifyin the underlying vascular injury related to radiation injury and linking it to neurocognitive impairment is therefore a critical consideration in caring for children with brain tumors who are likely to survive long after receiving therapy. Our strategy in this proposal is to employ 7T MRI for heightened sensitivity, and novel acquisition schemes and post-processing algorithms for detecting and quantifying structural parameters, in order to create a framework for evaluating radiation-related brain injury and its relationship to domain-specific cognitive function. The increased signal-to-noise ratio and enhanced susceptibility contrast of 7T MRI will be applied to more accurately detect subtle changes in microvasculature and structural connectivity. We hypothesize that 7T imaging techniques that can simultaneously depict arteries, veins, and CMBs on one image, semi-automatically quantify vascular metrics, and identify structural changes in brain architecture will facilitate our ability to link vascular injury to neurocognitive impairment.
Aim 1 : Investigating associations between CRT-induced CMBs & cognitive impairment A. Global changes: To determine whether the CMBs that are observed in children with medulloblastoma after CRT are related to neurocognitive outcome. B. Local changes: To establish whether CMB location impacts domain-specific cognitive impairment in these children using DTI fiber tractography to localize CMBs to specific white matter tracts.
Aim 2 : Evaluating the evolution of CMB properties in children who received prior CRT A. Longitudinal variations: To assess longitudinal changes in CMB formation and features including time of onset, number, burden, and location in brain, in children with medulloblastomas treated with prior CRT. B. Cross-sectional differences: To compare CMB characteristics in patients who received prior CRT for a medulloblastoma with age-matched controls who were not treated with CRT for a posterior-fossa tumor.
Aim 3 : Linking CMB features to imaging metrics that describe surrounding vasculature & brain tissue A. CMBs & Microvasculature: To investigate the effects of CMB presence on the surrounding vasculature. B. CMBs & Brain Tissue Structure: To explore the relationship between CMBs and other imaging parameters that can assess the structural integrity of potentially affected brain tissue. The proposed study will result in an understanding of the relationship between radiation-induced CMBs and associated structural changes in brain architecture with corresponding neurocognitive decline. Knowledge of the underlying mechanism of long-term cognitive impairment in these children will provide a tool for anticipatory cognitive rehabilitation and ultimately improve outcomes by tailoring radiation delivery to spare identified regions that are most susceptible to radiation injury.
The proposed research will address the current need for linking radiation-induced vascular injury to neurocognitive outcome and associated structural changes in brain architecture in children with brain tumors using enhanced imaging attributes from ultra-high field MRI. The framework developed in this proposal will facilitate the noninvasive characterization of radiation-related brain injury and its relationship to domain-specific cognitive function. Ultimately, successful completion of the proposal will help optimize radiation treatment planning and identify areas for neurocognitive rehabilitation strategies that are specifically tailored to microbleed location, resulting in more personalized patient care, enhanced quality of life, and overall improved outcome.
| Morrison, Melanie A; Payabvash, Seyedmehdi; Chen, Yicheng et al. (2018) A user-guided tool for semi-automated cerebral microbleed detection and volume segmentation: Evaluating vascular injury and data labelling for machine learning. Neuroimage Clin 20:498-505 |
| Chen, Yicheng; Villanueva-Meyer, Javier E; Morrison, Melanie A et al. (2018) Toward Automatic Detection of Radiation-Induced Cerebral Microbleeds Using a 3D Deep Residual Network. J Digit Imaging : |
| Roddy, Erika; Sear, Katherine; Felton, Erin et al. (2016) Presence of cerebral microbleeds is associated with worse executive function in pediatric brain tumor survivors. Neuro Oncol 18:1548-1558 |
