OT-01EFFECTIVENESS OF PERFUSION AND DIFFUSION IMAGING FOR GRADING PEDIATRIC BRAIN TUMORS
| Autor: | Melissa Prah, Sean M. Lew, Kathleen M. Schmainda, Sachin Jogal, Teresa G. Kelly, Mohit Maheshwari |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2015 |
| Předmět: |
Cancer Research
Imaging biomarker business.industry Perfusion scanning Text mining Oncology Cerebral blood flow Pediatric brain Medicine Neurology (clinical) business Nuclear medicine Grading (tumors) Perfusion Abstracts from the 3rd Biennial Conference on Pediatric Neuro-Oncology Basic and Translational Research Diffusion MRI |
| Popis: | PURPOSE: To prospectively evaluate the effectiveness of perfusion MRI (pMRI) and diffusion MRI (DWI/ADC) measurements in distinguishing slow-growing low-grade (LG) from fast-growing high-grade (HG) pediatric brain tumors. Brain tumors in children differ significantly from adult. Therefore, testing the utility of pMRI and DWI values as biomarkers for grading pediatric brain tumors is essential. MATERIALS AND METHODS: Following IRB approval, 40 children with new brain tumors who had signed consent and assent were prospectively recruited. MRI exams with severe artifact, lesions too small for accurate diagnosis or those that did not fall under a WHO category were excluded. Thirty-four children with untreated tumors remained (16M/18F, ages 0-23yrs). Perfusion imaging was added to the routine clinical MRI. DWI and ADC images were part of the routine clinical exam. Surgically resected tumors were classified as LG (WHO I/II) or HG (WHO III/IV). Pilocytic astrocytomas (PA) and pilocytic optic-pathway gliomas (OPG) were grouped together because they behave differently than other LG lesions. Non-tumor (NT) lesions were also evaluated. Using serial imaging, stable and non-resectable lesions were classified as LG, whereas rapidly enlarging lesions were classified as HG. Based on these criteria there were HG (n = 4), LG (n = 18), PA + OPG (n = 8), and NT (n = 4). Perfusion normalized (nRCBV) and standardized (sRCBV) relative cerebral blood volume and normalized cerebral blood flow (nRCBF) were calculated. RESULTS: The sRCBV is statistically different for HG vs LG (p = 0.022), but not for nRCBV or nRCBF. ADC can distinguish HG from PA + OPG (p = 0.012) and HG from LG (p = 0.035); however, when sRCBV and ADC are combined, there is greater statistical power for distinguishing these tumor categories (HG vs LG: p = 0.0029; HG vs PA + OPG: p = 0.0095). CONCLUSIONS: The results of this study represent the first step toward developing an imaging biomarker that will allow us to subclassify pediatric brain tumors and tailor therapy to each child. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|