Abstract #4847

Radiomics of MRI at Diagnosis is Predictive of Extreme Survival in Glioblastoma Multiforme

Olya Stringfield^1,2, Mahmoud Abdalah^1,2, Sandra Johnston^3,4, Nicolas Rognin², Yoganand Balagurunathan², John Arrington⁵, Kristin Swanson³, Kathleen M. Egan⁶, Robert A. Gatenby⁵, and Natarajan Raghunand²

¹Image Response Assessment Team Core, Moffitt Cancer Center, Tampa, FL, United States, ²Department of Cancer Imaging & Metabolism, Moffitt Cancer Center, Tampa, FL, United States, ³Department of Neurologic Surgery, Mayo Clinic, Phoenix, AZ, United States, ⁴Department of Radiology, University of Washington, Seattle, WA, United States, ⁵Department of Diagnostic & Interventional Radiology, Moffitt Cancer Center, Tampa, FL, United States, ⁶Department of Cancer Epidemiology, Moffitt Cancer Center, Tampa, FL, United States

We retrospectively analyzed pre-treatment MR scans in two cohorts diagnosed with Glioblastoma. The Long-Term Survival (LTS) group survived >36 months post-diagnosis, while the Short-Term Survival (STS) group survived ≤18 months. The discovery cohort included 22 LTS patients and 22 STS patients and the validation cohort consisted of 15 patients, each. Tumor voxels were clustered on post-contrast T1w and FLAIR sequences into 6 distinct “habitats”. Radiomic features were extracted from both sequences. The enhancement value on T1w and fraction of Habitat 6 (high signal on T1w and FLAIR) were significantly higher in the LTS groups compared to the STS groups.

This abstract and the presentation materials are available to members only; a login is required.

Join Here