Lipid Fraction Derived From MRI In- and Opposed-Phase Sequence as a Novel Biomarker for Predicting Survival Outcome of Glioma

Rationale and Purpose: Our study evaluated the capability of magnetic resonance imaging in- and opposed-phase (IOP) derived lipid fraction as a novel prognostic biomarker of survival outcome in glioma. Materials and Methods: We analyzed 46 histologically proven glioma (WHO grades II-IV) patients using standard 3T magnetic resonance imaging brain tumor protocol and IOP sequence. Lipid fraction was derived from the IOP sequence signal-loss ratio. The lipid fraction of solid nonenhancing region of glioma was analyzed, using a three-group analysis approach based on volume under surface of receiveroperating characteristics to stratify the prognostic factors into three groups of low, medium, and high lipid fraction. The survival outcome was evaluated, using Kaplan-Meier survival analysis and Cox regression model. Results: Significant differences were seen between the three groups (low, medium, and high lipid fraction groups) stratified by the optimal cut-off point for overall survival (OS) (p <= 0.01) and time to progression (p <= 0.01) for solid nonenhancing region. The group with high lipid fraction had five times higher risk of poor survival and earlier time to progression compared to the low lipid fraction group. The OS plot stratified by lipid fraction also had a strong correlation with OS plot stratified by WHO grade (R = 0.61, p < 0.01), implying association to underlying histopathological changes. Conclusion: The lipid fraction of solid nonenhancing region showed potential for prognostication of glioma. This method will be a useful adjunct in imaging protocol for treatment stratification and as a prognostic tool in glioma patients.