P15.04.a biological correlates of leakage effect metrics computed from dynamic sage-epi mri in enhancing and non-enhancing human gliomas

Abstract BACKGROUND Dynamic spin-and-gradient-echo echoplanar imaging (SAGE-EPI) can be used to compute novel quantitative maps derived from contrast leakage effects, which are thought to reflect tissue characteristics, while acquiring perfusion imaging. In this study, we explored leakage effect metrics and their biological correlates in human gliomas. METHODS 57 patients with histology-proven gliomas were studied (38 enhancing, 19 non-enhancing). The following quantitative leakage effect maps were computed: ΔR2,ss* (reflecting T2*-leakage effects, dependent on cytoarchitecture), ΔR1,ss (reflecting T1-leakage effects, dependent on blood-brain barrier permeability), and the transverse relaxivity at tracer equilibrium (TRATE, reflecting the balance between ΔR2,ss* and ΔR1,ss). In enhancing gliomas, these metrics were compared between patient subgroups (treatment-naïve [TN] vs recurrent [R]) and biological features (IDH status, Ki67 expression). In non-enhancing gliomas, qualitative and quantitative analyses were performed to evaluate the potential usefulness of leakage effects in the absence of gross contrast enhancement. RESULTS In enhancing IDH-wild-type gliomas (IDHwt), previous exposure to treatment determined higher ΔR1,ss (p=0.035) and lower TRATE (p=0.002). In IDH-mutant-gliomas (IDHm), previous treatment determined higher ΔR1,ss (p=0.026). In TN-gliomas, IDHwt status tended to determine higher ΔR1,ss and lower TRATE (p=0.13). ΔR1,ss values were correlated with Ktrans values (TN/R p=0.036/0.0024, r=0.68/0.59). TRATE values above 142 mM-1s-1 were exclusively seen in TN-IDHwt, and in TN-gliomas this cutoff had 89% sensitivity and 80% specificity as a predictor of Ki67>10%. Preliminary analyses on non-enhancing gliomas revealed that a subset of gliomas (n=11) showing a visual tumor-to- BACKGROUND contrast in leakage effect maps were higher in grade (p=0.049), and all grade 4 gliomas (n=2) displayed pronounced leakage effects. CONCLUSION s: In enhancing gliomas novel leakage effect metrics computed with dynamic SAGE-EPI may reflect cytoarchitecture and blood-brain barrier disruption. More in detail, TRATE was higher in IDHwt-gliomas and high-Ki67 gliomas, consistently with the hypothesis of TRATE values being proportional to cell volume fraction and cell size. ΔR1,ss was correlated to Ktrans and higher in IDHwt-gliomas and treated gliomas, consistently with the hypothesis of ΔR1,ss being a potential marker of blood-brain barrier disruption. Preliminary results on non-enhancing gliomas suggest that leakage effect metrics are more likely to be visualized in higher grades, and may be insightful in this subset despite the absence of gross contrast-enhancement. Overall, leakage effect metrics may provide insights regarding tissue characteristics with potential applications in differential diagnosis and treatment response assessment.