Quantitative Characterization of Tumoural Leakage Phenomena Using Dynamic Susceptibility Contrast Perfusion Imaging 

Abstract Purpose: Microvascular Leakiness varies between tumours. Extravascular leakage, hitherto the Achilles heel of Dynamic-Susceptibility-Contrast-Perfusion-Weighted-Imaging (DSC-PWI), has lately been quantified by leakage coefficient “K2”. To evaluate K2's diagnostic potential for differentiation of PCNSL, GBM and Mets and assess its relationships with other DSC-PWI metrics. Methods: Retrospective analysis of T2*-weighted DSC-PWI of PCNSL, Mets (n=10 each )and GBM n=16) was performed using a “Leakage Correction” model to generate K2 , uncorrected relative CBV (rCBVucor) and corrected rCBV (rCBVcor). Peak Signal Recovery (PSR) was calculated. Group-level statistical comparisons was made. Results: All PCNSL showed high-magnitude-positive K2 value with mean [SD] 846.1[462] while Mets showed high-magnitude-negative K2 value with mean [SD] -754[546]. In contrast, GBM showed varied values (overall with lower magnitude and positive) of K2 with mean [SD] 391.9[294.1]. The intergroup differences in K2 were statistically different (P 0.000). Lesion K2 (magnitude) exhibited a significant strong positive correlation with the magnitude of CBV correction (ρ 0.63) and with PSR (ρ 0.70). Conclusion: PCNSL had greater high-magnitude-positive-K2 compared with other tumour groups, while Mets were notable for a high-magnitude-negative-K2. K2 parameter obtained from T2*-weighted DSC technique characterizes and quantifies microvascular leakage of contrast and thus provides an alternative means of measuring permeability in tumours. K2 quantification adds adjunctive value to pre-operative discrimination of intra-axial neoplasms. Given the widespread acceptance that DSC-PWI has, K2 holds promise as an attractive alternative to (the methodologically challenging) DCE-PWI-derived permeability metric, K trans .