<i>In Vivo</i> Detection of EGFRvIII in Glioblastoma via Perfusion Magnetic Resonance Imaging Signature Consistent with Deep Peritumoral Infiltration: The <i>ϕ</i> -Index

Abstract

<b>Purpose:</b> The epidermal growth factor receptor variant III (<i>EGFRvIII</i>) mutation has been considered a driver mutation and therapeutic target in glioblastoma, the most common and aggressive brain cancer. Currently, detecting <i>EGFRvIII</i> requires postoperative tissue analyses, which are <i>ex vivo</i> and unable to capture the tumor's spatial heterogeneity. Considering the increasing evidence of <i>in vivo</i> imaging signatures capturing molecular characteristics of cancer, this study aims to detect <i>EGFRvIII</i> in primary glioblastoma noninvasively, using routine clinically acquired imaging.<b>Experimental Design:</b> We found peritumoral infiltration and vascularization patterns being related to <i>EGFRvIII</i> status. We therefore constructed a quantitative within-patient peritumoral heterogeneity index (PHI/φ-index), by contrasting perfusion patterns of immediate and distant peritumoral edema. Application of φ-index in preoperative perfusion scans of independent discovery (<i>n</i> = 64) and validation (<i>n</i> = 78) cohorts, revealed the generalizability of this <i>EGFRvIII</i> imaging signature.<b>Results:</b> Analysis in both cohorts demonstrated that the obtained signature is highly accurate (89.92%), specific (92.35%), and sensitive (83.77%), with significantly distinctive ability (<i>P</i> = 4.0033 × 10^-10, AUC = 0.8869). Findings indicated a highly infiltrative-migratory phenotype for <i>EGFRvIII⁺</i> tumors, which displayed similar perfusion patterns throughout peritumoral edema. Contrarily, <i>EGFRvIII</i>^- tumors displayed perfusion dynamics consistent with peritumorally confined vascularization, suggesting potential benefit from extensive peritumoral resection/radiation.<b>Conclusions:</b> This <i>EGFRvIII</i> signature is potentially suitable for clinical translation, since obtained from analysis of clinically acquired images. Use of within-patient heterogeneity measures, rather than population-based associations, renders φ-index potentially resistant to inter-scanner variations. Overall, our findings enable noninvasive evaluation of <i>EGFRvIII</i> for patient selection for targeted therapy, stratification into clinical trials, personalized treatment planning, and potentially treatment-response evaluation. <i>Clin Cancer Res; 23(16); 4724-34. ©2017 AACR</i>.