In vivo detection of egfr mutation in glioblastoma via mri signature consistent with deep peritumoral infiltration

Abstract

A method, including a computer-implemented method, is provided for in vivo detection of epidermal growth factor receptor (EGFR) mutation status within peritumoral edematous tissue of a patient. The method includes performing quantitative pattern analysis of magnetic resonance imaging (MRI) data corresponding to MRI of in vivo peritumoral edematous tissue to determine a level of spatial heterogeneity or similarity within the in vivo peritumoral edematous tissue. EGFR mutation status is assigned as one of negative or positive based on the level of spatial heterogeneity or similarity determined. A non-transitory computer-readable storage medium and a system are also provided.

We claim: 
     
         1 . A computer-implemented method for in vivo detection of an epidermal growth factor receptor (EGFR) mutation status within peritumoral edematous tissue, comprising executing on a processor the steps of:
 performing quantitative pattern analysis of magnetic resonance imaging (MRI) data corresponding to MRI of in vivo peritumoral edematous tissue to determine a level of spatial heterogeneity or similarity within the in vivo peritumoral edematous tissue; and   assigning EGFR mutation status as one of negative or positive based on the level of spatial heterogeneity or similarity determined during said performing step.   
     
     
         2 . The method according to  claim 1 , wherein the mutation is selected from EGFR variant III (vIII), a point mutation at EGFR A289V, a point mutation at EGFR variant G598, and/or a point mutation at EGFR variant R108, with reference to the residue numbering of SEQ ID NO:1. 
     
     
         3 . The computer-implemented method according to  claim 1 , wherein, during said performing step, spatial heterogeneity or similarity of perfusion temporal dynamics is determined within the in vivo peritumoral edematous tissue. 
     
     
         4 . The computer-implemented method according to  claim 1 , wherein, during said performing step, imaging data of at least separate first and second regions of interest (ROIs) within the in vivo peritumoral edematous tissue are analyzed and compared to determine the level of spatial heterogeneity or similarity therebetween. 
     
     
         5 . The computer-implemented method according to  claim 3 , wherein said first ROI within the in vivo peritumoral edematous tissue corresponds to a region of tissue adjacent an enhancing part of a tumor, and wherein said second ROI corresponds to a separate region of tissue within the in vivo peritumoral edematous tissue located at a location spaced farthest from the enhancing part of the tumor along a periphery of the in vivo peritumoral edematous tissue. 
     
     
         6 . The computer-implemented method according to  claim 5 , wherein said performing step includes applying a multi-variance statistical procedure to the MRI data to determine perfusion temporal dynamics of the first and second ROIs. 
     
     
         7 . The computer-implemented method according to  claim 6 , wherein the MRI data is dynamic susceptibility contrast material-enhanced magnetic resonance imaging (DSC-MRI). 
     
     
         8 . The computer-implemented method according to  claim 6 , wherein said multi-variance statistical procedure is Principal Component Analysis (PCA). 
     
     
         9 . The computer-implemented method according to  claim 6 , wherein said performing step includes measuring separability between the perfusion temporal dynamics determined for the first and second ROIs, and wherein said assigning step includes an assignment of EGFR-positive status when the separability is low and an assignment of EGFR-negative status when the separability is high. 
     
     
         10 . The computer-implemented method according to  claim 9 , wherein separability is measured via Bhattacharyya coefficient analysis. 
     
     
         11 . The computer-implemented method according to  claim 1 , wherein the MRI data is selected from the group consisting of dynamic susceptibility contrast material-enhanced magnetic resonance imaging (DSC-MRI) data, dynamic contrast enhanced (DCE) MRI perfusion image data, T1-weighted (pre- and post-contrast) data, T2-weighted (pre- and post-contrast) data, and T2-weighted fluid-attenuated inversion recovery (T2-FLAIR) data. 
     
     
         12 . A method of in vivo detection of epidermal growth factor receptor (EGFR) mutation status within peritumoral edematous tissue of a patient, comprising the steps of: acquiring MRI data corresponding to in vivo peritumoral edematous tissue of a patient; identifying separate, non-overlapping first and second ROIs within the peritumoral edematous tissue;
 analyzing the MRI data corresponding to the separate first and second ROIs to determine a level of heterogeneity or similarity therebetween; and   assigning EGFR mutation status as one of negative or positive based on the level of heterogeneity or similarity determined during said analyzing step.   
     
     
         13 . The method according to  claim 12 , wherein the mutation is selected from EGFR variant III (vIII), a EGFR variant at position A289, a point mutation at EGFR G598V, and/or a point mutation of EGFR variant position R108, with reference to the residue numbering of SEQ ID NO:1. 
     
     
         14 . The method according to  claim 12 , wherein said first ROI within the peritumoral edematous tissue corresponds to a region of tissue adjacent an enhancing part of a tumor, and wherein said second ROI corresponds to a separate region of tissue within the peritumoral edematous tissue located at a location spaced farthest from the enhancing part of the tumor along a periphery of the peritumoral edematous tissue. 
     
     
         15 . The method according to  claim 14 , wherein said first ROI is defined on a contrast-enhanced T1-weighted (T1-CE) MRI during said identifying, and wherein said second ROI is defined on a T2-weighted fluid-attenuated inversion recovery (T2-FLAIR) MRI during said identifying step. 
     
     
         16 . The method according to  claim 12 , wherein, during said analyzing step, heterogeneity or similarity of perfusion temporal dynamics between said first and second ROIs is determined via a time-series of MRI data. 
     
     
         17 . The method according to  claim 16 , wherein data from dynamic susceptibility contrast material-enhanced magnetic resonance imaging (DSC-MRI) is used during said analyzing step to determine heterogeneity or similarity of perfusion temporal dynamics between said first and second ROIs. 
     
     
         18 . The method according to  claim 16 , wherein said analyzing step includes applying a multi-variance statistical procedure to the MRI data to determine perfusion temporal dynamics of the first and second ROIs. 
     
     
         19 . The method according to  claim 18 , wherein said multi-variance statistical procedure is Principal Component Analysis (PCA). 
     
     
         20 . The method according to  claim 18 , wherein said analyzing step includes measuring separability between the perfusion temporal dynamics determined for the first and second ROIs, and wherein said assigning step includes an assignment of EGFR-positive mutation status when the separability is low and indicates that the perfusion temporal dynamics between the first and second ROIs are similar and an assignment of EGFR-negative mutation status when the separability is high and indicates that perfusion temporal dynamics between the first and second ROIs are heterogeneous. 
     
     
         21 . The method according to  claim 20 , wherein the separability is measured by Bhattacharyya coefficient analysis. 
     
     
         22 . The method according to  claim 14 , wherein, during said assigning step, EGFR-negative mutation status is assigned when differing levels of neovascularization is determined to exist between said first and second ROIs and EGFR-positive status is assigned when similar levels of neovascularization is determined to exist in said first and second ROIs. 
     
     
         23 . A non-transitory computer-readable storage medium comprising stored instructions which, when executed by one or more computer processors, cause the one or more computer processors to perform steps of:
 performing quantitative pattern analysis of MRI data corresponding to MRI of peritumoral edematous tissue to determine a level of spatial heterogeneity or similarity within the peritumoral edematous tissue; and   assigning EGFR mutation status as one of negative or positive based on the level of spatial heterogeneity or similarity determined during said analyzing step.   
     
     
         24 . A system for in vivo detection of epidermal growth factor receptor (EGFR) mutation status within peritumoral edematous tissue of a patient, comprising:
 at least one processor configured to perform quantitative pattern analysis of MRI data corresponding to MRI of in vivo peritumoral edematous tissue to determine a level of spatial heterogeneity or similarity within the peritumoral edematous tissue; and   
       said at least one processor being configured to assign EGFR mutation status as one of negative or positive based on the level of spatial heterogeneity or similarity determined. 
     
     
         25 . A method for targeted treatment of a patient having a neoplasm associated with an epidermal growth factor receptor (EGFR) mutation, the method comprising:
 (a) detecting an epidermal growth factor receptor variant (EGFR) mutation status within peritumoral edematous tissue of a patient according to the method of  claim 1 ; and   (b) treating a patient with a EGFR-targeting therapy.