Disorder in Pixel-Level Edge Directions on T1WI Is Associated with the Degree of Radiation Necrosis in Primary and Metastatic Brain Tumors: Preliminary Findings.

BACKGROUND AND PURPOSE: Co-occurrence of local anisotropic gradient orientations (COLLAGE) is a recently developed radiomic (computer extracted) feature that captures entropy (measures the degree of disorder) in pixel-level edge directions and was previously shown to distinguish predominant cerebral radiation necrosis from recurrent tumor on gadolinium-contrast T1WI. In this work, we sought to investigate whether COLLAGE measurements from posttreatment gadolinium-contrast T1WI could distinguish varying extents of cerebral radiation necrosis and recurrent tumor classes in a lesion across primary and metastatic brain tumors. MATERIALS AND METHODS: On a total of 75 gadolinium-contrast T1WI studies obtained from patients with primary and metastatic brain tumors and nasopharyngeal carcinoma, the extent of cerebral radiation necrosis and recurrent tumor in every brain lesion was histopathologically defined by an expert neuropathologist as the following: 1) "pure" cerebral radiation necrosis; 2) "mixed" pathology with coexistence of cerebral radiation necrosis and recurrent tumors; 3) "predominant" (>80%) cerebral radiation necrosis; 4) predominant (>80%) recurrent tumor; and 5) pure tumor. COLLAGE features were extracted from the expert-annotated ROIs on MR imaging. Statistical comparisons of COLLAGE measurements using first-order statistics were performed across pure, mixed, and predominant pathologies of cerebral radiation necrosis and recurrent tumor using the Wilcoxon rank sum test. RESULTS: COLLAGE features exhibited decreased skewness for patients with pure (0.15 ± 0.12) and predominant cerebral radiation necrosis (0.25 ± 0.09) and were statistically significantly different (P < .05) from those in patients with predominant recurrent tumors, which had highly skewed (0.42 ± 0.21) COLLAGE values. COLLAGE values for the mixed pathology studies were found to lie between predominant cerebral radiation necrosis and recurrent tumor categories. CONCLUSIONS: With additional independent multisite validation, COLLAGE measurements might enable noninvasive characterization of the degree of recurrent tumor or cerebral radiation necrosis in gadolinium-contrast T1WI of posttreatment lesions.

Computer-Extracted Texture Features to Distinguish Cerebral Radionecrosis from Recurrent Brain Tumors on Multiparametric MRI: A Feasibility Study.

BACKGROUND AND PURPOSE: Despite availability of advanced imaging, distinguishing radiation necrosis from recurrent brain tumors noninvasively is a big challenge in neuro-oncology. Our aim was to determine the feasibility of radiomic (computer-extracted texture) features in differentiating radiation necrosis from recurrent brain tumors on routine MR imaging (gadolinium T1WI, T2WI, FLAIR). MATERIALS AND METHODS: A retrospective study of brain tumor MR imaging performed 9 months (or later) post-radiochemotherapy was performed from 2 institutions. Fifty-eight patient studies were analyzed, consisting of a training (n = 43) cohort from one institution and an independent test (n = 15) cohort from another, with surgical histologic findings confirmed by an experienced neuropathologist at the respective institutions. Brain lesions on MR imaging were manually annotated by an expert neuroradiologist. A set of radiomic features was extracted for every lesion on each MR imaging sequence: gadolinium T1WI, T2WI, and FLAIR. Feature selection was used to identify the top 5 most discriminating features for every MR imaging sequence on the training cohort. These features were then evaluated on the test cohort by a support vector machine classifier. The classification performance was compared against diagnostic reads by 2 expert neuroradiologists who had access to the same MR imaging sequences (gadolinium T1WI, T2WI, and FLAIR) as the classifier. RESULTS: On the training cohort, the area under the receiver operating characteristic curve was highest for FLAIR with 0.79; 95% CI, 0.77-0.81 for primary (n = 22); and 0.79, 95% CI, 0.75-0.83 for metastatic subgroups (n = 21). Of the 15 studies in the holdout cohort, the support vector machine classifier identified 12 of 15 studies correctly, while neuroradiologist 1 diagnosed 7 of 15 and neuroradiologist 2 diagnosed 8 of 15 studies correctly, respectively. CONCLUSIONS: Our preliminary results suggest that radiomic features may provide complementary diagnostic information on routine MR imaging sequences that may improve the distinction of radiation necrosis from recurrence for both primary and metastatic brain tumors.

Chordoid Meningioma: Differentiating a Rare World Health Organization Grade II Tumor from Other Meningioma Histologic Subtypes Using MRI.

BACKGROUND AND PURPOSE: Meningiomas are very commonly diagnosed intracranial primary neoplasms, of which the chordoid subtype is seldom encountered. Our aim was to retrospectively review preoperative MR imaging of intracranial chordoid meningiomas, a rare WHO grade II variant, in an effort to determine if there exist distinguishing MR imaging characteristics that can aid in differentiating this atypical variety from other meningioma subtypes. MATERIALS AND METHODS: Ten cases of WHO grade II chordoid meningioma were diagnosed at our institution over an 11-year span, 8 of which had preoperative MR imaging available for review and were included in our analysis. Chordoid meningioma MR imaging characteristics, including ADC values and normalized ADC ratios, were compared with those of 80 consecutive cases of WHO grade I meningioma, 21 consecutive cases of nonchordoid WHO grade II meningioma, and 1 case of WHO grade III meningioma. RESULTS: Preoperative MR imaging revealed no significant differences in size, location, signal characteristics, or contrast enhancement between chordoid meningiomas and other meningiomas. There were, however, clear differences in the ADC values and normalized ADC ratios, with a mean absolute ADC value of 1.62 ± 0.33 × 10(-3) mm(2)/s and a mean normalized ADC ratio of 2.22 ± 0.47 × 10(-3) mm(2)/s in chordoid meningiomas compared with mean ADC and normalized ADC values, respectively, of 0.88 ± 0.13 × 10(-3) mm(2)/s and 1.17 ± 0.16 × 10(-3) mm(2)/s in benign WHO grade I meningiomas, 0.84 ± 0.11 × 10(-3) mm(2)/s and 1.11 ± 0.15 × 10(-3) mm(2)/s in nonchordoid WHO grade II meningiomas, and 0.57 × 10(-3) mm(2)/s and 0.75 × 10(-3) mm(2)/s in the 1 WHO grade III meningioma. CONCLUSIONS: Chordoid meningiomas have statistically significant elevations of ADC and normalized ADC values when compared with all other WHO grade I, II, and III subtypes, which enables reliable preoperative prediction of this atypical histopathologic diagnosis.

EGFR wild type antagonizes EGFRvIII-mediated activation of Met in glioblastoma.

Epidermal growth factor receptor (EGFR)vIII is the most common EGFR mutant found in glioblastoma (GBM). EGFRvIII does not bind ligand, is highly oncogenic and is usually coexpressed with EGFR wild type (EGFRwt). EGFRvIII activates Met, and Met contributes to EGFRvIII-mediated oncogenicity and resistance to treatment. Here, we report that addition of EGF results in a rapid loss of EGFRvIII-driven Met phosphorylation in glioma cells. Met is associated with EGFRvIII in a physical complex. Addition of EGF results in a dissociation of the EGFRvIII-Met complex with a concomitant loss of Met phosphorylation. Consistent with the abrogation of Met activation, addition of EGF results in the inhibition of EGFRvIII-mediated resistance to chemotherapy. Thus, our study suggests that ligand in the milieu of EGFRvIII-expressing GBM cells is likely to influence the EGFRvIII-Met interaction and resistance to treatment, and highlights a novel antagonistic interaction between EGFRwt and EGFRvIII in glioma cells.

An EGFR wild type-EGFRvIII-HB-EGF feed-forward loop regulates the activation of EGFRvIII.

EGFRvIII is a key oncogene in glioblastoma (GBM). EGFRvIII results from an in-frame deletion in the extracellular domain of EGFR, does not bind ligand and is thought to be constitutively active. Although EGFRvIII dimerization is known to activate EGFRvIII, the factors that drive EGFRvIII dimerization and activation are not well understood. Here we present a new model of EGFRvIII activation and propose that oncogenic activation of EGFRvIII in glioma cells is driven by co-expressed activated EGFR wild type (EGFRwt). Increasing EGFRwt leads to a striking increase in EGFRvIII tyrosine phosphorylation and activation while silencing EGFRwt inhibits EGFRvIII activation. Both the dimerization arm and the kinase activity of EGFRwt are required for EGFRvIII activation. EGFRwt activates EGFRvIII by facilitating EGFRvIII dimerization. We have previously identified HB-EGF, a ligand for EGFRwt, as a gene induced specifically by EGFRvIII. In this study, we show that HB-EGF is induced by EGFRvIII only when EGFRwt is present. Remarkably, altering HB-EGF recapitulates the effect of EGFRwt on EGFRvIII activation. Thus, increasing HB-EGF leads to a striking increase in EGFRvIII tyrosine phosphorylation while silencing HB-EGF attenuates EGFRvIII phosphorylation, suggesting that an EGFRvIII-HB-EGF-EGFRwt feed-forward loop regulates EGFRvIII activation. Silencing EGFRwt or HB-EGF leads to a striking inhibition of EGFRvIII-induced tumorigenicity, while increasing EGFRwt or HB-EGF levels resulted in accelerated EGFRvIII-mediated oncogenicity in an orthotopic mouse model. Furthermore, we demonstrate the existence of this loop in human GBM. Thus, our data demonstrate that oncogenic activation of EGFRvIII in GBM is likely maintained by a continuous EGFRwt-EGFRvIII-HB-EGF loop, potentially an attractive target for therapeutic intervention.