Abstract"Radiomics" refers to the extraction and analysis of large amounts of advanced quantitative imaging features with high throughput from medical images obtained with computed tomography (CT), positron emission tomography (PET) or magnetic resonance imaging (MRI). Importantly, these data are designed to be extracted from standard-of-care images, leading to a very large potential subject pool. Radiomic data are in a mineable form that can be used to build descriptive and predictive models relating image features to phenotypes or gene-protein signatures. The core hypothesis of radiomics is that these models, which can include biological or medical data, can provide valuable diagnostic, prognostic or predictive information. The radiomics enterprise can be divided into distinct processes, each with its own challenges that need to be overcome: (i) image acquisition and reconstruction (ii) image segmentation and rendering (iii) feature extraction and feature qualification (iv) databases and data sharing for eventual (v) ad hoc informatic analyses.Each of these individual processes poses unique challenges. For example, optimum protocols for image acquisition and reconstruction have to be identified and harmonized. Also, segmentations have to be robust and involve minimal operator input. Features have to be generated that robustly reflect the complexity of the individual volumes, but cannot be overly complex or redundant. Furthermore, informatics databases that allow incorporation of image features and image annotations, along with medical and genetic data have to be generated. Finally, the statistical approaches to analyze these data have to be optimized, as radiomics is not a mature field of study. Each of these processes will be discussed in turn, as well as some of their unique challenges and

The pH of solid tumors is acidic due to increased fermentative metabolism and poor perfusion. It has been hypothesized that acid pH promotes local invasive growth and metastasis. The hypothesis that acid mediates invasion proposes that H+ diffuses from the proximal tumor microenvironment into adjacent normal tissues where it causes tissue remodeling that permits local invasion. In the current work, tumor invasion and peritumoral pH were monitored over time using intravital microscopy. In every case, the peritumoral pH was acidic and heterogeneous and the regions of highest tumor invasion corresponded to areas of lowest pH. Tumor invasion did not occur into regions with normal or near-normal pHe. Immunohistochemical analyses revealed that cells in the invasive edges expressed the glucose transporter GLUT-1 and the sodium-hydrogen exchanger NHE-1, both of which were associated with peritumoral acidosis. In support of the functional importance of our findings, oral administration of sodium bicarbonate was sufficient to increase peritumoral pH and inhibit tumor growth and local invasion in a preclinical model, supporting the acid-mediated invasion hypothesis.

Purpose Many radiomics features were originally developed for non-medical imaging applications and therefore original assumptions may need to be reexamined. In this study, we investigated the impact of slice thickness and pixel spacing (or pixel size) on radiomics features extracted from Computed Tomography (CT) phantom images acquired with different scanners as well as different acquisition and reconstruction parameters. The dependence of CT texture features on gray level discretization was also evaluated. Methods and Materials A texture phantom composed of 10 different cartridges of different materials was scanned on eight different CT scanners from three different manufacturers. The images were reconstructed for various slice thicknesses. For each slice thickness, the reconstruction Field Of View (FOV) was varied to render pixel sizes ranging from 0.39 to 0.98 mm. A fixed spherical region of interest (ROI) was contoured on the images of the shredded rubber cartridge and the 3D printed, 20% fill, acrylonitrile butadiene styrene plastic cartridge (ABS20) for all phantom imaging sets. Radiomics features were extracted from the ROIs using an in-house program. Features categories were: shape (10), intensity (16), GLCM (24), GLZSM (11), GLRLM (11), and NGTDM (5), fractal dimensions (8) and first order wavelets (128), for a total of 213 features. Voxel size resampling was performed to investigate the usefulness of extracting features using a suitably chosen voxel size. Acquired phantom image sets were resampled to a voxel size of 1 × 1 × 2 mm3 using linear interpolation. Image features were therefore extracted from resampled and original data sets and the absolute value of the percent coefficient of variation (%COV) for each feature was calculated. Based on %COV values, features were classified in 3 groups: 1) features with large variations before and after resampling (%COV > 50); 2) features with diminished variation (%COV < 30) after resampling; and 3) features that had originally moderate variation (%COV < 50%) and were negligibly affected by resampling. Group 2 features were further studied by modifying feature definitions to include voxel size. Original and voxel-size normalized features were used for interscanner comparisons. A subsequent analysis investigated feature dependency on gray level discretization by extracting 51 texture features from ROIs from each of the 10 different phantom cartridges using 16, 32, 64, 128 and 256 gray levels. Results Out of the 213 features extracted, 150 were reproducible across voxel sizes, 42 improved significantly (%COV < 30, Group 2) after resampling, and 21 had large variations before and after resampling (Group 1). Ten features improved significantly after definition modification effectively removed their voxel size dependency. Interscanner comparison indicated that feature variability among scanners nearly vanished for 8 of these 10 features. Furthermore, 17 out of 51 texture features were found to be dependent on the number of gray levels. These features were redef...

We study the reproducibility of quantitative imaging features that are used to describe tumor shape, size, and texture from computed tomography (CT) scans of non-small cell lung cancer (NSCLC). CT images are dependent on various scanning factors. We focus on characterizing image features that are reproducible in the presence of variations due to patient factors and segmentation methods. Thirty-two NSCLC nonenhanced lung CT scans were obtained from the Reference Image Database to Evaluate Response data set. The tumors were segmented using both manual (radiologist expert) and ensemble (software-automated) methods. A set of features (219 three-dimensional and 110 two-dimensional) was computed, and quantitative image features were statistically filtered to identify a subset of reproducible and nonredundant features. The variability in the repeated experiment was measured by the test-retest concordance correlation coefficient (CCCTreT). The natural range in the features, normalized to variance, was measured by the dynamic range (DR). In this study, there were 29 features across segmentation methods found with CCCTreT and DR ≥ 0.9 and R(2) Bet ≥ 0.95. These reproducible features were tested for predicting radiologist prognostic score; some texture features (run-length and Laws kernels) had an area under the curve of 0.9. The representative features were tested for their prognostic capabilities using an independent NSCLC data set (59 lung adenocarcinomas), where one of the texture features, run-length gray-level nonuniformity, was statistically significant in separating the samples into survival groups (P ≤ .046).

Background This study retrospectively evaluated the capability of computed-tomography (CT) based radiomic features to predict EGFR mutation status in surgically-resected peripheral lung adenocarcinomas in an Asian cohort of patients. Materials and Methods 298 patients with surgically resected peripheral lung adenocarcinomas were investigated in this institutional review board-approved retrospective study with waived consent. 219 quantitative 3D features were extracted from segmented volumes of each tumor, and 59 of these which were considered as independent features were included in the analysis. Clinical and pathological information were obtained from the institutional database. Results Mutant EGFR was significantly associated with female gender (p=0.0005); never smoker status (p<0.0001), lepidic predominant adenocarcinomas (p=0.017), and low or intermediate pathologic grade (p=0.0002). Statistically significant differences were found in 11 radiomic features between EGFR mutant and wild type groups on univariate analysis. Mutant EGFR status could be predicted by a set of five radiomic features that fall in three broad groups: CT attenuation energy, tumor main direction and texture defined by wavelets and Laws (AUC 0.647). Multiple logistic regression model showed that adding radiomic features to a clinical model resulted in a significant improvement of predicting power, as the AUC increased from 0.667 to 0.709 (p<0.0001). Conclusions CT based radiomic features of peripheral lung adenocarcinomas can capture useful information regarding tumor phenotype, and the model we built can be useful to predict the presence of EGFR mutations in peripheral lung adenocarcinoma in Asian patients when mutational profiling is not available or possible.

This contribution is part of the special series of Inaugural Articles by members of the National Academy of Sciences elected in 2009.Contributed by V. Craig Jordan, September 14, 2011 (sent for review June 21, 2011) In laboratory studies, acquired resistance to long-term antihormonal therapy in breast cancer evolves through two phases over 5 y. Phase I develops within 1 y, and tumor growth occurs with either 17β-estradiol (E 2 ) or tamoxifen. Phase II resistance develops after 5 y of therapy, and tamoxifen still stimulates growth; however, E 2 paradoxically induces apoptosis. This finding is the basis for the clinical use of estrogen to treat advanced antihormone-resistant breast cancer. We interrogated E 2 -induced apoptosis by analysis of gene expression across time (2-96 h) in MCF-7 cell variants that were estrogen-dependent (WS8) or resistant to estrogen deprivation and refractory (2A) or sensitive (5C) to E 2 -induced apoptosis. We developed a method termed differential area under the curve analysis that identified genes uniquely regulated by E 2 in 5C cells compared with both WS8 and 2A cells and hence, were associated with E 2 -induced apoptosis. Estrogen signaling, endoplasmic reticulum stress (ERS), and inflammatory response genes were overrepresented among the 5C-specific genes. The identified ERS genes indicated that E 2 inhibited protein folding, translation, and fatty acid synthesis. Meanwhile, the ERS-associated apoptotic genes Bcl-2 interacting mediator of cell death (BIM; BCL2L11) and caspase-4 (CASP4), among others, were induced. Evaluation of a caspase peptide inhibitor panel showed that the CASP4 inhibitor z-LEVD-fmk was the most active at blocking E 2 -induced apoptosis. Furthermore, z-LEVD-fmk completely prevented poly (ADP-ribose) polymerase (PARP) cleavage, E 2 -inhibited growth, and apoptotic morphology. The up-regulated proinflammatory genes included IL, IFN, and arachidonic acid-related genes. Functional testing showed that arachidonic acid and E 2 interacted to superadditively induce apoptosis. Therefore, these data indicate that E 2 induced apoptosis through ERS and inflammatory responses in advanced antihormone-resistant breast cancer.aromatase inhibitor | antihormonal resistance | estrogen receptor | gene expression microarrays | selective estrogen receptor modulator E lucidation of the basic structure function relationships of synthetic estrogens based on either stilbene (1) or triphenylethylene (2) was a landmark achievement that continues to have major therapeutic implications to this day. The first successful chemical therapy for the treatment of any cancer was the use of high-dose synthetic estrogen for the treatment of metastatic breast cancer (3). Response rates for patients who were more than a decade beyond menopause were about 30%. Importantly, treatment near menopause was ineffective, and therefore, tumor responsiveness was related to the duration of estrogen deprivation. In 1970, Alexander Haddow commented that "the extraordinary extent of tumor regression observed in...

).q RSNA, 2016 Purpose:To retrospectively identify the relationship between epidermal growth factor receptor (EGFR) mutation status, predominant histologic subtype, and computed tomographic (CT) characteristics in surgically resected lung adenocarcinomas in a cohort of Asian patients. materials andMethods:This study was approved by the institutional review board, with waiver of informed consent. Preoperative chest CT findings were retrospectively evaluated in 385 surgically resected lung adenocarcinomas. A total of 30 CT descriptors were assessed. EGFR mutations at exons 18-21 were determined by using the amplification refractory mutation system. Multiple logistic regression analyses were performed to identify independent factors of harboring EGFR mutation status. The final model was selected by using the backward elimination method, and two areas under the receiver operating characteristic curve (ROC) were compared with the nonparametric approach of DeLong, DeLong, and Clarke-Pearson. Results:EGFR mutations were found in 168 (43.6%) of 385 patients. Mutations were found more frequently in (a) female patients (P , .001); (b)those who had never smoked (P , .001); (c)those with lepidic predominant adenocarcinomas (P = .001) or intermediate pathologic grade (P , .001); (e) smaller tumors (P , .001); (f) tumors with spiculation (P = .019), ground-glass opacity (GGO) or mixed GGO (P , .001), air bronchogram (P = .006), bubblelike lucency (P , .001), vascular convergence (P = .024), thickened adjacent bronchovascular bundles (P = .027), or pleural retraction (P , .001); and (g) tumors without pleural attachment (P = .004), a well-defined margin (P = .010), marked heterogeneous enhancement (P = .001), severe peripheral emphysema (P = .002), severe peripheral fibrosis (P = .013), or lymphadenopathy (P = .028). The most important and significantly independent prognostic factors of harboring EGFR-activating mutation for the model with both clinical variables and CT features were those who had never smoked and those with smaller tumors, bubblelike lucency, homogeneous enhancement, or pleural retraction when adjusting for histologic subtype, pathologic grade, or thickened adjacent bronchovascular bundles. ROC curve analysis showed that use of clinical variables combined with CT features (area under the ROC curve = 0.778) was superior to use of clinical variables alone (area under the ROC curve = 0.690). Conclusion:CT imaging features of lung adenocarcinomas in combination with clinical variables can be used to prognosticate EGFR mutation status better than use of clinical variables alone.q RSNA, 2016

Quantitative size, shape, and texture features derived from computed tomographic (CT) images may be useful as predictive, prognostic, or response biomarkers in non-small cell lung cancer (NSCLC). However, to be useful, such features must be reproducible, non-redundant, and have a large dynamic range. We developed a set of quantitative three-dimensional (3D) features to describe segmented tumors and evaluated their reproducibility to select features with high potential to have prognostic utility. Thirty-two patients with NSCLC were subjected to unenhanced thoracic CT scans acquired within 15 min of each other under an approved protocol. Primary lung cancer lesions were segmented using semi-automatic 3D region growing algorithms. Following segmentation, 219 quantitative 3D features were extracted from each lesion, corresponding to size, shape, and texture, including features in transformed spaces (laws, wavelets). The most informative features were selected using the concordance correlation coefficient across test-retest, the biological range and a feature independence measure. There were 66 (30.14%) features with concordance correlation coefficient ≥ 0.90 across test-retest and acceptable dynamic range. Of these, 42 features were non-redundant after grouping features with R (2) Bet ≥ 0.95. These reproducible features were found to be predictive of radiological prognosis. The area under the curve (AUC) was 91% for a size-based feature and 92% for the texture features (runlength, laws). We tested the ability of image features to predict a radiological prognostic score on an independent NSCLC (39 adenocarcinoma) samples, the AUC for texture features (runlength emphasis, energy) was 0.84 while the conventional size-based features (volume, longest diameter) was 0.80. Test-retest and correlation analyses have identified non-redundant CT image features with both high intra-patient reproducibility and inter-patient biological range. Thus making the case that quantitative image features are informative and prognostic biomarkers for NSCLC.