Search

To date, studies investigating radiomics-based predictive models have tended to err on the side of data-driven or exploratory analysis of many thousands of extracted features. In particular, spatial assessments of texture have proven to be especially adept at assessing for features of intratumoral heterogeneity in oncologic imaging, which likewise may correspond with tumor biology and behavior. These spatial assessments can be generally classified as spatial filters, which detect areas of rapid change within the grayscale in order to enhance edges and/or textures within an image, or neighborhood-based methods, which quantify gray-level differences of neighboring pixels/voxels within a set distance. Given the high dimensionality of radiomics datasets, data dimensionality reduction methods have been proposed in an attempt to optimize model performance in machine learning studies; however, it should be noted that these approaches should only be applied to training data in order to avoid information leakage and model overfitting. While area under the curve of the receiver operating characteristic is perhaps the most commonly reported assessment of model performance, it is prone to overestimation when output classifications are unbalanced. In such cases, confusion matrices may be additionally reported, whereby diagnostic cut points for model predicted probability may hold more clinical significance to clinical colleagues with respect to related forms of diagnostic testing.

ObjectivesTo evaluate the performance of machine learning-augmented MRI-based radiomics models for predicting response to neoadjuvant chemotherapy (NAC) in soft tissue sarcomas.MethodsForty-four subjects were identified retrospectively from patients who received NAC at our institution for pathologically proven soft tissue sarcomas. Only subjects who had both a baseline MRI prior to initiating chemotherapy and a post-treatment scan at least 2 months after initiating chemotherapy and prior to surgical resection were included. 3D ROIs were used to delineate whole-tumor volumes on pre- and post-treatment scans, from which 1708 radiomics features were extracted. Delta-radiomics features were calculated by subtraction of baseline from post-treatment values and used to distinguish treatment response through univariate analyses as well as machine learning-augmented radiomics analyses.ResultsThough only 4.74% of variables overall reached significance at p ≤ 0.05 in univariate analyses, Laws Texture Energy (LTE)-derived metrics represented 46.04% of all such features reaching statistical significance. ROC analyses similarly failed to predict NAC response, with AUCs of 0.40 (95% CI 0.22-0.58) and 0.44 (95% CI 0.26-0.62) for RF and AdaBoost, respectively.ConclusionOverall, while our result was not able to separate NAC responders from non-responders, our analyses did identify a subset of LTE-derived metrics that show promise for further investigations. Future studies will likely benefit from larger sample size constructions so as to avoid the need for data filtering and feature selection techniques, which have the potential to significantly bias the machine learning procedures.

There is no consensus regarding the radiomic feature terminology, the underlying mathematics, or their implementation. This creates a scenario where features extracted using different toolboxes could not be used to build or validate the same model leading to a non-generalization of radiomic results. In this study, the image biomarker standardization initiative (IBSI) established phantom and benchmark values were used to compare the variation of the radiomic features while using 6 publicly available software programs and 1 in-house radiomics pipeline. All IBSI-standardized features (11 classes, 173 in total) were extracted. The relative differences between the extracted feature values from the different software and the IBSI benchmark values were calculated to measure the inter-software agreement. To better understand the variations, features are further grouped into 3 categories according to their properties: 1) morphology, 2) statistic/histogram and 3)texture features. While a good agreement was observed for a majority of radiomics features across the various programs, relatively poor agreement was observed for morphology features. Significant differences were also found in programs that use different gray level discretization approaches. Since these programs do not include all IBSI features, the level of quantitative assessment for each category was analyzed using Venn and the UpSet diagrams and also quantified using two ad hoc metrics. Morphology features earns lowest scores for both metrics, indicating that morphological features are not consistently evaluated among software programs. We conclude that radiomic features calculated using different software programs may not be identical and reliable. Further studies are needed to standardize the workflow of radiomic feature extraction., Comment: 21 pages, 8 figures

Claudins, the integral tight junction (TJ) proteins that regulate paracellular permeability and cell polarity, are frequently dysregulated in cancer; however, their role in neoplastic progression is unclear. Here, we demonstrated that knockout of Cldn18, a claudin family member highly expressed in lung alveolar epithelium, leads to lung enlargement, parenchymal expansion, increased abundance and proliferation of known distal lung progenitors, the alveolar epithelial type II (AT2) cells, activation of Yes-associated protein (YAP), increased organ size, and tumorigenesis in mice. Inhibition of YAP decreased proliferation and colony-forming efficiency (CFE) of Cldn18-/- AT2 cells and prevented increased lung size, while CLDN18 overexpression decreased YAP nuclear localization, cell proliferation, CFE, and YAP transcriptional activity. CLDN18 and YAP interacted and colocalized at cell-cell contacts, while loss of CLDN18 decreased YAP interaction with Hippo kinases p-LATS1/2. Additionally, Cldn18-/- mice had increased propensity to develop lung adenocarcinomas (LuAd) with age, and human LuAd showed stage-dependent reduction of CLDN18.1. These results establish CLDN18 as a regulator of YAP activity that serves to restrict organ size, progenitor cell proliferation, and tumorigenesis, and suggest a mechanism whereby TJ disruption may promote progenitor proliferation to enhance repair following injury.

Claudins, the integral tight junction (TJ) proteins that regulate paracellular permeability and cell polarity, are frequently dysregulated in cancer; however, their role in neoplastic progression is unclear. Here, we demonstrated that knockout of Cldn18, a claudin family member highly expressed in lung alveolar epithelium, leads to lung enlargement, parenchymal expansion, increased abundance and proliferation of known distal lung progenitors, the alveolar epithelial type II (AT2) cells, activation of Yes-associated protein (YAP), increased organ size, and tumorigenesis in mice. Inhibition of YAP decreased proliferation and colony-forming efficiency (CFE) of Cldn18-/- AT2 cells and prevented increased lung size, while CLDN18 overexpression decreased YAP nuclear localization, cell proliferation, CFE, and YAP transcriptional activity. CLDN18 and YAP interacted and colocalized at cell-cell contacts, while loss of CLDN18 decreased YAP interaction with Hippo kinases p-LATS1/2. Additionally, Cldn18-/- mice had increased propensity to develop lung adenocarcinomas (LuAd) with age, and human LuAd showed stage-dependent reduction of CLDN18.1. These results establish CLDN18 as a regulator of YAP activity that serves to restrict organ size, progenitor cell proliferation, and tumorigenesis, and suggest a mechanism whereby TJ disruption may promote progenitor proliferation to enhance repair following injury.

Here we detail the design, fabrication, and use of a microfluidic device to evaluate the deformability of a large number of individual cells in an efficient manner. Typically, data for ~10(2) cells can be acquired within a 1 hr experiment. An automated image analysis program enables efficient post-experiment analysis of image data, enabling processing to be complete within a few hours. Our device geometry is unique in that cells must deform through a series of micron-scale constrictions, thereby enabling the initial deformation and time-dependent relaxation of individual cells to be assayed. The applicability of this method to human promyelocytic leukemia (HL-60) cells is demonstrated. Driving cells to deform through micron-scale constrictions using pressure-driven flow, we observe that human promyelocytic (HL-60) cells momentarily occlude the first constriction for a median time of 9.3 msec before passaging more quickly through the subsequent constrictions with a median transit time of 4.0 msec per constriction. By contrast, all-trans retinoic acid-treated (neutrophil-type) HL-60 cells occlude the first constriction for only 4.3 msec before passaging through the subsequent constrictions with a median transit time of 3.3 msec. This method can provide insight into the viscoelastic nature of cells, and ultimately reveal the molecular origins of this behavior.