Entwicklungen in der Analyse von Tumorheterogenität in computertomografischen Aufnahmen

Zusammenfassung

Lange beschränkte sich die bildmorphologische Beurteilung von Tumoren auf deren Größe einschließlich der Größendynamik, das Kontrastierungsmuster sowie deren visuell fassbaren Verhalten zum umliegenden Gewebe. In der letzten Dekade rückten zunehmend weitere Aspekte der Bildauswertung von in der klinischen Routine angefertigten Bilddatensätzen in den Fokus, welche dank stetig steigender Prozessorleistung und entsprechend neuer Anwendungsmöglichkeiten erstmals in größerem Umfang analysiert werden konnten. Begriffe wie Tumorheterogenität, „machine learning“ und „big data“ fanden sich immer häufiger in den Überschriften der Publikationen. Es ist gemeinhin anerkannt, dass Tumoren biologisch in der überwiegenden Zahl der Fälle keine homogene Masse darstellen, sondern sowohl auf makroskopischer als auch auf mikroskopischer und genetischer Ebene heterogene Gewebe darstellen. Diese histopathologischen und immunhistochemischen Erkenntnisse mit Tumorarealen unterschiedlicher Zelldichte, Angioneogenese und nekrotischen Anteilen sollten ein entsprechend quantifizierbares Korrelat in den bildgebenden Verfahren aufweisen. Die Heterogenitätsanalyse von Geweben in computertomografischen Datensätzen findet entsprechend in der Onkologie ein breites Anwendungsspektrum, aber auch nicht maligne Erkrankungen stellen einen möglichen Anwendungsbereich für diese Art der Bilddatenauswertung dar. Der nachfolgende Artikel soll eine Übersicht über bereits erfolgte Auswertungen von CT-Datensätzen bei verschiedenen Tumorentitäten und nicht-onkologischen Fragestellungen liefern und die Herausforderungen für die weitere Anwendung von Heterogenitätsanalysen aufzeigen.

Ein wichtiges Ziel stellt hierbei die Identifikation von möglichen bildgebenden Biomarkern zur Therapieresponseevaluation dar, um mit entsprechenden Rückschlüssen die Fortführung oder Umstellung der therapeutischen Maßnahmen zu untermauern.

Abstract

For a long time, the morphological assessment of tumours in imaging data was restricted to the evaluation of their size including the size dynamics, the attenuation pattern as well as their visually comprehensible behavior to the surrounding tissue. In the last decade, different aspects of image analysis of data sets produced in the clinical routine came into focus, which could be analyzed for the first time on a larger scale thanks to constantly increasing processor performance and new applications. Terms such as tumour heterogeneity, „machine learning“ and „big data“ were increasingly found in the headlines of the publications. It is generally accepted that tumours rather do not represent a biologically homogenous mass in the vast majority of cases, than a heterogeneous tissue both on a macroscopic as on a microscopic and genetic level. These histopathological and immunohistochemical findings with tumour areas of different cell density, angioneogenesis and necrotic proportions should have a correspondingly quantifiable correlate in the imaging datasets. The heterogeneity analysis of tissues in computed tomography images accordingly finds a wide range of applications in oncology, but also non-malignant diseases represent a possible area of ​​application for this type of image data analysis. The following article is intended to provide an overview of already performed evaluations of CT datasets regarding tissue heterogeneity in various tumour entities and non-oncological issues and identify the challenges for the further applications of heterogeneity analysis.

An important goal is the identification of possible imaging biomarkers for therapeutic response evaluation in order to underpin the continuation or conversion of the therapeutic strategies with appropriate conclusions.

Schlüsselwörter

Tumorheterogenität - Texturanalyse - Computertomografie - bildgebende Biomarker

Keywords

tumour heterogeneity - texture analysis - computed tomography - imaging biomarker

• Literatur

• 1 Minardi D, Lucarini G, Mazzucchelli R. et al. Prognostic role of Fuhrman grade and vascular endothelial growth factor in pT1a clear cell carcinoma in partial nephrectomy specimens. J Urol 2005; 174: 1208-1212
• 2 Ganeshan B, Miles KA. Quantifying tumour heterogeneity with CT. Cancer Imaging 2013; 13: 140-149
• 3 Feng Z, Shen Q, Li Y. et al. CT texture analysis: a potential tool for predicting the Fuhrman grade of clear-cell renal carcinoma. Cancer Imaging 2019; 19: 6
• 4 Skogen K, Ganeshan B, Good C. et al. Measurements of heterogeneity in gliomas on computed tomography relationship to tumour grade. J Neurooncol 2013; 111: 213-219
• 5 Castellano G, Bonilha L, Li LM. et al. Texture analysis of medical images. Clin Radiol 2004; 59: 1061-1069
• 6 Miles KA, Ganeshan B, Hayball MP. CT texture analysis using the filtration-histogram method: what do the measurements mean?. Cancer Imaging 2013; 13: 400-406
• 7 Lubner MG, Smith AD, Sandrasegaran K. et al. CT Texture Analysis: Definitions, Applications, Biologic Correlates, and Challenges. Radiographics 2017; 37: 1483-1503
• 8 Raman SP, Chen Y, Schroeder JL. et al. CT texture analysis of renal masses: pilot study using random forest classification for prediction of pathology. Acad Radiol 2014; 21: 1587-1596
• 9 Yan L, Liu Z, Wang G. et al. Angiomyolipoma with minimal fat: differentiation from clear cell renal cell carcinoma and papillary renal cell carcinoma by texture analysis on CT images. Acad Radiol 2015; 22: 1115-1121
• 10 Hodgdon T, McInnes MD, Schieda N. et al. Can Quantitative CT Texture Analysis be Used to Differentiate Fat-poor Renal Angiomyolipoma from Renal Cell Carcinoma on Unenhanced CT Images?. Radiology 2015; 276: 787-796
• 11 Takahashi N, Takeuchi M, Sasaguri K. et al. CT negative attenuation pixel distribution and texture analysis for detection of fat in small angiomyolipoma on unenhanced CT. Abdom Radiol (NY) 2016; 41: 1142-1151
• 12 Leng S, Takahashi N, Gomez Cardona D. et al. Subjective and objective heterogeneity scores for differentiating small renal masses using contrast-enhanced CT. Abdom Radiol (NY) 2017; 42: 1485-1492
• 13 Canellas R, Mehrkhani F, Patino M. et al. Characterization of Portal Vein Thrombosis (Neoplastic Versus Bland) on CT Images Using Software-Based Texture Analysis and Thrombus Density (Hounsfield Units). AJR Am J Roentgenol 2016; 207: W81-W7
• 14 Hanania AN, Bantis LE, Feng Z. et al. Quantitative imaging to evaluate malignant potential of IPMNs. Oncotarget 2016; 7: 85776-85784
• 15 Hu Y, Liang Z, Song B. et al. Texture Feature Extraction and Analysis for Polyp Differentiation via Computed Tomography Colonography. IEEE Trans Med Imaging 2016; 35: 1522-1531
• 16 Song B, Zhang G, Lu H. et al. Volumetric texture features from higher-order images for diagnosis of colon lesions via CT colonography. Int J Comput Assist Radiol Surg 2014; 9: 1021-1031
• 17 Pooler BD, Lubner MG, Theis JR. et al. Volumetric Textural Analysis of Colorectal Masses at CT Colonography: Differentiating Benign versus Malignant Pathology and Comparison with Human Reader Performance. Acad Radiol 2019; 26: 30-37
• 18 Feng C, Lu F, Shen Y. et al. Tumor heterogeneity in gastrointestinal stromal tumors of the small bowel: volumetric CT texture analysis as a potential biomarker for risk stratification. Cancer Imaging 2018; 18: 46
• 19 Fried DV, Tucker SL, Zhou S. et al. Prognostic value and reproducibility of pretreatment CT texture features in stage III non-small cell lung cancer. Int J Radiat Oncol Biol Phys 2014; 90: 834-842
• 20 Dou TH, Coroller TP, van Griethuysen JJM. et al. Peritumoral radiomics features predict distant metastasis in locally advanced NSCLC. PLoS One 2018; 13: e0206108
• 21 Ganeshan B, Abaleke S, Young RC. et al. Texture analysis of non-small cell lung cancer on unenhanced computed tomography: initial evidence for a relationship with tumour glucose metabolism and stage. Cancer Imaging 2010; 10: 137-143
• 22 Ganeshan B, Goh V, Mandeville HC. et al. Non-small cell lung cancer: histopathologic correlates for texture parameters at CT. Radiology 2013; 266: 326-336
• 23 Kim HC, Lee JM, Kim SH. et al. Small gastrointestinal stromal tumours with focal areas of low attenuation on CT: pathological correlation. Clin Radiol 2005; 60: 384-388
• 24 Chong DY, Kim HJ, Lo P. et al. Robustness-Driven Feature Selection in Classification of Fibrotic Interstitial Lung Disease Patterns in Computed Tomography Using 3D Texture Features. IEEE Trans Med Imaging 2016; 35: 144-157
• 25 Ginsburg SB, Zhao J, Humphries S. et al. Texture-based Quantification of Centrilobular Emphysema and Centrilobular Nodularity in Longitudinal CT Scans of Current and Former Smokers. Acad Radiol 2016; 23: 1349-1358
• 26 Daginawala N, Li B, Buch K. et al. Using texture analyses of contrast enhanced CT to assess hepatic fibrosis. Eur J Radiol 2016; 85: 511-517
• 27 Simpson AL, Adams LB, Allen PJ. et al. Texture analysis of preoperative CT images for prediction of postoperative hepatic insufficiency: a preliminary study. J Am Coll Surg 2015; 220: 339-346
• 28 Kotze CW, Rudd JH, Ganeshan B. et al. CT signal heterogeneity of abdominal aortic aneurysm as a possible predictive biomarker for expansion. Atherosclerosis 2014; 233: 510-517
• 29 Orlhac F, Frouin F, Nioche C. et al. Validation of A Method to Compensate Multicenter Effects Affecting CT Radiomics. Radiology 2019; 291: 53-59