Franca Wagner1,2,3,4, Yahya Ali Hakami5, Geoffrey Warnock5,6, Gabriel Fischer7,8, Martin W Huellner5,6,9, Patrick Veit-Haibach5,6,10. 1. Department of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Zurich, Switzerland. franca.wagner@insel.ch. 2. University of Zurich, Zurich, Switzerland. franca.wagner@insel.ch. 3. Department of Diagnostic and Interventional Neuroradiology, University hospital of Berne, University of Berne, Berne, Switzerland. franca.wagner@insel.ch. 4. Institute for Diagnostic and Interventional Neuroradiology, Inselspital, Bern University Hospital, Freiburgstrasse 4, 3010, Bern, Switzerland. franca.wagner@insel.ch. 5. Department of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Zurich, Switzerland. 6. University of Zurich, Zurich, Switzerland. 7. Institute for Mathematical Statistics and Actuarial Science, University of Bern, Bern, Switzerland. 8. significantis GmbH, Herzwil b. Köniz, Berne, Switzerland. 9. Department of Neuroradiology, University Hospital of Zurich, Zurich, Switzerland. 10. Department of Nuclear Medicine, University Hospital Zurich, University of Zurich, Zurich, Switzerland.
Abstract
PURPOSE: Heterogeneity is a key feature of malignancy associated with adverse tumor biology. Texture analysis with different imaging modalities has been shown to reflect tumor heterogeneity. The current study analyzed and compared pre-treatment (contrast-enhanced) ceCT and positron emission tomography/computed tomography (PET/CT) tumor heterogeneity using the parameters skewness and kurtosis in patients with primary colorectal cancer with and without liver metastases. PROCEDURES: The histogram-derived texture parameters skewness and kurtosis were retrospectively analyzed obtained on PET/CT and ceCT images from 50 patients with histopathology proven primary colorectal cancer (32 patients without liver metastases and 18 with liver metastases). All patients underwent pretreatment ceCT and a 2-deoxy-2-[18F]Fluoro-D-glucose PET/CT of the whole body for initial staging before treatment. The following parameters were analyzed: (1) skewness ceCT and PET/CT combined, (2) skewness ceCT, (3) skewness PET/CT, (4) kurtosis ceCT and PET/CT combined, (5) kurtosis ceCT, and (6) kurtosis PET/CT. For each parameter, the following comparisons were analyzed: (A) colon tumors in 32 patients without liver metastasis versus colon tumors in 18 patients with liver metastasis, (B) colon tumors (all 50 patients, with and without liver metastases) versus 56 liver metastases of patients with metastatic disease to the liver, (C) colon tumors in 32 patients without liver metastasis versus 56 summed metastasis of the 18 patients with metastatic disease to the liver, and (D) colon tumors in 18 patients with liver metastasis versus summed 56 liver metastasis. Furthermore, ceCT parameters were compared to PET/CT parameters (separately for skewness and kurtosis), and the correlation of the clinical stage with skewness and kurtosis was analyzed. RESULTS: Skewness and kurtosis were not statistically significantly different in primary colon cancer lesions between patients with or without liver metastases (p > 0.05). However, there were significant differences found between primary colon cancer and hepatic metastases (p < 0.05). Thus, skewness and kurtosis derived by ceCT and PET/CT showed significantly different heterogeneities between primary colon cancer lesions and hepatic metastases. The clinical stage was not associated with skewness or kurtosis (r = 0.06 and r = 0.05). CONCLUSION: Texture analysis with relatively simple first-order parameters skewness and kurtosis on ceCT and on PET/CT was able to differentiate between primary colon cancer and hepatic metastases. No differentiation was observed between primary colon tumor lesions in patients with or without hepatic metastases. Overall, skewness and kurtosis measured with ceCT and PET/CT seem to have complementary roles in the evaluation of heterogeneity.
PURPOSE: Heterogeneity is a key feature of malignancy associated with adverse tumor biology. Texture analysis with different imaging modalities has been shown to reflect tumor heterogeneity. The current study analyzed and compared pre-treatment (contrast-enhanced) ceCT and positron emission tomography/computed tomography (PET/CT) tumor heterogeneity using the parameters skewness and kurtosis in patients with primary colorectal cancer with and without liver metastases. PROCEDURES: The histogram-derived texture parameters skewness and kurtosis were retrospectively analyzed obtained on PET/CT and ceCT images from 50 patients with histopathology proven primary colorectal cancer (32 patients without liver metastases and 18 with liver metastases). All patients underwent pretreatment ceCT and a 2-deoxy-2-[18F]Fluoro-D-glucose PET/CT of the whole body for initial staging before treatment. The following parameters were analyzed: (1) skewness ceCT and PET/CT combined, (2) skewness ceCT, (3) skewness PET/CT, (4) kurtosis ceCT and PET/CT combined, (5) kurtosis ceCT, and (6) kurtosis PET/CT. For each parameter, the following comparisons were analyzed: (A) colon tumors in 32 patients without liver metastasis versus colon tumors in 18 patients with liver metastasis, (B) colon tumors (all 50 patients, with and without liver metastases) versus 56 liver metastases of patients with metastatic disease to the liver, (C) colon tumors in 32 patients without liver metastasis versus 56 summed metastasis of the 18 patients with metastatic disease to the liver, and (D) colon tumors in 18 patients with liver metastasis versus summed 56 liver metastasis. Furthermore, ceCT parameters were compared to PET/CT parameters (separately for skewness and kurtosis), and the correlation of the clinical stage with skewness and kurtosis was analyzed. RESULTS:Skewness and kurtosis were not statistically significantly different in primary colon cancer lesions between patients with or without liver metastases (p > 0.05). However, there were significant differences found between primary colon cancer and hepatic metastases (p < 0.05). Thus, skewness and kurtosis derived by ceCT and PET/CT showed significantly different heterogeneities between primary colon cancer lesions and hepatic metastases. The clinical stage was not associated with skewness or kurtosis (r = 0.06 and r = 0.05). CONCLUSION: Texture analysis with relatively simple first-order parameters skewness and kurtosis on ceCT and on PET/CT was able to differentiate between primary colon cancer and hepatic metastases. No differentiation was observed between primary colon tumor lesions in patients with or without hepatic metastases. Overall, skewness and kurtosis measured with ceCT and PET/CT seem to have complementary roles in the evaluation of heterogeneity.
Authors: Balaji Ganeshan; Vicky Goh; Henry C Mandeville; Quan Sing Ng; Peter J Hoskin; Kenneth A Miles Journal: Radiology Date: 2012-11-20 Impact factor: 11.105
Authors: M Höckel; C Knoop; K Schlenger; B Vorndran; E Baussmann; M Mitze; P G Knapstein; P Vaupel Journal: Radiother Oncol Date: 1993-01 Impact factor: 6.280
Authors: Marian G W Scheer; Thamar H Stollman; Wouter V Vogel; Otto C Boerman; Wim J G Oyen; Theo J M Ruers Journal: J Nucl Med Date: 2008-05-15 Impact factor: 10.057
Authors: Jacques Ferlay; Isabelle Soerjomataram; Rajesh Dikshit; Sultan Eser; Colin Mathers; Marise Rebelo; Donald Maxwell Parkin; David Forman; Freddie Bray Journal: Int J Cancer Date: 2014-10-09 Impact factor: 7.396