Author of 1 Presentation
SS 2.8 - Prediction value of CT texture analysis to differentiate V-Ki-ras-2 Kirsten rat sarcoma viral oncogene homolog mutation status in colorectal cancer
Abstract
Purpose
To investigate the value of CT texture analysis in the prediction of V-Ki-ras-2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutation status in patients with lung metastases from colorectal cancer.
Material and methods
Eighteen patients with pathologically proven lung metastases from colorectal cancer were retrospectively enrolled. All patients underwent contrast-enhanced CT before the resection of metastases and KRAS mutation testing was performed on surgical specimen. Each metastasis was manually segmented from portal venous-phase CT images by an expert radiologist and analyzed with a dedicated software (TexRAD Ltd, Somerset, UK), which extrapolated the following texture parameters: mean, standard deviation of the pixel histogram (SD), skewness, kurtosis, mean value of positive pixel (MPP) and entropy. Mean value of texture parameters was calculated for each spatial spacing factor (SSF 0-6) and compared with KRAS mutation status. P values <0.05 were considered statistically significant.
Results
Nine patients (50%) had mutant KRAS and nine patients (50%) had wild-type KRAS. SD, entropy, MPP and kurtosis resulted significantly different (SD: p = 0.0016; entropy: p = 0.042; MPP, p = 0.0025; kurtosis, p = 0.05) between mutant and wild-type patients at medium and high filter levels (SSF 4-6). Mean and skewness showed no significant differences between the two groups of patients.
Conclusion
Texture parameters are significantly different between mutant and wild-type patients; texture analysis, providing a quantitative assessment of tumor microenvironment, may represent a non-invasive tool in the early prediction of KRAS mutation status in patients with metastatic colorectal cancer, allowing to customize treatment according to the predicted outcome.
Video-on-demand
Presenter of 1 Presentation
SS 2.8 - Prediction value of CT texture analysis to differentiate V-Ki-ras-2 Kirsten rat sarcoma viral oncogene homolog mutation status in colorectal cancer
Abstract
Purpose
To investigate the value of CT texture analysis in the prediction of V-Ki-ras-2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutation status in patients with lung metastases from colorectal cancer.
Material and methods
Eighteen patients with pathologically proven lung metastases from colorectal cancer were retrospectively enrolled. All patients underwent contrast-enhanced CT before the resection of metastases and KRAS mutation testing was performed on surgical specimen. Each metastasis was manually segmented from portal venous-phase CT images by an expert radiologist and analyzed with a dedicated software (TexRAD Ltd, Somerset, UK), which extrapolated the following texture parameters: mean, standard deviation of the pixel histogram (SD), skewness, kurtosis, mean value of positive pixel (MPP) and entropy. Mean value of texture parameters was calculated for each spatial spacing factor (SSF 0-6) and compared with KRAS mutation status. P values <0.05 were considered statistically significant.
Results
Nine patients (50%) had mutant KRAS and nine patients (50%) had wild-type KRAS. SD, entropy, MPP and kurtosis resulted significantly different (SD: p = 0.0016; entropy: p = 0.042; MPP, p = 0.0025; kurtosis, p = 0.05) between mutant and wild-type patients at medium and high filter levels (SSF 4-6). Mean and skewness showed no significant differences between the two groups of patients.
Conclusion
Texture parameters are significantly different between mutant and wild-type patients; texture analysis, providing a quantitative assessment of tumor microenvironment, may represent a non-invasive tool in the early prediction of KRAS mutation status in patients with metastatic colorectal cancer, allowing to customize treatment according to the predicted outcome.
Video-on-demand
Poster Author of 1 e-Poster
SE-080 - CT-Texture Analysis of liver metastases in PNETs versus NPNETs: correlation with histopathological findings
Author of 1 Presentation
SE-080 - CT-Texture Analysis of liver metastases in PNETs versus NPNETs: correlation with histopathological findings
Abstract
Purpose
To compare CT Texture Analysis and CT features (necrosis and delta enhancement) of liver metastases in pancreatic Neuroendocrine Tumors (PNETs) and in non-pancreatic Neuroendocrine Tumors (NPNETs) according to tumor grading and risk of dying.
Material and methods
We assessed baseline CT images of 23 patients with liver metastases in PNETs and 25 patients with liver metastases in NPETs. The lesions were G1 and G2 according to WHO classification. Texture parameters (Mean, Standard Deviation, Entropy, Kurtosis, Skewness, Mean of Positive Pixel and Tx_sigma) at different spatial scale image filtration (SSF) were evaluated in CT arterial and portal phase using a dedicated software for volumetric analysis. CT images were scanned before the beginning of medical treatment.
Results
The following significant results (P< 0,05) were found: value of Skewness in arterial phase between PNETs G2 versus NPNETs G2; in portal phase between PNETs versus NPNETs, PNETs G1 versus NPNETs G1, PNETs G2 versus NPNETs G2; value of Mean in portal phase in PNETs vs NPNETs. Evaluating PNETs, P< 0,05 was found in: Kurtosis and high risk of dying; Skewness and low risk of dying. Evaluating NPNETs, P< 0,05 was found in: Entropy and high risk of dying.
Conclusion
This study shows that CT Texture features are significantly different in PNETs from NPNETs. Moreover, we highlight that some textural features have a significantly correlation with high risk of dying. The evidence of various expressions in term of textural parameters could help the physicians to plan appropriate follow-up programs, therapeutic strategies and to evaluate the response to therapy.
Author of 1 Presentation
SS 2.8 - Prediction value of CT texture analysis to differentiate V-Ki-ras-2 Kirsten rat sarcoma viral oncogene homolog mutation status in colorectal cancer (ID 992)
Abstract
Purpose
To investigate the value of CT texture analysis in the prediction of V-Ki-ras-2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutation status in patients with lung metastases from colorectal cancer.
Material and methods
Eighteen patients with pathologically proven lung metastases from colorectal cancer were retrospectively enrolled. All patients underwent contrast-enhanced CT before the resection of metastases and KRAS mutation testing was performed on surgical specimen. Each metastasis was manually segmented from portal venous-phase CT images by an expert radiologist and analyzed with a dedicated software (TexRAD Ltd, Somerset, UK), which extrapolated the following texture parameters: mean, standard deviation of the pixel histogram (SD), skewness, kurtosis, mean value of positive pixel (MPP) and entropy. Mean value of texture parameters was calculated for each spatial spacing factor (SSF 0-6) and compared with KRAS mutation status. P values <0.05 were considered statistically significant.
Results
Nine patients (50%) had mutant KRAS and nine patients (50%) had wild-type KRAS. SD, entropy, MPP and kurtosis resulted significantly different (SD: p = 0.0016; entropy: p = 0.042; MPP, p = 0.0025; kurtosis, p = 0.05) between mutant and wild-type patients at medium and high filter levels (SSF 4-6). Mean and skewness showed no significant differences between the two groups of patients.
Conclusion
Texture parameters are significantly different between mutant and wild-type patients; texture analysis, providing a quantitative assessment of tumor microenvironment, may represent a non-invasive tool in the early prediction of KRAS mutation status in patients with metastatic colorectal cancer, allowing to customize treatment according to the predicted outcome.
Video-on-demand
Presenter of 1 Presentation
SS 2.8 - Prediction value of CT texture analysis to differentiate V-Ki-ras-2 Kirsten rat sarcoma viral oncogene homolog mutation status in colorectal cancer (ID 992)
Abstract
Purpose
To investigate the value of CT texture analysis in the prediction of V-Ki-ras-2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutation status in patients with lung metastases from colorectal cancer.
Material and methods
Eighteen patients with pathologically proven lung metastases from colorectal cancer were retrospectively enrolled. All patients underwent contrast-enhanced CT before the resection of metastases and KRAS mutation testing was performed on surgical specimen. Each metastasis was manually segmented from portal venous-phase CT images by an expert radiologist and analyzed with a dedicated software (TexRAD Ltd, Somerset, UK), which extrapolated the following texture parameters: mean, standard deviation of the pixel histogram (SD), skewness, kurtosis, mean value of positive pixel (MPP) and entropy. Mean value of texture parameters was calculated for each spatial spacing factor (SSF 0-6) and compared with KRAS mutation status. P values <0.05 were considered statistically significant.
Results
Nine patients (50%) had mutant KRAS and nine patients (50%) had wild-type KRAS. SD, entropy, MPP and kurtosis resulted significantly different (SD: p = 0.0016; entropy: p = 0.042; MPP, p = 0.0025; kurtosis, p = 0.05) between mutant and wild-type patients at medium and high filter levels (SSF 4-6). Mean and skewness showed no significant differences between the two groups of patients.
Conclusion
Texture parameters are significantly different between mutant and wild-type patients; texture analysis, providing a quantitative assessment of tumor microenvironment, may represent a non-invasive tool in the early prediction of KRAS mutation status in patients with metastatic colorectal cancer, allowing to customize treatment according to the predicted outcome.