1 April 2015 in Annals of Diagnostic Pathology
5 November 2020 in Diagnostic and Interventional Radiology
Lymphovascular invasion (LVI) and perineural invasion (PNI) are associated with poor prognosis in gastric cancers. In this work, we aimed to investigate the potential role of computed tomography (CT) texture analysis in predicting LVI and PNI in patients with tubular gastric adenocarcinoma (GAC) using a machine learning (ML) approach. Sixty-eight patients who underwent total gastrectomy with curative (R0) resection and D2-lymphadenectomy were included in this retrospective study. Texture features were extracted from the portal venous phase CT images. Dimension reduction was first done with a reproducibility analysis by two radiologists. Then, a feature selection algorithm was used to further reduce the high-dimensionality of the radiomic data. Training and test splits were created with 100 random samplings. ML-based classifications were done using adaptive boosting, k-nearest neighbors, Naive Bayes, neural network, random forest, stochastic gradient descent, support vector machine, and decision tree. Predictive performance of the ML algorithms was mainly evaluated using the mean area under the curve (AUC) metric. Among 271 texture features, 150 features had excellent reproducibility, which were included in the further feature selection process. Dimension reduction steps yielded five texture features for LVI and five for PNI. Considering all eight ML algorithms, mean AUC and accuracy ranges for predicting LVI were 0.777–0.894 and 76%–81.5%, respectively. For predicting PNI, mean AUC and accuracy ranges were 0.482–0.754 and 54%–68.2%, respectively. The best performances for predicting LVI and PNI were achieved with the random forest and Naive Bayes algorithms, respectively. ML-based CT texture analysis has a potential for predicting LVI and PNI of the tubular GACs. Overall, the method was more successful in predicting LVI than PNI.
28 June 2009 in Surgery Today
17 February 2020 in Molecular Imaging and Radionuclide Therapy
Objective: Prognostic effect of KRAS mutation and side of tumor in colorectal cancer is a highly controversial subject. Therefore, we evaluated the association between FDG uptake pattern in F-18-fluoro-2-deoxy-glucose positron emission tomography/computed tomography (F-18-FDG PET/CT) imaging and KRAS mutation and tumor localization in patients with a diagnosis of colon cancer and assessed the effects of these three factors on prognosis and survival. Methods: Eighty-three patients with colorectal cancer were retrospectively induded in this study. F-18-FDG PET/CT study was performed for pretreatment staging. The maximum standardized uptake value (SUVmax) of the primary tumor and survival data of patients were compared between groups. KRAS mutations were detected with the help of real-time Polymerase Chain Reaction technique through genomic DNA extracted from paraffin-embedded tumor tissue blocks. Tumor lesions with potential KRAS mutations were classified as mutant KRAS and wild type. Results: Twenty five patients were female while 58 were male. The mean age of the patients was 59.8 +/- 11.3 years. Mean follow-up was 35.5 +/- 18.9 months. Primary tumor was localized in the left colon in 83.1% of patients and in the right colon in 16.9%. KRAS mutation was detected in 54.2% (n=45) of patients. Mean SUVmax of patients with primary tumor was estimated to be 21.1+9.1 (range= 6.0-47.5). Mean tumor SUVmax of patients with a KRAS mutation (24.0 +/- 9.0) was found to be significantly higher than those without KRAS mutation (17.7 +/- 8.2) (p=0.001). Mean survival was significantly shorter in patients with locoregional nodal metastasis than in patients without locoregional nodal metastasis as well as in patients with distant nodal metastasis than in patients without distant nodal metastasis and in patients with organ metastasis in initial PET/CT than in patients without organ metastasis. Also, mean survival was nearly statistically-significantly shorter in patients with tumors located in left colon (34.2 +/- 19.4) than in right colon (43.2 +/- 14.6) (p=0.059). However, we found no significant impact of KRAS mutation on survival. Conclusion: In our study, we found that tumor localization had no significant effect on prognosis in patients with colon cancer. On the other hand, FDG uptake was observed to be higher in the presence of KRAS mutation and it was concluded that coexistence of KRAS mutation with higher SUVmax is a negative prognostic factor.
17 February 2020 in Molecular Imaging and Radionuclide Therapy
The aim of this study was to evaluate 18F-fluoro-2-deoxy-glucose (FDG) uptake patterns in primary tumors and metastatic lesions, and also to assess the diagnostic contribution of positron emission tomography/computed tomography (PET/CT) in the initial staging of gastric cancer (GC). The total number of 341 patients with GC were included in this study. All 18F-FDG PET/CT imagings were performed for initial staging. The maximum standardized uptake value (SUVmax) of primary tumor, obtained from 18F-FDG PET/CT imaging was compared between subtypes of GC. Mean SUVmax of 339 patients’ primary tumor was 12.9±8.6. The highest mean SUVmax was detected in patients with medullary subtype GC (17.8±9.9) while the lowest mean SUVmax (9.7±7.6) was seen in signet ring cell carcinoma (SRCC). The primary mean SUVmax was found statistically higher in adenocarcinoma (AC) group than SRCC group (p3 cm) in patients with AC (p=0.03). Primary tumor SUVmax was found statistically higher in regional lymph node (RLN) positive patients than in RLN negative patients in AC and SRCC groups (p<0.001 and p=0.012, respectivelly). Also, in patients with SRCC, SUVmax was significantly higher in the distant metastatic group than in the group without metastasis (p=0.025). Increased primary tumor SUVmax was associated with some of clinical parameters such as age and RLN metastasis in patients with AC. However, there was no relationship between distant metastatic state and primary tumor 18F-FDG uptake in AC. However, high SUVmax of primary tumor in SRCC was associated with regional and distant metastasis, and primary tumor 18F-FDG uptake may be a prognostic value for this subgroup.
5 March 2020 in Japanese Journal of Radiology