CT texture analysis for the prediction of KRAS mutation status in colorectal cancer via a machine learning approach.

PURPOSE: This study aimed to investigate whether a machine learning-based computed tomography (CT) texture analysis could predict the mutation status of V-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS) in colorectal cancer. METHOD: This retrospective study comprised 40 patients with pathologically confirmed colorectal cancer who underwent KRAS mutation testing, contrast-enhancement CT, and 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) before treatment. Of the 40 patients, 20 had mutated KRAS genes, whereas 20 had wild-type KRAS genes. Fourteen CT texture parameters were extracted from portal venous phase CT images of primary tumors, and the maximum standard uptake values (SUVmax) on 18F-FDG PET images were recorded. Univariate logistic regression was used to develop predictive models for each CT texture parameter and SUVmax, and a machine learning method (multivariate support vector machine) was used to develop a comprehensive set of CT texture parameters. The area under the receiver operating characteristic (ROC) curve (AUC) of each model was calculated using five-fold cross validation. In addition, the performance of the machine learning method with the CT texture parameters was compared with that of SUVmax. RESULTS: In the univariate analyses, the AUC of each CT texture parameter ranged from 0.4 to 0.7, while the AUC of the SUVmax was 0.58. Comparatively, the multivariate support vector machine with comprehensive CT texture parameters yielded an AUC of 0.82, indicating a superior prediction performance when compared to the SUVmax. CONCLUSIONS: A machine learning-based CT texture analysis was superior to the SUVmax for predicting the KRAS mutation status of a colorectal cancer.

[Survival of glottic carcinoma patients treated with definitive radiation therapy: value of the Sixth edition of the UICC TNM staging system].

PURPOSE: To evaluate the survival of glottic carcinoma patients treated with radiation therapy (RT). The predictive value of the new sixth edition of the UICC staging system was also evaluated. METHODS AND MATERIALS: 193 patients with T1-2N0 glottic squamous cell carcinoma, classified according to the fifth edition of the UICC staging system, were treated with definitive RT. Of them, 130 patients evaluated with pretreatment radiological examinations (CT and/or MR) were retrospectively classified according to the UICC sixth edition. RESULTS: According to the UICC fifth edition, 132 lesions were staged as T1 and 61 as T2. Thirty lesions were categorized as stage T3 according to the UICC sixth edition. Of all patients, 13 died of glottic carcinoma, 25 of second malignancy, and 20 of intercurrent disease. Second primary sites included lung (n=8), esophagus (n=4), liver (n=4), pancreas (n=3), and others (n=6). The 10-year overall survival rate was 66%, and cause-specific survival rates of glottic carcinoma and second malignancy were 91% and 86%, respectively. Multivariate analysis confirmed T-stage in the UICC sixth edition as an independent predictor for death from glottic carcinoma. Although there were no significant factors for second malignancy, there was no death from second malignancy in non-smoking patients. CONCLUSION: Second malignancy was the most frequent cause of death, and an association with smoking was suggested. The UICC sixth edition appears to correctly identify patients with T3 lesions as a high-risk group not only for local failure but also for survival.

[Clinical evaluation of esophageal cancer complicated with head and neck cancer].

We prospectively investigated endoscopic findings of the upper digestive tract in patients with head and neck (H & N) cancer, from 1992 to 2001. Of 687 patients with H & N cancer, esophageal cancers were found to affect 74 patients (10.8%). The 74 patients consisted of 49 (66.2%) with the superficial type and 25 (33.8%) with the advanced type. Other additional cancers were detected in 32 cases (4.7%), including 21 gastric cancers. The incidence was highest in patients with hypopharyngeal cancer (32.4%), whereas the incidences in those with oral floor cancer and mesopharyngeal cancer were 14.3% and 13.1%, respectively. The incidence of stage I cancers was lower than that of stage II, III, or IV cancers. Therapy for superficial esophageal cancers consisted of trisection for endoscopic mucosal resection (EMR), surgery, and no treatment. Surgery, radiation therapy, or no treatment was selected in advanced type. For double cancers of H & N and esophagus, treatment should be selected in consideration of the prognosis of the disease. These findings suggest that endoscopy with the Lugol-spraying method should be performed in H & N cancers for early detection of esophageal cancers.

[Clinical evaluation of gastric diverticulum on indirect radiograph for gastric mass survey].

The aim of this study was to evaluate the clinical conditions of gastric diverticulum. Fifty-four patients with gastric diverticulum (20 men and 34 women among 34,314 patients who underwent medical check-ups) were evaluated on indirect radiographs, for an incidence of 0.16% among the total number of examined cases, a rate lower than that of previous reports. Almost all cases were asymptomatic, had a single diverticulum, and showed a saccular shape. The age distribution indicated higher frequencies in the 5th and 6th decades, and the posterior wall of the fornix was the most common location. Size ranged from 0.6 cm to 12 cm, and 41 cases (75.9%) were between 1.0 cm and 4.0 cm in size. This entity should be kept in mind when reading radiographs of upper gastrointestinal series as well as recognition of pseudodiverticulum and aberrant pancreas as noted for the stomach in several past report. Diverticulum on the cardia, which was previously classified as gastric diverticulum, should be excluded because of the possibility of normal variation.