Texture analysis in contrast enhanced CT: New method to predict prognosis of small cell lung cancer treated with platinum-based chemotherapy

Background: Recently the texture analysis has been applied to help predict the clinical outcome and treatment response of various tumor types. This study investigates the relationship between the contrast enhanced computed tomography (CECT) texture features of small cell lung cancer (SCLC) tumors tissue, and the survival of the patients. Methods: Eighty-eight patients diagnosed with unresectable SCLC (Extended stage, 57; Limited stage, 31), and treated with platinum-based chemotherapy at West China hospital between January 2010 and 2015 were enrolled. All the patients are followed for at least 18 months or until death. The texture features of tumor tissues were extracted from CECT images taken prior to treatment. Based on the receiver operating characteristics curve analysis, the optimal cutoff value of each texture parameter was obtained for further calculation. The Kaplan-Meier analysis was performed to compare the eighteen-month overall survival (OS) rates and six-month event-free survival (EFS) rates of different patient subgroups. The multivariate Cox regression analysis was performed to determine the prognostic value of texture features on CECT images in patients with unresectable SCL C. Results: Of all CT texture features analyzed, six (GLCM-Contrast, GLCM-Dissimilarity, HISTO-Energy, HISTO-Entropy, HISTO-Skewness and Std-value) were shown to be significantly related to 18-month OS and could be considered as independent predictors of patients' survival. Two features (GLCM-Energy and GLCM-Entropy) were shown to be significantly correlated with the 6-month EFS but could not be taken as independent predictors according to the multivariate Cox regression analysis. Conclusions: Texture features of contrast-enhanced computed tomography could potentially serve as prognostic predictors for SCLC patients taken platinum-based chemotherapy. In view of the present study and previous reports, the CECT-based teture analysis atlas can be recommended in the clinical practice of SCLC. Legal entity responsible for the study: West China Hospital, Sichuan University. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest.