Exploration of Imaging Biomarkers for Predicting Survival of Patients With Advanced Non-Small Cell Lung Cancer Treated With Antiangiogenic Chemotherapy.

Abstract

The objective of this study was to compare imaging biomarkers, including (18)F-FDG PET, CT perfusion (CTP), and CT texture analysis (CTTA), in predicting the survival of patients with advanced non-small cell lung cancer (NSCLC) treated with antiangiogenic chemotherapy. A total of 35 patients (17 men and 18 women; median age, 64.0 years) with advanced NSCLC treated with antiangiogenic chemotherapy were evaluated. CTP and FDG PET were performed before the therapy, and blood flow, blood volume, mean transit time, and the maximum standardized uptake value (SUV max) of the tumor were measured. Texture parameters, including the mean value of pixels with positive values (MPP) and entropy (a measure of irregularity), were also measured on pretherapeutic unenhanced CT images, using CTTA software with a medium texture scale filtration. The best percent change in the tumor burden was also measured. These image-derived tumor parameters were then compared with progression-free survival (PFS) and overall survival (OS). In univariate Cox regression analysis, MPP and entropy were significantly correlated with PFS (p = 0.01 and p = 0.01, respectively), whereas SUV max, MPP, and entropy were significantly correlated with OS (p = 0.03, p = 0.04, and p = 0.0008, respectively). In Kaplan-Meier analysis, high MPP and low entropy were significantly associated with favorable PFS (p < 0.0001 and p = 0.03, respectively) and OS (p = 0.0009 and p = 0.005, respectively), and low SUV max was significantly associated with favorable OS (p = 0.01). CTP parameters and the best change in the tumor burden had no associations with survival. In multivariate analysis, only entropy was identified as an independent prognostic factor for OS (p = 0.02). CTTA is the optimal imaging biomarker for predicting the survival of patients with advanced NSCLC treated with antiangiogenic chemotherapy.