AUTHOR=Yang Bin , Zhong Jian , Zhong Jing , Ma Lu , Li Ang , Ji Hengshan , Zhou Changsheng , Duan Shaofeng , Wang Qinggen , Zhu Chaohui , Tian Jiahe , Zhang Longjiang , Wang Feng , Zhu Hong , Lu Guangming 

TITLE=Development and Validation of a Radiomics Nomogram Based on 18F-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography and Clinicopathological Factors to Predict the Survival Outcomes of Patients With Non-Small Cell Lung Cancer

JOURNAL=Frontiers in Oncology

VOLUME=Volume 10 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2020.01042

DOI=10.3389/fonc.2020.01042

ISSN=2234-943X

ABSTRACT=Purpose: 
This study aimed toIn this study, we developed and validated a radiomicsradiomics nomogram by combining the radiomic features extracted from 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) images and clinicopathological factors to evaluate the overall survival (OS) of patients with non-small cell lung cancer (NSCLC).
Patients and methods: 
A total of 315 consecutive patients with NSCLC (221 in the training cohort and 94 in the validation cohort) were enrolled in this study. A total of 840 radiomic features were extracted from the CT and PET images, respectively. Three radiomic scores (rad-scores) were calculated using the least absolute shrinkage and selection operator (LASSO) Cox regression based on subsets of CT, PET, and PET/CT radiomic features subsets. A multivariate Cox regression analysis was performed for each rad-score combined with the clinicopathological factors to determine the independent risk factors. The OS nomogram was constructed based on the PET/CT rad-score and the independent clinicopathological factors. Validation and calibration were conducted to evaluate the performance of the model in the training and validation cohorts, respectively.
Results
Results: A total of 144 (45.71%) women and 171 (54.29%) men with NSCLC were enrolled in this study. The PET/CT rad-score combined with the clinical model had the best C-index (0.789). Ageage (hazard ratio [HR], 1.03), distant metastasis (HR, 2.24) , stage (HR, 2.99) , CEA (HR, 1.10), and targeted therapy (HR, 0.36) were the independent risk factors for patients with NSCLC. The validation curve showed that the OS nomogram had a strong predictive power in patients’ survival. The calibration curve showed that the predicted survival time was significantly close to the observed one.   
Conclusion: 
A radiomicradiomics nomogram based on 18F-FDG PET/CT rad-score and clinicopathological factors had a good predictive performance for the survival outcome, offering feasible and practical guidance for the individualized management of patients with NSCLC.