Tunali I, et al. - In 228 non-small cell lung cancer (NSCLC) patients treated with immune-checkpoint blockades, researchers sought clinical and computational image-based predictors of quick disease progression phenotypes by developing quick disease progression phenotypes (including hyperprogressive disease) using time-to-progression (TTP) and/or tumor growth rates. Models including radiomics and clinical data were capable of predicting patients that responded to checkpoint blockades. A subset of patients with hyperprogressive disease displayed extremely poor outcomes. More information was able to be taken from radiomic features extracted from peritumoral regions of lung tumors.