CT-based deep learning radiomics may help determine NSCLC treatment

Thursday, November 30 | 11:10 a.m.-11:20 a.m. | R4-SSCH10-2 | Room S405 A CT-based deep-learning radiomics biomarker appears to characterize the level of expression of programmed cell death ligand 1 (PD-L1) in non-small cell lung cancers (NSCLCs), according to study results that will be presented Thursday morning.The findings could help clinicians set treatment for lung cancer patients, since tumors that express 50% or greater levels of PD-L1 tend to respond well to immunotherapy, wrote a team led by Jingshan Gong, MD, of Shenzhen Second People's Hospital in China.The study included 259 patients with pathology-confirmed NSCLC; these patients were divided into radiomics training and validation cohorts at a seven-to-one ratio. Gong's group culled radiomics and deep-learning features from preoperative, noncontrast CT exams that had strong association with PD-L1 expression in lung tumors. The researchers then calculated a radiomics score and a deep-learning score and evaluated the predictive value of PD-L1 using the area under the curve (AUC) measure.Gong and colleagues reported that the AUCs of the radiomic scores for PD-L1 expression were 0.79 and 0.7 in the training cohort and validation cohort, respectively, and that the AUCs of the deep learning scores for PD-L1 expression were 0.98 and 0.94 in the training cohort and validation cohort, respectively.The study results offer further evidence for the use of immune checkpoint inhibitors for treatment of NSCLC, according to the...
Source: AuntMinnie.com Headlines - Category: Radiology Authors: Tags: 2023 CT Preview Source Type: news