Site Editor
Gregory J. Riely, MD, PhD
Gregory J. Riely, MD, PhD
Posted: Monday, March 24, 2025
Combined radiomic and RNA-sequencing data together may prove to be able to better predict survival patterns for patients with brain metastases from stage IV non–small cell lung cancer (NSCLC). These findings, published in the journal Advanced Science, suggest that these radiogenomic data may help to personalize treatment strategies for this patient population.
“The relationship between radiomic features and immune infiltration suggests that tumors with specific radiomic characteristics may harbor a more dynamic immune microenvironment,” stated Rongrong Zhou, MD, of Xiangya Hospital, Central South University, China, and colleagues. “Understanding this interplay between radiomics biomarkers and immune dynamics bears significant clinical implications.”
The researchers looked at data of patients with NSCLC and brain metastases from Xiangya Hospital as a training set (n = 41) and patients from Yale New Haven Hospital as a test set (n = 86). They incorporated MRI data and RNA sequencing to create a radiogenomic model for overall survival predictions. Both the training and test sets showed that the model had a strong predictive performance as well as generalizability, according to the investigators.
Based on the results of the model, patients were stratified into high- and low-risk groups. In a group of 22 patients who were treated with immunotherapy, a statistically significant association was found between clinical response and risk status (P = .01). The low-risk population had a higher percentage of complete and partial responses compared with the high-risk population (P = .01).
“Future research should aim to include larger, more diverse cohorts and extend the follow-up period to further validate and refine our predictive models,” suggested the study authors.
Disclosure: The study authors reported no conflicts of interest.
JNCCN Spotlights
