The tumor immunosuppressive microenvironment can influence treatment response and outcomes. A previously validated immunosuppression scoring system (ISS) assesses multiple immune checkpoints in gastric cancer (GC) using tissue-based assays. We aimed to develop a radiological signature for non-invasive assessment of ISS and treatment outcomes. A total of 642 patients with resectable GC from three centers were divided into four cohorts. Radiomic features were extracted from portal venous-phase CT images of GC. A radiomic signature for predicting ISS (RISS) was constructed using the least absolute shrinkage and selection operator (LASSO) regression method. Moreover, we investigated the value of the RISS in predicting survival and chemotherapy response. The RISS, which consisted of 10 selected features, showed good discrimination of immunosuppressive status in three independent cohorts (area under the curve=0.840, 0.809, and 0.843, respectively). Multivariate analysis revealed that the RISS was an independent prognostic factor for both disease-free survival (DFS) and overall survival (OS) in all cohorts (all p<0.05). Further analysis revealed that stage II and III GC patients with low RISS exhibited a favorable response to adjuvant chemotherapy (OS: hazard ratio [HR] 0.407, 95% confidence interval [CI] 0.284-0.584); DFS: HR 0.395, 95% CI 0.275-0.568). Furthermore, the RISS could predict prognosis and select stage II and III GC patients who could benefit from adjuvant chemotherapy independent of microsatellite instability status and Epstein-Barr virus status. The new, non-invasive radiomic signature could effectively predict the immunosuppressive status and prognosis of GC. Moreover, the RISS could help identify stage II and III GC patients most likely to benefit from adjuvant chemotherapy and avoid overtreatment.
Read full abstract