ES SCLC has faced limited success with standard chemotherapy (chemo) for over three decades. Immunotherapy introduces new hope, but only a minority benefits. This study investigates pre-treatment CT radiomic features to predict the added benefit of chemoimmunotherapy (chemoIO) over chemo alone in ES SCLC. The study involved 142 pt with ES SCLC across three institutions. Annotations were made for lung lesions, and 860 intra- and peri-tumoral radiomic features were extracted. The training dataset (St) included 50 pt treated with chemo at University Hospital (UH). Validation was conducted on a combined dataset (Sv) with 47 pt treated with chemo from UH and 45 pt treated with chemoIO at Roswell Park Cancer Institute and Emory Healthcare. A least absolute shrinkage and selection operator (LASSO) Cox regression model identified prognostic features for overall survival (OS) in St. The radiomic risk score (RRS) was generated using the selected features and coefficients. To validate predictive performance, pt were stratified into high and low-risk groups based on their median RRS in St. The added benefit of chemoIO was assessed using RRS by comparing OS between pt who received chemo alone and those who underwent chemoIO in Sv. The median overall survival (OS) was 5.57 (95% CI: 5.3-7.4) for chemo and 5.9 (95% CI: 5.6-7.9) for chemoIO-treated pt, with no significant difference (P = 0.92). The RRS stratified chemoIO pt into low-risk (25) and high-risk (20) groups based on the median RRS, showing no significant OS difference between them. However, among chemo-alone pt, a statistically significant OS difference was observed between low and high-risk groups in Sv (HR=1.4, 95% CI: 1.1–1.7, p=0.02). High-risk patients tended to have prolonged survival with chemoIO (HR=1.3, 95% CI: 1.08–1.4, P = .006), while the low-risk groups showed no survival improvement with chemoIO (P > 0.05). Radiomic features show promise as predictors of the added benefit of chemoIO compared to chemo alone in ES SCLC pt. These findings hold potential implications for refining treatment decisions and enhancing outcomes through improved patient stratification.
No takes yet. Share an insight, caveat, or question.
Khorrami et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: