Fig. 1

ROS1 mutation is an independent biomarker for favorable outcomes in pan-cancer immunotherapy. (A) Kaplan–Meier survival analysis stratified by ROS1 mutation status in 1610 cancer patients with 10 tumor types treated with ICIs in the discovery cohort. (B) Association between ROS1 mutation and OS in 1395 patients with 7 tumor types in the validation cohort. (C-D) Comparison of OS (C) and ORR (D) between patients with ROS1 mutant tumors and patients with ROS1 non-mutant tumors in 3005 subjects with 12 tumors treated with ICIs. (E-F) Univariate (E) and multivariate (F) Cox analysis of the association between ROS1 mutation and OS in 3888 patients with 12 tumors treated with ICIs. (G) Nomogram to predict the 12- and 24-month survival. (H) Calibration plots for validation of the 12- and 24-month survival from the nomogram in the discovery cohort. The average predicted probability (X axis) was plotted against the observed Kaplan-Meier estimate in the subgroup (Y axis, 95% CIs of the estimates are presented as vertical lines). Continuous line is the reference line, indicating what an optimal nomogram would be. (I-J) Based on the optimal cutoff value derived from nomogram, low-score was associated with favorable OS in both discovery cohort (I) and validation cohort (J). CI, confidence interval; CR, complete response; HR, hazard ratio; ICI, immune checkpoint inhibitor; ORR, objective response rate; OS, overall survival; PD, progressive disease; PR, partial response; SD, stable disease