Fig. 4: Association between gene alterations and outcomes in the MDACC primary cohort (n = 735 with genomic data).

a, b Volcano plot from univariate logistic regression depicting odds ratio (x-axis) versus −log 10 (P-value) (y-axis) for 3-month progression in patients treated immune checkpoint inhibitors (ICI) as a monotherapy (ICI-mono) or b with chemotherapy (ICI-chemo); logistic regression models with unadjusted effect were applied to calculate the odds ratio and p values. c Treatment effect analyses of gene subgroups on 3-month progression. Data are presented as the treatment effect estimates with error bars showing the lower and upper bounds of the 95% confidence interval. d, e Volcano plot from univariate cox regression depicting hazard ratio (x-axis) versus −log 10 (P value) (y-axis) for overall progression-free survival (PFS) in patients treated with d ICI-mono or e ICI-chemo; cox proportional hazards regression models with unadjusted effects were applied to calculate the hazard ratio and p values. f Treatment effect analyses of gene subgroups on PFS. Treatment effect estimates and the lower and upper bounds of the 95% confidence interval are shown as dots and whiskers. Source data are provided as a Source Data file.