Extended Data Fig 5: Association of TCRA T cell fraction with prognosis.

a, Kaplan-Meier curves for the multi-sample TRACERx100 cohort for LUAD (top) and LUSC (bottom) divided by the number of cold samples in the tumour. Immune-hot and immune-cold samples were defined by using the median of all the tumour samples (0.0736) as a threshold. In each Kaplan-Meier curve the included patients were restricted to those with total samples greater than the number of immune-cold samples used in defining the threshold. b, Kaplan-Meier curves for overall and progression free survival in the TCGA LUAD cohort, dividing the cohort into immune-hot and immune-cold groups using the mean of the TCGA LUAD cohort (0.109) as a threshold. c, Kaplan-Meier curves for the TCGA LUSC, and TCGA LUAD & LUSC cohorts for overall and progression free survival using the mean of the TCGA LUAD cohort (0.109) as a threshold for distinguishing hot and cold tumours. d, Log2(Hazard ratios) from Kaplan-Meier plots for the TCGA separating the tumour samples into immune-hot and immune-cold based on different thresholds from 0 to 0.16 in steps of 0.0025 for overall and progression free survival. e, Hazard ratios of separate Cox regression models relating disease free survival to different multi-sample measures related to the TCRA T cell fraction in the entire TRACERx100 cohort as well as the LUAD and LUSC patients separately. TCRA divergence score is defined as the maximum divided by the upper 95% confidence interval of the minimum. f, Hazard ratios of separate Cox regression models for TCRA T cell fraction for the TCGA LUAD and LUSC cohort for both overall survival (OS) and progression free survival (PFS).