Extended Data Fig. 3: Factors influencing the proportion of patients receiving CGP-guided genomic biomarker-linked therapies.
From: Real-world clinical utility of comprehensive genomic profiling in advanced solid tumors
a, Fraction of patients receiving active treatment according to the C-CAT evidence level in patients with pre-CGP treatment regimen and post-CGP treatment strategy information available. b, KaplanMeier survival curves for overall survival in 15,201 patients with survival data and pre-CGP treatment regimen and post-CGP treatment strategy information available, stratified by post-CGP treatment strategies. P values were derived from two-sided Wald tests in multivariable Cox proportional hazard models, without adjustments for multiple comparisons. c–f, Fraction of patients enrolled in clinical trials (red) and receiving off-label drugs (blue) in patients who received active treatment after CGP and whose highest level was A (unapproved) or B–D. Patients were stratified by age at CGP (c), ECOG PS (d), presence of double cancer (e) and clinical trial availability in Japan at the time of CGP (f). The dotted line represents the average fraction of patients harboring alterations with level E or no evidence (11.4%). Two-sided Fisher’s exact test. g, Fraction of patients with clinically actionable alterations (with evidence levels A–D) according to year of CGP testing. Two-sided Cochran–Armitage trend test. a,c–g, Numbers in parentheses indicate the number of evaluated patients.
