Fig. 7. Multiple mechanisms of TKI resistance in EGFR-mutant NSCLC cells.

(A and B) Viability as determined by CTG of osimertinib-resistant HCC827AZR (A) or PC9AZR (B) cells plated in RPMI10 containing 10 μg/mL rhIGFBP5, rhIGFBP6, or rhIGFBP7 and treated 24 hours later with 100 nM osimertinib for 72 hours. n=3 experiments. (C) Viability as determined by CTG of erlotinib-resistant HCC827ER4 cells plated in RPMI10 containing 10 μg/mL rhIGFBP5 or rhIGFBP6 and treated 24 hours later with DMSO (n=6 experiments), 100 nM osimertinib, 250 nM crizotinib, or both (each n=3 experiments) for 72 hours. 100% viability was set to total luminescence in DMSO-treated cells plated in RPMI10 containing PBS as buffer control. Each experiment (n) was performed as technical triplicates, which were averaged before determining the mean ± SD and significance across all biological replicates, determined by unpaired t test with single pooled variance. (D) Viability as determined by CTG of HCC827ER4 cells plated in RPMI10 and treated 24 hours later with osimertinib in combination with DMSO or the stated concentrations of linsitinib, ceritinib, crizotinib, or a combinations thereof for 72 hours. 100% viability was set to total luminescence in DMSO-only treated cells. Data are mean ± SD of at least three biological replicates, each performed as technical triplicates. Significance of comparison of the triple-combination curves (purple) to the crizotinib curve (blue; concentration marked by the arrow), was determined by unpaired t test with single pooled variance and Holm-Sidak’s multiple comparison test. (E) Cartoon depicting the effects that different CAF subsets (generically labelled CAF-A and CAF-B) secreting varying relative amounts of pro- (HGF, IGF) and anti- (IGFBP) tumorigenic proteins may have on cancer cell survival upon TKI treatment.