EGFR
ligands (e.g.,
EGF and TGFA) have been shown to be clinically associated with poor survival in
lung cancer. Since TGFA itself initiates autochthonous
tumors in liver, breast, and pancreas but not in the lung in transgenic mice in vivo, it would appear that an EGFR
ligand may not initiate but rather promote
lung cancer. However, it has not been proven in vivo whether
lung cancer is promoted by an EGFR
ligand. Using transgenic mouse models conditionally expressing EGFRL858R or KrasG12D with TGFA (an EGFR
ligand) in lung epithelium, we determined that TGFA promoted the growth of EGFRL858R-lung
tumors in airway regions but not that of KrasG12D-lung
tumors. Analysis of TCGA datasets identified ΔNp63 and AGR2 as potential key
tumor-promoting regulators, which were highly induced in the TGFA-induced EGFRL858R-lung
tumors. The expression of AGR2 was positively correlated with the expression of TGFA in human EGFR-mutant
lung adenocarcinomas. The expression of TGFA in human EGFR-mutant
lung adenocarcinomas but not in the EGFR wild-type
lung adenocarcinoma was associated with poor survival. These results suggest that targeting EGFR
ligands may benefit patients who carry EGFR-mutant lung
tumors but will not benefit patients with KRAS-mutant lung
tumors.