To assess whether the described
To assess whether the described cooperation between the EGFR and FGFR receptors was relevant at the clinical level, we determined the effect on progression-free survival of FGFR1 expression in patients with adenocarcinoma who were receiving erlotinib or gefitinib. As we had observed EGFR-FGFR1 cooperation taking place not only with mutated EGFR but also with its wild-type variant, we included wild-type and mutated EGFR tumors in this analysis and found shorter times to progression for patients with high FGFR1 expression. Interestingly, concordant results were obtained from analysis of an extended cohort in which all NSCLC histologic types were represented, which suggests that these findings may be applicable beyond adenocarcinoma tumors. In line with this, NSCLC cells from all histologic types with high FGFR1 expression levels and EGFR activation showed increased resistance to EGFR inhibitors. Upregulation of some FGFRs, such as FGFR1 or fibroblast growth factor receptor 2, and fibroblast growth factor receptor 3, mutations have been reported as mechanisms of acquired resistance to anti-EGFR therapy,22, 23, 24, 36, 37 and combined EGFR and FGFR inhibition has been proposed in this setting. Furthermore, fibroblast growth factor receptor 2 inhibition has been linked to increased erlotinib sensitivity in several in vitro models of lung cancer. These results certainly suggest a relationship between both EGFR and FGFR receptors in acquired resistance to anti-EGFR inhibition. However, we are the first to report a reciprocal activating interaction between FGFR1 and EGFR leading to intrinsic anti-EGFR resistance. Our data provide clinical evidence that tumors may exhibit high levels of FGFR1 expression before anti-EGFR therapy, which would make EGFR inhibition less effective: patients bearing tumors with these characteristics may benefit from dual inhibition of EGFR and FGFR, which would overcome this primary resistance.
Prooncogenic cooperation among receptor tyrosine kinases, similar to the FGFR1-EGFR interaction reported here for lung cancer, occurs in diverse malignancies. Insulin-like growth factor 1 receptor (IGF1R) has been found to interact with the insulin receptor in gastric and hepatocellular carcinoma cell lines, and the treatment of these cell lines with an antibody targeting this interaction reduces STAT3 and AKT activation. In Heme Colorimetric Assay Kit cancer, it has been reported that EGFR can physically interact with platelet derived growth factor receptor beta and induce p42/p44 activation and resistance to anti-EGFR therapy.
In the context of lung cancer, the heterodimerization of EGFR with IGF1R has also been described as a mechanism of erlotinib resistance, which could be bypassed through IGF1R inhibition.
EGFR activation induces FGFR1 inhibition resistance in head and neck carcinoma cell lines, suggesting that the FGFR1-EGFR interaction may be of relevance in other tumor types as well. To date, most targeted therapeutic approaches in the lung cancer setting have focused on concrete driver genetic alterations, but all of those data, along with the results of the present work, highlight the importance of a more comprehensive molecular characterization of tumors that harbor not just one but numerous molecular aberrations. The study of the coactivation of diverse signaling pathways may be of clinical relevance for predicting primary resistance to targeted therapies and the identification of efficacious combination therapies.
Acknowledgments Dr. Paz-Ares was funded by ISCIII (grants PI14/01964, PIE15/00076, PI17/00778 and DTS17/00089) and CIBERONC (CD16/12/00442) and cofunded by FEDER from Regional Development European Funds (European Union). Dr. Carnero was supported by grants from the Spanish Ministry of Economy and Competitiveness Plan Estatal de IþDþI 2013-2016, ISCIII (PI15/00045) and CIBERONC (CD16/12/00275) and cofunded by FEDER from Regional Development European Funds (European Union). Dr. Molina-Pinelo is funded by Ministry of Health and Social Welfare of Junta de Andalucía (PI-0046-2012, Nicolas Monardes Program C-0040-2016), Mutua Madrileña Foundation (2014) and ISCIII (PI17/00033). Dr. Ferrer is funded by AECC (AIO2015) and Ministry of Equality, Health and Social Policies of the Junta de Andalucía (PI-0029-2013), Comunidad de Madrid (B2017/BMD3884), and ISCIII (PI16/01311) and cofunded by FEDER from Regional Development European Funds (European Union). Dr. Quintanal-Villalonga is funded by ISCIII (FI12/00429). Dr. Ojeda-Márquez is funded by Ministry of Education, Culture and Sports (FPU13/02595). We thank those in the Spanish National Cancer Research Center Confocal Microscopy Unit for their specialized assistance and support. The authors thank the donors and the 12 de Octubre Hospital Biobank and University Hospital Virgen del Rocío Biobank for the human specimens used in this study.