1322 - Predictive biomarkers in NSCLC patients treated with erlotinib after chemotherapy: EGFR expression or mutations?

Date 28 September 2012
Event ESMO Congress 2012
Session Publication Only
Topics Biomarkers
Non-Small-Cell Lung Cancer, Metastatic
Presenter Alessandro Inno
Authors A. Inno1, E. Maci2, M. Martini3, V. Arena3, V.P. Di Noia2, G. Schinzari2, L.M. Larocca3, A. Cassano2, C. Pozzo4, C. Barone2
  • 1Medical Oncology, Università Cattolica del Sacro Cuore, 00168 - Roma/IT
  • 2Oncologia Medica, Università Cattolica del Sacro Cuore, 00168 - Roma/IT
  • 3Istituto Di Patologia, Università Cattolica del Sacro Cuore, Roma/IT
  • 4Dept. Medical Oncology, Università Cattolica del Sacro Cuore, 00168 - Roma/IT



Retrospective subgroup analysis from randomized trials did not show a significant association between activating EGFR mutations and benefit from erlotinib in patients with NSCLC previously treated with chemotherapy, so whether EGFR mutational status should be used as a tool to select patients for erlotinib in the second-line setting is debatable. Novel predictive biomarkers would be helpful to choose the most appropriate therapeutic strategy.


We correlated retrospectively the mutational status of EGFR and KRAS and also the IHC expression of EGFR, cMET, IGF1R and HER2, with the outcome of 51 patients with metastatic NSCLC treated with erlotinib as second or third-line at our institution from 2009 to 2012.


EGFR and KRAS activating mutations were mutually exclusive and were found in 5 and 7 patients. IHC score for EGFR, IGF1R, cMET and HER2 was 3+ in 11, 8, 7 and 1 patients, respectively. RR was 20% and 13% and PFS was 9 and 2 months for EGFR mutant and wt patients, respectively, but the difference was not statistically significant. Similarly, KRAS mutational status did not significantly affect the outcome, although none of KRAS mutant patients achieved an objective response. RR and PFS were not related to IGF1R and cMET expression. Nine out of 11 patients (81.8%) with EGFR overexpression responded to treatment, compared with 13 out of 40 patients (32.5%) with a lower EGFR level. EGFR overexpression was also associated with a longer PFS (8 vs 2 months, p = 0.05). Interestingly, in our study the outcome of patients was not affected by gender, performance status, histology or smoking history.


cMET and IGF1R were not related to the efficacy of erlotinib as second or third-line treatment for metastatic NSCLC patients. No conclusions can be drawn about HER2 since only one case of overexpression was found. EGFR mutations are associated with a longer PFS, even if not significantly. The IHC expression of EGFR seems to be predictive of a better outcome, whereas KRAS mutations may represent a mechanism of resistance. Given the retrospective nature of our study, however, those findings should be confirmed within prospective clinical trials.


A. Inno: Speaker at educational meetings sponsored by Merk-Serono and Amgen.

All other authors have declared no conflicts of interest.