Including total EGFR staining in scoring improves EGFR mutations detection by mutation-specific antibodies and EGFR TKIs response prediction.

Abstract

Epidermal growth factor receptor (EGFR) is a novel target for therapy in subsets of non-small cell lung cancer, especially adenocarcinoma. Tumors with EGFR mutations showed good response to EGFR tyrosine kinase inhibitors (TKIs). We aimed to identify the discriminating capacity of immunohistochemical (IHC) scoring to detect L858R and E746-A750 deletion mutation in lung adenocarcinoma patients and predict EGFR TKIs response. Patients with surgically resected lung adenocarcinoma were enrolled. EGFR mutation status was genotyped by PCR and direct sequencing. Mutation-specific antibodies for L858R and E746-A750 deletion were used for IHC staining. Receiver operating characteristic (ROC) curves were used to determine the capacity of IHC, including intensity and/or quickscore (Q score), in differentiating L858R and E746-A750 deletion. We enrolled 143 patients during September 2000 to May 2009. Logistic-regression-model-based scoring containing both L858R Q score and total EGFR expression Q score was able to obtain a maximal area under the curve (AUC: 0.891) to differentiate the patients with L858R. Predictive model based on IHC Q score of E746-A750 deletion and IHC intensity of total EGFR expression reached an AUC of 0.969. The predictive model of L858R had a significantly higher AUC than L858R intensity only (p = 0.036). Of the six patients harboring complex EGFR mutations with classical mutation patterns, five had positive IHC staining. For EGFR TKI treated cancer recurrence patients, those with positive mutation-specific antibody IHC staining had better EGFR TKI response (p = 0.008) and longer progression-free survival (p = 0.012) than those without. In conclusion, total EGFR expression should be included in the IHC interpretation of L858R. After adjusting for total EGFR expression, the scoring method decreased the false positive rate and increased diagnostic power. According to the scoring method, the IHC method is useful to predict the clinical outcome and refine personalized therapy.