EGFR-activating mutations predict responsiveness to EGFR tyrosine kinase inhibitors (TKIs) in non-small cell lung cancer (NSCLC) patients. Mutation screening is crucial to support therapeutic decisions and is commonly conducted using dideoxy sequencing, although its sensitivity is suboptimal in clinical settings. To evaluate the diagnostic performance of pyrosequencing and dideoxy sequencing, we examined EGFR mutation status in a retrospective cohort of 53 patients with NSCLCs clinically selected for TKI therapy and whose clinical outcome was available. Moreover, pyrosequencing quantitative results were compared with EGFR amplification data. EGFR mutations were investigated by pyrosequencing and by dideoxy sequencing. Detection rates of both methods were determined by titration assays using NCI-H1975 and HCC-827 cell lines. Increased EGFR copy number was assessed by fluorescence in situ hybridization (FISH). Pyrosequencing showed a higher detection rate than dideoxy sequencing. Tumor control rate of cases with mutant and wild-type EGFR was 86% and 29%, respectively. EGFR amplification was significantly associated with EGFR mutation and a positive correlation between high percentages of mutant alleles and clinical response to TKI was observed. We concluded that pyrosequencing is more sensitive than dideoxy sequencing in mutation screening for EGFR mutations. Detection rate of dideoxy sequencing was suboptimal when low frequencies of mutant alleles or low tumor cell contents were observed. Pyrosequencing enables quantification of mutant alleles that correlates well with increased EGFR copy number assessed by FISH. Pyrosequencing should be used in molecular diagnostic of NSCLC to appropriately select patients who are likely to benefit from TKI therapy.