The active mutation of epidermal growth factor receptor (EGFR) and clinical characteristics are significant biomarkers for chemotherapy selection in non-small cell lung cancer (NSCLC). Although docetaxel is a key agent in second-line therapy for NSCLC, predictive biomarkers for assessing its efficacy have yet to be determined. To assess the clinical efficacy of docetaxel in second-line therapy for NSCLC according to NSCLC histology and the therapeutic effect of EGFR-tyrosine kinase inhibitors (EGFR-TKIs), we retrospectively reviewed 454 NSCLC patients treated with docetaxel between April 2002 and April 2009. In total, 239 patients with advanced NSCLC treated with docetaxel as second-line therapy following failure of platinum-based chemotherapy were analyzed in this study. A total of 59 (25%) patients had squamous cell carcinoma. The overall response rate and median progression-free survival time in the squamous cell group were significantly inferior to those in the non-squamous cell group (p=0.031 and p=0.005, respectively). Following the failure of docetaxel, 91 non-squamous patients were treated with EGFR-TKIs. The patients that achieved clinical benefit from EGFR-TKIs (n=32) demonstrated a significantly better response rate and longer progression-free survival compared to the other group (p<0.001 and p=0.027, respectively). In the univariate and multivariate analysis, the favorable therapeutic effect of EGFR-TKIs had an independent effect on progression- free survival (HR 1.484, p=0.0464). In conclusion, this retrospective study suggests that non-squamous histology and favorable therapeutic effect from EGFR-TKIs are useful markers for predicting the efficacy of docetaxel in second-line therapy for NSCLC.