Gefitinib, an epidermal growth factor receptor tyrosine kinase inhibitor, has shown both anti-proliferative and anti-tumoral activity in breast cancer. This study was designed to determine the effect of adding gefitinib to neoadjuvant epirubicin and cyclophosphamide (EC) on tumor response rates. Women with unilateral, primary operable, estrogen receptor negative invasive breast cancer = 2 cm were eligible for inclusion. Randomized patients were to receive four cycles of neoadjuvant EC plus 12 weeks of either gefitinib (250 mg daily) or placebo. Primary endpoint was pathologic complete response (pCR), and secondary endpoints were complete response (CR) and overall objective response (OR). 181 patients were randomized. A pCR was observed in 17% (12/71) of patients treated with gefitinib and in 12% (9/73) of patients treated with placebo (4.57% difference, 95% CI -7.19 to 6.33; P = 0.44). CR was observed in 10% of patients in both the gefitinib (7/71) and the placebo group (7/73) (0.27% difference, 95% CI -9.6 to 10.2; P = 0.96). There was no significant difference in OR (5.96%; 95% CI -9.9 to 21.9; P = 0.45) between the two groups. Post hoc subgroup analysis showed a significant difference in pCR between triple negative breast cancer (TNBC) and non-TNBC tumors (P = 0.03). More patients in the gefitinib arm had hematological toxicity (P = 0.15) and discontinued treatment (9/94 vs. 2/86) because of adverse events (AE). Tumor response rates were similar in the two groups. A significantly higher pCR rate was observed post hoc in TNBC versus non-TNBC independent of treatment. More patients in the gefitinib group discontinued treatment because of AE.