Purpose: To evaluate the in vivo effect of adding gefitinib to preoperative chemotherapy on the EGFR-dependent p42/44 MAPK in operable breast cancer (BC) patients. Secondary aims: to evaluate EGFR, (p)-EGFR, Ki67, apoptotic index (TUNEL test) and VEGFR2 expression from baseline to surgery, percentage of pathologic complete response (pCR), and toxicity. Patients and Methods: 90 patients with stage II-IIIA BC have been randomized to receive epirubicin 90 mg/sqm and paclitaxel 175 mg/sqm on day 1 plus: gefitinib 250 mg daily from day 5 to 16 (Arm A, intermittent), gefitinib 250 mg daily from day 1 to 21 (Arm B, continuous), or placebo (Arm C). Treatment plan: 4 courses every 3 weeks, followed by surgery. Results: After preoperative therapy, 86/90 patients underwent surgery; 46 patients (51%) received breast conservative surgery. A pCR was observed in 4 patients. No significant differences in the expression of p42/44 MAPK, EGFR, (p)-EGFR, VEGFR2, proliferation index and apoptosis were observed comparing the combined Arms A + B vs C, and comparing Arm A vs B. Hematologic toxicities were not significantly different comparing Arms A + B vs Arm C, and comparing Arm A vs B. Significantly higher skin and mucosal toxicities were observed when comparing the two gefitinib Arms (A + B) vs Arm C (32% vs 9.6%, P = 0.018; 57% vs 29%, P = 0.009 respectively), while no significant differences were observed comparing Arm A vs B. Conclusion: Adding gefitinib to chemotherapy did not result in different effects on the EGFR-dependent pathway, proliferation, apoptosis and VEGFR2 expression as compared to placebo, while enhancing skin and mucosal toxicity. The two schedules of gefitinib (intermittent vs continuous) did not result in different biologic effects. © 2007 Springer Science+Business Media, LLC.