Résumé : Background Preclinical data suggest that epidermal growth factor receptor (EGFR) inhibitors (e.g. gefitinib) can delay endocrine resistance in breast cancer. A double-blind, placebo-controlled, phase II trial investigated whether adding gefitinib (G) to anastrozole (A) would improve outcome in advanced breast cancer (ABC). Methods Postmenopausal pre-treated hormone receptor-positive ABC patients (locally recurrent or metastatic) were 1:1 randomized to A (1 mg/d) plus G 250 mg/d or plus placebo (P). Patients who had prior treatment with an aromatase inhibitor in metastatic setting or with trastuzumab, anti-EGFR or anti-VEGF agents were excluded. Treatment was given until disease progression, unacceptable toxicity or patient withdrawal. Progression-free survival (PFS) rate at 1 year was assessed according to Response Evaluation Criteria in Solid Tumours, version 1.0. Results Of 108 planned patients, 71 were recruited (36 in A/G and 35 in A/P). The trial closed prematurely due to slow recruitment; 31 patients had prior chemotherapy and 53 prior endocrine therapy (all except one received tamoxifen); 60% in adjuvant and 16% in metastatic setting received tamoxifen; 59 patients had visceral disease. Median follow-up was 18 months. PFS rate at 1 year was 35% for A/G and 32% for A/P arm. Objective responses were six (22%) in the A/G and nine (28%) in the A/P arm. Median duration of response was 13.8 and 18.6 months in the A/G and A/P arms, respectively. Fatigue (35%), diarrhoea (31%), rash (32%), dry skin (27%), and arthralgia/myalgia (27%) were the commonest adverse events in the A/G arm. Conclusions This phase II study, although prematurely closed, did not show a signal that adding G to A improves PFS at 1 year and its use is not supported. Gastrointestinal and skin toxicities were more pronounced with G resulting in premature therapy interruption in almost 1 in 3 patients (ClinicalTrials.gov number, NCT00066378).