The aim of endocrine therapy is to reduce the estrogenic stimulus for tumour growth. After failure of tamoxifen--the standard "first-line" hormonotherapy for advanced breast cancer (ABC)--the most frequently prescribed endocrine therapies are progestins and aromatase inhibitors (AIs) ("second-line" hormonotherapy). Estrogen deprivation through AIs provides effective treatment of hormone-dependent ABC in postmenopausal (PMP) women. Over the past few years, the goals of our research programme were to develop more potent, more selective and better tolerated AIs than our first AI, aminoglutethimide (AG). Lentaron (4-hydroxyandrostenedione, formestane), a highly selective steroidal compound was the first of the new AIs to be used in clinical trials. It is a substrate analogue, administered as an i.m. injection every 2 weeks. It is effective in the treatment of ABC with an objective response rate (ORR) similar to tamoxifen and megestrol acetate (MA) and is generally well tolerated; rare instances of injection site reactions have been reported. Afema (fadrozole HCl), a non-steroidal AI is active orally, and effectively suppresses estrogen levels in PMP women, but it is not completely selective for aromatase when administered at higher than therapeutic doses. At the therapeutic dose of 1 mg twice a day it has an anti-tumour efficacy similar to MA, a good tolerability and no clinically relevant effects on other hormones of the endocrine system. Letrozole is the fourth of our AIs in clinical development. It is a non-steroidal, achiral, orally active, potent and highly selective competitive AI. Clinical endocrine studies have shown that the dose of 0.5 mg a day is the lowest dose achieving maximum estrogen suppression. Over a wide dose range, a lack of clinically relevant effects on other hormones of the endocrine system has confirmed its high selectivity. In four phase Ib/IIa studies in PMP patients with ABC who failed previous therapy, letrozole produced ORRs of 9, 31, 33 and 36%. One phase IIb/III study has been completed and two others are ongoing, comparing two doses of letrozole with MA or AG to confirm the anti-tumour efficacy of letrozole in the treatment of ABC in PMP women after treatment with anti-estrogens. Preliminary results from the completed trial show that letrozole 2.5 mg is superior to 0.5 mg in terms of ORR, time to progression and time to treatment failure, and is superior to the standard dose of MA in terms of ORR and tolerability. Therefore letrozole 2.5 mg can be recommended as a first choice for the treatment of PMP patients with hormone receptor-positive or unknown ABC after anti-estrogen therapy.