Retinoids have been studied as chemopreventive agents in clinical trials due to their established role in regulating cell growth, differentiation and apoptosis in preclinical models. Experimental evidence suggests that retinoids affect gene expression both directly, by activating and/or repressing specific genes, and indirectly, by interfering with different signal transduction pathways. Induction of apoptosis is a unique feature of fenretinide, the most widely studied retinoid in clinical trials on breast cancer chemoprevention due to its selective accumulation in breast tissue and to its favourable toxicological profile. In a phase III breast cancer prevention trial, fenretinide showed a durable trend to a reduction of second breast malignancies in premenopausal women. This pattern was associated with a favourable modulation of circulating IGF-I and its main binding protein (IGF-binding protein-3, IGFBP-3), which have been associated with breast cancer risk in premenopausal women in different prospective studies. In a subsequent biomarker study on premenopausal women who had participated in the phase III trial, high IGF-I and low IGFBP-3 baseline levels were found to predict second breast cancer risk, although the magnitude of their changes during treatment did not fulfil the requirements for suitable surrogate end-point biomarkers. In postmenopausal women, fenretinide did not reduce second breast cancer incidence, nor did it induce significant modulation of the IGF system. Similarly, fenretinide was not found to affect risk biomarkers significantly in early postmenopausal women on hormone replacement therapy, who are at increased risk of developing breast cancer. Biomarker studies of fenretinide alone or in combination with different nuclear receptor ligands are being conducted. In particular, clinical trials of fenretinide and tamoxifen have proved to be feasible, and this combination appears to be safe and well tolerated in high-risk women, especially when low-dose tamoxifen is employed. Novel retinoid X receptor-selective retinoids, or rexinoids, have been shown to suppress the development of breast cancer in several animal models with minimal toxicity, and are being intensively studied either alone or in combination with selective oestrogen receptor modulators, both in vitro and in vivo. The rexinoid, bexarotene, has recently been approved for the treatment of patients with cutaneous T-cell lymphoma, and a biomarker trial with bexarotene in women with high breast cancer risk is currently underway.