Worldwide, breast cancer is the second leading cause of cancer death among women and the third most common cancer. Although our understanding of the molecular basis of this fatal disease has improved, this malignancy remains elusive. Melatonin (Mel), retinoic acid (RA) and Nigella sativa (NS) are substances with anticancer effects. To date, our understanding of the mechanisms of therapeutic effects of these products in mammary cancer is still marginal. To look at the preventive and therapeutic values of these products, we carried out this investigation. An animal model formed of 80 rats was established. The animals were divided into eight groups of 10 animals each: (a) control group injected with the same vehicle used for treatments in the relevant dosages and routes; (b) carcinogen group injected with the known carcinogenic substance 7,12-di-methylbenz(a)anthracene (DMBA) that induces mammary carcinoma; (c) three prophylactic (Pro) groups (Mel-Pro, RA-Pro and NS-Pro) injected with test substances (Mel, RA and NS, respectively) 14 days before the intake of the carcinogenic substance DMBA and then continued until the end of the experiments; and (d) three treated (Tr) groups (Mel-Tr, RA-Tr and NS-Tr) injected with the vehicles after the intake of DMBA. In both the Pro and Tr groups, the drugs were daily administered for 3 months. The animals were killed, and their serum and tissues were evaluated for (a) markers of tumorigenicity [serum levels of total sialic acid (TSA) and lipid-bound sialic acid (LSA)], (b) markers of endocrine derangement (serum prolactin, estradiol and progesterone levels), (c) apoptotic changes [serum tumour necrosis factor (TNF)-alpha, tissue caspase-3 activity, percentage of DNA fragmentation and ultrastructural features of apoptosis] and (d) markers of oxidative stress (tissue levels of lipid peroxides and nitric oxide). Carcinoma was absent both in the control and in the NS-Pro groups. Mammary carcinoma occurred in DMBA and other Pro and Tr groups. The frequency of mammary carcinoma was high in the carcinogen DMBA group (60%), followed by the Tr (56%) and finally the Pro groups (33%). These tumours included papillary, comedo and cribriform carcinomas. As compared with the control group, the development of carcinoma in the carcinogen DMBA group was associated with increased levels of (a) markers of tumorigenicity (77.0 +/- 3.3 vs. 209.0 +/- 5.6 and P < 0.05 for TSA; 28.7 +/- 1.7 vs. 41.8 +/- 1.2 and P < 0.01 for LSA), (b) markers of endocrine derangement (2.5 +/- 0.1 vs. 3.6 +/- 0.3 and P < 0.05 for prolactin; 39.6 +/- 1.3 vs. 24.8 +/- 2.1 and P < 0.01 for progesterone and 31.0 +/- 0.7 vs. 51.1 +/- 3.4 and P < 0.01 for estradiol) and (c) markers of oxidative stress (2.3 +/- 0.2 vs. 5.2 +/- 0.7 and P < 0.01 for lipid peroxides and 4.4 +/- 0.2 vs. 7.6 +/- 0.8 and P < 0.01 for nitric oxide). Also, it was associated with decreased levels of markers of apoptotic activity (20.8 +/- 1.1 vs. 13.4 +/- 0.7 and P < 0.01 for caspase-3; 29.0 +/- 1.7 vs. 20.9 +/- 1.3 and P < 0.05 for percentage of DNA fragmentation; and 9.4 +/- 0.8 vs. 52.1 +/- 3.3 and P < 0.01 for TNF-alpha). When compared with the carcinogen DMBA group, the development of carcinoma in the Pro and Tr groups was associated with decreased levels of (a) markers of tumorigenicity, (b) markers of endocrine derangement and (c) markers of oxidative stress. Alternatively, carcinogenicity was associated with statistically significant (P < 0.01) increased levels of markers of apoptotic activity. To conclude, the administration of Mel, RA and NS reduced the carcinogenic effects of DMBA, suggesting a protective role. The possible underlying mechanisms of these effects await further investigations.