Chronic caloric restriction (CCR) prevents mammary tumorigenesis in rodents, but a protective effect for intermittent caloric restriction (ICR) is less well documented. We recently reported that ICR reduced mammary tumor (MT) incidence of mouse mammary tumor virus-transforming growth factor (MMTV-TGF)-alpha mice to a greater extent than did CCR. Here, we repeated this protocol and obtained serum and tissue samples. Ad libitum (AL) MMTV-TGF-alpha mice were fed AIN-93M diet. Beginning at 10 weeks of age, ICR mice received isocaloric AIN-93M-mod diet (2-fold increases in protein, fat, vitamins, and minerals) at 50% of ad libitum for 3 weeks followed by 3 weeks refeeding with AIN-93M diet. CCR mice were pair-fed AIN-93M:AIN-93M-mod (2:1) matching intakes for restriction/refeeding cycles. Mice were sacrificed for MT size, at 79 (end of 12th restriction) or at 80 (1 week after 12th refeeding) weeks of age. AL and ICR-80 mice had heavier body weights than ICR-79 and CCR mice (P < 0.0001). Cumulative food intakes of ICR and CCR mice were reduced 12% and 15% versus AL mice (P < 0.0001). However, ICR mice consumed significantly (P < 0.0001) more food than did AL mice during refeeding. MT incidence was 84%, 13%, and 27% for AL, ICR, and CCR mice, respectively. MT weight (P < 0.0011) and number (P < 0.01) were higher for AL mice compared with ICR and CCR mice. AL and ICR-80 mice had similar serum IGF-I levels, but only AL values were higher than those of ICR-79 and CCR mice (P < 0.0017). ICR mice had more MT DNA breaks compared with AL and CCR mice, suggesting enhanced apoptosis (P < 0.02). AL mice had higher mammary fat pad ObR and ObRb leptin receptor mRNA expression than did ICR and CCR mice (P < 0.001), but there was no effect on MTs. These results confirm that ICR prevents development of MTs to a greater extent than does CCR, although "overeating" during refeeding may compromise this protection.