The association between phthalate exposure and breast cancer remains controversial. We performed a prospective patient cohort design to explore the interaction between creatinine-corrected urinary phthalate metabolites and hormone receptors as well as body mass index (BMI) on recurrent breast cancer. In this follow-up study, 636 female breast cancer patients and 45 new recurrent cases diagnosed for a total of 1576.68 person-years of follow-up were recruited. Mono-(2-ethyl-5-oxohexyl) phthalate (MEOHP) was negatively associated with breast cancer recurrence, with adjusted hazard ratio (aHR) 3rd vs. 1st quartile of 0.15 (95% CI 0.04-0.51). The MEOHP presented as a non-monotonic dose-response (NMDR) curve, being U-shaped. In the stratification of hormone receptors, MEOHP still exhibited a U-shaped dose-response curve. The third quartile of MEOHP showed significant lowest recurrent risk in the status of ER-positive (aHR 0.18, 95% CI 0.05-0.66), PR-negative (aHR 0.14, 95% CI 0.03-0.63), and HER2-negative (aHR 0.24, 95% CI 0.08-0.76). Whether in BMI < 25 or in BMI ≥ 25, the third quartile of MEOHP was negatively associated with recurrent breast cancer, and there was a negative interaction on an additive scale between MEOHP and BMI (p  = 0.042). The association between MEOHP and recurrent breast cancer was modified by hormone receptors and BMI.