Incidence, time course, and determinants of menstrual bleeding after breast cancer treatment: a prospective study.

OBJECTIVE

To assess ovarian function using the surrogate of monthly bleeding after breast cancer treatment in premenopausal women.

METHODS

Five hundred ninety-five US women age 20 to 45 years were accrued from January 1998 to July 2002 within 8 months of diagnosis with stages I to III breast cancer (median follow-up 45 months). Daily bleeding records were obtained prospectively, as well as extensive clinical, demographic, quality of life, and treatment data. Repeated measures logistic regression was used to assess which variables were predictive of monthly bleeding.

RESULTS

Significantly different proportions of women had monthly bleeding depending on their age (P < .001), chemotherapy program (P < .001), and time since treatment regimen. In the month after the standard course of doxorubicin and cyclophosphamide (AC), whether or not followed by paclitaxel or docetaxel, approximately 16% had monthly bleeding compared with the cyclophosphamide, methotrexate, fluorouracil (CMF) group, in which 48% bled (P < .001). Following any AC regimen, there was a slow recovery phase of about 9 months followed by a plateau, during which almost half continued monthly bleeding for the remainder of the follow-up period compared with after CMF in which there was no recovery phase and a continual decline in monthly bleeding to approximately 18% of women at study end (P < .001). Tamoxifen use decreased bleeding between months 12 and 24 after chemotherapy with 15% fewer women having bleeding.

CONCLUSIONS

Using daily menstrual bleeding records, it is demonstrated that age, the specific chemotherapy regimen received, and tamoxifen use impact ovarian function.