Risks of second malignancies after breast cancer treatment: Long-term results.

ABSTRACT

PURPOSE: To estimate the long-term risk of second malignancies after breast cancer treatment in a large homogeneous cohort from a single institution. PATIENTS AND METHODS: All patients in this study were treated for non-metastatic breast cancer at the Curie institute, Paris, between 1981 and 2000. We calculated the cumulative incidence of second malignancies and the risk of developing each type of second malignancies over a period of 10 to 15 years. The observed crude incidence rates in the entire patient population were then compared to the expected incidence in the general population of French women, as provided by age-standardized data. A standardized incidence ratio (SIR) was calculated for all second malignancies. We also calculated second malignancies standardized incidence ratios for patients who underwent adjuvant therapy for breast cancer. RESULTS: The study cohort included a total of 17,745 women. The median follow-up since diagnosis was 13.4 years (range 2-29 years). The 15-year cumulative incidence of second malignancies was 1.807 per 100,000 (CI 1.729-1.884). A total of 2370 second malignancies were observed during follow-up, 2010 in the radiotherapy arm and 360 in the no radiotherapy arm (relative risk [RR] 1.15 [1.03-1.28], P=0.0134). Crude incidence rates were significantly higher in our cohort than in the general population for contralateral breast cancer (SIR 2.96 [confidence interval (CI) 2.82-3.12], P<0.0001), sarcomas (SIR 8.48 [CI 6.41-11.22], P<0.0001), leukaemia (SIR 2.37 [CI 1.85-3.04], P<0.0001), lung cancer (SIR 1.39 [CI 1.13-1.72], P<0.0022) and gynaecological cancer (SIR 1.31 [CI 1.15-1.50], P=0.0001). Among patients treated for breast cancer, those who received radiotherapy was associated with an excess risk of sarcoma as compared to those have not had (RR 5.59 [CI 1.35-23.17], P<0.001). CONCLUSIONS: Women treated for breast cancer had a significantly increased risk of several kinds of second malignancies compared to the general population.