P1-08-13: Determinants of Risk, Characteristics and Prognosis of Breast Cancer Occurring after Hodgkin Lymphoma.

Abstract

Abstract Background Several studies have evaluated the excess of breast cancer (BC) risk after Hodgkin lymphoma (HL) but none has reported of all the aspects of BC occurring after HL including determinants of risk, comparison with other BC profile, treatment and outcome. This is the aim of our study. Patients and Methods We used data from the Surveillance, Epidemiology and End Results (SEER) Programme of the National Cancer Institute. To assess the risk and determinants of BC after HL we included all 9'620 women with first primary HL diagnosed between 1973 and 2007 and calculated age and period standardized incidence ratio (SIR) of BC as compared with general population. We also compared tumour characteristics and treatment between BC after HL (n= 316) vs. all other first primary BC occurring during the same period (n=450'413) by logistic regression. Finally, we compared breast cancer specific mortality between the two groups by Cox model. Results Overall, HL patients had an increased risk of developing BC (SIR: 2.4, 95% confidence interval [CI]: 2.2−2.7); the risk decreased with increasing age at HL, and disappeared when HL occurred after 50 years. BC risk was the same among those treated with or without radiotherapy until 15 years after HL and then an excess appeared in the irradiated group and persisted over 30 years after diagnosis of HL. Compared with other BC, BC after HL occurred at younger age (Adjusted odds ratio [OR] for age 40–59 vs. < 40 years 0.25, 95% CI: 0.20−0.33), was diagnosed at an early stage (OR for regional extension vs. localized 0.63, 95% CI: 0.42−0.89), expressed less frequently hormone receptors (OR for negative vs. positive status: 1.34, 95% CI: 0.99−1.81), was located more frequently in external quadrants (OR for internal vs. external: 0.61, 95% CI: 0.42−0.89), and received less frequently radiotherapy (OR for use vs. non use: 0.30, 95%CI: 0.22-042). The last two results were limited to patients who had received radiotherapy for HL. BC after HL presented higher breast cancer mortality (Hazard ratio: 1.36, 95%CI: 1.05−1.76) after adjustment for patients, tumour and treatment characteristics. Conclusion: This study provides valuable results on various aspects of BC occurring after HL. In particular, we found earlier stage at diagnosis of BC probably linked to higher surveillance, more aggressive tumours probably due to iatrogenic effect of HL treatment, more frequent occurrence in outer quadrants less protected during mantel radiation. Also, this study confirms the poorer prognostic of those BC not explained by differences in patient, tumour or treatment characteristics. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-08-13.