For classification of breast cancer (BC), tumor-node-metastasis (TNM) staging has been considered state of the art for more than 50 years. The T category is well defined, and in multicentric and multifocal tumors, tumor size is assessed by the largest tumor focus. The aim of this study was to compare multicentric/multifocal tumor spread in breast cancer with unifocal disease and to evaluate the diagnostic relevance of multifocality. A retrospective analysis was performed on survival related events in a series of 5,691 breast cancer patients between 1963 and 2007. By matched-pair analysis, patients were entered into two comparable groups of 288 patients after categorizing them as having multifocal/multicentric or unifocal breast cancers. Matching criteria were tumor size, grading, and hormone receptor status, which were equally distributed between both groups (P = 1.000 each). Disease free survival and the occurrence of relapse or of metastatic disease were evaluated. Cox's regression analysis was used for multivariate analysis. In the unifocal group, the mean breast cancer-specific survival time was 221.6 months as opposed to 203.3 months in the multicentric/multifocal group (P < 0.001, log-rank test). The occurrence of local relapse and distant metastasis was significantly increased in the multifocal group in comparison to the unifocal equivalent group (P < 0.001 and P < 0.003, respectively). Cox regression analysis for multivariate analyses demonstrated focality and centricity to be highly significant predictors for reduced overall survival (P = 0.016), local relapse (P = 0.001) and distant metastasis (P = 0.038). Tumor size, histopathological grading, hormone receptor status, and staging of lymph nodes are well-established prognostic parameters. Additionally, the number of foci should be considered as an independent prognostic parameter, which is currently not reflected in the TNM classification. We conclude that multicentric/multifocal BC is an independent BC risk factor and should be included in the risk assessment by re-evaluating the current TNM classification of the UICC.

#2075 Background:&#x2028; For classification of breast cancer, the TNM-staging is standard since more than fifty years and the prognostic relevance of the T-category is well established. However, tumor size in multicentric and multifocal tumors is assessed by the largest tumor focus only, and does not reflect the cumulative tumor volume. Aim of this study was to compare the prognosis of multicentric/-focal tumors vs unifocal tumors with identical tumor sizes according to TNM-Staging.&#x2028; Methods:&#x2028; Overall, 576 patients between 1963 and 2007 were examined. Based on a consecutive patient cohort (n=5.691 pts), two equivalent groups of 288 early breast cancer patients with multicentric/-focal vs unifocal tumors were selected by matched-paired analysis. Match criterias were tumor size, histopathologic grading as well as hormone receptor status, which were equally distributed in both groups (p=1,0, each). Disease free survival and breast cancer specific survival were evaluated. For the multivariate analysis the Cox regression system was used. The median follow-up was 70 month (5,8 years, range 19 - 312 mts)&#x2028; Results:&#x2028; The Kaplan-Meier estimate for mean breast cancer specific survival was 221.6 mon (206.7 – 236.4 95%CI) in patients with unifocal and 203.3 mon (184.1 – 222.5, 95% CI, p&lt; .0001, log rank test) in patients with multicentric/-focal tumors. Furthermore, in patients with a unifocal tumor the relapse-free survival was 205.9 mon (190.7 – 221.2 95%CI) and 169.6 mon (150.8 – 188.5, 95% CI, p&lt; .0001, log rank test) in patients with multicentric/-focal disease. In multivariate Cox regression analysis, allowing for tumor size, grading and nodal status, multicentricity/-focality was an independent significant predictor for reduced breast cancer specific survival (p&lt; .0001) as well as reduced relapse-free survival.&#x2028; Conclusion:&#x2028; In breast cancer, multicentric/-focal disease is associated with a worse prognosis compared to unifocal disease with identical tumor size. The authors conclude that the tumor biology of multicentric/-focal tumors is not reflected adequately by the current TNM-classification. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 2075.