Can Tumor Cellularity Predict Outcomes in Primary Non-Treated Breast Carcinoma?
Emily S Reisenbichler, Omar Hameed. Brigham and Women's Hospital, Boston, MA; Vanderbilt University, Nashville, TN
Background: Tumor cellularity (TC) is used to calculate residual cancer burden in breast carcinoma (BC) following neoadjuvant therapy, with the latter being shown to predict distant relapse-free survival (RFS). It is not clear however, whether TC can help predict outcomes in non-treated BC. The goal of this study was to evaluate the prognostic value of TC in this particular setting.
Design: Following a detailed histologic review and after excluding foci of necrosis and in-situ carcinoma, TC (%) was evaluated in the primary excision of 366 cases of BC. Tumor size (TS), histological type and grade, lymph node status, RFS and overall survival (OS) were also recorded. The TC was then multiplied by size (mm) to derive the primary tumor burden (PTB). A receiver operating curve (ROC) was then used to determine the best PTB cutoff point to segregate the cohort into 2 groups for survival analysis.
Results: Mean patient age was 58 yr (range, 21-91) and median follow-up was 87 mo (range, 0.7-165). Invasive ductal carcinoma of no special type constituted 80% of cases, invasive lobular carcinoma 10%, and other special types of carcinoma, 10%. Nottingham grades I, II and III, represented 25%, 41% and 32% of the cases, respectively (unknown in 4). TC ranged from 2-99% (mean 47.6%) and PTB from 1-64 (mean, 10). A PTB cutoff of 16.9 (determined by receiver operative curve analysis) was able to stratify patients into high and low risk groups with significantly different RFS (Fig 1) and OS (Fig 2) rates. Additionally, a high PTB was predictive of a worse outcome (hazard ratio, 5.5; 95% confidence interval, 2.2-13.7) in patients with lymph node negative disease (P=.0002).
Conclusions: Tumor cellularity appears to be a prognostic marker in primary breast carcinoma. Additional statistical analysis is being performed to determine whether incorporating this along with other established markers can better define outcomes.
Tuesday, March 20, 2012 1:00 PM
Poster Session IV # 6, Tuesday Afternoon