Is tumor cellularity in primary invasive breast carcinoma of prognostic significance?

Many methods of analysis to predict survival of invasive mammary carcinoma in the post-neoadjuvant setting utilize tumor cellularity alone or in combination with other tumor features. The goal of this study was to evaluate the prognostic value of tumor cellularity in primary non-treated carcinoma. We used 366 cases of invasive breast carcinoma to determine invasive tumor cellularity (%) by reviewing a representative excisional tumor section and correlated this with breast cancer recurrence (BCR) and overall mortality (OM). Mean patient age was 58 years (range, 21–91) and median follow-up was 87 months (range, 0.7–165). Of the cases, 25% were Nottingham grades I, 41% grade II, and 32% grade III. The Nottingham Prognostic Index (NPI) ranged from 2.06 to 6.8 (mean 3.93). Estrogen receptor was positive in 66% and negative in 25% of cases. Cellularity ranged from 2 to 99% (mean 47.6%). The OM hazard ratio increased by 1.73 for every unit increase in NPI (P < 0.00005; 95% confidence interval 1.45–2.05) The BCR hazard ratio increased by 2.011 for every unit increase in NPI BCR (P < 0.00005; 95% confidence interval 1.62–2.50). Cellularity, unadjusted for other covariates, was not significantly associated with either OM or BCR. When adjusted for NPI, cellularity still showed no significant relation to OM or BCR. The same analysis performed on estrogen receptor-positive and estrogen receptor-negative subgroups continued to show no relation between cellularity and OM or BCR. In conclusion, despite its utility in the neoadjuvant setting, we were unable to show that cellularity is predictive of survival in primary breast carcinomas.