Differences between male and female breast cancer. III. Prognostic features

The aim of this study was direct comparison of prognostic features in male and female breast cancer patients. The study included 100 male and 500 female patients with pathohistologically confirmed breast cancer, who were treated at the University Hospital for Tumors in Zagreb, Croatia, between 1970 and 1990. Stage-adjusted univariate analysis of differences in five-year overall survival revealed similar prognosis in both sexes when the disease was in TNM-stages I and II (males: 95% and 80%; females: 90% and 79%, respectively). However, the study showed that in TNM stages III and IV, male patients had a significantly worse five-year overall survival (males: 39% and 5%; females: 61% and 27%, respectively; p < 0.05). The impact of ten clinicopathologic characteristics on survival in males and females was also analyzed through the multivariate statistical analysis, in this case by multiple regression. Analyzed characteristics included patient age at diagnosis, delay in treatment, tumor size, number of affected lymph nodes, TNM-stage, histologic grade, affected breast, estrogen and progesterone receptor values, and type of surgery. An extremely intriguing finding was that the percent of survival variance explained by ten selected predictors (R2) amounted to as much as 78.7% in males, in comparison to only 23.1% in females. Three predictors that explained most of the survival variance in males were TNM-stage (21.0%), number of affected lymph nodes (19.5%) and delay in treatment (16.2%). The most important corresponding prognostic factors in females were number of affected lymph nodes (12.7%), histologic grade (6.6%) and tumor size (5.9%). The authors concluded that, from the prognostic point of view, male and female breast cancer represent two quite different diseases: in males, the outcome can be very precisely predicted from the basic clinicopathologic characteristics of the disease, while in females there is probably a wide spectrum of other factors with a significant influence on the prognosis, which require additional studies.