Purpose: The association between HER-2-positivity, and prognostic variables and survival have been addressed in many studies with still controversial results because of the small series analyzed.
Experimental design: A series of 1928 primary breast carcinomas was analyzed for the prognostic potential of HER-2 overexpression.
Results: In our series, HER-2-positivity was not associated with nodal status, unless the number of infiltrated nodes was considered, whereas it was strongly associated with large tumors (P < 10(-4)), grade III tumors (P < 10(-4)), lymphoid infiltration (P < 10(-4)), and absence of hormone receptor expression (P < 10(-4)). HER-2 overexpression was a strong prognostic indicator in N+ patients (P < 10(-7)), whereas its prognostic impact was weak and not statistically significant in the N- patients. Analysis of the hazard ratio of relapse in relation to time from surgery indicates that the poor prognosis associated with HER-2 positivity in N+ patients was found to be attributable to a peak of relapses in the first 3-4 years from surgery. Multivariate analysis of different prognostic factors in HER-2+ and HER-2- subsets indicated that grade is the most important factor followed by nodal status, lymphoid infiltration, and tumor size in HER-2-negative breast carcinomas, whereas nodal status was the most important prognostic factor, with tumor size showing only borderline significance, in the HER-2-positive group.
Conclusions: Together, the results indicate that HER-2-positive breast carcinomas represent a particular subset of tumors with peculiar clinical and pathological behaviors. Thus, conclusions drawn from clinical trials, which serve as the basis for clinical management of breast carcinomas, might not always be valid for this low-frequency subset.