[168] HER2 Over-Expression and Histological Features Correlation in Argentinean Breast Cancer Patients
I Frahm, G Acosta Haab, S Sarancone, V Caceres, HER2 Group. Sanatorio Mater Dei, Buenos Aires, Argentina; Hospital Marie Curie, Buenos Aires, Argentina; Laboratorio Quantum, Rosario, Santa F
, Argentina; Productos Roche, Tigre, Buenos Aires, Argentina
Background: HER2 over-expression was found to be an independent overall survival and time to relapse predictor; furthermore, correlates with poor prognostic features variables. HER2 National Program enabled access to HER2 testing country-wide with standardized and replicate technique. Internal as well as independent and external quality control has assured accurate results. Aim: To correlate HER2 over-expression in our program population with age and classic histological features, such as: histological type and grade, tumor size and hormone receptors status.
Design: 1702 H&E-stained, formalin fixed paraffin embedded invasive breast carcinoma tissue samples with required criteria were collected. Routine histological parameters were assessed according to WHO Tumor classification. HER2 analysis was performed using policlonal antibody anti Her 2 (DAKO), microwave antigenic recovery, detection system EnVision (Dako) and developed with diaminobenzidine. Results were interpreted as hercepTest
guideline's. ER and PR were screened by IHC analyses and interpreted as positive when more than 10% of tumor cells showed positive nuclear staining.
Results: HER2 over-expression in all samples (1702): Score 0-1+:1450 (85,19 %), 2+: 74 (4,34 %), 3+: 178 (10,45 %). Age and histological features in HER2 - (0-1) versus HER2 + (3) patients: (Table 1)
| HER2-(n1450) | HER2+(n178) | p value(univariate analysis) | |
| Age( median-range) | 55 (25-97) | 55 (27-88) | 0,365 |
| T size cm.(median) | 1,8 cm | 2,00 cm | 0,349 |
| ER+ (%) | 1052 (73) | 105 (59) | <0,001 |
| PR+ (%) | 968 (67) | 105 (59) | 0,039 |
| GH3 (%) | 671 (46) | 115(65) | <0,001 |
| GN3 (%) | 518 (36) | 120 (67) | <0,001 |
| GM3 (%) | 133 (9) | 29 (16) | <0,001 |
With univariate analysis: histological, nuclear and high mitotic rate were statistically linked with HER2+ tumors and ER+ were more frequent in HER 2 - tumors. Using logistic regression for multivariate analysis high histological and nuclear grade and less percentage of ER + were independent predictors for HER2 + disease.
Conclusions: HER2 over-expression was positively associated with high histological and nuclear grade and less ER+ in our population in a logistic regression model.
Category: Breast
Tuesday, March 10, 2009 9:30 AM
Poster Session III # 48, Tuesday Morning