Evaluation of the predictive value of Her-2/neu overexpression and p53 mutations in high-risk primary breast cancer patients treated with high-dose chemotherapy and autologous stem-cell transplantation
Y. Nieto et al., Evaluation of the predictive value of Her-2/neu overexpression and p53 mutations in high-risk primary breast cancer patients treated with high-dose chemotherapy and autologous stem-cell transplantation, J CL ONCOL, 18(10), 2000, pp. 2070-2080
Purpose: To ascertain the predictive value of Her-2/ neu overexpression and
p53 mutations, assessed by immunohistochemistry, in high-risk primary brea
st cancer (HRPBC) treated with high-dose chemotherapy (HDCT).
Patients and Methods: We obtained paraffin-embedded tumor blocks from 146 H
RPBC patients previously enrolled at our program onto clinical trials of HD
CT for four to nine involved axillary lymph nodes, greater than or equal to
10 involved axillary nodes, or inflammatory carcinoma. All patients receiv
ed the same HDCT regimen, with cyclophosphamide, cisplatin, and carmustine
(STAMP-I), followed by autologous stem-cell transplantation, Median follow-
up was 42 months (range, 5 to 90 months). The same pathologist, blinded to
clinical outcome, reviewed all immunostained slides.
Results: Positive results for Her-2/neu and p53 were found in 44.5% and 34%
of the patients, respectively, Positivity for Her-2/neu was significantly
associated with increased risk of relapse and death, No correlation was fou
nd between p53 mutations and relapse-free survival (RFS) or overall surviva
l (OS), Multivariate analyses included Her-2/neu overexpression and the fol
lowing variables previously identified as independent predictors of outcome
in this population: tumor size, nodal ratio (number of involved nodes/numb
er of dissected nodes), and hormone receptor status, All four variables had
independent valve.
Conclusion: Her-2/neu overexpression is an independent negative predictor o
f RFS and OS in HRPBC treated with HDCT. Its inclusion in our previously de
scribed predictive model increases the predictive capacity of this model fo
r the low-risk subgroup. In contrast, p53 mutations lack predictive value i
n this setting, J Clin Oncol 18:2070-2080, (C) 2000 by American Society of
Clinical Oncology.