A gene-expression signature as a predictor of survival in breast cancer

Area of Science:

• Genomics
• Oncology
• Biomarkers

Background:

• Accurate breast cancer prognostication is crucial for optimizing adjuvant systemic therapy selection.
• Current prognostic systems rely on clinical and histologic criteria, which may lack precision for certain patient groups.

Purpose of the Study:

• To evaluate a previously established 70-gene prognosis profile for its predictive power in early-stage breast cancer.
• To compare the efficacy of this gene-expression signature against standard prognostic criteria.

Main Methods:

• Microarray analysis was used to classify 295 primary breast carcinoma patients (Stage I/II, <53 years) into good or poor prognosis groups based on a 70-gene signature.
• Univariable and multivariable statistical analyses, including Cox regression, were employed to assess the profile's predictive value.

Main Results:

• The 70-gene profile identified 180 patients with a poor prognosis and 115 with a good prognosis.
• Ten-year overall survival rates were 54.6% for the poor-prognosis group and 94.5% for the good-prognosis group.
• The hazard ratio for distant metastases was significantly higher in the poor-prognosis group (5.1), indicating its strong predictive capability independent of lymph node status.

Conclusions:

• The studied 70-gene expression profile is a more powerful predictor of disease outcome in young breast cancer patients.
• This genomic approach offers superior prognostication compared to conventional clinical and histologic methods.