In previous research, scientists at the Netherlands Cancer Institute in Amsterdam identified 231 genes associated with patient outcomes in a group of patients 55 years or younger with breast tumors under 5 cm. The 231 genes were reduced to a core group of 70 that make up a genetic signature to predict outcomes in a separate group of patients from the same institution.
To determine whether the 70-gene signature performed better than traditional risk classifiers, scientists from the international consortium TRANSBIG assigned 307 patients to high- and low-risk groups based on scores from the 70-gene signature and on traditional risk assessment with the Adjuvant Online! software. The patients were followed for 13.6 years and assessed for disease recurrence and death.
The authors found that the 70-gene signature was a more accurate predictor of disease outcomes than assessment with the Adjuvant Online! software. Moreover, statistical analysis indicated that most of the prognostic information provided by traditional risk classifiers was included in the gene classifier.
The authors write, "These results indicate that the gene signature adds independent prognostic information to that provided by a risk assessment based solely on clinico-pathologic factors."
In an accompanying editorial, Richard Simon, D.Sc., of the National Cancer Institute in Rockville, Md., discusses the strengths and limitations of the study by Buyse et al. He suggests improvements that might have been made in the study's design, and writes that the research, "Illustrate[s] many desirable features of gene expression profiling studies for optimizing treatment selection for individual patients."
The 70-gene signature will be pros
Source:Journal of the National Cancer Institute