A team of researchers led by Jonas Bergh from the Karolinska Institutet in Stockholm, Sweden, analysed the gene expression profiles of 159 breast cancer patients using DNA microarray analysis. From these samples they identified the genetic signatures shown by 38 patients who had a poor prognosis - defined as relapse or death from any cause within 5 years. The remaining 121 patients were defined as the 'good prognosis' group. The researchers also used gene expression profiling to separate patients who did well with and without adjuvant therapy, and those whose tumours failed to respond to treatment.
An analysis of the genes expressed in the tumours of all 159 patients showed that 64 genes were used to separate the patients with good and poor prognoses. The researchers then tested the predictive value of the group of 64 genes compared with three currently used clinical markers. Using the expression patterns of the 64 genes identified by the researchers gave significantly better (P=0.007) prediction rates than histological grading, tumour stage and age - which are all accepted prognostic markers for breast cancer.
The present lack of criteria to help tailor breast cancer treatment to individual patients indicates a need to develop new techniques for better prediction of how patients wi
Contact: Prof Per Hall