Dr. Liu and colleagues from the University of Michigan, Stanford University and Oncomed published the work. The researchers have identified that a population of cancer cells characterized by CD44 expression but low or undetectable levels of CD24 have a high tumorigenic capacity when injected into immunodeficient mice. The clinical translation of this finding is that tumors bearing these cells have a high potential to invade and metastasize. This observation was applied to human tumors and correlated with prognostic and outcomes variables. A total of 186 genes that are differentially expressed in tumorigenic breast cancer cells generated the gene signature, which was substantially different from previously reported gene signatures. As it represented genes involved in invasion and metastasis, it was called the invasiveness gene signature (IGS). The IGS was made by isolation of tumorigenic breast cancer cells and normal breast epithelium. RNA amplification, microarray analysis, real-time polymerase chain reaction, normalization of microarray data was used to produce the IGS. The IGS was compared to other reported gene signatures and was clearly different. A Pearson correlation coefficient was used for the correlation between the expression data for each patient and the average expression of the IGS. Patients were grouped according to correlation values. The methods were also applied to a cohort of patients with medulloblastoma, lung cancer and prostate cancer. A positive correlation was associated with reduced metastasis-free survival and reduced overall survival in breast cancer patients (the univariate hazard ratio for metastasis or death was 1.3). The estimated 10-year rates of overall survival and metastasis-free survival were 98% and 82%, respectively, among the group with a low correlation coefficient (good prognosis), and 62% and 54%, respectively in the group with a higher correlation coefficient (poor prognosis). The IGS was also applied to a data set from Erasmus Medical Center for 109 of the 186 genes in the IGS. This showed that the risk of metastasis was higher in the group with a higher vs. lower correlation coefficient, and relapse rates of 43% and 16% were found in the two groups, respectively. Regarding prostate cancer, 21 patients were analyzed. The overall survival or relapse-free survival was higher among patients with a tumor that was negatively correlated with the IGS (>80% at 5 years) vs. less than 50% in those whose tumors correlated closely to the IGS. This work suggests that an IGS can be applied to human tumors to stratify risk of metastasis and death. Clinical application, for example would be to then put high-risk patients in clinical trials for adjuvant therapy. Rui Liu, Ph.D., Xinhao Wang, Ph.D., Grace Y. Chen, M.D., Ph.D., Piero Dalerba, M.D., Austin Gurney, Ph.D., Timothy Hoey, Ph.D., Gavin Sherlock, Ph.D., John Lewicki, Ph.D., Kerby Shedden, Ph.D., and Michael F. Clarke, M.D.

N Engl J Med 2007;356(3) :217-26.

read other UroToday.com Editors Picks

UroToday.com Prostate Cancer Section

Written by Christopher P. Evans, MD, a Contributing Editor with UroToday.

Please log-in or register in order to submit comments.

Powered by AkoComment!