Deciphering the Challenge of the Genomic Era in Breast Cancer

Daniel F Hayes, MD

“Luminal A genetic signature in node-negative, estrogen receptor- (ER-) positive breast cancer predicts a very high disease-free survival rate of more than 95% at 8.5 years with adjuvant hormonal therapy alone, and thus, in general, should not be offered adjuvant chemotherapy” said Daniel F. Hayes, MD, Stuart B. Padnos Professor of Breast Cancer Research at University of Michigan at this year’s Annual Meeting of the American Society of Clinical Oncology.

Hayes, who is also a co-director of the Breast Care Center and the clinical director of the Breast Cancer Program at University of Michigan, was discussing research presented by Aleix Prat, MD from University of North Carolina.

Hayes added that among the various gene signatures available, most are similar in identifying either the low-risk or the high-risk breast cancers. However, some differences among various gene signatures are seen in predicting breast cancers with intermediate risk for relapse.

Prat, the lead investigator of the research paper, stated that among the ER+ breast cancer, all of the different intrinsic molecular subtypes are represented, although the luminal subtypes predominate. His group evaluated the prognostic ability and biologic significance of 6 independent gene-expression signatures, including Oncotype DX, Mamma print, PAM50, Rotterdam 76-gene signature, endocrine sensitivity index (SET), and the estrogen-induced gene set. Each signature was evaluated as a continuous variable, as well as grouped according to respective published literature. A total of 1417 patients, comprising 686 node-negative and 731 node-positive, were identified and contributed to the study. All patients were evaluated with a combination of gene signature, and had early stage ER+ tumors treated only with adjuvant tamoxifen. As expected, approximately 85% of patients had the luminal subtypes, with 46% being luminal A and 39% being luminal B.

As per Prat, most signatures discriminated the relatively good prognosis Luminal A tumors from the poorer prognosis Luminal B or the HER2-enriched as well as the basal-like tumors. All the signatures were prognostic in all patients as well as in patients with node-negative or node-positive disease. When the signatures were combined together, there was a small increase in prognostic capacity for node-negative disease, while the increase in prognostic capacity for node-positive disease with the combination of various signatures was minimal.

Prat concluded that despite tracking similar biology, a group of patients with very high disease-free survival of more than 95% was identified only in node-negative disease, thus suggesting that clinical features are still important in addition to the molecular features. He also stated that though different gene signatures are similar, they are not exactly the same, and differences are especially seen in patients predicted to have intermediate risk. It is in this group of patients that combination of gene signatures may provide additional advantage in accurately predicting the outcomes. Prat also added that the performance of gene signatures in node-positive appears minimal, and thus from a clinical perspective, these signatures should specifically aid in identifying patients with node-negative Luminal A breast cancers that might be considered for adjuvant endocrine therapies alone.

Prat A, Parker JS, Fan C, et al. Concordance among gene-expression-based predictors for ER-positive breast cancer treated with adjuvant tamoxifen. J Clin Oncol. 2011. (suppl; abstr 502).