A gene signature identified using a new approach has the potential to be used in the clinic to predict which patients with estrogen receptor-positive breast cancer will benefit from tamoxifen therapy after surgery, according to data published in Cancer Research, a journal of the American Association for Cancer Research.

"We have used a very innovative approach to identify genes that help foretell whether a patient will respond to tamoxifen, and we showed that this gene signature performed well in two large patient groups," said René Bernards, PhD, professor and head of the Division of Molecular Carcinogenesis at the Netherlands Cancer Institute in Amsterdam.

"About one-third of women with hormone receptor-positive breast cancer experience a relapse after adjuvant treatment with tamoxifen. Median overall survival in these patients, even with further treatment, is around 30 to 45 months," explained Bernards. "It has been very difficult to identify patients whose tumors lack a proper response to tamoxifen, the most frequently used drug in breast cancer.

"By using CHIPseq and RNAseq technologies in combination with RNA expression data from publically available data sets, we were able to translate results from cell culture experiments into a diagnostic tool that has the potential to help identify breast cancer patients that will benefit from tamoxifen," added Bernards.

To find out whether loss of function of certain genes is involved in resistance to tamoxifen, Bernards and colleagues took human hormone receptor-positive breast cancer cells and infected them with thousands of small pieces of RNA called shRNAs, each designed to silence a specific gene. By studying the shRNAs that survived despite treatment with tamoxifen, they identified a gene called USP9X, whose loss of function in breast cancer cells resulted in tamoxifen resistance.

Next, the researchers identified other genes that were altered during treatment with tamoxifen, in the absence of USP9X. Then, using publicly available data sets, they studied the expression of these specific genes in patients who were treated with tamoxifen after surgery, and whose treatment outcomes were known. They found that the data split into two groups, one with a gene signature with a good outcome, and the other with a signature showing bad outcome, after treatment with tamoxifen.

The gene signature they identified could not predict treatment outcomes in patients who did not receive tamoxifen therapy, suggesting that this signature is specific to outcomes with tamoxifen treatment.

Collectively, the researchers used data from about 680 patients from four different data sets to test the utility of the gene signature in predicting responses to tamoxifen therapy in hormone receptor-positive breast cancer patients.

"We are currently validating our promising results using the data from a prospective randomized controlled trial, and we expect to complete this validation by the end of this year," said Bernards. "If this is successful, clinical implementation is a logical next step."

This study was funded by grants from the Dutch Cancer Society. Bernards declares no conflicts of interest.