Genomic Assay May Pinpoint Who Can Skip RT After Breast-Conserving Surgery

An investigational genomic assay may be able to identify women who don't stand to benefit from adjuvant radiation therapy (RT) following breast-conserving surgery, researchers reported.

Using tumor samples from a completed phase III trial in Sweden, investigators found no difference in 10-year locoregional recurrence (LRR) among patients classified as low-risk on the 16-gene signature who were treated with or without adjuvant RT (5% vs 6%; HR 1.1, 95% 0.39-3.4), according to Per Karlsson, MD, PhD, of the University of Gothenburg in Sweden, and colleagues.

When the gene signature, dubbed POLAR (Profile for the Omission of Local Adjuvant Radiation), was evaluated in a group of patients from a second randomized trial, those categorized as low-risk had a 10-year LRR of 13% with RT and 7% without it (HR 1.5, 95% CI 0.14-16), the group detailed in the Journal of Clinical Oncology.

While these results should be interpreted "with caution," due to the low number of patients who were categorized as low risk in both cohorts, this appears to be the first molecular signature found to identify low-risk patients who may be candidates for omission of RT after breast-conserving surgery for invasive breast cancer, noted Karlsson and colleagues.

Their rationale for developing POLAR was that while RT following breast-conserving surgery for invasive breast cancer is effective in reducing LRR, most women will not experience a LRR even without RT, and "tools to guide selection of patients who can safely omit RT are lacking."

"POLAR has the potential to identify patients whose disease is locally indolent without benefit from adjuvant RT," Karlsson and his colleagues concluded. "When considering the cost and morbidity of such therapy, this signature has the potential to improve patient outcomes, emotional well-being, and overall quality of life without compromising cancer control."

SweBCG91-RT was a randomized trial conducted in Sweden that assigned 1,178 patients with node-negative, stage I-IIA breast cancer to either RT or no RT following breast-conserving surgery. This analysis included 597 samples from patients with estrogen receptor (ER)-positive, HER2-negative tumors, who had not been treated with systemic adjuvant therapy and had gene expression data and LRR information available. Of these 597 patients, 243 were split into a training cohort, and the remainder into a validation cohort.

The second cohort of patients came from a Canadian randomized trial that assigned 769 women 50 years or older with T1 or T2 node-negative breast cancer to adjuvant whole-breast RT or no RT following breast-conserving surgery. All of these patients were treated with tamoxifen. The current analysis included 132 patients who had ER-positive, HER2-negative tumors, complete LRR information, and gene expression data available, who were used as a validation cohort.

In addition to the analysis of low-risk patients, the authors determined that among those patients categorized as high-risk by POLAR, adjuvant RT resulted in a significantly lower risk of LRR at 10 years in both the Swedish cohort (8% vs 19% without RT; HR 0.43, 95% CI 0.24-0.78) and the Canadian cohort (8% vs 22%; HR 0.25, 95% CI 0.07-0.92).

Karlsson and his colleagues noted that further validation of POLAR in prospective trials "would further strengthen these findings." Additionally, they pointed out that patients in the two trials analyzed in this study were enrolled and treated in the 1990s when axillary dissection was routinely used rather than sentinel-node biopsy.

"Thus, the LRR rates may be expected to be lower in contemporary cohorts," they observed. "That said, this makes the need to identify current patients in whom RT can be safely omitted even more compelling."

In an accompanying editorial, Mylin Torres, MD, of Emory University in Atlanta, noted that while Karlsson and his colleagues call POLAR a signature of radiation omission, another potential use is to identify patients who can forego both adjuvant radiation and hormone therapy.

She pointed out that none of the SweBCG91-RT patients in this study received hormone therapy. Thus, Torres suggested, "one could also argue that this signature appears to identify a group of individuals who do not receive appreciable local control benefit from adjuvant hormone therapy either if classified as low risk."

Torres agreed that further validation of POLAR in a prospective trial in the modern treatment era is necessary. However, she added, "POLAR is an important step in the development of genomic assays to assist clinicians with radiation and potentially hormone therapy decisions particularly when LRR is the primary concern."

Comment