San Antonio – as well as those who can be safely spared from breast radiation following breast-conserving surgery, an international team of investigators said.
In combined data from three independent randomized trials grouped into a meta-analysis, patients who had low scores on the messenger RNA–based signature, dubbed “Profile for the Omission of Local Adjuvant Radiotherapy” (POLAR), derived only minimal benefit from radiotherapy following breast-conserving surgery. In contrast, patients with high POLAR scores had significant clinical benefit from adjuvant radiotherapy, reported Per Karlsson, MD, chief physician with the Sahlgrenska Comprehensive Cancer Center and the University of Gothenburg (Sweden). Dr. Karlsson reported his findings at the San Antonio Breast Cancer Symposium.
Dr. Per Karlsson
“To our knowledge, POLAR is the first genomic classifier that is not only prognostic but also predictive of radiotherapy benefit, showing a significant interaction between radiotherapy and the classifier,” he said. “These important retrospective findings warrant further investigation, including in contemporary clinical studies.”
Investigators with the Swedish SweBCG91RT trial (Swedish Breast Cancer Group 91 Radiotherapy), the Scottish Conservation (radiotherapy) Trial (SCT), and a trial from the Princess Margaret Cancer Hospital in Toronto, collaborated on improving and validating the POLAR signature, which was originally developed for use in the SweBCG91RT trial in patients with lymph node–negative breast cancer who underwent breast-conserving surgery. The patients were randomized to whole breast irradiation or no radiotherapy.
To develop the signature, researchers collected tumor blocks from 1,004 patients, and extracted RNA from the samples. Gene expression data were obtained from primary tumors of 764 patients. The subset of 597 patients with estrogen receptor–positive, HER2-negative tumors (ER+/HER2–) who did not receive systemic therapy were divided into a training set with 243 patients, and a validation cohort with 354 patients.
They identified a total of 16 genes involved in cellular proliferation and immune response, and then validated the signature using retrospective data from three clinical trials of patients randomized to radiotherapy or no radiation following breast-conserving surgery.
Of 623 patients with node-negative ER+/HER2– tumors who were included in the meta-analysis, 429 patients were found to have high POLAR scores. These patients benefited from adjuvant radiation therapy after breast-conserving surgery with a 10-year cumulative incidence of low risk of locoregional recurrence ranging from 15% to 26% for those who were not treated with radiation therapy, compared with only 4%-11% percent for those who received radiation therapy (hazard ratio, 0.37; P < .001).
In contrast, among the 194 patients whose tumors had POLAR low scores, there was no apparent benefit from radiation therapy with a nonsignificant HR of 0.92 (P = .832).
In Cox proportional hazard models for time to locoregional recurrences for 309 patients who did not undergo radiation, POLAR scores were significantly prognostic for recurrence, with a HR of 1.53 (P < .001) in univariable analysis, and 1.43 (P = .005) in multivariable analysis controlling for age, tumor size, tumor grade and molecular groupings.
New modalities may make findings less relevant
Alphonse Taghian, MD, PhD, a breast radiation oncologist with Mass General Cancer Center, Boston, who was not involved in the study, said there have been major changes in radiation therapy since the studies used for development of the POLAR signature were performed. For example, the Scottish Conservation Trial ran from 1985 to 1991, while the SweBCGR91RT trial and Princess Margaret trial were both conducted in the 1990s.
He noted that patients in those studies would likely experience more morbidities from radiation than patients treated with more recent modalities such as intensity modulated radiation therapy, and that patients treated 30 years ago would have to put up with lengthy fractionation schedules that required daily trips to the hospital over as long as 6 weeks, whereas a majority of patients can now be treated with hypofractionated radiation that can be performed in a much shorter time and with minimal comorbidities.
He acknowledged, however, that “it will help to have a signature proved, confirmed, or validated retrospectively with a different set of data.”
Dr. Taghian also said that it would be helpful to have more data about the age of patients, because omitting radiation is more common for elderly patients than it is for younger patients.
“It will maybe be beneficial to look at this signature in patients that we think might not need radiation,” he said.
The study was supported by the Swedish Cancer Society, Swedish Research Council, King Gustav 5 Jubilee Clinic Foundation, the ALF Agreement of the Swedish government, PFS Genomics, and Exact Sciences. Dr. Karlsson has pending patents with and receives royalties from Exact Sciences and PreludeDX. Dr. Taghian reported having no relevant disclosures.