New UH Seidman Radiation Oncologist Leads Team Developing Genomic Signatures to Aid in Breast Cancer Decision-Making

Clinical utility studies of new signatures beginning across the U.S. this fall

Innovations in Cancer | Fall 2022

The advent of personalized medicine in oncology has revolutionized the care of women with breast cancer. Medical oncologists today, for example, can recommend specific endocrine therapy or chemotherapy agents tailored to the specific genomic signature of the tumor. Radiation oncologists, on the other hand, have had relatively fewer genomic tools at their disposal to guide their treatment recommendations for these patients – but now that is changing.

Corey Speers, MD, PhD

Publishing in the Journal of Clinical Oncology, an international research team has now analyzed tumor samples from previous breast cancer clinical trials with patients randomized to receiving or not receiving radiotherapy and has correlated these findings with patient outcomes data. The result is two validated, multi-gene genomic signatures indicating which patients with early-stage invasive breast cancer stand to benefit most from radiotherapy – and which patients can safely omit it entirely.

“Our colleagues in in in the systemic therapy space have been ahead of us for quite some time at developing these kinds of genomically based signatures that predict benefit of treatment,” says radiation oncologist Corey Speers, MD, PhD, Hennessy Hyland Master Clinician in Immunotherapy and New Drug Development at UH Seidman Cancer Center and one of the leaders of this new effort. “But over the past several years, we’ve sought to develop signatures for two early-stage breast cancer patient populations – the 70 percent of women who are going to be cured with surgery alone and don’t need radiation therapy and the 10 percent of women who are unfortunately going to experience a recurrence of their cancer despite radiation. Ten years in the making, we finally have signatures that have been developed and have validated in several prospectively run studies that really were designed to address the role of radiation therapy in women with breast cancer. We’ve validated these signatures in those cohorts that were designed appropriately to answer the question. We are extremely excited that these signatures appear to do exactly what they were designed to do, be not just prognostic of recurrence with and without radiation, but more importantly predict who will respond to radiation and who will not.”

International Effort

For this ambitious project, Dr. Speers and his colleagues across dozens of institutions in the U.S., Canada, England, Scotland and Sweden worked with tumor samples and data from the SweBCG91-RT, Scottish Breast Conservation Trial, and Princess Margaret clinical trials, which randomized women with early-stage, node-negative invasive breast cancer to either receive or not receive adjuvant radiation therapy after breast-conserving surgery. The logistics were daunting.

“We needed to get access to those tissue specimens from those trials from all the different hospitals across the world and get them centrally processed with pathology,” Dr. Speers says. “We needed to get sequencing and RNA expression profiling done in a consistent manner by a CLIA-certified lab. And then we needed to do the validation studies to show that we could reproducibly do that. That was no small whole feat, and it took 10+ years.”

The result of this decade of work is a validated 16-gene genomic signature named POLAR (Profile for the Omission of Local Adjuvant Radiation) to aid in decisions about adjuvant radiotherapy after breast-conserving surgery and another named ARTIC (Adjuvant Radiotherapy Intensification Classifier (ARTIC), comprising 27 genes and patient age, showing patients most likely to benefit from radiation therapy.

Signatures Validated

Analysis shows that patients categorized as POLAR low-risk without radiation therapy had a 10-year locoregional recurrence of 6% and 7% in SweBCG91-RT and Princess Margaret cohorts, respectively, and that there was no significant benefit from radiation therapy in POLAR low-risk patients. On the other hand, patients categorized as POLAR high-risk had a significant decreased risk of locoregional recurrence with radiation therapy.

Results show that ARTIC, too, was highly prognostic in predicting locoregional recurrence in patients treated with radiation therapy. Patients with low ARTIC scores had a large benefit from radiation therapy, with 10-year cumulative incidence of locoregional recurrence at 6%, vs. 21% for those not receiving radiation therapy. At the same time, those patients with high ARTIC scores benefited less from radiation therapy.

“It turns out those these signatures do exactly what we thought they would do,” Dr. Speers says. “They predicted those that were likely to benefit, and we predicted those that we thought had very low risk of recurrence without getting radiation.”

A meta-analysis and validation study is scheduled to be presented as a plenary session at the Annual AACR San Antonio Breast Cancer Symposium this December.

Moving into Clinical Practice

The next step for Dr. Speers and his colleagues is building these genomic signatures into clinical practice. The ARTIC signature, for example, is being used in a randomized, prospective trial in Scandinavia as a means of stratifying breast cancer patients.

“This trial will randomize women to more intensive radiation therapy vs. less intensive radiation therapy based on their ARTIC score,” he says. “This is a several-thousand-patient, randomized trial being run across multiple countries in Scandinavia to look at whether this signature, in a prospective fashion, helps effectively refine radiation treatment decision.”

Clinical utility studies of the genomic signatures are also beginning this fall at between 30 and 40 cancer centers across the U.S.

“We’ll start sending these test score results back to patients along with their chemotherapy benefit scores,” Dr. Speers says. “They'll have information about likelihood of benefit from chemotherapy, but now also from radiation therapy, so medical oncologists and radiation oncologists can use these scores to discuss treatment options and expected benefits with patients.”

Dr. Speers says he’s optimistic about the difference these genomic signatures can make for women with breast cancer who are trying to make an informed decision about radiation therapy.

“We think this information refines what has been a somewhat crude process such that we can more accurately define the group of patients for whom radiation doesn't provide them any benefit and also those who need more intensive therapy – as well as that middle group that that is appropriately getting the benefits of radiation and for whom it reduces the risk of the cancer coming back. Until this, we had nothing that we could use clinically to help patients. This will really be the first test that that is predictive of radiation benefit for women with breast cancer.”

Contributing Expert:
Corey Speers, MD, PhD
Hennessy Hyland Master Clinician in Immunotherapy and New Drug Development
UH Seidman Cancer Center
Vice Chair of Research and Director of the Phase I Clinical Trials Program
Department of Radiation Oncology
Co-director of the Breast Oncology Program
UH Seidman Cancer Center
Associate Professor of Medicine
Case Western Reserve University School of Medicine