At a median follow-up of almost 3 years, rates of disease-free survival, distant metastasis-free survival, and overall survival were similar in the everolimus and hormone therapy-alone arms.
These findings were presented at the inaugural ESMO Virtual Plenary and published in Annals of Oncology.
The UNIRAD results contrast results from prior studies of everolimus in the advanced breast cancer setting. In the BOLERO-2 and BOLERO-4 studies, the mTOR inhibitor provided a progression-free survival benefit when added to hormone therapy.
“There clearly is rationale for targeted therapy in early ER+, HER2- breast cancer,” said Rebecca Dent, MD, of the National Cancer Center in Singapore, who chaired the ESMO Virtual Plenary in which the UNIRAD findings were presented.
“Patients with high-risk luminal breast cancer clearly have an unmet need. We probably still underestimate the risk of early and late recurrences, and chemotherapy is not necessarily the answer,” Dr. Dent said.
She observed that a lot has been learned about the mTOR pathway, including how complicated it is and its role in endocrine resistance. Since mTOR inhibition was standard care in the metastatic setting, “it really is appropriate now to test in early breast cancer,” she added.
Study details
The aim of the UNIRAD study was to compare the efficacy and safety of everolimus plus standard adjuvant hormone therapy to hormone therapy alone in women with ER+, HER2- early breast cancer who had a high risk of recurrence. High risk was defined as having more than four positive nodes, having one or more positive nodes after neoadjuvant chemotherapy or hormone therapy, or having one or more positive nodes and an EPclin score of 3.3 or higher.
The trial enrolled 1,278 patients. At baseline, their median age was 54 years (range, 48-63), 65.8% were postmenopausal, and 52.7% had four or more positive nodes.
The patients were randomized 1:1 to receive 2 years of everolimus plus hormone therapy or placebo plus hormone therapy. The type of hormone therapy was investigor’s choice.
Investigator Thomas Bachelot, MD, PhD, of Centre Leon Berard in Lyon, France, noted that the study started in 2013 and underwent several protocol amendments, first for accrual problems and then because of toxicity. This led to dropping the starting dose of everolimus from 10 mg to 5 mg.
“Acceptability was a concern; 50% of our patients stopped everolimus before study completion for toxicity or personal decision,” Dr. Bachelot acknowledged.
Grade 3 or higher adverse events were more frequent in patients taking everolimus (29.9%) than placebo (15.9%) in combination with hormone therapy. The rates of serious adverse events were a respective 11.8% and 9.3%.
Mucositis was one of the main adverse events, occurring in more than half of all patients treated with everolimus (33.8% grade 1, 25.4% grade 2, and 7.4% grade 3/4). The success of managing this side effect with a dexamethasone mouthwash was not known at the time of the UNIRAD trial design.
The study also showed no benefit of everolimus over placebo for the following efficacy outcomes:
- Disease-free survival – 88% and 89%, respectively (hazard ratio, 0.95; 95% confidence interval, 0.69-1.32; P = .78)
- Distant metastasis-free survival – 91% and 90%, respectively (HR, 0.88; 95% CI, 0.62-1.25)
- Overall survival – both 96% (HR, 1.09; 95% CI, 0.62-1.92).
With the exception of patients who had received tamoxifen rather than an aromatase inhibitor, a preplanned subgroup analysis suggested there was no population of patients who benefited from the addition of everolimus.
Problems interpreting data
There are several problems in interpreting the UNIRAD data, observed study discussant Peter Schmid, MD, PhD, of St. Bartholomew’s Hospital & Barts Cancer Institute in London.
For one, “whether we like it or not,” the trial was underpowered, he said. This was because the trial had been halted early for futility at the first interim analysis when about two-thirds of the intended study cohort had been accrued.
In addition, Dr. Schmid said, this is clearly not a trial that included patients with primary endocrine resistance. In all, 43% of patients had received less than 1 year of endocrine treatment, 42% had received 2-3 years, and 15% had received more than 3 years of endocrine treatment.
Dr. Schmid said that the starting dose of everolimus had to be lowered because of toxicity or poor acceptance. “As a result, two-thirds of patients received a 5-mg dose, and we don’t know whether that had an impact on efficacy,” he said.
Furthermore, the median time on treatment was less than half of what was initially planned, and 53% of patients had to stop everolimus before the end of the study.
Dr. Schmid noted that this discontinuation rate is higher than that seen in trials of CDK4/6 inhibitors added to endocrine treatment. Dropout rates were 19% in the negative Penelope-B trial with palbociclib, 42% in the negative PALLAS trial with palbociclib, and 27% in the positive monarchE trial with abemaciclib.
With such a high discontinuation rate in the UNIRAD trial, “we’re not sure whether we can ultimately evaluate really whether this trial did work,” Dr. Schmid said.
“Could the results change over time?” he asked. “I personally think it is unlikely, as the trial clearly has already an adequate follow-up for what it was supposed to show.”
Looking at whether the trial’s hypothesis is still valid, he added: “I think that is unclear to all of us, and we need to work out whether these compounds are cytostatic in nature, or cytotoxic. And that is something we need to learn over time.”
Commending the study overall, Dr. Schmid observed: “I think it was an excellent trial design based on what we knew at that time,” and the design “was changed in a pragmatic way because of recruitment challenges.”
Something for the future would be to select patients for such trials based on the tumor biology rather than risk status, Dr. Schmid suggested. “That is something we may have to take into consideration with our increasing knowledge around primary and secondary resistance and what treatments we want to introduce to target-resistant clones,” he said.
The UNIRAD study was sponsored by UNICANCER, with funding and support from the French Ministry of Health, Cancer Research UK, Myriad Genetics (which provided Endopredict tests), and Novartis (which provided everolimus and placebo). Dr. Bachelot, Dr. Schmid, and Dr. Dent disclosed relationships with Novartis and several other pharmaceutical companies.