SAN ANTONIO – The immunologic checkpoint inhibitor avelumab showed only modest single-digit efficacy in an unselected group of metastatic breast cancer patients, but the results were sevenfold better in the subset of JAVELIN trial participants with strong expression of PD-L1 by immune cells within the tumor, Dr. Luc Y. Dirix reported at the San Antonio Breast Cancer Symposium.
Avelumab is an investigational fully human IgG1 monoclonal antibody that selectively binds to PD-L1 (programmed death–ligand 1). In addition, it’s believed that avelumab elicits antibody-dependent cellular cytotoxicity as a secondary mechanism for its antitumor activity, according to Dr. Dirix of the University of Antwerp, Belgium.
JAVELIN is a multipronged phase Ib clinical trial which to date has enrolled more than 1,000 participants with various types of advanced cancer. Avelumab has shown antitumor activity in patients with lung, gastric, bladder, ovarian, and other cancers. Phase III randomized trials are underway evaluating avelumab in patients with advanced non–small cell lung cancer and gastric cancer.
Dr. Dirix reported on the 168 JAVELIN participants with locally advanced or metastatic breast cancer refractory to standard therapy, including anthracycline and a taxane. Their median time since diagnosis of metastatic disease was 21.6 weeks at the time they went on avelumab at 10 mg/kg every 2 weeks until disease progression occurred. Roughly half of participants had already undergone three or more regimens for their advanced malignancy.
At a median duration of 10 months follow-up, 1 of the 168 patients had shown a complete response and 7 others had a partial response, for an overall response rate of 4.8%. Of note, 5 of the 8 responders were among the 58 women with triple-negative breast cancer (TNBC). Median time to response was 11.4 weeks, with a relatively long 28.7-week median duration of response. The response was ongoing in 5 of 8 patients at the time of Dr. Dirix’s presentation.
Of the 136 patients for whom data on PD-L1 expression level was available, 4 of the 12 with PD-L1 expression by at least 10% of immune cells within the tumor had a clinical response, for a 33% rate. Nine patients with TNBC had a PD-L1 response which rose to this level, and 4 of these 9 (44%) had a clinical response.
In contrast, patients whose immune cells within the tumor were PD-L1–negative had a clinical response rate of less than 3%. But expression of PD-L1 by tumor cells was not predictive of benefit from avelumab; indeed, even when 25% or more of a patient’s tumor cells expressed PD-L1, the clinical response rate was in the single digits.
Ten of 58 patients with TNBC (17%) experienced tumor shrinkage by 30% or more during the course of treatment.
Dr. Dirix characterized avelumab’s safety profile as acceptable for patients with metastatic breast cancer. Only 8 patients (4.8%) discontinued participation because of treatment-related adverse events. Potentially treatment-related immune toxicity occurred in 8 patients who became hypothyroid, 3 with autoimmune hepatitis, 3 with pneumonia, and 2 with thrombocytopenia. Three-quarters of these complications were Grade 1/2.
Far more common treatment-related adverse events included Grade 1/2 fatigue, which occurred in 19% of patients, nausea in 13%, and diarrhea in 9%. Infusion reactions occurred in 14% of patients; however, the incidence was halved after mandatory pretreatment with an antihistamine and antipyretic was instituted.
Session moderator Dr. Nicholas Turner of Royal Marsden Hospital in London observed that in lung cancer and other malignancies included in the JAVELIN program, PD-L1 expression by tumor cells predicted benefit with avelumab. “Why not in metastatic breast cancer?” he asked.
Dr. Dirix replied that he and his coinvestigators are looking into that question but don’t have an answer yet.