SAN ANTONIO – Maintenance therapy with the immune checkpoint inhibitor durvalumab (Imfinzi) did not improve outcomes compared with chemotherapy for patients with metastatic breast cancer, although there was a trend toward benefit for patients with triple-negative breast cancer and those whose tumors were positive for programmed death ligand 1 (PD-L1), a substudy of SAFIR-02 Breast
Median progression-free survival for 131 patients with metastatic or locally advanced, HER2-negative, hormone receptor–negative or endocrine-resistant breast cancer who received durvalumab maintenance following chemotherapy in thewas 2.7 months, compared with 4.4 months for 68 patients who received maintenance chemotherapy, reported Florence Dalenc, MD, of Institut Claudius Regaud, Toulouse, France.
“For hormone receptor–positive metastatic breast cancer, exploratory analyses suggest that anti-PD-L1 as a single agent is less effective than chemotherapy as maintenance,” she said at the San Antonio Breast Cancer Symposium.
The rationale for investigating an anti-PD-L1 agent for maintenance in patients with metastatic breast cancer is that the tumor is (presumably) not growing when immunotherapy is introduced and disease burden, a mechanism of immune suppression, has been reduced by chemotherapy. In addition, chemotherapy attracts tumor infiltrating lymphocytes (TILs) to the tumor bed, and response to chemotherapy may be a marker for immunogenic tumors, she said.
Durvalumab is a human immunoglobulin G1 kappa monoclonal antibody that blocks the interaction of the programmed death protein 1 with PD-L1. In the, the addition of durvalumab to anthracycline-/taxane-based neoadjuvant chemotherapy increased the pathologic complete response (pCR) rate particularly in patients treated with durvalumab alone before the start of chemotherapy.
To see whether durvalumab could offer a benefit over chemotherapy in the maintenance setting, Dr. Dalenc and colleagues studied patients who were enrolled in SAFIR-02 Breast, which was designed to assess the efficacy of genomic analysis as a therapeutic decision tool in patients with metastatic breast cancer.
In the trial, patients with HER2-negative metastatic breast cancer who have a complete or partial response or stable disease following six to eight chemotherapy cycles (or 4 cycles if chemotherapy was stopped for toxicities) are randomized based on the presence or absence of targetable molecular alterations. Patients with identified alteration are then randomized to either targeted therapy matched to genomics, or to maintenance chemotherapy.
Patients without molecular alterations – the population included in the SAFIR02-IMMUNO substudy reported by Dr. Dalenc and colleagues – are stratified by one or two prior lines of chemotherapy and by response type, and then stratified on a 2:1 basis to maintenance with either durvalumab 10 mg/kg every 14 days or standard maintenance chemotherapy for up to 2 years.
In each arm, the median patient age was 56, and approximately 58% of patients had Eastern Cooperative Oncology Group (ECOG) performance status of 0. Approximately 43% of patients in each arm had three or more metastatic sites, and roughly half had liver metastases, while slightly more than one-fourth had metastases to lung.
As noted before, median PFS in the overall substudy population, the primary endpoint, was 4.6 months for patients on maintenance chemotherapy, compared with 2.7 months with durvalumab. This difference translated into an adjusted hazard ratio for progression with durvalumab of 1.40 (P = .047).
Also as noted, there were nonsignificant trends toward benefit with durvalumab among patients with triple-negative breast cancer or PD-L1-positive tumors. For all other subgroups, however, maintenance chemotherapy was favored, with significant benefit seen among patients with disease that progressed after second-line chemotherapy, patients with a clinical response at the time of re-randomization, those 50 years and older, patients with non-triple-negative disease, patients with liver metastases, and those whose tumors were negative for PD-L1.
Median overall survival (OS) was numerically longer with durvalumab (21.7 vs. 17.9 months), but this difference was not statistically significant.
In exploratory analyses, median OS for patients with triple-negative disease was 21 vs. 14 months, respectively, with an unadjusted hazard ratio favoring durvalumab of 0.54 (P = .0377). Among patients with PD-L1 positive tumors, the median OS was 26 months with durvalumab vs. 12 months with chemotherapy, with an unadjusted hazard ratio of 0.42, which just missed statistical significance (P = .0552).
Dr. Dalenc said that the study generates the hypothesis that single-agent durvalumab could improve outcomes in metastatic triple-negative breast cancer, but also raises questions about which disease parameters limit the response to anti-PD-L1 therapy.
The study was sponsored by UNICANCER with support from Fondation ARC and AstraZeneca. Dr. Dalenc reported research grants, travel, and/or advisory board activity for AstraZeneca and others.
SOURCE: Dalenc F et al. SABCS 2019. .