Background: Metastatic breast cancer is incurable, and novel treatment (tx) strategies are needed. Single agent immune checkpoint inhibitor (ICI) tx has limited efficacy; combination approaches may yield better results. Multiple mechanisms of glucocorticoid-induced immunosuppression have been proposed, including the suppression of the cytotoxic Th1 cytokine response via glucocorticoid receptor (GR) activation. We hypothesize that pretreatment with a GR antagonist shifts the cytotoxic immune response toward Th1, thus enhancing response to ICIs. We present the safety and efficacy of the combination of pembrolizumab and mifepristone in advanced HER2-negative breast cancer. Methods: This (NCT03225547) was a phase II non-randomized study of pembrolizumab and mifepristone in patients (pts) with advanced HER2-negative breast cancer. Pts with prior ICI tx were excluded. Pts received mifepristone 300mg daily and pembrolizumab 200mg every 3 weeks; mifepristone was started 7 days prior to pembrolizumab. The primary endpoint was objective response rate (ORR). Secondary endpoints included progression-free survival (PFS), overall survival (OS), disease control rate (DCR), and safety and tolerability. Results: 18 patients were enrolled from March 2018 – November 2020; 10 patients with TNBC (cohort 1) and 8 patients with HR+ BC (cohort 2). Mean age was 53 yrs (range 33-74) and 33% of patients were self-reported Black. In cohort 1, 50% of patients were tx naive in this advanced setting, and 90% of patients had < 2 prior lines of tx. In cohort 2, 25% of patients had not received prior chemo in the metastatic setting. The overall ORR was 6%. One pt with TNBC had a complete response (CR). She received 36 cycles of tx; tx was discontinued due to mucositis and rash, and she remains in CR at >48 mos from start of tx. No patients in cohort 2 had a partial or complete response. The most common treatment-related adverse events were rash (61%), thyroid abnormality (17%), and pruritis (17%). 45% of patients experienced a grade 3 or 4 adverse event; all except one (hypokalemia) were due to rash. Rash was described as maculopapular dermatitis. The DCR at 18 weeks was 33.3% in the overall population; it was 44.4% in cohort 1 and 16.7% in cohort 2. Overall median PFS was 1.9 months [95% CI 1.8 – 3.6]. Overall median OS was 12.9 months [95% CI 5.8 – 23.0]. Conclusions: While this regimen demonstrated efficacy in a small subset of patients, including one pt with long-term CR, given the non-randomized design, we cannot determine if mifepristone enhanced the efficacy of ICI. In addition, a higher than anticipated rate of skin toxicity, possibly related to mifepristone enhancing the activity of ICI, was observed in the study, leading to early closure. While the benefit/risk analysis of this combination does not support further evaluation, additional investigation of alternative chemotherapy-free, immunomodulatory strategies is warranted. Clinical trial information: NCT03225547.