A phase II study of nivolumab plus ipilimumab and androgen receptor antagonist bicalutamide to stimulate thymic T cell generation in HER2-negative metastatic breast cancer

Background: Systemic chemotherapy is used commonly in breast cancer and is associated with lymphopenia, potentially limiting efficacy of immune checkpoint blockade. The androgen receptor (AR) serves as a negative regulator of thymic function. AR antagonists, such as bicalutamide, may stimulate thymic production of naïve T-cells, which may help to restore lymphocyte diversity and augment immunotherapy response. The AR is expressed in 50% of hormone receptor negative (HR-) metastatic breast cancer (MBC) and >75% of HR+ MBC, and bicalutamide is clinically active with stable disease as best response. We hypothesize that bicalutamide could be combined with dual immune checkpoint blockade (anti-programmed death 1, nivolumab; plus anti-cytotoxic T-lymphocyte antigen 4, ipilimumab), stimulating T-cell production and resulting in durable clinical response. Methods: This is a phase II Simon 2-stage trial of nivolumab (240 mg IV q2w), ipilimumab (1 mg/kg IV q6w), and bicalutamide (150mg oral daily) for first- or second-line treatment of HER2-negative MBC (NCT03650894). The primary endpoint is iRECIST 24-week clinical benefit rate (CBR), evaluated separately for HR+ MBC (n=15) and AR+/HR- MBC (n=15). An optimum Simon 2-stage design (80% power, 5% one-sided alpha) is employed to evaluate an alternative hypothesis of >50% CBR compared to historical control of < 30% CBR for conventional chemotherapy, with expansion of each cohort planned if ≥6/15 responses are observed. Secondary outcomes include best overall objective response rate (ORR), progression-free and overall survival (PFS,OS), and safety. Exploratory endpoints include assessment of peripheral T-cell counts and T cell receptor excision circles (TREC PCR, Mayo Laboratories, a surrogate of thymic T-cell production) over time. Results: Durable responses were observed, with an ongoing complete response at 41+ months in a patient who discontinued due to toxicity (AR+/HR- cohort, PD-L1 CPS score 3%). Therapy was tolerated with a toxicity profile consistent with prior studies of ipilimumab & nivolumab, with five subjects (19%) requiring treatment discontinuation related to toxicity. Outcomes by cohort and PDL1 are summarized in Table 1. Using mixed effects linear models, therapy was associated with a significant expansion of CD8+ T cells (23/mcL/month, p< .01) but not CD3+ T cells (7/mcL/month, p=.59) or CD4+ T cells (4/mcL/month, p=.65). Baseline TREC counts were higher in younger participants (mean: < 50y 2156/mcL; 50-65y 961/mcL; >65y 311/mcL). TRECs did not increase globally, however TREC expansions were observed in several younger participants. Conclusions: The regimen of ipilimumab, nivolumab, plus bicalutamide is safe and clinically active in 1st/2nd-line HER2-negative MBC. Neither cohort crossed the Simon futility barrier for cohort expansion, however durable responses were observed including in PD-L1-negative patients, highlighting the unmet need for biomarkers to identify candidates for chemotherapy-sparing checkpoint blockade. Further research is indicated to evaluate the impact of AR antagonists on T-cell function and thymic stimulation in MBC. Acknowledgements: Drug and study support was provided by Bristol-Myers Squibb and The BMS International Immuno-Oncology Network (II-ON).