A single-arm, phase II study assessing the efficacy of pembrolizumab (pembro) plus radiotherapy (RT) in metastatic triple negative breast cancer (mTNBC).
14 Background: Overall response rates of 13-19% have been reported with checkpoint inhibitor monotherapy in chemotherapy-resistant, PD-L1-positive mTNBC. RT is frequently used to enhance local control in mTNBC and has been reported to induce distant (abscopal) tumor responses when combined with immunotherapy. In this study, we evaluate the safety and efficacy of RT combined with a programmed cell death protein 1 (PD-1) inhibitor, pembro, in a single-arm, two-stage, phase II study in mTNBC. Methods: Eligible women had biopsy-proven mTNBC, ECOG performance status 0-2, and ≥2 measurable sites of metastatic disease with at least one site requiring RT. A total RT dose of 3000 cGy was delivered in 5 daily fractions. Pembro 200 mg was given intravenously within 3 days of first RT fraction, then every 3 weeks +/-3 days until disease progression. The primary endpoint was overall response rate at week 13 in the non-irradiated lesions by RECIST v1.1. Secondary endpoints included safety and overall survival. Tumor biopsies were obtained at baseline and at week 7. PD-L1 expression was not required for study entry. Results: Of the 17 women enrolled, the median age was 52 y (range 37-73y). and the median number of prior chemotherapies received for metastatic disease was 3 (range 0 to 8). Of the 8 women not evaluable at 13 weeks: 5 died secondary to disease-related complications (at weeks 2, 6, 7, 8, and 9) and 3 came off study due to disease progression prior to week 13. Of the 9 women evaluable at week 13, 3 (33%) had a partial response, 1 (11%) had stable disease and 5 (56%) had disease progression. The 3 partial responses represented 60%, 54%, and 34% decreases in tumor burden by RECIST v1.1 and were durable for 31, 21, and ongoing at 22 weeks, respectively. The stable disease response was durable for 22 weeks. Common toxicities were mild and included fatigue, myalgia and nausea. Conclusions: The combination of pembro and RT is well-tolerated. This is a poor prognosis population with 5/17 (29%) of patients dying within 12 weeks of study entry. However, durable responses were observed outside of the RT field in 3/9 (33%) patients who were unselected for PD-L1 expression and evaluable at 13 weeks. Clinical trial information: NCT02730130.