Research in context
Evidence before this study
We searched PubMed up to May 5, 2017, with the terms “metastatic sarcoma”, “clinical trials”, “PD-1”, “PD-L1”, “immune checkpoint blockade”, “immune response”, “pembrolizumab”, and “nivolumab” for articles published in English. Patients with metastatic soft-tissue sarcoma or bone sarcoma have a poor prognosis and low survival. First-line treatment of soft-tissue sarcoma with single-agent chemotherapy remains the standard of care and leads to a response in 10–15% of patients. Combination therapies have little or no effect on overall survival in patients with soft-tissue sarcoma, while consistently increasing the incidence of adverse events. Treatment of bone sarcoma with combination chemotherapy in the neoadjuvant and adjuvant setting can be curative despite the substantial toxicity. However, in patients with advanced bone sarcomas, no existing therapeutic options offer a survival advantage. Sarcoma has not traditionally been considered an immunogenic tumour; however, several studies showed PD-L1 to be expressed in up to 30–40% of certain sarcoma subtypes. Anecdotal reports and retrospective series studies have described responses to immune checkpoint blockade in select sarcomas. One phase 2 study of nivolumab in leiomyosarcoma was stopped early for futility, but no prospective studies have been published to date in other subtypes of soft-tissue sarcoma or bone sarcoma. Treatment with antibodies targeting the PD-1/PD-L1 axis led to durable responses in several cancers.
Added value of this study
To our knowledge, this prospective, multicentre, phase 2 study is the first to investigate the use of immune checkpoint blockade with anti-PD-1 antibodies in the treatment of soft-tissue sarcoma and bone sarcoma. We considered the heterogeneity of sarcomas and limited the study to specific histological subtypes to identify potential subtype-specific efficacy signals. We observed an objective response in seven (18%) of 40 patients with soft-tissue sarcoma, and the benefit was limited to patients with undifferentiated pleomorphic sarcoma and dedifferentiated liposarcomas. The characteristics of the responses were consistent with the durable benefit observed with checkpoint blockade in other cancers. We observed little to no benefit in patients with synovial sarcoma or leiomyosarcoma. Similarly, we observed infrequent responses to therapy in patients with bone sarcomas, suggesting the need for combination approaches in this population.
Implications of all the available evidence
These findings suggest that pembrolizumab is clinically active in patients with undifferentiated pleomorphic sarcoma or dedifferentiated liposarcoma. Expanded cohorts of these subtypes, which represent more than 30% of all soft-tissue sarcomas, are ongoing. Further investigation is required to determine the utility of predictive biomarkers for response and to understand the mechanisms of resistance in the other subtypes of soft-tissue sarcoma and bone sarcoma, in which rational combination therapies could be considered.