Abstract 5872: Pediatric Preclinical Testing Consortium (PPTC) of eribulin in osteosarcoma (OS) patient-derived xenograft (PDX) models

Introduction: Eribulin is a synthetic analogue of halichondrin B and inhibits cancer cell proliferation via blockade of microtubule function. It is FDA approved for patients with breast cancer with two prior chemotherapy regimens for the treatment of metastatic disease and for previously treated patients with inoperable or metastatic liposarcoma. In vivo testing of eribulin against osteosarcoma (OS) Pediatric Preclinical Testing Program (PPTP) PDX models, developed from primary tumors, demonstrated significant single-agent activity. However, a Children's Oncology Group (COG) phase 2 trial of eribulin in patients with relapsed OS did not demonstrate significant antitumor activity of single-agent eribulin. In the current study the efficacy of eribulin in PDX models generated from primary and relapsed tumors, as well as dose-response, were assessed. Methods: Eribulin was evaluated in 6 PPTC models, 4 from primary OS specimens (OS46, OS51, OS55, OS56) and 2 from relapsed specimens (OS39R, OS60). Eribulin was administered via intraperitoneal (IP) injection at a dose of 1mg/kg on days 1 and 4, repeated every 21 days. Response to treatment was determined based on PPTP-established endpoints (1). In addition, 3 dose levels of eribulin were evaluated in 3 PPTP models developed from primary tumors (OS1, OS17, OS33) and 1 model developed from a relapsed tumor (OS60). In the dose-response studies, eribulin was administered IP at 1mg/kg, 0.5mg/kg and 0.25mg/kg on days 1 and 4, repeated every 21 days with follow-up through day 42. Results: Eribulin at a dose level of 1mg/kg induced stable disease (SD) or partial response (PR) in 4/9 and 3/9 OS xenografts, respectively. SD/PR responses were observed in PDX models generated from primary tumors (5/7) as well as PDX models generated from relapsed tumors (2/2). Eribulin demonstrated maximal antitumor activity at the highest dose level, 1mg/kg (SD/PR in 3/4). At 0.5 mg/kg 1/4 models achieved SD/PR, and at 0.25mg/kg all 4 models showed progressive disease within the 6-week testing period. Minimum relative tumor volumes also showed a dose-response effect from 0.25 to 1.0 mg/kg. Conclusions: Eribulin demonstrated antitumor activity against OS PDX models generated from primary and relapsed tumors at a dose level of 1mg/kg. Eribulin showed a clear dose-response effect in terms of its ability to induce SD/PR responses, highlighting the importance of dose selection. This dose-response effect suggests that an inability to achieve comp