Single-agent expansion cohort of lenvatinib (LEN) and combination dose-finding cohort of LEN + etoposide (ETP) + ifosfamide (IFM) in patients (pts) aged 2 to ≤25 years with relapsed/refractory osteosarcoma (OS).

11527 Background: LEN is a multikinase inhibitor of VEGFR1-3, FGFR1-4, PDGFRα, KIT, and RET. In human pediatric OS xenograft models, LEN enhanced the antitumor activity of IFM + ETP. In a phase (Ph) 2 study of sorafenib in pts with unresectable high-grade OS, progression-free survival at 4 months (PFS-4) was 46%. We report emerging data from a Ph 2 single-agent LEN expansion cohort and Ph 1b combination dose-finding cohort of LEN + ETP + IFM in pts with relapsed/refractory OS. Methods: Pts were age 2 to ≤25 years, had relapsed OS, and had < 2 prior VEGF-targeted therapies. The recommended dose (RD; 14 mg/m2) from Ph 1 was used in Ph 2. In Ph 1b, a 20% reduction of LEN (11 mg/m2/day) was used as starting dose in combination with IFM 3000 mg/m2 + ETP 100 mg/m2 daily/3 days. Ph 2 and Ph 1b primary endpoints were PFS-4 per RECIST 1.1 and RD of the combination. Secondary objectives included best overall response (BOR), objective response rate, PFS, safety, and PK. Data cutoff was October 6, 2017. Results: In Ph 2, 16 pts received LEN 14 mg/m2. Median (min, max) number of cycles received was 3.0 (1, 9). Most common any-grade TEAEs were diarrhea, hypothyroidism, and proteinuria (43.8% each). Most common Grade (G) 3 or 4 TEAEs were back pain and dyspnea (12.5% each). One pt discontinued due to TEAEs (myelodysplastic syndrome); 5 of 15 evaluable pts achieved PFS-4 and BOR was PR (n = 1) and SD (n = 7). In Ph 1b, 7 pts received LEN 11 (LEN 11 mg/m2 + IFM + ETP) and 6 pts received LEN 14 (LEN 14 mg/m2 + IFM + ETP). 1 pt had a DLT at LEN 11 (G3 thrombocytopenia) and 2 pts had DLTs at LEN 14 (G4 thrombocytopenia with G3 bleeding in 1 pt; G2 oral dysesthesia, G2 back pain, and G1 muscle spasms in 1 pt). G ≥3 TEAEs occurred in 85.7% and 66.7% of LEN 11 and LEN 14 pts, respectively. No pts discontinued due to TEAEs. BOR was PR (n = 1) and SD (n = 5). AUC at steady state of LEN and distributions of interindividual variability of oral clearance were comparable between LEN single agent and the combination. Conclusions: In Ph 2, LEN at 14 mg/m2 achieved PFS-4 in 5 pts; BOR was PR in 1 pt and SD in 7 pts. In Ph 1b, LEN + chemotherapy had a manageable safety profile. The study is ongoing and updated data will be presented. Clinical trial information: NCT02432274.