(P 174) SELISARC: A SPANISH GROUP FOR RESEARCH IN SARCOMA (GEIS) PHASE I/II TRIAL OF SELINEXOR PLUS GEMCITABINE IN SELECTED SARCOMA SUBTYPES: RESULTS OF THE PHASE I PART

Objective: Selinexor (KPT-330) (S), a small-molecule inhibitor of the nuclear export mediator exportin-1 (XPO-1), is able to reduce mRNA and protein expression of DNA damage repair (DDR) gene products, being synergistic with DNA damage agents as gemcitabine (G) in preclinical experiments. We hypothesized that selinexor would have a synergistic effect with gemcitabine in patients with selected sarcoma subtypes.

Methods: Adult patients, with progressing sarcoma (based in RECIST), ECOG 0-1, with up to 2 previous systemic therapies for advanced disease (localized unresectable/metastatic) with centrally confirmed diagnosis of undifferentiated pleomorphic sarcoma (UPS), leiomyosarcoma (LMS), alveolar soft part sarcoma (ASPS), or osteosarcoma (OS) were eligible. Selinexor (days 1, 8, 15) and Gemcitabine (days 1, 8) were given in four dose levels, L1: S 60 mg + G 1,000 mg/m2 in 30 min, L2: S 60 mg + G 1,000 mg/m2 (10 mg/m2/min), L3: S 60 mg + G 1,200 mg/m2 (10 mg/m2/min), and L4: S 80 mg + G 1,200 mg/m2 (10 mg/m2/min). A -1 level was defined with S 60 mg + G 800 mg/m2 (30 min). A classic 3+3 design was used to determine the recommended phase 2 dose (RP2D) based on Dose-limiting toxicities (DLTs) observed during the first 21-day cycle. In vitro research was performed in LMS and OS cell lines to test the synergy of the combination (MTS and flow cytometry assays).

Results: Between November 2020 and June 2022, 17 patients (Male/Female 9/8), ECOG 0/1 (14/3), median age 50 years (22-71) were recruited. The median of previous lines was 1 (1-2). Diagnosis was: 9 LMS (52.9%), 6 OS (35.3%), 1 ASPS and 1 synovial sarcoma (SS) (5.9%). Three patients were treated in each of the first 3 levels and 8 patients in L4 (2 patients in that cohort were not evaluable for DLT due to improper dosing). Only one DLT was observed in L4 (G4 thrombocytopenia) and this level was selected as RP2D. G3/4 toxicity: Neutropenia (52.9%), thrombocytopenia (41.2%), febrile neutropenia (11.8%), anemia, nausea, asthenia, vomiting, alopecia, and lipase increased (5.9% each). There were 3 RECIST partial responses (PR), 7 stable disease (SD), and 7 progressive disease (PD). Median progression-free survival (PFS) for LMS was 7 months (95% CI: 3-11). No relevant clinical activity was observed in OS. In vitro cell viability studies shown that S+G was synergistic in the majority of LMS cell lines tested (combination index of 0.789 for CP0024, 0.791 for SK-UT-1, 0.604 for AA and 1.186 for IEC005). However, our results indicated antagonism of S+G in OS cell lines, with values for the combination index of 1.67 for MG63, 1.54 for SAOS2 and 1.23 for U2OS. Apoptosis assays by flow cytometry confirmed these observations in LMS and OS cells.

Conclusion: At the RP2D, S 80 mg + G 1,200 mg/m2 (10 mg/m2/min) is a feasible scheme with manageable toxicity. S+G has shown promising clinical activity for LMS, which warrants further investigation in a phase II. Preclinical studies shown the synergy and the antagonism of S+G in LMS and OS, respectively. Currently, a single-arm, Phase II study is currently being implemented.