Background: A defining feature of malignant rhabdoid tumors (MRTs), epithelioid sarcoma (ES), and poorly differentiated chordomas (PDC) is loss of integrase interactor 1 (INI1) expression, which induces dependence on enhancer of zeste homolog 2 (EZH2). Tazemetostat (TAZ) is a selective EZH2 inhibitor approved by the US Food and Drug Administration for patients (pts) aged ≥16 y with metastatic or locally advanced ES ineligible for complete resection. Pediatric dose escalation and interim efficacy and safety data for the dose expansion were previously reported. Updated efficacy, safety, subgroup analyses, and translational results are reported here. Methods: NCT02601937 is a phase 1 multicenter study evaluating TAZ monotherapy ≤2400 mg/m² daily in pediatric pts with relapsed or refractory INI1⁻ tumors (MRT, atypical teratoid rhabdoid tumor [ATRT], select tumors with rhabdoid features, other INI1⁻ tumors, and SS18-SSX synovial sarcoma). Primary endpoint for dose expansion was objective response rate (ORR). Secondary endpoints included safety/tolerability, duration of response (DOR), progression-free survival (PFS), and overall survival. Selected exploratory flow data using 250,000 cells from peripheral blood were available for up to 25 pts. Results: Phase 1 enrolled 109 pts (escalation, n=46; expansion, n=63; Table) with mean ages of 5.4 y in the dose escalation and 7.4 y in the dose expansion. ORRs were 7% (3/46) in the dose escalation and 14% (9/63) in the dose expansion. Per tumor category in the dose expansion, ORRs were 24% (5/21) ATRT, 33% (2/6) PDC, 22% (2/9) ES, and 0% in non–CNS RTs (n=21) and other tumors (n=6). ORR for pts with prior radiotherapy was 14% (11/80) vs 3% (1/29) with prior radiotherapy (P>0.05). In the dose expansion, median PFS was 8 wk (95% CI: 8–13), OS was 21 wk (95% CI: 13–38), and DOR was 35 wk (95% CI: 24–121). Grade 3–4 treatment-related treatment-emergent adverse event rates were 15% (7/46) in the dose escalation and 22% (14/63) in dose expansion. Differences between responders (R) and nonresponders (NR) at C1D1 for neutrophil counts were significant (P<0.05). All R (6/6) had <125,000 total neutrophils at C1D1 vs 42% (8/19) of NR; 50% (3/6) of R had >50,000 total CD3 T cells vs 0% (0/19) of NR. Conclusions: TAZ showed promising antitumor activity in ATRT, ES, and PDC. Potential synergism between prior radiotherapy and TAZ requires further investigation. TAZ was generally well tolerated. The biological/clinical significance of differences in peripheral blood cell counts between R and NR at C1D1 needs further investigation. Clinical trial information: NCT02601937.

Dosed (mg/m²) twice daily unless otherwise noted. ^aThree times daily.