Phase II Trial of Ixazomib combined with Gemcitabine and Doxorubicin in Patients with SMARCB1-Deficient Renal Medullary Carcinoma

SMARCB1-deficient renal medullary carcinoma (RMC) is an aggressive kidney cancer lacking mechanism-directed therapies. We conducted a single-center, single-arm, phase II study (NCT03587662) testing the oral proteasome inhibitor ixazomib combined with gemcitabine and doxorubicin. Thirty consecutive patients (median age 34.5 years old, 90% Black, 83.3% metastatic at diagnosis) were treated between September 2018 and June 2022. Ixazomib 5.5 mg, gemcitabine 756 mg/m² and doxorubicin 42 mg/m² were given every 2 weeks for up to 13 cycles, followed by ixazomib plus gemcitabine maintenance. Co-primary endpoints were objective response rate (ORR) and 28-week disease-control rate (DCR) versus historical gemcitabine plus doxorubicin controls. Secondary endpoints included progression-free survival (PFS), overall survival (OS) and safety. The posterior ORR was 36% (95% credible interval [CrI], 21–54 %) with a 91.4% probability of exceeding historical doublet therapy. However, the 28-week DCR was 17 % (95% CrI 6–31%), crossing the predefined futility boundary. Median PFS and OS were 3.5 months (95% confidence intervals [CI] 1.9–4.9) and 7.4 months (95% CI 2.1-4.9), respectively. Grade ≥ 3 toxicities were predominantly hematologic (thrombocytopenia 20 %, leukopenia 17%) and manageable. No treatment-related deaths occurred. Single-cell and bulk multi-omics from 11 patients revealed that immune-inflamed tumors enriched for T/NK cells, plasmacytoid and conventional dendritic cells, and S100A8/IL-1ß-skewed monocytes correlated with response, whereas stromal-myeloid niches and proliferative or neuroendocrine-squamous plastic epithelial states were associated with resistance. Resistant lesions upregulated unfolded-protein-response and Hippo pathways. These results pave the way for future biomarker-guided patient selection and rational combination or microenvironment-targeted therapies.