Abstract
Background: There is a significant need for effective therapies to treat recurrent/metastatic (R/M) adenoid cystic carcinoma (ACC). This study evaluated the multi-targeted, vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitor (TKI) regorafenib in patients with R/M ACC. Methods: Patients with progressive R/M ACC were treated with regorafenib until disease progression, consent withdrawal, or excessive toxicity. The primary endpoints were best overall response (BOR) and 6-month progression-free survival (PFS). Genomic and transcriptomic biomarker analyses were performed in tumors from trial participants. Results: Thirty-eight patients were enrolled, including 7 (18%) patients with prior VEGFR-TKIs. No objective responses were observed. Six-month PFS was 45%, and median PFS was 7.2 months (95%CI 5.2-11.9 months). Presence of either activating NOTCH1 (22%) or KDM6A alterations (24%) was associated with decreased PFS (HR 2.6, 95%CI 1.1-6.1, p=0.03). Bulk RNA sequencing of pre-treatment tumors revealed that regorafenib clinical benefit (CB; PFS≥6 months; n=11) was associated with native enrichment of immune-related signatures. Immune deconvolution revealed a greater degree of macrophage and T-cell infiltration in CB tumors. Tumors from patients with no clinical benefit (NCB; PFS<6 months; n=9) had greater expression of signatures related to cell cycle progression (E2F targets, G2/M checkpoint). Conclusion: The trial failed to meet the pre-specified 6-month PFS and BOR targets. We hypothesize that TKI efficacy may be reliant upon an interplay between kinase inhibition and the ACC immune microenvironment, while programs promoting cell cycle progression may contribute to TKI resistance. These observations suggest that trials evaluating CDK4/6 inhibition plus a VEGFR-TKI should be considered.