A phase II trial of riluzole, an antagonist of metabotropic glutamate receptor 1 (GRM1) signaling, in patients with advanced melanoma

Abstract

Studies demonstrate that GRM, expressed by >60% of human melanomas, may be a therapeutic target. We performed a phase II trial of 100 mg PO bid of riluzole, an inhibitor of GRM1 signaling, in patients with advanced melanoma with the primary endpoint of response rate. Thirteen patients with GRM1-positive tumors were enrolled. No objective responses were observed, and accrual was stopped. Stable disease was noted in six (46%) patients, with one patient on study for 42 weeks. Riluzole was well tolerated, with fatigue (62%) as the most common adverse event. Downregulation of MAPK and PI3K/AKT was noted in 33% of paired tumor biopsies. Hypothesis-generating correlative studies suggested that downregulation of angiogenic markers and increased leukocytes at the active edge of tumor correlate with clinical benefit. Pharmacokinetic analysis showed interpatient variability consistent with prior riluzole studies. Future investigations should interrogate mechanisms of biologic activity and advance the development of agents with improved bioavailability.

Keywords: angiogenic factors and receptors; clinical trials; glutamate; melanoma; riluzole; skin cancers.

Conflict of interest statement

Disclosure statement: The authors have no potential conflicts of interest related to this work to disclose.

© 2018 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.

Figures

Figure 1

Western blots showing signaling response in paired tumor samples for ERK, pERK, pAKT308, total AKT and CD31. Tumor tissue protein lysates were prepared as described (Yip et al., 2009) and 10 μg of total protein lysate were resolved on 4-20% SDS-PAGE gels. Resolved proteins were blotted to polyvinylidene difluoride membranes by standard procedures and probed with antibodies as indicated in the figure legends. Probed and chemiluminated protein bands were exposed and captured by a Syngene G: Box Chemiluminescence imaging system with the auto-exposure mode, which is determined by the strongest band intensity on each blot. The band intensity quantities were obtained by the supporting software Gene Tools from SynGene. (A) Results from eight pairs (pre-treatment A, post-treatment B) of tumor samples from patients treated with riluzole are shown. Patients 1, 5 and 11 experienced down regulation of pERK and p-AKT and stable disease at first re-evaluation. CD31 expression was decreased in these patients as well. (B, C, and D) Quantification of activated ERK, activated AKT and CD31 from the band intensities shown in (A).

Figure 2

IHC staining of paired tumor samples. Slide processing and immunohistochemistry (IHC) staining was performed by the Histopathology and Imaging Shared Resource of Rutgers Cancer Institute of New Jersey. Slides were scanned by the Biomedical Imaging Center for Biomedical Imaging and Informatics. Three fields were selected from each slide for IHC staining index quantification by the online software at the site of http://153.1.200.58:8080/immunoratio. (A) Column A displays pre-treatment tumor tissue and column B displays post-treatment tumor tissue. Pre-treatment and post-treatment tumor specimens from patients 1, 5, and 11, who showed a trend of disease stabilization and decreased pERK, pAKT and CD31 on Western blot, were stained for pERK, cleaved caspase-3 and CD31. While staining intensity decreased for pERK and CD31, it increased for cleaved caspase-3 in post-treatment specimens. (B) IHC quantification of pERK, cleaved caspase-3 and CD31 (mean + SD).

Figure 3

IHC quantification of (A) CD45+ cells and (B) CD68+ cells in tumor, at the tumor-stromal interface, and stroma in pre- and post-treatment samples from four patients. (C) CD45+ staining at the tumor-stromal interface increased in the post-treatment samples from patient #1 (1A, 1B) and 11 (11A, 11B). Tissue from paired biopsy pre-treatment and post-treatment samples was stained for tumor-associated leukocytes and macrophages. Intensity was graded on a 1-3 scale in a blinded manner by a clinical pathologist.