Biomarkers of treatment effect in a phase II trial of velimogene aliplasmid in patients with metastatic melanoma.

Abstract

e19058 Background: Velimogene aliplasmid (Allovectin, A) is a plasmid-based immunotherapeutic that encodes human leukocyte antigen-B7 and β-2 microglobulin. A is designed to induce allogeneic and tumor-antigen-specific T-cell responses against tumor cells. In a phase II trial, 127 patients with stage III or IV recurrent metastatic melanoma were treated in cycles of 6 weekly intralesional injections of 2 mg/dose. Complete (CR) or partial responses (PR) were observed in 15 patients (11.9%, 95% CI: 6.2-17.4), stable disease (SD) in 32 (25.2%), and progressive disease (PD) in 80 (63.0%). Methods: Serum collected before treatment and 3 weeks after each cycle from 29 patients who showed clinical benefit (CB: 15 CR or PR patients and 14 patients with SD for >4 months) and 62 with PD (progressed by the end of the first cycle) were evaluated for 45 inflammation biomarkers in a quantitative multiplex immunoassay. Changes in analyte concentrations relative to baseline were evaluated by paired t-test or Wilcoxon signed-rank test. Results: Overall, 52% of patients had decreases in serum IL-10 after treatment; however, decreases were more prevalent in CB patients (71.4% for CB, 44.4% for PD; p=0.036; Chi-square test). At week 8, mean serum IL-10 concentration within a group was reduced relative to baseline in CB (p=0.008) but not PD patients suggesting greater treatment effect in CB patients. In CB patients, IL-10 remained at lower levels relative to baseline through week 26. CCL-3 decreased in CB patients by week 26 (p=0.021) and von Willebrand factor (a biomarker of vascular endothelium dysfunction) was increased at week 26 (p=0.007). In PD patients, by week 8, several biomarkers associated with metastatic melanoma (TIMP-1, VCAM-1, and TNF-R2) or inflammation (haptoglobin) increased (p=0.001, 0.022, <0.001, 0.004, respectively) consistent with tumor progression, whereas others (IL-8 and stem cell factor) decreased (p=0.007, 0.013, respectively). Conclusions: Consistent with A’s proposed mechanism of action, treatment was associated with a decrease in the production of the T-cell suppressive cytokine, IL-10. Furthermore, biomarker profiles in the two groups were markedly different and analogous with treatment outcomes.