Presentation Abstract

Abstract Number: 1684
Presentation Title: Inflammatory signaling genes as predictive markers of vorinostat sensitivity in multiple myeloma
Presentation Time: Monday, Apr 19, 2010, 9:00 AM -12:00 PM
Location: Exhibit Hall A-C, Poster Section 27
Poster Section: 27
Poster Board Number: 24
Author Block: Venugopalan Cheriyath, Melissa Kuhns, Matt Kalaycio, Ernest C. Borden. The Cleveland Clinic, Cleveland, OH
Abstract Body: Purpose: The objective of this preclinical study is to identify a gene signature that can predict therapeutic responses to vorinostat (SAHA, SuberoylAnilide Hydroxamic Acid), an inhibitor of histone deacetylases (HDACi), which is currently under investigation in multiple myeloma. This is essential for identifying responding patients from nonresponders (intrinsic resistant) to vorinostat who may benefit from alternative treatment.
Experimental Design: To determine the antimyeloma effects of vorinostat, bone marrow aspirates from 24 myeloma patients were treated with increasing concentrations of vorinostat (0 - 1 μM) for 72 hrs. The percentage reduction and loss of myeloma cell (CD138+) viability was assessed by Flow Cytometry after staining with FITC conjugated anti-CD138 antibody and 7AAD. CD138+ cells from sensitive (IC50 ≤1 μM) and resistant (IC50 >1 μM groups were enriched to >80% by magnetic separation. Basal gene expression profiles of sensitive and resistant groups were determined using a 48 k Illumina expression array. TM4-MeV software was used to analyze the gene array data to identify genes that were differentially expressed in sensitive and resistant groups, which was confirmed by qRT-PCR and western blotting.
Results: Among 24 fresh myeloma samples analyzed, vorinostat achieved IC50 in 10 (41.7%) and IC75 in 5 (20.8%) samples. Nine samples (37.5%) were vorinostat resistant. Non-parametric analysis of gene expression array results identified differential expression of 118 genes (>2x increase in median expression with a P ≤0.05) between sensitive (70 genes) and resistant (48 genes) groups. Interestingly, Ingenuity Pathway Analysis suggested a correlation between differential activation of inflammatory signals and vorinostat sensitivity of myeloma. 17 genes for interferon (IFN) signaling were constitutively expressed in the sensitive group whereas 13 genes related to TNF/IL-6 signaling were predominant in the resistant group. Based upon these results, we hypothesized that IFN-α2b pretreatment would sensitize myeloma cells to vorinostat. As predicted, IFN-α2b and sub-IC50 concentrations of vorinostat combinations synergistically (combination indices of <1) reduced the viability of myeloma cell lines.
Conclusions: This study identified 118 genes as potential predictors of patient derived fresh myeloma cell sensitivity to vorinostat. After gene data reduction, a signature of IFN stimulated genes was generated for vorinostat sensitivity and genes related to TNF/IL-6 inflammatory signals for vorinostat resistance. In myeloma cell lines, pretreatment of IFN-α2b augmented antimyeloma effects of vorinostat. Further demonstration of thissynergy on fresh myeloma cells and in mouse xenograft models will provide a rationale for combining these two agents for myeloma therapy.