Alterations in Metabolism and Mitochondrial Respiration Are Associated with Response to Targeted Inhibition of WNT Signaling in Lymphomas

Abstract Abstract 778 Background Wnt signaling is a highly conserved pathway with diverse roles in proliferation, self-renewal and differentiation in multiple tissue types. Aberrant Wnt signaling and deletions/mutations of Wnt regulatory genes have been found in a variety of malignancies including lymphoma. The factors which govern this pathway and the precise role of Wnt signaling in mediating tumor proliferation are unknown. We previously identified recurrent mutations in Wnt pathway genes including APC and WIF1 in diffuse large B cell lymphoma (DLBCL) using exome sequencing and thus sought to investigate if targeting of Wnt with molecular agents would be a viable treatment approach in NHL. Methods Alamar blue proliferation assays were used to determine IC50 values for 2 small molecule inhibitors of Wnt signaling: ICG001 (a selective inhibitor of TCF/LEF1 mediated transcription) and salinomycin (an ionophore which has been shown to target Wnt) in a panel of 65 cell lines comprising subtypes of lymphoma (including 7 ABC DLBCL, 16 GCB DLBCL, 10 Burkitt lymphoma, 5 mantle cell lymphoma and 5 primary effusion lymphoma cell lines) and leukemia. Both molecules inhibited proliferation of cell lines in the screening panel at physiologically achievable concentrations (IC50 range for salinomycin 0.11–6.23 μM, IC50 range for ICG001 0.96–28.9 μM). IC50s values for salinomycin and ICG001 correlated well (Pearson r=0.56) suggesting a common mechanism of action between the 2 agents. Furthermore, we found lower IC50 values in primary effusion lymphoma (PEL) cell lines (0.35 μM vs. 1.37 μM for other cell line histologies, p=0.01 for salinomycin and 3.25 μM vs. 6.22 μM, p=0.13 for ICG001), which is a lymphoma subtype associated with aberrant Wnt signaling. We did not note significant differences in cytoxicity among other subsets. Wnt activity was confirmed using TOPFLASH luciferase assays and correlated with sensitivity to ICG001 and salinomycin. We then used gene expression profiles derived from Affymetrix Gene 1.0 arrays from each cell line and a Cox proportional hazards model to identify genes associated with sensitivity or resistance to both agents. A gene list consisting of 367 genes associated with either sensitivity or resistance at p<0.05 for both agents was derived from this analysis. Hierarchical clustering was used to define 2 groups of cell lines based upon these genes and analyzed with gene set enrichment (GSEA). GSEA comparisons of sensitive and resistance groups derived from this method revealed that genes associated with mitochondrial pathways were highly associated with response to both ICG001 and salinomycin (FDR=0.036, p=0.03). Specifically, expression of genes involved in glycolysis and mitochondrial respiration including FBG1, CS, UBQRB5, OXA1L, LDHA and ATP5B was higher in ICG001 and salinomycin-sensitive cell lines. Conversely, we noted higher IC50s in cell lines with higher expression of mitochondrial genes having opposing functions, including PC and CPT1A. This overall suggests that tumor cells utilizing glycolytic and oxidative pathways are sensitive to Wnt inhibition. Remarkably, the GCB DLBCL cell line Pfeiffer also was noted to contain an inactivating mutation in APC and had the lowest noted IC50 among DLBCL cell lines. Conclusions The Wnt pathway appears an important oncologic driver in at least a subset of NHLs and we found ICG001 and salinomycin have significant in vitro activity in NHLs. These agents may be most useful in subsets of tumors with high mitochondrial mass and/or oxidative phosphorylation. This work serves as a starting point to characterize the importance of the Wnt pathway in NHL and identifies inhibition of this pathway as a therapeutic opportunity in lymphomas. Disclosures: No relevant conflicts of interest to declare.