Therapeutic Targeting of the Bcl-6: p53 Axis with Histone Deacetylase Inhibitors in Diffuse Large B-Cell Lymphoma,

Abstract 3733

Diffuse large B-cell lymphoma (DLBCL) is the most common non-Hodgkin's Lymphoma. Despite advances in treatment, 1/3 of patients die from their disease. Gene expression profiling has delineated three subtypes with different genetic features known to be prognostic: the Activated B-cell (ABC), Germinal Center (GC), and grey zone types. For example, ABC DLBCL is addicted to NFkB over-expression. The oncogene, BCL6, encodes a transcription factor that functions as a transcriptional repressor within normal germinal center B-cells. Constitutive activation of Bcl-6 leads to GC-type DLBCL by turning off genes expressing cell cycle dependent kinase inhibitors, and essential tumor suppressor genes, like p53. There is a critical inverse relationship between Bcl-6 and p53, the functional status of which is linked to each transcription factor's degree of acetylation. Deacetylation of Bcl-6 is required for maintaining its effects as a transcriptional repressor. Conversely, acetylation of p53 is activating when class III histone deacetylases (HDAC), also known as sirtuins, are inhibited by drugs such as niacinamide. HDAC inhibitors are presently approved for T-cell lymphoma and may require the targeting of additional pathways to be effective in B-cell lymphomas. Trichostatin A and niacinamide modulate Bcl-6 in lymphoma cell lines. One therapeutic strategy that could favorably shift the relationship between oncogenes and tumor suppressors is the pharmacologic modification of Bcl-6 and p53 using HDAC inhibitors.

Eight DLBCL cell lines were screened (4 ABC: Su-DHL2, HBL-1, OCI-Ly10, RIVA; 4 GC:OCI-Ly1, OCI-Ly7, Su-DHL6, Su-DHL4) with four class I/II HDAC inhibitors (romidepsin, vorinostat, panobinostat and belinostat) in combination with niacinamide (sirtuin inhibitor) at two dose levels each at three time points. Cell growth inhibition was measured by luminescence cell viability and apoptosis flow cytometry assays. Synergy was measured by the relative risk ratio (RRR) calculation where values <1 represent synergy. Synergy was achieved in significantly greater number and intensity in the GC versus ABC cell lines. Specifically, romidepsin in combination with niacinamide achieved the greatest synergy. To analyze mechanism of action, DLBCL cell lines were treated with combinations of class I/II HDAC inhibitors and niacinamide. Cells of both GC and ABC subtypes treated with the combination resulted in increased acetylation of p53, and increased p21 and BLIMP-1 content compared to controls. These results did not correlate with cytotoxicity as the ABC cell lines did not achieve the same synergy as the GC cells. GC cells treated with the same combinations resulted in acetylation of Bcl-6 compared with controls as measured by immunoprecipitation and Western blotting assays; ABC cells do not express Bcl-6. This finding correlated with cytotoxicity implying that a rational second pathway must be targeted to shift the balance between oncogene and tumor suppressor activity to achieve effective cell kill. p300 content was also increased suggesting that treatment with HDAC inhibitors recruit or upregulate its production and activity leading to increased acetylation. Using a novel double transgenic mouse model of aggressive spontaneous B-cell lymphoma (l-myc overexpressing crossed with CD19-tagged mCherry luciferase), in vivo effects of the drug combination were studied. These mice express equal basal amounts of Bcl-6 and p53 as GC cell lines. Mice treated with niacinamide 20 mg/kg and romidepsin 2.3mg/kg IP for 5 hours achieved increased acetylation of Bcl-6 and p53, and accumulation of p21 and BLIMP1 compared with controls. Importantly, mice treated with the combination of niacinamide 40 mg/kg and romidepsin 2.3 mg/kg IP achieved decreased tumor burden as measured by bioluminescence signal intensity compared to mice treated with each drug alone and controls.

Presently, we are translating these concepts and observations in a proof-of-principle phase I trial evaluating the safety of vorinostat plus niacinamide in lymphoid malignancies. By targeting the specific pathogenetic features of DLBCL, it may be possible to tailor future treatment platforms for discrete subtypes of DLBCL.

View largeDownload slide

View largeDownload slide

Close modal

View largeDownload slide

View largeDownload slide

Close modal