Sirtuin Inhibition in Combination with Histone Deacetylase (HDAC) Inhibition Is Effective Therapy for Aggressive B-Cell Lymphomas in Both Pre-Clinical and Clinical Studies of Disease.

Abstract 2725

There is a critical inverse relationship between Bcl6 and p53, the functional status of which is linked to each transcription factor's degree of acetylation. Deacetylation of Bcl6 is required for its transcriptional repressor effects allowing for the oncogene to drive lymphomagenesis. Conversely, acetylation of p53 is activating when class III HDACs, or sirtuins, are inhibited by agents such as niacinamide. One therapeutic strategy for lymphomas addicted to Bcl6 overexpression is the pharmacologic modification of Bcl6 and p53 using HDAC inhibitors. Here we report on the combination of sirtuin inhibitor, niacinamide (NIA), plus HDAC inhibitors in preclinical models, and a phase Iclinical trial in patients with relapsed or refractory lymphoma.

Cytotoxicity was measured in 8 diffuse large B-cell lymphoma (DLBCL) cell lines (4 ABC and 4 GC) with 4 HDAC inhibitors (romidepsin, vorinostat, panobinostat and belinostat) in combination with NIA. Synergy was achieved predominantly in GC vs ABC cell lines, with romidepsin plus NIA having the greatest synergy. Acetylation of Bcl6 was observed in cells treated with HDAC inhibitors, or the combination, compared with control as measured by immunoprecipitation. Cells treated with the combination had increased acetylated-p53, p21 and BLIMP-1 content. In vivo effects of the drug combination were studied in a double transgenic mouse model of aggressive spontaneous B-cell lymphoma (l-myc overexpressing crossed with CD19-tagged mCherry luciferase). These mice express equal basal levels of Bcl6 and p53 as GC cell lines. Mice treated with NIA and romidepsin for 5 hrsachieved increased acetylation of Bcl6 and p53, and accumulation of p21 and BLIMP1. Mice treated with the combination exhibited decreased tumor burden achieving complete responses (CR=30%) compared to single agents and control (CR=0) after 3 weeks of treatment. The combination was well tolerated.

Based on this rationale, a phase I clinical trial of vorinostat plus NIA was conducted in patients with relapsed or refractory lymphoid malignancies. Twenty-five patients enrolled between 3/2009 and 3/2011. Median age was 43 (range:25–75), 44% female, and 84% white. This was a group of heavily pretreated patients, median number of therapies was 4; with 16 auto- and 4 allo-transplants. Patients were treated on a 14/21 day cycle. Vorinostatwas given orally as a fixed dose of 400 mg daily. NIA was given orally, daily, in escalating doses from 20–100 mg/kg. A range of 1–18 cycles were given.

The most common toxicities included fatigue (84%), nausea (80%), diarrhea (72%), and anorexia (56%). There was a DLT in cohort 4 (vorinostat 400 mg/NIA 80 mg/kg) of grade 3 infection. Two DLTs occurred at dose level 5 (vorinostat 400mg/NIA 100 mg/kg), a grade 4 transaminitis, not otherwise explained; and grade 4 hypotension. These events led to the determination of dose cohort 4 as the MTD. In total, there were 12 different grade 3–4 toxicities including neutropenia, infection, and transaminitis that occurred in 11 patients. There were two deaths that occurred within safety follow-up period in patients who had fulminant disease progression (POD); none of these events were related to the study drugs. Overall this was a very well tolerated treatment regimen and the responses seen were with the weakest HDAC and sirtuininhibitors available given to extremely heavily pretreated patients.

The overall response rate was 24%, with 2 CRs and 3 partial responses (PR). CRs were maintained for 13 to 18 weeks. All PRs were in HL patients with 6–10 prior regimens including auto- and allo-transplants, and maintained for 4–13 weeks. Twelve patients achieved stable disease (SD) (57%), 2 of whom achieved near-PR. Of the 7 patients with germinal center derived lymphomas (4 DLBCL + 3 FL), there was 1 CR, 3 SD, and 3 POD.

The preclinical and clinical data establish proof-of-principle that the Bcl6: p53 axis may be therapeutic targets, and that acetylation of these transcription factors could potentially serve as biomarkers for activity. Currently we are exploring new, more potent sirtuin inhibitors with second generation HDAC inhibitors, and working to further define mechanism of action. By targeting discrete molecular subtypes of DLBCL it may be possible to increase complete response rates in this challenging disease.