American Society of Hematology

Inhibiting EZH2 in Lymphoma: Breaking the Silence of the Genes

Peter Johnson, MD
Cancer Research UK Centre, University of Southampton, UK

Published on: January 01, 2013

Dr. Johnson indicated no relevant conflicts of interest.

McCabe MT, Ott HM , Ganji G, et al. EZH2 inhibition as a therapeutic strategy for lymphoma with EZH2-activating mutations. Nature. 2012;492:108-112.

The discovery that a large proportion of germinal center lymphomas have mutations in epigenetic regulator genes has been one of the notable revelations from genome sequencing in recent years. This finding has rapidly led to the testing of small molecule inhibitors that might reverse the defective processes caused by these mutations, and this paper from the group at GlaxoSmithKline laboratories describes the results for one such inhibitor, GSK126. The target of the inhibitor is EZH2, an enzyme that is affected by gain of function mutations in up to a quarter of cases of diffuse large B-cell lymphomas (DLBL) and follicular lymphoma. As the enzymatic component of the polycomb repressor complex, EZH2 is involved in transcriptional repression through chromatin condensation by methylation of histone H3 on lysine 27 (H3K27). Thus, EZH2 functions as a histone-lysine methyltransferase.

GSK126 was identified from a high throughput screen of compounds that inhibited the EZH2 catalytic subunit. Following optimization of structure, GSK126 was found to be active at low nanomolar concentration, inhibiting both wild-type and mutant EZH2 with a high degree of selectivity. Fifty-percent reduction of H3K27 methylation in DLBL cells lines was observed at GSK126 concentrations between 7 and 252 nM. Although the suppression of methylation was not different between EZH2 wild-type and EZH2 mutant lines, proliferation assays showed that six of the seven most sensitive lines bore EZH2 activating mutations. The effects were independent of the presence of BCL2 translocations or p53 mutations and appeared to be mediated by caspase cleavage and apoptosis in some lines but not others, reflecting different degrees of dependence on EZH2 activity. Studies of gene expression indicated de-repression of EZH2 targets by GSK126 in sensitive lines, with the specificity of this effect confirmed using shRNA to knockdown EZH2 expression. The gene de-repression profile was significantly different between cell lines, with sensitive lines showing alterations in pathways involved in cell-cycle regulation, cell death, and control of biologic/cellular processes. Cell lines in which both high EZH2 expression and high levels of H3K27 methylation were present together were the most sensitive to GSK126.

The investigators went on to test the effects of GSK126 in xenograft models, where they showed that the cell lines that had been sensitive in vitro were also responsive in vivo. Mice treated with 50 mg/kg daily showed tumor growth arrest, although larger doses of 150 to 300 mg/kg daily were required to prevent regrowth following withdrawal of the drug. Further studies of weekly schedules suggested that the effective half-life of the drug would permit intermittent dosing. The treatment appeared to be well tolerated at the doses described, and interestingly, there were no signs of bone marrow suppression, despite the many effects of EZH2 on normal hematopoiesis.

Modifiers of epigenetic regulation such as the histone deacetylase inhibitors have so far yielded interesting but modest results in the treatment of B-cell lymphoma, despite some evidence in preclinical studies of synergy with rituximab. As single agents, their activity is limited, and few combination studies have been performed to date. Nonetheless, there is mounting evidence of the fundamental role that epigenetic dysregulation plays in the pathophysiology of germinal center lymphomas, with almost all follicular lymphomas and a substantial proportion of DLBL showing mutations in genes that are involved in histone modification (acetylation or methylation). Mutation of epigenetic regulators appears to be an early event in disease pathogenesis, making the resulting aberrant process a tempting target for specific therapy. The preclinical studies reported by McCabe and colleagues can be seen as proof of concept; however, questions such as whether the presence of EZH2 mutations is essential for response in vivo, or whether looking for specific patterns of methylation in the tumor may be predictive of response, remain to be answered. The results of the first-in-man studies are eagerly awaited.

back to top