Antitumor Activity Of The Glutaminase Inhibitor CB-839 In Hematological Malignances

Many tumor cells are dependent on glutamine (Gln) for growth and survival. Gln is the most abundant amino acid in plasma and tumor cells can utilize Gln for energy production and generation of building blocks for the synthesis of macromolecules. A key step in Gln utilization is its conversion to glutamate (Glu) by the mitochondrial enzyme glutaminase (GLS). Suppressing GLS activity via either siRNA knockdown or small molecule inhibitor treatment has anti-proliferative activity in several tumor types. GLS mRNA levels are elevated in several hematological malignances, particularly those derived from the B-cell lineage including multiple myeloma (MM), acute lymphocytic leukemia (ALL), chronic lymphocytic leukemia (CLL) and several types of non-Hodgkin’s lymphoma (NHL). Interestingly, these B-cell derived tumors express low levels of glutamine synthetase (GLUL), an enzyme that reverses the action of GLS by converting Glu to Gln. A high GLS/low GLUL mRNA expression pattern may be indicative of tumor-dependence on extracellular Gln for growth. This expression pattern is observed in triple negative breast cancer (TNBC), a tumor type that we and others have shown to be highly dependent on exogenous Gln for cell proliferation and sensitive to GLS inhibitor treatment. CB-839 is a novel, potent, orally bioavailable GLS inhibitor that inhibits recombinant human GLS with an IC₅₀ of 15 nM. The anti-proliferative activity of CB-839 on a panel of breast cell lines is correlated with a high ratio of GLS to GLUL expression, a pattern associated with the TNBC subtype. Given the similarity of the GLS/GLUL expression pattern observed in TNBC with some hematologic malignancies, we evaluated the anti-tumor activity of CB-839 in cell culture and xenograft models of hematological malignancies predicted to be dependent on Gln for growth. Across a panel of 18 cell lines derived from MM, ALL and diffuse large B-cell lymphoma (DLBCL), Gln deprivation resulted in a minimum of 70% reduction in growth for 16 cell lines and induced cell death in 5 cell lines. Across the same panel of cell lines, the majority were sensitive to CB-839 treatment with 16 cell lines exhibiting an antiproliferative IC₅₀ of<100nM (range 2-72 nM). Importantly, there was a strong correlation between sensitivity to Gln withdrawal and CB-839 sensitivity (r=0.63, p=0.005), suggesting that the dominant pathway by which Gln supports tumor cell growth and survival is through its conversation to Glu by GLS. Consistent with this notion, CB-839 treatment caused a decrease in the intracellular pool of Glu and an increase in the intracellular pool of Gln. The extent of decrease in Glu was highly correlated with the effect of CB-839 on cell proliferation (r=-0.64, p=0.006). The in vivo antitumor activity of CB-839 was also evaluated in a xenograft model using the multiple myeloma RPMI-8226 model. Oral, twice daily dosing of CB-839 resulted in a significant decrease in tumor volume (70% tumor growth inhibition). As observed in vitro, CB-839 dosing caused an increase in tumor Gln levels and a decrease in the level of tumor Glu. In a dose dependent fashion, there was a correlation between tumor Gln accumulation, tumor GLS inhibition and antitumor activity. Efficacious doses of CB-839 were well tolerated with no effect on hematological cell counts or body weight. These data suggest that GLS inhibition with CB-839 may provide therapeutic benefit in hematological malignances and motivate the evaluation of this compound in clinical trials that include patients with B-cell malignancies.