GB3226, a novel, orally active, first-in-class small molecule inhibitor of ENL-yeats and FLT3 kinase for the treatment of Relapsed/Refractory Acute Myeloid Leukemia

Abstract

Abstract Background: The YEATS domain of ENL is an epigenetic reader protein that binds to acetylation sites on histones and exerts transcriptional control over genes that determine leukemogenesis and leukemia stem cell fate, including MEIS1, HOXA9 and MYC. Preclinically, genetic deletion or pharmacological inhibition of ENL-YEATS has demonstrated reduced leukemia burden and prolonged survival in both cell line-derived (CDX) and patient-derived xenograft (PDX) AML models, underscoring its potential as a therapeutic target. FLT3 is a receptor tyrosine kinase that normally plays a key role in controlling the survival, proliferation and differentiation of hematopoietic cells. Activating mutations in FLT3 (FLT3-ITD, TKD) are found in approximately 30% of adult AML patients where it acts as a key oncogenic driver. Dual targeting of ENL-YEATS and FLT3 may offer superior therapeutic benefit by inhibiting epigenetic and upstream oncogenic drivers versus FLT3 inhibition alone. Here, we describe the preclinical pharmacology and activity of GB3226, a first-in-class, orally bioavailable, small molecule, dual inhibitor targeting the ENL-YEATS domain and FLT3. Methods: GB3226 was developed through structure-based drug design (SBDD) as an optimized analog of a selective ENL-YEATS tool compound. In ligand binding assays, GB3226 inhibited ENL-YEATS (IC50=185 nM), FLT3 wild-type (IC50=7 nM), and mutant FLT3 D835Y (IC50<0.2 nM). GB3226 exhibited similar potency in inhibiting AF9, a structurally similar YEATS family member. GB3226 was inactive in a bromodomain panel. In a kinase panel of 370 kinases, GB3226 was active vs c-KIT, CLK1, CLK4, MELK, and TRKC (IC50=40-60 nM) and was inactive vs the remaining kinases. In cell-based assays, GB3226 potently inhibited proliferation of both MV4;11 (IC50=18 nM) and MOLM13 (IC50=32 nM) cell lines and was additive/synergistic when combined with standard-of-care agents, including venetoclax, gilteritinib, and revumenib. In vitro testing in AML patient samples showed that GB3226 is active across multiple genotypes, including AXSL1+, FLT3+, c-KIT+, NPM1c+ and TET2+. In a subcutaneous xenograft model where MV4;11 cells were implanted, oral administration of GB3226 (25, 50 and 100 mg/kg, BID, or 100 mg/kg or QD) resulted in significant, rapid tumor regression and suppression of MEIS1, HOXA9, and MYC. In a disseminated MV4;11 xenograft model, GB3226 dramatically reduced leukemia burden and significantly prolonged survival at comparable doses. GB3226 displayed favorable pharmacokinetic properties, good oral bioavailability across multiple species, no hERG inhibition (IC50>50 uM), and no QTc prolongation in canine cardiovascular safety studies. GB3226 was well tolerated in pilot 14-day repeat-dose toxicology studies in rats and dogs. Conclusion: We have developed GB3226, a novel, oral, first-in-class, dual inhibitor of ENL-YEATS and FLT3 for the treatment of AML. Preclinical studies demonstrate that GB3226 exhibits potent anti-leukemic activity and has the potential for combination use with multiple therapeutic agents across diverse AML genotypes. GB3226 has shown robust in vitro and in vivo efficacy, along with a favorable preclinical safety profile. These findings support further clinical development of GB3226 for the treatment of patients with relapsed or refractory AML.