AUTX703, a Novel and Potent KAT2A and KAT2B Protein Degrader, Induces Differentiation and Offers Survival Advantage in a Primary Human AML Xenograft Model

Phenotypic plasticity, or the ability of cancer cells to switch between cell states, has recently been recognized as a key hallmark of cancer. Cancer cells can hijack normal developmental pathways, preventing terminal differentiation which pushes cells to a more proliferative, plastic cell state similar to those found in early development. To identify therapeutic targets driving tumor plasticity, we utilized our proprietary AI/ML platform to create a high-resolution atlas of normal human development. Tumors were mapped to the atlas using our CLASSIFY algorithm to define their altered cell state and our DRIVE algorithm was used to identify the genes driving the oncogenic cell state. Our platform identified KAT2A, a lysine acetyltransferase, as a key driver in AML.

KAT2A and its paralog lysine acetyltransferase 2B (KAT2B) are epigenetic enzymes involved in normal human development and have been shown to play a critical role in regulating somatic cell fate decisions. Histone acetylation can activate transcriptional programs necessary to drive cell fate. KAT2A expression has been shown to homogenize transcriptional programs to sustain stem cell phenotypes, while loss of KAT2A has been shown to promote differentiation.

Genetic validation of KAT2A as a therapeutic target was performed in the AML cell line, MOLM-13, where knockdown resulted in significant cell growth inhibition and induction of cell state changes consistent with a more mature differentiated phenotype. Following traditional drug discovery efforts, AUTX-703 was discovered as a highly potent, selective and orally bioavailable protein degrader of KAT2A and KAT2B. In MOLM-13 cells degradation of KAT2A and KAT2B has a half maximal degradation concentration (DC₅₀) of <0.1nM and 90% degradation (DC₉₀) of 1nM. Treatment at 1nM resulted in growth inhibition (GI₅₀) over 7 days and significant increases in the monocytic marker, CD86. By RNA-Seq, significant increases in myeloid and granulocytic differentiation markers were also observed. In ex vivo treatment of CD34+ blasts from multiple AML patients, AUTX-703 potently inhibited growth with a GI₅₀ <1nM and induced monocytic differentiation as measured by CD11b, as well as by morphology, over 11 days of culture.

AUTX-703 was dosed in a primary patient AML xenograft model. NSG mice were engrafted and then treatment initiated once peripheral blasts reached ~ 3%. Three dosing regimens were administered in addition to vehicle control. After 4 weeks of dosing, vehicle mice had a significant increase in peripheral blast cells and began to succumb to disease burden. Over 8 weeks, a dose dependent survival was seen, and a significant survival advantage remained at 8 weeks in two of the three dose groups. Monocytic differentiation was observed in bone marrow and AUTX-703 was well tolerated at all doses.

These data provide evidence that degradation of KAT2A and KAT2B can reverse the block in cellular differentiation in primary patient AML cells which could provide therapeutic benefit to patients. A Phase 1 clinical trial testing the safety and tolerability of AUTX-703 is planned in AML.