Resistance Mechanisms to SYK Inhibition in AML

Alterations in signaling pathways are critical to the pathogenesis of acute myeloid leukemia (AML) and are often driven by aberrant kinases. We previously identified spleen tyrosine kinase (SYK), a non-receptor, cytoplasmic tyrosine kinase, as a druggable target in AML and a critical regulator of FLT3, the most commonly mutated receptor tyrosine kinase in this disease. SYK was initially described as having an important role in B-cell development. More recent studies showed a broader relevance of SYK in hematopoietic signaling. Multiple orally bioavailable SYK inhibitors, including entospletinib (GS-9973) and TAK-659, are currently in Phase I/II clinical trials in patients with AML, with promising results. Targeted therapy, however, is typically associated with the emergence of resistance, and combination therapy is almost always needed for a durable therapeutic response. The aim of our study, therefore, was to identify resistance mechanisms to SYK inhibition in AML and to identify new synergistic drug combinations to overcome them.

We conducted a genome-scale, pooled open reading frame (ORF) library screen to investigate the mechanisms of drug resistance to SYK inhibition. We screened the AML cell lines MV4;11 and MOLM-14, which are both highly sensitive to SYK inhibitors, with entospletinib after transduction with the ORF library. Expression of KRAS and HRAS ORFs, as well as other RAS signaling pathway-related proteins, emerged as conferring the strongest resistance to entospletinib in both cell lines. Upon overexpression of NRAS and KRAS in MV4;11 and MOLM-14 cells, both wildtype for RAS, we observed i) upregulation of phospho ERK 1/2, indicating activated RAS signaling, and ii) we confirmed markedly reduced sensitivity to entospletinib. We next screened 12 AML cell lines for sensitivity to entospletinib and found that the presence of activating mutations of the RAS signaling pathway was associated with innate resistance to entospletinib.

To further study the mechanism of resistance to SYK inhibition, we generated entospletinib resistant cell lines by exposing MV4;11 cells to increasing concentrations of the drug for five months. In these cells, the entospletinib IC₅₀ shifted from 500 nM to 5 μM, and these cells were also cross resistant to the SYK inhibitor PRT062607. RNA-sequencing showed increased expression of NRAS, KRAS and related pathway genes in entospletinib resistant compared to sensitive parental cells. Accordingly, gene set enrichment analysis revealed significant upregulation of RAS signaling-related gene sets. These findings were confirmed by qPCR and western blot analysis, supporting that the activation of the RAS/MAPK signaling pathway also confers acquired resistance to SYK inhibition in AML.

Because RAS mutant AML cells have been reported to be sensitive to MEK inhibition, we next tested whether MEK inhibition using PD0325901 would recapture response to SYK inhibition in RAS/MAPK activated AML cells. Indeed, entospletinib was synergistic with PD0325901 in the MV4;11 entospletinib resistant cells. Moreover, entospletinib was synergistic with PD0325901 in five different AML cell lines, including AML cell lines with RAS pathway mutations. Additionally, this drug combination was synergistic when tested in a patient sample with NRAS G12D, KRAS G12D and PTPN11 G60V mutations.

In conclusion, we found that RAS pathway activation results in resistance to entospletinib in AML that may be overcome with the combination of a SYK inhibitor with a MEK inhibitor.