Background

Somatic NRAS mutations occur in 3-5% of human cancers and are prevalent in acute myeloid leukemia (AML) and other hematologic malignancies. NRAS mutations are key drivers of de novo and adaptive resistance to BCL-2, FLT3, and IDH1/2 inhibitors. Recent retrospective analysis of pediatric AML clinical trial data has found that cooperating NRAS mutations confer subgroup-specific poor prognosis, increasing the urgency to address this clinical need. There are no approved targeted therapies for NRAS-mutant cancers.

N-Ras subcellular localization and activity depends on a series of tightly regulated post-translational modifications, including a dynamic cycle of palmitoylation by palmitoyl-acyl transferases and depalmitoylation by ABHD17 serine hydrolases. Disruption of either side of this cycle leads to N-Ras mis-localization. However, the K-Ras4b isoform does not undergo palmitoylation, suggesting an exploitable therapeutic index for palmitoylation cycle inhibition in NRAS-mutant cancers. Consistent with this hypothesis, prior in vitro experiments using ABD957, a pyrazole urea-based tool compound that inhibits ABHD17 enzymes, revealed growth inhibition of NRAS- but not KRAS-mutant AML cells. Here, we evaluated the in vitro and in vivo efficacy of ABD778, an analog of ABD957 with improved drug-like pharmacologic properties, as a single agent and in combination with the allosteric MEK inhibitor PD0325901 (PD901).

Results

Exposing AML cells to ABD778 blocked N-Ras depalmitoylation in 17-octadecynoic acid pulse-chase experiments (EC50 ~53 nM), revealed selective target engagement of ABHD17 serine hydrolases by activity-based protein profiling, and showed sustained in vivo inhibition of ABHD17 activity for up to 24 hours after dosing.

ABD778 showed nanomolar potency, but incomplete effect in the NRAS-mutant AML cell lines OCI-AML3, HL60, and THP1 (Emax ~50%). Consistent with in vitro studies of prior ABHD17 inhibitors, ABD778 had no effect on KRAS-mutant NB4, SKM1, and NOMO1 AML cells. ABD778 and PD901 synergistically and selectively inhibited the growth of NRAS-mutant AML cell lines as assessed by Bliss independence and Chou-Talay methods, which correlated with enhanced suppression of ERK and S6 phosphorylation. We confirmed these functional and biochemical data in an isogenic AML cell line model engineered to express physiologic levels of oncogenic N-Ras or K-Ras proteins.

To extend these in vitro data, we performed preclinical trials in congenic recipient mice transplanted with 5 independent primary mouse AMLs harboring a NrasG12D (n = 3) or KrasG12D (n = 2) driver mutation. Mice continuously received ABD778 (60 mg/kg), PD901 (2.5 mg/kg), this combination, or vehicle by oral gavage. The combination was well tolerated. ABD778 modestly prolonged survival in Nras-mutant AMLs (median survival of 14 days and 11 days, respectively, p=0.0168). Recipient mice treated with ABD778 and PD901 had double the median survival of mice dosed with PD901 alone (median survival of 40 days and 20 days, respectively, p<0.0001). We observed no significant single-agent or combination activity of ABD778 in Kras-mutant AMLs.

Mice treated with ABD778 and PD901 eventually died from progressive leukemia. Re-transplanting NrasG12D AMLs from mice that received ABD778 and PD901 confirmed intrinsic drug resistance upon re-treatment. Subsequent whole exome sequencing uncovered on-pathway candidate resistance mutations that included KrasA146T and BrafG466E. Expressing the respective mutant proteins in NRAS-mutant OCI-AML3 cells caused adaptive resistance to the ABD778/PD901 combination that was associated with persistent ERK phosphorylation.

Conclusions

These studies validate ABHD17 enzymes and - more broadly - the N-Ras palmitoylation cycle as a therapeutic target in AML. Post-translational N-Ras processing is a novel druggable vulnerability in AML that may extend to other cancers harboring oncogenic NRAS mutations. We are currently interrogating ABD778 in patient derived xenograft AML models and exploring combinations with other Ras pathway inhibitors.

Disclosures

Niphakis:Lundbeck: Current Employment. Cravatt:BMS: Research Funding; Pfizer: Research Funding; Janssen: Consultancy, Research Funding; Eisai: Consultancy; Vividion (Bayer): Consultancy, Research Funding.

This content is only available as a PDF.
Sign in via your Institution