KRAS G12D Specific Inhibitors Relieve Erythroid Differentiation Block and Modulate Inflammatory Pathways at the Single Cell Level in Myeloid Malignancies

Introduction:

Myelodysplastic Syndrome (MDS) and Acute Myeloid Leukemia (AML) are characterized by a hematopoietic differentiation block that leads to anemia, a major source of morbidity and mortality. These myeloid malignancies and clonal hematopoiesis (CH) are associated with a myeloid bias and heightened inflammation, but the mechanisms of these phenomenon are not well elucidated. KRAS G12D (KRAS^G12D) mutation is seen in ~5% of MDS/AML and associated with a poor prognosis, currently with no approved precision medicine strategies. Novel KRAS^G12D inhibitors are being developed for solid tumors, but no data exists testing their efficacy in myeloid malignancies. Thus, we examined the impact of KRAS^G12D mutation and KRAS inhibition on human hematopoietic differentiation.

Results:

To first define the impact of KRAS^G12D on human hematopoiesis, we performed single-cell RNA-seq coupled with mutational calling via Genotyping of Transcriptomics (GoT) on primary KRAS^G12D samples. GoT of a CH sample with isolated KRAS^G12D revealed that the KRAS^G12D mutated (MT) cells were overrepresented in the monocytic compartment. MT cells also displayed an inflammatory signature, including elevated IL-8 expression.

GoT analysis of KRAS^G12D AML samples (N=3) revealed that the MT cells were enriched in a distinct AML subpopulation, defined by a stem/progenitor-like expression profile, with upregulated RUNX1, ETV6, JARID2, and ZEB2. Additionally, CD83, an immune checkpoint molecule, was elevated in this AML subgroup. Differential gene expression analysis between CD83⁺ MT vs. CD83⁺ WT cells revealed overexpression of IL-6, FCN1, CXCL3 in MT cells, suggesting an inflamed phenotype. Consistent with CD83 upregulation in the stem/progenitor-like subgroup, sorted CD83⁺ cells from KRAS^G12D primary MDS samples create more colonies in vitro when compared to sorted CD83^- cells.

To test the impact of KRAS-targeted therapy on isolated KRAS^G12D clones (without confounding co-occurring mutations), we treated the KRAS^G12D CH sample with a clinically valid KRAS^G12D inhibitor MRTX1133 and a pan-KRAS inhibitor BI-2865. Treatment with both inhibitors decreased total monocytic population while increasing erythrocytes at the single cell level. We validated this finding by treating primary KRAS^G12D mutant CH, MDS and AML samples (N=3) with both KRAS-inhibitors in clonogenic assays. Inhibition of KRAS^G12D led to increased erythroid colonies, associated with increased Glycophorin A and CD71 expression by flow cytometry. This demonstrates that KRAS-inhibition may rescue erythroid differentiation block.

As we observed that KRAS^G12D induced inflammatory signaling, we hypothesized that KRAS inhibition may rescue erythroid differentiation by suppressing inflammation. Consistently, the KRAS^G12D-induced inflammatory signature, including IL-8, was downregulated upon treatment with MRTX1133 and BI-2865. We therefore tested whether IL-8 antibody treatment may rescue erythroid differentiation. Indeed, IL-8 antibody in vitro treatment led to increased Glycophorin A and CD71 expression, suggesting that KRAS inhibition may rescue KRAS^G12D -mediated anemia via downregulation of IL-8-mediated inflammation.

Conclusion:

These data demonstrate that KRAS^G12D induces a monocytic differentiation bias associated with inflammation in CH. In AML, KRAS mutant cells are enriched in a novel stem progenitor-like inflamed population, characterized by CD83 positivity. We demonstrate the efficacy of novel, clinically relevant KRAS^G12D inhibitors in relieving IL8-mediated inflammation and erythroid differentiation block in KRAS^G12D neoplasms. These data support investigations of KRAS^G12D inhibitors in myeloid malignancy clinical trials. Ongoing studies test the efficacy of in vivo targeting of KRAS^G12D clones in murine and xenograft models.