ABSTRACT: Myeloid cells, in particular monocytes and macrophages, are well-described to suppress the antitumor immune response. However, the molecular mechanisms controlling immunosuppressive myeloid cell states are ill- defined, hampering the development of myeloid-targeted therapies for cancer1,2. Here, we transcriptionally profiled over 500,000 tumor infiltrating leukocytes in non-small cell lung cancer (NSCLC) patients and in the KrasG12DTp53-/- (KP) murine lung adenocarcinoma model. In both species, the Type 2 cytokine IL-4 was predicted to be the primary driver of tumor infiltrating monocyte-derived macrophage (mo-mac) phenotype, despite the fact that lung tumors were largely devoid of IL-4 producing cells. Using a panel of conditional knockout mice, we found that only deletion of IL-4Ra within early myeloid progenitors in bone marrow (BM) reduced lung tumor burden, while deletion of this receptor in downstream mature myeloid cells had no effect. Detailed transcriptional analysis followed by mechanistic studies in vivo revealed an essential role for local BM IL-4 signaling in reprograming myelopoiesis in cancer. Mechanistically, basophils and eosinophils within BM upregulated IL4 production upon sensing distal tumor cues; local BM IL-4 acted on granulocyte-monocyte progenitors to transcriptionally program the development of immunosuppressive myeloid cells, which then homed to the tumor and promoted tumor growth. Consequentially, specific depletion of basophils, which were enriched in BM but absent from the tumor, profoundly reduced tumor burden and normalized myelopoiesis in experimental lung tumors. Prompted by these results, we designed and initiated the first clinical trial of dupilumab, a humanized IL- 4Ra blocking antibody commonly used for atopic disease, given in conjunction with PD-(L)1 blockade in relapsed/refractory NSCLC patients who had progressed on standard chemoimmunotherapy combinations. Consistent with our preclinical mouse data, dupilumab drove a reduction in circulating monocytes. This was coupled with an expansion of circulating and tumor-infiltrating CD8 T cells. Notably, one out of the six patients who enrolled in Phase 1b of this trial exhibited significant decrease in his tumor burden after two months of dupilumab treatment, and the clinical response deepened further to a near complete response even after cessation of dupilumab, suggestive of successful reprogramming of the antitumor response. Collectively, our study defines a role for IL-4 signaling in lung tumor progression, identifies a central axis controlling immunosuppressive myelopoiesis in cancer, and highlights a novel combination therapy for immune checkpoint blockade in humans.
