Roles of immunosuppressive myeloid states in colorectal cancer checkpoint inhibitor non-response: single-cell and spatial proteomics, and reprogramming approaches.
Yu Dai, Xinyu Luo, Li Zhang, Jin Yan
Abstract
Open AccessColorectal cancer (CRC) exhibits a striking dichotomy in response to immune checkpoint inhibitors (ICIs), with durable benefit largely restricted to dMMR/MSI-H disease while most pMMR/MSS tumors remain refractory. Central to this resistance is an immunosuppressive myeloid ecosystem-dominated by SPP1high tumor-associated macrophages, neutrophils/PMN-MDSCs, and LAMP3+ mregDCs-that enforces chemokine-driven exclusion, attenuates antigen presentation, and sustains metabolite-mediated T-cell suppression. Despite advances in checkpoint blockade and biomarker stratification, these microenvironmental circuits constitute a major therapeutic hurdle. Moreover, single-cell and spatial proteomic platforms (e.g., CITE-seq, CODEX, imaging mass cytometry) now resolve the composition and topology of suppressive neighborhoods and highlight their utility for patient stratification and pharmacodynamic monitoring. This mini review summarizes cellular and spatial mechanisms by which myeloid states drive ICI non-response in CRC, emphasizing stromal TGF-β-coupled SPP1+ TAM programs, granulocytic chemokine axes (CXCL1/2-CXCR2; IL-8-CXCR1/2), and mregDC-mediated co-inhibition. We outline current and emerging myeloid-reprogramming strategies-including PI3Kγ and CSF1-CSF1R targeting, TREM2 antagonism, COX-2-PGE2 blockade, and adenosine-axis inhibition-and their integration with PD-(L)1 therapy, alongside single-cell/spatial endpoints to quantify on-treatment remodeling. The purpose of this mini-review is to provide a mechanistic and technology-informed framework to reference rational trial design and clinical translation for overcoming checkpoint resistance in CRC.