Work

A single-cell-informed framework maps context-dependent myeloid damage-response states across cancers

Year: 2026

Type: article

Abstract: Tumor-infiltrating myeloid cells occupy heterogeneous functional states that are not adequately represented by macrophage abundance or fixed polarization labels. We developed the myeloid damage response index (MDRi), a single-cell-informed framework comprising injury, resolution and antigen-presentation/interferon (APC/IFN) programs, together with a derived injury–resolution axis. MDRi was established in an immune checkpoint blockade-treated multi-cancer atlas containing 47,750 myeloid cells from 192 samples across eight cancer types and assessed separately in an independent multi-cancer myeloid dataset. The programs organized recurrent macrophage and monocyte states along related but non-identical functional dimensions. Non-overlapping antigen-presentation and interferon-response signatures provided gene-independent support for the APC/IFN dimension. Root-sensitivity analyses showed stable state connectivity but root-dependent pseudotime directionality, supporting transcriptional topology rather than a universal developmental sequence. Exploratory response analyses showed concurrent elevation of injury, resolution and APC/IFN scores in post-treatment non-responders, whereas cancer- and regimen-adjusted analyses did not support an independent predictive effect. Complementary T/NK-cell analyses indicated that response-associated immune differences were more evident in pretreatment samples. In primary-tumor TCGA cohorts summarized at the patient level, joint multivariable Cox models containing MDR injury, MDR resolution and MDR APC/IFN identified MDR injury as an adverse factor in selected cancers, whereas resolution and APC/IFN showed cancer- and endpoint-dependent associations. The derived injury–resolution axis was excluded from these joint models because it is mathematically dependent on injury and resolution. Benchmarking against immune-deconvolution and tumor-associated macrophage signatures demonstrated partial but non-uniform overlap. Nasopharyngeal carcinoma Visium and multi-cancer Xenium analyses further revealed platform- and cancer-dependent spatial distributions of MDRi-related programs. MDRi Explorer provides an open-source implementation for scoring, reference comparison and exploratory clinical analysis. MDRi therefore represents a reusable, hypothesis-generating framework for investigating context-dependent myeloid functional organization rather than a universal prognostic signature or clinically validated predictor.

Source: Cancer Immunology Immunotherapy

DOI: 10.1007/s00262-026-04543-4

Authors: Ran Ding, Weihong Zheng, Zhiqing Long, Zhangqi Cao, Jianye Liang, Qi Quan

Institutions: South China University of Technology, Sun Yat-sen University, Sun Yat-sen University Cancer Center, State Key Laboratory of Oncology in South China

Language: English

Topic: Immune cells in cancer

Subfield: Immunology

Field: Immunology and Microbiology

Domain: Life Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $4,590