Abstract

Solid tumours rewire systemic haematopoiesis, skewing bone marrow and splenic myeloid progenitors toward monocyte and granulocyte fates that seed the tumour microenvironment with immunosuppressive macrophages and myeloid-derived suppressor cells. Cytokines, growth factors, metabolites, hypoxia, and tumour-derived extracellular vesicles cooperate to imprint epigenetic and metabolic programmes in progenitors and their progeny. The resulting tumour-associated macrophages (TAMs) suppress T cell and NK cell function, remodel extracellular matrix, promote angiogenesis and metastasis, and blunt responses to immunotherapy. Here we outline key mechanisms of progenitor dysregulation, the phenotypic spectrum of suppressive macrophages, clinical implications, and therapeutic strategies to disrupt this axis or reprogramme myeloid cells toward antitumour activity.

Background: from progenitors to tumour-associated macrophages

In healthy adults, haematopoietic stem cells (HSCs) in the bone marrow give rise to common myeloid progenitors (CMPs), which differentiate into downstream lineages such as granulocyte–monocyte progenitors (GMPs) and monocyte–dendritic cell progenitors (MDPs). These progenitors generate circulating monocytes and tissue macrophages. Under inflammatory stress, “emergency myelopoiesis” accelerates this process to meet demand.

Tumours exploit this flexibility. By secreting myelopoietic cytokines and danger signals, cancers induce a chronic, systemic emergency programme that expands myeloid-biased progenitors. The output is disproportionately channelled into monocytes and granulocytes with a high potential to adopt immunosuppressive phenotypes once they reach the tumour or pre-metastatic sites.

How tumours dysregulate myeloid progenitors

• Cytokine and growth-factor signalling: Tumour-derived IL-6, GM-CSF (CSF2), M-CSF (CSF1), G-CSF (CSF3), VEGF, and TGF-β stimulate expansion and skewing of CMPs, GMPs and MDPs. Downstream transcription factors including STAT3, C/EBPβ, PU.1 and IRF families enforce monocyte/granulocyte fate at the expense of lymphoid and dendritic lineages.
• Metabolic and hypoxic cues: Lactic acid, adenosine, and hypoxia (via HIF-1α) modulate progenitor and myeloid-cell metabolism, often favouring oxidative phosphorylation and fatty-acid oxidation programs linked to tolerogenic macrophage states. Systemic tumour-induced cachexia can further alter myelopoietic niches.
• Extracellular vesicles and alarmins: Tumour-released exosomes carrying microRNAs and proteins (for example, S100A8/A9) reach the bone marrow and spleen, reprogramming progenitors toward suppressor trajectories and priming pre-metastatic niches.
• Niche remodelling and migration: Disruption of the CXCL12–CXCR4 axis facilitates HSC/progenitor egress from bone marrow to blood and spleen, where extramedullary haematopoiesis amplifies myeloid output. Sympathetic nervous system inputs and stromal cell changes reinforce this shift.
• Epigenetic imprinting: Chronic cytokine exposure establishes persistent chromatin states in progenitors (for example, via HDAC, DNMT, and TET pathways), creating a “memory” that biases their descendants toward immunosuppressive phenotypes even outside the tumour.

The spectrum of immunosuppressive macrophages in tumours

The simplistic M1/M2 dichotomy does not capture the heterogeneity of TAMs. Single-cell multi-omics reveal diverse, spatially organized macrophage states that coexist and interconvert. Common immunosuppressive features include:

• Checkpoint and inhibitory ligand expression: PD-L1, PD-L2, VISTA, and Galectin-9 dampen T cell activation; SIRPα engages CD47 to inhibit phagocytosis.
• Metabolic restriction of T cells: Arginase-1 depletes arginine; indoleamine 2,3-dioxygenase (IDO) disrupts tryptophan metabolism; ectonucleotidases (CD39/CD73) generate adenosine, suppressing effector functions.
• Immunoregulatory cytokines: IL-10 and TGF-β drive T cell exhaustion and promote regulatory T cell expansion.
• Antigen handling and presentation: High scavenger receptor activity (for example, MARCO, MerTK) with reduced co-stimulatory signals yields tolerogenic antigen presentation.
• Tissue remodelling and angiogenesis: VEGF, MMPs, and osteopontin foster abnormal vasculature, ECM deposition, and tumour cell invasion.

A related population, myeloid-derived suppressor cells (monocytic and granulocytic), shares many suppressive mechanisms and can differentiate into TAMs within tumours.

Systemic consequences and disease progression

• Pre-metastatic niche formation: Tumour cues mobilize and instruct myeloid cells to seed distant organs, where they remodel stroma, suppress local immunity, and facilitate tumour cell colonization.
• Resistance to therapy: TAMs can sequester or inactivate drugs, repair therapy-induced damage, and counteract T cell–mediated killing, reducing responses to chemotherapy, targeted therapy, and immune checkpoint blockade.
• Blood biomarkers: Elevated circulating myeloid progenitor signatures, increased monocytic/granulocytic MDSCs, high IL-6/GM-CSF levels, and a high monocyte-to-lymphocyte ratio often correlate with poor outcomes.

Mechanistic hallmarks of progenitor-to-TAM reprogramming

• STAT3-centric signalling: Persistent STAT3 activation in progenitors and descendants enforces suppressive gene programmes, including PD-L1, ARG1, and IL-10.
• C/EBPβ and IRF balance: C/EBPβ promotes emergency myelopoiesis and suppressor functions, while loss of IRF8 biases granulocyte/monocyte output at the expense of dendritic cells.
• Metabolic rewiring: FAO and oxidative phosphorylation support suppressive macrophage states; hypoxia/hyperlactatemia stabilize HIF-1α, reinforcing immunoregulation and angiogenesis.
• Epigenetic persistence: Cytokine-driven chromatin accessibility at enhancers primes progenitors so that even neutral environments yield immunosuppressive macrophages upon differentiation.

Therapeutic strategies: where to intervene

Therapies can target different tiers of the axis: the tumour signals that skew progenitors, the progenitors themselves, the recruitment of their progeny, or the suppressive state of TAMs in situ.

1) Blunt tumour-driven myelopoiesis

• Neutralize upstream cytokines such as IL-6, GM-CSF, or G-CSF in tumours that overproduce them.
• Modulate the bone marrow niche (for example, CXCR4 agonism to retain HSPCs) cautiously to avoid haematologic toxicity.
• Epigenetic modulators (for example, HDAC inhibitors, hypomethylating agents) at carefully titrated doses to reset progenitor programming.

2) Block myeloid cell recruitment

• CCR2/CCL2 axis inhibitors reduce monocyte influx; CCR5 and CXCR2 antagonists limit recruitment of suppressive monocytes and neutrophils.
• Target integrins and adhesion molecules involved in myeloid extravasation in select tumour types.

3) Deplete or reprogramme TAMs in tumours

• CSF1R blockade to reduce survival of CSF1-dependent macrophages; efficacy may require combination with checkpoint inhibitors.
• PI3Kγ inhibitors, CD40 agonists, TLR or STING agonists to convert TAMs into pro-inflammatory, antigen-presenting cells.
• Adenosine pathway inhibitors (A2A/A2B), arginase inhibitors, and COX-2/PGE2 blockers to relieve metabolic suppression.
• Antibodies against SIRPα or CD47 to enhance phagocytosis of tumour cells.
• Differentiation inducers such as all-trans retinoic acid to mature MDSCs and reduce suppressive potency.

4) Rational combinations with immunotherapy

• Pair PD-1/PD-L1 or CTLA-4 blockade with TAM reprogrammers (for example, PI3Kγ inhibitors or CD40 agonists) to unlock cytotoxic T cell activity.
• Coordinate timing with chemotherapy or radiation to avoid rebound myelopoiesis that replenishes suppressive myeloid pools.

Safety remains central: systemic interference with myelopoiesis risks neutropenia and infection. Biomarkers that track progenitor skewing and TAM states can guide dosing, timing, and patient selection.

Enabling technologies and translational biomarkers

• Single-cell multi-omics: scRNA-seq, scATAC-seq, CITE-seq and spatial profiling resolve progenitor and TAM states, lineage biases, and cell–cell interactions across tumour regions.
• Lineage tracing and barcoding: In vivo barcoding and fate mapping track progenitor contributions to TAM pools and reveal reprogramming bottlenecks.
• Circulating signatures: Blood-based cell-free RNA, vesicle cargo, cytokine panels, and flow cytometry of myeloid subsets provide minimally invasive readouts of myelopoiesis.
• Functional assays: Ex vivo T cell suppression assays and metabolic flux analyses quantify the impact of therapies on TAM function beyond surface markers.

Open questions

• Which tumour genotypes and microenvironmental contexts most strongly drive progenitor dysregulation?
• How stable is the epigenetic imprinting of progenitors, and can short therapeutic pulses durably reset it?
• What are the minimal biomarker sets that reliably predict benefit from TAM-targeted combinations?
• How do comorbidities such as chronic infection or clonal haematopoiesis alter the progenitor–TAM axis in cancer patients?
• Can engineered macrophages (for example, CAR-M therapies) overcome suppressive niches without being reprogrammed back to tolerance?

Clinical implications

A practical takeaway is that many patients labelled as “non-responders” to checkpoint blockade may harbour a dominant myeloid barrier rather than purely T cell–intrinsic resistance. Measuring and modulating the progenitor–TAM axis offers routes to:

• Improve response rates to immunotherapy by lifting myeloid suppression.
• Delay or prevent metastasis by disrupting pre-metastatic niche conditioning.
• Reduce relapse by preventing therapy-induced rebound myelopoiesis from re-seeding suppressive macrophages.

Conclusion

Tumour-driven dysregulation of myeloid progenitors creates a self-reinforcing pipeline that populates cancers with immunosuppressive macrophages. This axis acts at systemic and local levels to shape disease course and treatment response. Dissecting and intercepting the signals that skew progenitors, the routes that recruit their progeny, and the programmes that lock macrophages into tolerant states is central to next-generation immuno-oncology. With thoughtful combinations, biomarker guidance, and attention to haematologic safety, targeting the progenitor–TAM continuum can convert hostile microenvironments into ones that support durable antitumour immunity.