At the recent CIMR International Immunology Symposium, Professor Florent Ginhoux was invited to deliver a featured lecture entitled "Harnessing Tumor Macrophage Heterogeneity". Professor Ginhoux has long worked on the origin, differentiation, and tissue-specific functions of the monocyte-macrophage system. In his lecture, he focused on the dynamic behavior of tumor-associated macrophages in pancreatic ductal adenocarcinoma and highlighted how time and location influence their differentiation within tumors.

Background

Macrophages were first recognized as professional phagocytes, cells that remove pathogens and debris and help maintain tissue integrity. Over time, this view expanded considerably (Fig 1). Rather than serving only as innate immune scavengers, macrophages are now understood as highly adaptable cells that participate in tissue development, homeostasis, repair, and disease. As Siamon Gordon famously noted, the macrophage system can be thought of as an ancient and dispersed homeostatic network, on par with the nervous and endocrine systems.

This broader perspective emerged gradually. Early studies treated macrophages as a relatively uniform population, but the field has since moved toward a much more refined understanding of their diversity. The use of lineage tracing and single-cell technologies has revealed that macrophages are not all generated in the same way. Some are derived from embryonic progenitors and seed tissues before birth, while others arise from adult bone marrow monocytes as circulating monocytes and are recruited into tissues later in life, where they differentiate into macrophages (Fig 2). This distinction between tissue-resident macrophages and monocyte-derived macrophages has become central to modern macrophage biology.

In healthy organs, tissue-resident macrophages are long-lived inhabitants that adapt to local cues and support tissue-specific functions. They help regulate organ development, control vascular and stromal homeostasis, and contribute to the maintenance of local physiological balance. During inflammation or cancer, however, circulating monocytes are recruited into tissues and differentiate into macrophages under the influence of the local environment. These cells often acquire phenotypes that are distinct from resident macrophages and may either support or restrain disease progression.

The role of macrophages in cancer became especially prominent with the recognition of tumor-associated macrophages, or TAMs. By the time TAMs were established as major components of the tumor microenvironment, it had become clear that they were often linked to tumor progression, angiogenesis, immune suppression, and therapeutic resistance. Yet TAMs were often treated as a relatively homogeneous population. As the field matured, this simplification proved insufficient. Tumors are not uniform tissues, and the macrophages within them are not uniform either. Their behavior depends on developmental origin, local niche, and the stage of tumor evolution.

Research

This is where the work of Ginhoux and colleagues offers important insight. In a 2024 study in Science Immunology, their team examined how monocytes enter pancreatic ductal adenocarcinoma (PDAC) and how they progress into distinct TAM states. PDAC is a particularly informative setting because it is strongly shaped by dense stroma, hypoxia, necrosis, and immune exclusion. These features create a highly structured microenvironment in which macrophage fate can be traced with unusual clarity.

Using monocyte fate-mapping, time-stamping, single-cell transcriptomics, high-dimensional flow cytometry, and spatial analyses, the team showed that monocytes do not convert directly into mature TAMs in PDAC model. Instead, they first pass through a transient intermediate state, referred to as intermediate TAMs, or IntTAMs (Fig 3). These cells then give rise to at least two more stable TAM populations. This finding is important because it shows that macrophage heterogeneity in tumors is not only a matter of distinct end states, but also of developmental trajectory.

The study further showed that time matters. When monocytes enter the tumor at different stages, they contribute differently to the TAM pool. The intermediate population emerged soon after monocyte entry and represented a transient differentiation stage, whereas the more differentiated TAM states accumulate as the tumor develops. This temporal structure suggests that the tumor microenvironment does not simply contain macrophages at different activation levels. It actively guides them through a sequence of differentiation steps, creating specialized cell states over time.

Location matters as well. Spatial transcriptomics and multiplex imaging revealed that the TAM populations occupy different niches within the tumor (Fig 4). One subset was enriched in hypoxic and necrotic areas, while another was more abundant outside the necrotic core. This spatial segregation indicates that local microenvironments shape macrophage identity. Hypoxia, necrosis, stromal remodeling, and cell-cell interactions likely contribute to the emergence of distinct macrophage programs. In other words, TAMs are not only products of lineage and time, but also of geography within the tumor.

The human relevance of these findings is equally notable. Analysis of human PDAC datasets identified macrophage populations resembling the murine intermediate and terminal TAM states (Fig 5-A). These human macrophage subsets also showed spatial organization similar to that seen in mice. Importantly, some of the TAM signatures associated with hypoxic or necrotic niches correlated with worse clinical outcome, underscoring their potential translational significance (Fig 5-B). This suggests that the developmental logic observed in mouse PDAC may reflect a conserved biological principle rather than a mouse-specific phenomenon.

The broader implication is that TAM biology must be understood as a dynamic process rather than a fixed classification. For many years, therapeutic strategies in oncology focused on broadly suppressing macrophages or blocking their recruitment. While such approaches remain of interest, they may be too blunt if the cells being targeted are heterogeneous in origin, stage, and function. A more precise understanding of when and where TAMs arise may help identify more selective therapeutic windows. Targeting the transition from monocytes to intermediate TAMs, or selectively reprogramming macrophages in specific tumor niches, may prove more effective than global depletion.

This perspective also returns attention to tissue-resident macrophages, which remain essential for a complete picture of tumor immunity. Although monocyte-derived TAMs often dominate in advanced tumors, resident macrophages are already present in tissues before malignant transformation begins. Their long-term residence, close contact with stromal and parenchymal cells, and participation in tissue organization suggest that they may shape the earliest stages of tumor development. In some settings, resident macrophages may help establish permissive niches, influence vascular or stromal remodeling, or regulate inflammatory thresholds that later affect tumor progression. Their contribution may be less visible than that of recruited TAMs, but it is likely no less important.

Conclusion

Taken together, recent work has transformed macrophages from a broadly defined immune population into a highly dynamic system whose properties are governed by origin, timing, and location. The study from Ginhoux's group illustrates this shift particularly well. It shows that monocytes entering PDAC pass through an intermediate state before becoming distinct TAM subsets, and that these populations are organized by the local tumor landscape. At the same time, it reminds us that tissue-resident macrophages remain critical actors in tissue physiology and disease.

For cancer research, this means that macrophages can no longer be viewed as a single therapeutic target. They are a family of cells with different developmental histories and distinct roles within the tumor ecosystem. Understanding how these cells are shaped by their niches may ultimately help move macrophage-based therapy from broad intervention toward precise immunological engineering.

Reference

1. Cassetta L, Pollard JW. A timeline of tumour-associated macrophage biology. Nat Rev Cancer. 2023 Apr;23(4):238-257. doi: 10.1038/s41568-022-00547-1

2. Nahrendorf M, Ginhoux F, Swirski FK. Immune system influence on physiology. Science. 2025 Aug 7;389(6760):594-599. doi: 10.1126/science.adx4380

3. Park MD, Silvin A, Ginhoux F, Merad M. Macrophages in health and disease. Cell. 2022 Nov 10;185(23):4259-4279. doi: 10.1016/j.cell.2022.10.007

4. Shakiba M, Tuveson DA. Macrophages and fibroblasts as regulators of the immune response in pancreatic cancer. Nat Immunol. 2025 May;26(5):678-691. doi: 10.1038/s41590-025-02134-6

5. Dunsmore G, Guo W, Li Z, Bejarano DA, Pai R, Yang K, Kwok I, Tan L, Ng M, De La Calle Fabregat C, Yatim A, Bougouin A, Mulder K, Thomas J, Villar J, Bied M, Kloeckner B, Dutertre CA, Gessain G, Chakarov S, Liu Z, Scoazec JY, Lennon-Dumenil AM, Marichal T, Sautès-Fridman C, Fridman WH, Sharma A, Su B, Schlitzer A, Ng LG, Blériot C, Ginhoux F. Timing and location dictate monocyte fate and their transition to tumor-associated macrophages. Sci Immunol. 2024 Jul 26;9(97):eadk3981. doi: 10.1126/sciimmunol.adk3981