Beyond the Tumor Cell: The Shift to Ecosystem Profiling
Traditionally, oncology treated tumors as homogeneous clumps of mutating cells. Staging and treatment selection relied heavily on tumor-intrinsic features, such as specific genetic mutations or Tumor Mutational Burden (TMB). However, a tumor does not exist in a vacuum; it thrives within a complex ecological niche known as the tumor microenvironment (TME), populated by heterogeneous immune cells, fibroblasts, and signaling molecules.
A tumor's surrounding "soil" often exerts a greater influence over disease progression and therapeutic response than the tumor "seed" itself. For instance, in low-mutation cancers where neoantigen loads are sparse, Dean A. Fennell et al. demonstrated in Constitutive inflammation and epithelial-mesenchymal transition dictate sensitivity to nivolumab in CONFIRM: a placebo-controlled, randomised phase III trial that the pre-existing inflammatory architecture of the TME—such as B-cell-rich tertiary lymphoid structures—is a primary determinant of whether a patient responds to immunotherapy, completely overshadowing traditional genomic biomarkers. By shifting the analytical lens from single tumor mutations to the collective transcriptomic signatures of non-tumor cells, researchers can stratify patients into highly precise risk categories across different organ systems, as shown by Lin Zhou in Diagnosis and prognosis prediction model for digestive system tumors based on immunologic gene sets.
The Paradox of Infiltration and Functional Polarization
If immune cells are equipped to eliminate abnormal cells, one might assume that higher immune infiltration always correlates with a better prognosis. This is the classic distinction between "hot" (highly infiltrated) and "cold" (immune-excluded) tumors, a framework utilized by Nan Hu et al. in An Immune-Related Prognostic Signature Predicts Overall Survival in Stomach Adenocarcinomas. However, clinical observations present a paradox: highly infiltrated tumors can sometimes exhibit the most aggressive clinical trajectories and resistance to therapy.
The resolution to this paradox lies in functional polarization. Immune cells are highly plastic and can be subverted by the tumor to act as immunosuppressive agents. Consider tumor-associated macrophages (TAMs). In gastric cancer, Jinqing Li et al. found in Tumor-infiltrating macrophages express interleukin-25 and predict a favorable prognosis in patients with gastric cancer after radical resection that TAMs secreting Interleukin-25 (IL-25) actively recruit cytotoxic "killer" T cells, driving a favorable anti-tumor response. Conversely, Guohe Song et al. mapped the single-cell landscape of liver cancer in Global immune characterization of HBV/HCV-related hepatocellular carcinoma identifies macrophage and T-cell subsets associated with disease progression and identified a distinct $CCL18^+ CREM^+$ macrophage subset that actively suppresses immunity, facilitating tumor escape and rapid progression.
Therefore, microenvironment profiling works not merely by counting cells, but by decoding their specific functional states. As a pan-cancer census by Wei Chen et al. in Tumor-associated macrophage clusters linked to immunotherapy in a pan-cancer census proved, the ratio between pro-inflammatory and pro-tumor macrophage gene signatures yields vastly superior predictive accuracy for immunotherapy outcomes compared to traditional tumor mutational metrics.
Spatial Microlocalization and Structural Barriers
Knowing which cells are present within a biopsy is only half the battle; we must also understand where they are physically positioned. A tumor can be highly infiltrated on its periphery (the stroma) while remaining completely impenetrable at its core.
The spatial architecture of the TME dictates cellular interactions and therapeutic access. In bladder cancer, Hao Huang et al. demonstrated in Microlocalization and clinical significance of stabilin-1+ macrophages in treatment-naïve patients with urothelial carcinoma of the bladder that while Stabilin-1+ macrophages were highly abundant in the surrounding stroma, only the subpopulation that successfully penetrated intratumorally served as a significant predictor of poor survival and recurrence.
Furthermore, high-risk molecular profiles often correlate with dense, physical stromal structures that act as physical barricades. As noted by Zhihui Bai et al. in N6-Methylandenosine-related lncRNAs as potential biomarkers for predicting prognosis and the immunotherapy response in pancreatic cancer, high-risk pancreatic tumors frequently develop dense stromal walls that physically obstruct effector cells like B cells and Natural Killer (NK) cells from reaching the malignant core, transforming an otherwise "hot" genetic profile into an immune-excluded fortress.
Cellular Alliances and Coordinated Ecotypes
Immune and stromal cells do not operate in isolation; they form complex, self-sustaining networks or "alliances" that collectively dictate the microenvironment's state.
By applying machine learning and multi-modal transcriptomics to the TME, researchers can identify these coordinated cellular communities, often termed "ecotypes." For example, in pancreatic ductal adenocarcinoma, Wu Jun et al. uncovered in Integrating multi-modal transcriptomics identifies cellular subtypes with distinct roles in PDAC progression two opposing cellular axes: a tumor-promoting stromal axis composed of $POSTN^+$ fibroblasts and $SPP1^+$ macrophages, and a protective immune axis consisting of $CCL4^+$ CD8+ T cells and $IGHG1^+$ plasma cells. These cells actively signal to one another to reinforce their respective pro- or anti-tumor environments.
Mapping these multi-cellular assemblies provides a far more robust and stable biomarker framework than tracking individual cell types or single gene expressions. As Can Zhang et al. demonstrated in Multi-dimensional characterization of cellular states reveals clinically relevant immunological subtypes and therapeutic vulnerabilities in ovarian cancer, these macro-level ecotypes serve as highly accurate predictors of patient survival and drug sensitivity, offering a comprehensive blueprint for designing therapies that disrupt entire cellular networks rather than single targets.
Key Foundational Work
Tumor-infiltrating macrophages express interleukin-25 and predict a favorable prognosis in patients with gastric cancer after radical resection (2016) — contributed 1 concept elements to this direction.
Microlocalization and clinical significance of stabilin-1+ macrophages in treatment-naïve patients with urothelial carcinoma of the bladder (2019) — contributed 1 concept elements to this direction.
Golgi-Related Proteins GOLPH2 (GP73/GOLM1) and GOLPH3 (GOPP1/MIDAS) in Cutaneous Melanoma: Patterns of Expression and Prognostic Significance (2016) — contributed 1 concept elements to this direction.