A High-Dimensional Window into the Micro-Environment of Triple Negative Breast Cancer

Abstract

Providing effective personalized immunotherapy for triple negative breast cancer (TNBC) patients requires a detailed understanding of the composition of the tumor microenvironment. Both the tumor cell and non-tumor components of TNBC can exhibit tremendous heterogeneity in individual patients and change over time. Delineating cellular phenotypes and spatial topographies associated with distinct immunological states and the impact of chemotherapy will be necessary to optimally time immunotherapy. The clinical successes in immunotherapy have intensified research on the tumor microenvironment, aided by a plethora of high-dimensional technologies to define cellular phenotypes. These high-dimensional technologies include, but are not limited to, single cell RNA sequencing, spatial transcriptomics, T cell repertoire analyses, advanced flow cytometry, imaging mass cytometry, and their integration. In this review, we discuss the cellular phenotypes and spatial patterns of the lymphoid-, myeloid-, and stromal cells in the TNBC microenvironment and the potential value of mapping these features onto tumor cell genotypes.

Keywords: breast cancer; immune profiling; single cell; spatial profiling; tumor evolution.

Figure 1

The spatial lymphocytic phenotypes of immune infiltration in triple negative breast cancer (TNBC). (A) A schematic illustration of a poorly-infiltrated (left-side) and inflamed (right-side) tumor with key markers and processes. (I) Hypoxia in the tumor promotes the production of pro-angiogenic factors, including TGF-β and vascular endothelial growth factor (VEGF), which can both modulate and inhibit effector T cells; (II) Low infiltration of lymphocytes occurs due to their impaired extravasation, e.g., through endothelial expression of FAS-ligand (FAS-L); (III) CAF mediation, e.g., via CAF-S1 fibroblasts, inhibit T cells directly and via the modulation of regulatory T cells; (IV) Effector T cell restriction through tumor associated macrophages (TAMs) and immuno-inhibitory molecules, such B7H4. B7H4 can be expressed on e.g., TAMs or tumor cells to inhibit effector T cells; (V) When T cells are present mainly in the stroma, this is characterized by the presence of TAMs without the expression of CD206 and neutrophils that express IL-17. Phenotypically, the stroma-restricted T cells express PD-1 and CD69; (VI) When tumors harbor also intratumoral infiltration, this is accompanied by activated TAMs with CD206 expression, inflammatory markers like granzyme B (GZMB) and interferon-g (IFN γ). Phenotypically, the intratumoral T cells express CD69, CD103 and multiple immune checkpoints, including PD-1, CTLA-4, and TIGIT. (B) In TNBC, tumors are often divided in poorly infiltrated (excluded and immune deserted, left panels) and inflamed (stroma restricted (SR) or stromal+intratumoral (SI), right panels). "Immune excluded" tumors show a lack of lymphocytes in the tumor, but lymphocytes are present at the invasive margin. Characteristics: Expression of TGF-β and immuno-inhibitory molecules, such as B7H4 exclude the immune cells. "Immune deserted" tumors show a total lack of lymphocytes at the invasive margin. Characteristics: Impaired extravasation of lymphocytes, CAF-S1 and hypoxia. "Inflamed-SR" tumors show lymphocytic infiltration in the stroma, but not intratumorally. Characteristics: PD-1 expression and CD69 on T cells, an increase in macrophages compared to the poorly infiltrated tumors, and neutrophil infiltration with IL-17 expression. "Inflamed-SI" tumors show lymphocytic infiltration in the stroma and intratumorally. Characteristics: High macrophage levels with an activated phenotype (CD206+), inflammatory molecules including GZMB and IFNγ. T cell expression of PD-1, CTLA-4, TIGIT and GITR, CD69 and CD103 (T_RM). The question mark between immune-excluded and inflamed-SR emphasizes the uncertainty whether these are actual biological distinct phenotypes or are similar but with a different scale or magnitude of infiltration.

Figure 2

Evolution of tumor cells and the tumor micro-environment (TME) in TNBC. (A) An inflamed-SI tumor with four tumor clones, stroma, and immune cells illustrated. The evolution of the compartments is visualized below (evolution figure adapted from Sun, Zheng, and Curtis, 2018) and spans from early disease, during therapy to distant spread. The primary tumor consists of 3 tumor clones. The clones are surrounded by stroma and immune cells. Typically, these compartments are described separate since methods to reconstruct tumor evolution are based on genomic heterogeneity amongst tumor subclones and lack direct information regarding immune evolution. (B) Here, the same tumor is illustrated, however tumor subclones and the surrounding stroma and immune cells are instead clustered into ecological niches, revealing their topography within the tumor-immune microenvironment. Such spatially-resolved cellular maps may provide insight into why some subclones expand, whereas others are spatially restricted, thereby informing patterns of metastatic "spread" and "therapy response" in TNBC and will be enabled by scalable spatial genomics techniques. Spread: Clone 1 disseminates to distant sites and form a macro metastatic lesion. Under (Niche A) clone 1 can be found in both the primary tumor and metastasis. Under (Niche B) the same tumor clone ("1") escapes but now we can observe that the tumor microenvironmental niche in the metastasis changes quickly (from Niche A in light grey, to Niche B in dark grey), indicating that the clone is the same but the micro-environment drastically changes, potentially leading to a more aggressive phenotype. Therapy response: In the primary tumor clone 2 and 3 both show different microenvironmental niches, which respond differently to, e.g., therapy. For example, tumor clone 2 (Panel A) is divided in Niche C and D (Panel B), where the niches show differential response to therapy. Niche D rapidly decreases during therapy, whilst Niche C is less sensitive and persists longer. Consideration of the tumor-immune microenvironment during tumor progression may enhance our understanding of the evolutionary dynamics and drivers of tumor evolution.