OncoImmunology, vol. 6, issue 3 (2017) p. e1286436

Detailed resolution analysis reveals spatial T cell heterogeneity in the invasive margin of colorectal cancer liver metastases associated with improved survival.

Anna Berthela, Inka Zoernigb, Nektarios A. Valousa, Christoph Kahlertc, Fee Kluppd, Alexis Ulrichd, Juergen Weitzc, Dirk Jaegera,b and Niels Halamab*


aClinical Cooperation Unit “Applied Tumor Immunity”, National Center for Tumor Diseases (NCT) and German Cancer Research Center (DKFZ), Heidelberg, Germany;

bDepartment of Medical Oncology, National Center for Tumor Diseases (NCT) and University Hospital Heidelberg, Heidelberg, Germany;

cDepartment of Surgery, University Hospital Dresden, Dresden, Germany;

dDepartment of Surgery, University Hospital Heidelberg, Heidelberg, Germany

*Contact: Dr. Niels Halama, National Center for Tumor Diseases, Im Neuenheimer Feld 460, 69120 Heidelberg, Germany. Email: niels.halama@nct-heidelberg.de



Localization and high densities of T cells in colorectal cancer primary tumors and liver metastases as well are prognostic for patient survival and are associated with response to therapy. On tissue level, different patterns of distributions of T cells exist. Key questions are: are all T cells in the tumor microenvironment contributing to the improved survival? Or is there a spatial profile that is having a greater influence. T cell infiltration at higher resolutions in the direct vicinity of the tumor was therefore for the first time investigated. In order to delineate the effect of localization of the T cells, CD3+ T cell immunostainings of patient tissues were divided into 10 µm regions between the tumor border and adjacent normal liver following cell quantification. In the subsequent analysis, a decrease of T cells at 20 to 30 µm distance to the tumor could be detected. While cytotoxic Granzyme B was found to be significantly reduced, tumor-associated macrophages were increased within this region and identified to express immunosuppressive PD-L1. Therefore, our data suggests a regional physical or functional T cells border. Individual survival analysis of each distance class revealed improved overall survival in patients with high T cells numbers at the direct tumor border. Surprisingly, the decreased T cell numbers in the 20 to 30 µm region were also found to be significantly associated with improved survival. Consequently, the high-resolution analysis represents new insights into relevant heterogenous T cell distributions and those relations to clinical responses. The paradoxical observation of localization-dependent prognostic relevance of high and low T cell densities is only detectable by detailed spatial analyses and shows for the first time the spatial relevance of the presence of specific immune cell subsets. Therefore, the high resolution analysis is suggested as new prediction method for survival and response to therapies.

Keywords: T cell distribution, tumor microenvironment, colorectal cancer liver metastases, immunosuppression, tumor immunology, prognosis

PMID: 28405518



Metastatic colorectal cancer (CRC) accounts for more than 55,000 deaths per year in the United States alone. About 70% of the patients with liver metastases (LM) cannot undergo surgical resection leading to an overall survival rate of only 24 month under standard therapies.1 Interactions of the immune system with the tumor have an important influence on the clinical course of patients and high densities of tumor infiltrating T cells were shown to be correlated with improved survival in CRC-LM patients.2 Nevertheless, only few T cells can be found in direct contact with the tumor being generally engaged in direct anti-tumor effects, while more distant T cells might be blocked by immunosuppressive factors and exploited by the tumor.3 As novel immunotherapy approaches that target immune checkpoints to abrogate immunosuppression and reactivate T cells were less successful in CRC-LM patients so far,1 it is of high interest to further analyze T cells, their distribution and function in these tumors. Therefore, a more precise high resolution analysis of T cell distribution in the invasive margin of CRC-LM patients was performed within this study.

In brief, immunostainings of CRC-LM patient tissues were divided into ten consecutive 10 µm regions (distance classes) from the direct tumor border into the adjacent normal liver following quantification of CD3, a pan T cell marker, within each region (Fig. 1).



Figure 1: (A) Representative invasive margin of a CRC liver metastasis with annotations (CD3 positive lymphocytes appear dark brown). (B) The invasive margin of liver metastases is analyzed separately, encompassing 100 µm into normal adjacent liver from the tumor epithelium. Shaded areas highlight distance classes of 10 µm in relation to the tumor epithelium.



Figure 2: Boxplots of identified CD3 T cell densities in different distance classes. Statistically significant differences between adjacent distance classes are marked with asterisks.


We observed that in the 20-30 µm distance class CD3 cell densities significantly decreased, whereas in the distance classes closer than 20 µm and further away from the tumor epithelium higher CD3 cell densities were found (Fig. 2). To get a more detailed picture of this distribution pattern, we did a similar analysis in representative patient samples for the enzyme Granzyme B (GranB), which is primarily produced by the cytotoxic subtype of CD3 T cells (CD8) and involved in anti-tumor effects (Fig. 3a). Also GranB was found to be significantly decreased at the distance of 20-30 µm following a significant increase further away from the tumor. From these results we concluded that a physical or functional barrier for T cells might be present in 20-30 µm distance to the tumor.

The new observations prompted us to further investigate the possible factors involved in this heterogenous T cell distribution patterns and T cell blockade. As immunosuppressive regulatory T cell numbers are generally low in CRC-LM,2 we decided to repeat our spatial distribution analysis to examine the localization of CD163 macrophages. This tumor-associated macrophage (TAM) population is characterized by a tumor-promoting and immunosuppressive phenotype and associated with poor clinical outcome in many different tumors.4 In our exploratory analyses, we noticed an increase of CD163 cells at the 20-30 µm region, whereas no relevant differences between the other examined distances classes were observed (Fig. 3b). This opposite distribution of CD163 to the T cells in the 20-30 µm distance class led us to the assumption that CD163 cells fill the gap between the tumor distant and tumor proximal T cells thereby forming the physical barrier, which only a specific T cell subset can overcome. Even more important, this barrier could have functional impact, because it also marks the border, where cytotoxic T cell function (GranB) significantly decreased at the tumor margin. Indeed, we found by immunofluorescent double staining that the majority of CD163 cells in the invasive margin express the immune inhibitory checkpoint molecule programmed death-ligand 1 (PD-L1), which can directly suppress T cell responses through the immune checkpoint receptor PD-1 (Fig).4 From these data we concluded that TAM directly suppress T cells and consequently GranB production via immune checkpoints at 20-30 µm distance to the tumor.

Besides the factors involved in the observed T cell distribution, we also wanted to know, if the different distribution pattern in the distance classes have clinical impact. Therefore, we performed survival analyses comparing patients with high and low CD3 T cell densities within each distance class. We found that patients with high CD3 T cell densities within the >10 µm distance class had a significant survival benefit compared to patients with low T cells (Fig. 4). Interestingly, the opposite effect was found at the 20-30 µm distance to the tumor. Here, low CD3 T cell numbers had significant positive effects on patient survival. Consequently, despite the T cells barrier, the observed pattern of T cell distribution at 20-30 µm distance obviously is beneficial for patients. When considering this observation from the other side, high T cell numbers within the 20-30 µm region unexpectedly might have tumor promoting effects. Our own previous work illustrated the exploitation of T cells that produce the chemokine CCL5 thereby supporting CCL5 receptor expressing tumor cells.3 Hence, low T cell numbers could also be less tumor promoting. The results of this study indicate that T cell induced anti-tumor effects can only be conducted by T cells that have engaged the tumor margin at the 0-10 µm distance class, which is indeed also true for the cytotoxic CD8 subtype, while T cells further away remain blocked. Nevertheless, how low T cells numbers are involved in anti-tumor effects remains to be investigated. The here presented high resolution analysis reveals new insights into T cell distribution, function as well as influence of T cell localization on clinical responses and introduces possible spatial immunosuppressive hurdles. As paradoxical observations of localization-dependent prognostic relevance of T cell densities is only decipherable by detailed spatial analyses we suggest this method for CRC-LM patients regarding prediction of survival and response to therapies.


Summary and Outlook

For the first time, systematic high—resolution imaging and quantification is utilized specifically to analyze the positional effect of immune cells in the microenvironment. This analysis revealed the relevance of infiltrating T cells in direct contact with the tumor epithelium for an improved prognosis. Surprisingly, T cells at a distance 20 to 30 µm away from the tumor epithelium appear to have a detrimental effect on prognosis, which can be reconciled by recent work on the role of CCL5 produced by T cells in the microenvironment. Together the data implies an important role for the use of spatial profiling and quantification of immune cells in the microenvironment. With this methodology, higher precision for prognostic and predictive evaluation appears to be possible.



Figure 3:  Boxplots of identified GranB and CD163 cell densities in different distance classes. Statistically significant differences between adjacent distance classes are marked with asterisks.



Figure 4: Kaplan–Meier plots for estimated overall survival probabilities of CD3 T cell high and CD3 T cell low patient groups within all distance classes. Graphs indicate cumulative survival (y-axis) and survival in months (x-axis). P-values for statistically significant differences between groups are shown.



  1. Ahn DH, Goldberg RM. Colorectal clinical trials: what is on the horizon? Futur Oncol 2016; 12:525–31.
  2. Halama N, Michel S, Kloor M, Zoernig I, Benner A, Spille A, Pommerencke T, Von Knebel Doeberitz M, Folprecht G, Luber B, et al. Localization and density of immune cells in the invasive margin of human colorectal cancer liver metastases are prognostic for response to chemotherapy. Cancer Res 2011; 71:5670–7.
  3. Halama N, Zoernig I, Berthel A, Kahlert C, Klupp F, Suarez-Carmona M, Suetterlin T, Brand K, Krauss J, Lasitschka F, et al. Tumoral Immune Cell Exploitation in Colorectal Cancer Metastases Can Be Targeted Effectively by Anti-CCR5 Therapy in Cancer Patients. Cancer Cell 2016; 29:587–601.
  4. Ruffell B, Coussens LM. Macrophages and therapeutic resistance in cancer. Cancer Cell2015; 27:462–72.



The images and abstract are derived from an article published in Oncoimmunology 2017 by Taylor & Francis avaible freely online at http://www.tandfonline.com/doi/full/10.1080/2162402X.2017.1286436 .