Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster Display session

244 - Unsupervised analysis of the extent, organization and phenotype of tumor-infiltrating lymphocytes in breast cancer identifies two major clusters


14 Dec 2018


Poster Display session


Cinzia Solinas


Annals of Oncology (2018) 29 (suppl_10): x1-x10. 10.1093/annonc/mdy493


C. Solinas1, F. Richard2, S. Garaud1, P. De Silva1, A. de Wind3, G. Van Den Eyden1, C. Gu-Trantien1, M. Langouo Fontsa1, G. Noël1, A. Boisson1, C. Naveaux1, H. Duvillier1, L. Craciun3, D. Larsimont3, K. Willard-Gallo1

Author affiliations

  • 1 Molecular Immunology Unit, Institute Jules Bordet, 1000 - Brussels/BE
  • 2 Breast Cancer Translational Research Laboratory, Institute Jules Bordet, 1000 - Brussels/BE
  • 3 Anatomic Pathology Department, Institute Jules Bordet, 1000 - Brussels/BE


Abstract 244


Heterogeneous expression of immune checkpoint molecules by tumor-infiltrating lymphocytes (TIL) and within tertiary lymphoid structures (TLS) was previously demonstrated in breast cancer (BC). This study analyzed primary breast tumors for: TIL and TLS; LAG3 and TIM3 expression; and other clinicopathological parameters.


A cohort of untreated primary BC (N = 95) was prospectively analyzed. Hierarchical unsupervised clustering was performed on: the % of CD4+ and CD8+ TIL expressing LAG3 and TIM3 in fresh BC tissues (flow cytometry, FC); the % of stromal (str), intratumoral and global TIL, CD3+, CD4+, CD8+, CD20+ cells, and number of TLS scored on matched tissue sections dual-stained by immunohistochemistry (IHC; CD3/CD20; and CD4/CD8); age; estrogen receptor (ER) status; and BC clinical subtype. FC data was analyzed using Kaluza™ software. IHC stained tissues were evaluated by two pathologists blinded to the clinical data. Tumors were categorized as TILneg (<10% str-TIL), TILint (>10% and <50% str-TIL) and TILhi (>50% str-TIL) on IHC slides. Continuous marker data was scaled and z-scores were computed on log2transformed data. Clustering was based on Euclidian distances and Wald criteria.


Tumors in this cohort contain a majority of luminal BC (66%). Analysis of all parameters using the gap statistical method identified two optimal clusters, which were classified as TIL-poor and TIL-rich BC. The TIL-rich group (24%) contains the majority of TILhi tumors with increased numbers of TLS and fewer patients >70 years old, compared to the predominant TIL-poor group. LAG3 and TIM3 are expressed on T cells in a small fraction of BC in both TIL-rich and TIL-poor clusters, but rarely co-expressed by T cell subpopulations in the same tumors. LAG3 expression tends to be more frequently associated with TILint triple negative BC (TNBC). A higher proportion of TIL-rich tumors was found to express TIM3.


Our study identified TIL-poor and TIL-rich immune clusters in primary BC. TIL-rich BC has heterogeneous TIM3 and sporadic LAG3 expression. TILint TNBC from the TIL-poor cluster frequently express LAG3. These data suggest that LAG3 and TIM3 screening may be important for BC immunotherapy.

Editorial acknowledgement

Clinical trial identification

Legal entity responsible for the study

Molecular Immunology Unit, Institut Jules Bordet (Brussels).


Les Amis de l\'Institut Jules Bordet.


All authors have declared no conflicts of interest.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.