The immune microenvironment in hormone positive, HER2 negative breast cancer: A focus on tumour infiltrating lymphocytes, spatial phenotypes and tertiary lymphoid structures.

Abstract

2522 Background: Hormone positive, HER 2 negative breast cancer (ER+/HER2-) has traditionally been considered an immune cold tumour subtype. Research to date indicates that TILs in ER+/HER2- breast cancer may be a negative prognostic marker. However, there is a paucity of data in this area regarding immune subpopulations and spatial phenotypes. Tertiary lymphoid structures (TLS) have been associated with improved prognosis and response to immunotherapy in other solid organ malignancies. We aim to explore the immune microenvironment in early stage ER+/HER2- breast cancer with a focus on tumour infiltrating lymphocytes, spatial phenotypes and tertiary lymphoid structures (TLS). Methods: Full face sections of archival tissue samples from patients with early stage ER+/HER2- breast cancer were stained with CD45 antibody (DAKA clone M0701) via the automated Leica Bond III system. Open source digital pathology analysis software, Qupath, was utilised to identify positive cells and calculate the percentage of positive CD45 staining cells (CD45%) within the tumour area. Spatial analysis was done to divide tumour area into three spatial phenotypes; Immune cold -lacking immune infiltrate;Immune excluded- immune cells restricted to the tumour periphery and immune hot - immune cells infiltrating the tumour core. A tissue micro array (TMA) was created and stained manually with a multiplex chromogenic immunohistochemistry panel for CD20/CD3/DCLamp to identify immune subpopulations associated with tertiary lymphoid structures. Results: A total of 409 patient samples were included in the analysis. 89% (n=367) of samples had a CD45% of 0-10%. CD45% stratification by 21-gene-recurrence score(RS) into Low, Intermediate and High showed a statistically higher CD45% in those with a High RS compared to Low RS (p value= 0.0015). We identified a trend towards worse disease free survival (DFS) in the CD45%high group which did not reach statistical significance. The immune hot spatial phenotype was associated with a worse DFS (p value = 0.034) with no statistically significant difference in DFS between immune cold and immune excluded phenotypes. The majority of samples had minimal immune infiltrate. Staining with a TLS panel revealed that samples with a significant infiltrate could broadly be divided into T cell only infiltrate or those with the presence of TLS, the significance of this is unclear. Conclusions: The majority of ER+/HER2- breast cancer samples had low numbers of TILs, however there is clear heterogeneity within this subgroup. Spatial elements and immune cell subpopulations may further guide the value of TILs in prognosis in this breast cancer subtype.