1804 - Immunological differences between immune-rich estrogen receptor-positive and -negative breast cancers

Background

A subset of estrogen receptor-positive (ER+) breast cancers have significant tumor infiltrating lymphocytes (TILs), similar to triple-negative breast cancer (TNBC). We examined differences in the immune microenvironment of immune-rich ER+ and immune-rich TNBC to find out if similar or different immunotherapy strategies are appropriate for these distinct disease types.

Methods

ER+/HER2-negative and TNBC cases were obtained from The Cancer Genome Atlas (TCGA) (n = 818 RNA Seq) and METABRIC (n = 1465 microarray). An immune gene expression score, as surrogate for TILs, was calculated for each case (Danaher et al). Signature scores were correlated with histologic TILs (R = 0.44, p < 0.001) for available cases. All cases in the top 25% of signature scores were considered immune-rich. We compared 22 immune cell populations between immune-rich TNBC (n = 86) and ER + (n=119) using CIBERSORT (Student’s t-test, FDR adjusted). We examined differential expression of 770 immune-related genes and 137 immuno-oncology (IO) drug targets. Macrophage abundance was measured by quantitative immunofluorescence (QIF) using pan-macrophage marker CD68 in 169 TNBC and 87 ER+ FFPE tissues.

Results

In TCGA and METABRIC, CIBERSORT showed more M0 (p = 0.015, p = 0.0043) and M1 macrophages (p = 9.4e-08, p = 6.24e-11) in immune-rich TNBC compared to ER+. Mast cells (p = 0.0093, p = 4.09e-15) and M2 macrophages (p = 4.4e-05, p = 0.053) were more abundant in immune-rich ER+. QIF confirmed higher macrophage content in TNBC (p = 0.0001). In both datasets, 36 IO targets were higher expressed in TNBC and 15 in ER+ cases (Table). Notably, coordinated high expression of TGFb pathway members TGFb3, TGFb-R2, and LRRC32 was seen in ER+ cancers and correlated positively with M2 and negatively with M1 macrophage content across all cases.Table:

185PD

Conclusions

IO drugs targeting TGFb, M2 macrophages and mast cells are attractive therapeutic candidates in immune-rich ER+ breast cancer based on the expression characteristics of these targets.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Howard Hughes Medical Institute Medical Fellows Grant.

Disclosure

V. Pelenkanou: Full / Part-time employment: Sanofi, US. D. Rimm: Advisory / Consultancy: Amgen; Advisory / Consultancy, Research grant / Funding (institution): AstraZeneca; Advisory / Consultancy: Agendia; Advisory / Consultancy: Biocept; Advisory / Consultancy: BMS; Advisory / Consultancy: Cell Signaling Technology; Advisory / Consultancy, Research grant / Funding (institution): Cepheid; Advisory / Consultancy: Daiichi Sankyo; Advisory / Consultancy: GSK; Advisory / Consultancy: InVicro/Konica/Minolta; Advisory / Consultancy: Merck; Advisory / Consultancy, Research grant / Funding (institution): NanoString; Advisory / Consultancy, Research grant / Funding (institution): Perkin Elmer; Advisory / Consultancy: PAIGE.AI; Advisory / Consultancy, Research grant / Funding (institution): Ventana; Advisory / Consultancy, Research grant / Funding (institution): Ultivue; Shareholder / Stockholder / Stock options: PixelGear; Research grant / Funding (institution): Navigate/Novartis; Research grant / Funding (institution): NextCure; Research grant / Funding (institution): Lilly. L. Pusztai: Honoraria (self), Advisory / Consultancy: Merck; Honoraria (self), Advisory / Consultancy: AstraZeneca; Honoraria (self), Advisory / Consultancy: Novartis; Honoraria (self), Advisory / Consultancy: Seattle Genetics; Honoraria (self), Advisory / Consultancy: Pfizer; Honoraria (self), Advisory / Consultancy: Almac. All other authors have declared no conflicts of interest.