Abstract 216P
Background
A previous pan-cancer single-cell study on the tumor microenvironment (TME), particularly tumor-infiltrating lymphocytes (TILs), highlighted both the heterogeneity and dynamics of various T cell subtypes. However, the spatial heterogeneity of TILs in breast cancer has not been extensively studied. We aimed to investigate the spatial heterogeneity of the TME using spatial transcriptomics.
Methods
We retrospectively collected FFPE samples from five breast cancer cases (2 triple-negative breast cancer [TNBC] with high TILs, 2 TNBC with low TILs, 1 hormone receptor-positive[HR+] breast cancer with TIL high TILs) and performed the Visium Spatial Gene Expression method (10x Genomics). Subsequently, using public TILs single-cell data from a pan-cancer study, we predicted and annotated cell types in each Visium spot (single cell deconvolution). We studied the distribution of T cell subtypes across three layers at 3mm intervals from the most peripheral to the central direction of the tumor.
Results
The distribution of 24 CD4+ subtypes and 17 CD8+ subtypes was examined. In all five cases, the CD8.c.12.Tex.CXCL13 subtype showed a statistically significant increase in cell numbers at the tumor center compared to the periphery. The CD4.c20.Treg.TNFRSF9 subtype displayed a similar trend, with higher cell counts in the tumor center and lower counts in the periphery. Conversely, the CD8.c10.Trm.ZNF683 subtype, a key subtype in the previous pan-cancer study, exhibited varying distribution patterns between cases.
Conclusions
TILs in breast cancer consist of diverse subtypes. Numerical heterogeneity was observed among T cell subtypes, along with spatial heterogeneity between the tumor’s central and peripheral regions. Furthermore, heterogeneity was evident between individual tumors. However, some consistent distribution patterns, such as the CD8.c.12.Tex.CXCL13 population, were also observed across cases. Further studies are needed to determine the clinical significance of this spatial heterogeneity.
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
124P - AI-designed cancer vaccines: Antigens from the dark genome are promising cancer vaccine targets
Presenter: Daniela Kleine-Kohlbrecher
Session: Poster Display session
125P - Enhancing the efficacy of neoantigen tumor vaccines in melanoma treatment through different administration times
Presenter: Kai Xiao
Session: Poster Display session
126P - Combination of oncolytic viruses, radiation therapy, and immune checkpoint inhibitor treatment in a breast cancer model
Presenter: Olga Bezborodova
Session: Poster Display session
127P - Combining chemotherapy and checkpoint inhibitors with an engineered oncolytic adenovirus encoding a human vIL-2 cytokine to treat pancreatic ductal adenocarcinoma
Presenter: Nea Ojala
Session: Poster Display session
128P - Non-coding DNA lipid nanoparticles elicit antitumor immune responses and synergize with anti-PDL1 antibodies in mouse models of hepatocellular carcinoma
Presenter: Alba Rodriguez Garcia
Session: Poster Display session
129P - Computational approaches for enhancing the efficacy of cancer immunotherapy
Presenter: Byungho Lim
Session: Poster Display session
130P - Neoadjuvant treatment with a bispecific antibody cadonilimab in dMMR/MSI-H locally advanced colorectal cancer: Preliminary results from a phase II trial
Presenter: Caifeng Gong
Session: Poster Display session
131P - The efficacy and safety of cadonilimab with or without trastuzumab in combination with SOX as first-line (1L) treatment for advanced gastric (G) or gastroesophageal Junction adenocarcinoma (GEJA)
Presenter: Wenhui Yang
Session: Poster Display session
132P - An open-label, prospective phase II study of cadonilimab in combination with neoadjuvant chemotherapy for patients diagnosed with advanced ovarian cancer (AK104-IIT-003)
Presenter: Jie Tang
Session: Poster Display session