Abstract A040: Cellular ecosystem of breast tumor microenvironment in Chinese breast cancer patients
Hela Koka,Xin Li,Difei Wang,Difei Wang,Kristine Jones,Kristine Jones,Aurelie Vogt,Aurelie Vogt,Bin Zhu,Amy Hutchinson,Amy Hutchinson,Belynda Hicks,Belynda Hicks,Priscilla Ming Yi Lee,Lap Ah Tse,Xiaohong R Yang
DOI: https://doi.org/10.1158/1538-7445.advbc23-a040
IF: 11.2
2024-02-02
Cancer Research
Abstract:Background: Current evidence has demonstrated the important role of tumor microenvironment (TME) in cancer initiation and progression. However, the dynamic interactions among various cells and their communities in TME remain largely unknown. Recently, EcoTyper, was developed to discover different cell states and their ecosystems in silico using bulk gene expression data. The application of this tool in the Pan-Cancer analysis identified distinct cell states and ecotypes (co-association patterns between cell states) that act as strong classifiers for cancer types and subtypes. In this study, we aimed to discover the TME ecosystems in breast tumors from an East Asian population and associate them with well annotated clinical features and breast cancer (BC) risk factors. Methods: We applied the EcoTyper framework to tumor RNA-Seq data from 242 Hong Kong BC patients to identify carcinoma ecotypes (CEs) and associated them with intrinsic subtype (PAM50), driver gene mutations (based on whole genome/exome sequencing analyses), and well-known BC risk factors. Additionally, we used The Cancer Genome Atlas BC (TCGA-BRCA) data, which was part of the initial EcoTyper Pan-Cancer analysis, as a quality control and comparison dataset. The TCGA patients were predominantly of European ancestry and had similar age distribution (median=58 years) compared to HK patients (median=60 years). The distribution of PAM50 subtype was also similar between the two datasets, with HK having a slightly higher proportion of luminal A (38% vs. 33% in TCGA) and a lower proportion of basal (11% vs. 20% in TCGA). Results: Among all ten ecotypes, the frequency of each CE was very similar between HK and TCGA. In both datasets, the prevalence of these CEs varied by tumor subtype. CE1 and CE6 were enriched in luminal A and rare in basal, while CE9 was rarely found in luminal A but showed high prevalence in HER2-enriched and basal subtypes. CE7 was enriched in luminal B (especially in HK, where among tumors with CE7, 82% of them were luminal B), CE10 was enriched in HER2-enriched (especially in HK, where 41% of tumors with CE10 were HER2-enriched), and CE2 was enriched for basal subtype. CE8 was enriched for luminal A in HK and luminal B in TCGA. The presence of CE2, CE9, or CE10 was associated with higher prevalence of TP53 somatic mutations, particularly in HK. Among the BC risk factors examined, we did not find any statistically significant associations with any CEs, probably due to the small sample size. The work of assessing the associations with the abundance of each cell state and CE is currently ongoing. Conclusions: We replicated cellular ecosystems and their associations with tumor subtypes and driver gene mutations in an East Asian population, demonstrating its generalizability and potential utility in refining BC subtypes. Next, we plan to extend the analysis to paired adjacent normal tissue to further assess the role of these ecosystems in shaping TME heterogeneity and tumor evolution. Citation Format: Hela Koka, Xin Li, Difei Wang, Difei Wang, Kristine Jones, Kristine Jones, Aurelie Vogt, Aurelie Vogt, Bin Zhu, Amy Hutchinson, Amy Hutchinson, Belynda Hicks, Belynda Hicks, Priscilla Ming Yi Lee, Lap Ah Tse, Xiaohong R Yang. Cellular ecosystem of breast tumor microenvironment in Chinese breast cancer patients [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Breast Cancer Research; 2023 Oct 19-22; San Diego, California. Philadelphia (PA): AACR; Cancer Res 2024;84(3 Suppl_1) nr A040.
oncology