The role of cellular heterogeneity in thyroid cancer: a multi-modal analysis approach with Mendelian randomization

Abstract Background Thyroid cancer pathogenesis involves complex interactions between genetic predisposition and alterations in the tumor microenvironment. The causal relationships between inflammatory gene variants and thyroid cancer risk remain poorly understood, as does the cellular heterogeneity...

Full description

Saved in:
Bibliographic Details
Main Authors: Yichun Qian, Wei Chen, Xinyuan Chen, Shuai Cheng, Fangzhou Liu
Format: Article
Language:English
Published: Springer 2025-08-01
Series:Discover Oncology
Subjects:
Online Access:https://doi.org/10.1007/s12672-025-03348-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849764080411213824
author Yichun Qian
Wei Chen
Xinyuan Chen
Shuai Cheng
Fangzhou Liu
author_facet Yichun Qian
Wei Chen
Xinyuan Chen
Shuai Cheng
Fangzhou Liu
author_sort Yichun Qian
collection DOAJ
description Abstract Background Thyroid cancer pathogenesis involves complex interactions between genetic predisposition and alterations in the tumor microenvironment. The causal relationships between inflammatory gene variants and thyroid cancer risk remain poorly understood, as does the cellular heterogeneity within the tumor ecosystem. This study aimed to investigate the causal associations between inflammatory protein genes and thyroid cancer risk, and to characterize the cellular composition and differentiation trajectories within the thyroid cancer microenvironment. Methods We employed a two-pronged approach combining Mendelian randomization (MR) analysis and single-cell RNA sequencing (scRNA-seq). MR analyses were conducted using genetic variants associated with the expression of inflammatory proteins (4EBP1_EIF4EBP1, ADA_ADA, ARTN_ARTN, AXIN1_AXIN1, and Beta-NGF_NGF) as instrumental variables to assess their causal effects on thyroid cancer risk. Multiple MR methods (MR Egger, weighted median, inverse variance weighted, simple mode, and weighted mode) were used to enhance robustness. For the cellular characterization, scRNA-seq was performed on thyroid cancer samples, followed by dimensionality reduction, clustering analysis, cell type annotation, and pseudotime trajectory inference. Results MR analyses revealed a significant positive causal association between AXIN1_AXIN1 expression and thyroid cancer risk (weighted median: OR = 1.396, p < 0.05; inverse variance weighted: OR = 1.291, p < 0.05), while ADA_ADA showed protective effects (simple mode: OR = 0.731, p < 0.05). The scRNA-seq analysis identified six major cell populations within the thyroid cancer microenvironment: epithelial cells, T cells, natural killer cells, fibroblasts, stromal cells, and macrophages. Pseudotime analysis revealed distinct differentiation trajectories with natural killer cells and macrophages appearing in early pseudotime, while epithelial cells and fibroblasts demonstrated multiple developmental states. Gene expression profiling identified four distinct cellular states with unique molecular signatures, including immune/inflammatory, stromal, and vascular components. Conclusion Our findings suggest that inflammatory protein genes, particularly AXIN1, have causal effects on thyroid cancer risk, providing potential targets for risk prediction and intervention.
format Article
id doaj-art-d46062be992e4c338fb48288fe2428fb
institution DOAJ
issn 2730-6011
language English
publishDate 2025-08-01
publisher Springer
record_format Article
series Discover Oncology
spelling doaj-art-d46062be992e4c338fb48288fe2428fb2025-08-20T03:05:14ZengSpringerDiscover Oncology2730-60112025-08-0116111510.1007/s12672-025-03348-8The role of cellular heterogeneity in thyroid cancer: a multi-modal analysis approach with Mendelian randomizationYichun Qian0Wei Chen1Xinyuan Chen2Shuai Cheng3Fangzhou Liu4Department of Head and Neck Surgery, Jiangsu Cancer Hospital & Jiangsu Institute of Cancer Research & The Affiliated Cancer Hospital of Nanjing Medical UniversityDepartment of Head and Neck Surgery, Jiangsu Cancer Hospital & Jiangsu Institute of Cancer Research & The Affiliated Cancer Hospital of Nanjing Medical UniversityDepartment of Head and Neck Surgery, Jiangsu Cancer Hospital & Jiangsu Institute of Cancer Research & The Affiliated Cancer Hospital of Nanjing Medical UniversityDepartment of Head and Neck Surgery, Jiangsu Cancer Hospital & Jiangsu Institute of Cancer Research & The Affiliated Cancer Hospital of Nanjing Medical UniversityDepartment of Head and Neck Surgery, Jiangsu Cancer Hospital & Jiangsu Institute of Cancer Research & The Affiliated Cancer Hospital of Nanjing Medical UniversityAbstract Background Thyroid cancer pathogenesis involves complex interactions between genetic predisposition and alterations in the tumor microenvironment. The causal relationships between inflammatory gene variants and thyroid cancer risk remain poorly understood, as does the cellular heterogeneity within the tumor ecosystem. This study aimed to investigate the causal associations between inflammatory protein genes and thyroid cancer risk, and to characterize the cellular composition and differentiation trajectories within the thyroid cancer microenvironment. Methods We employed a two-pronged approach combining Mendelian randomization (MR) analysis and single-cell RNA sequencing (scRNA-seq). MR analyses were conducted using genetic variants associated with the expression of inflammatory proteins (4EBP1_EIF4EBP1, ADA_ADA, ARTN_ARTN, AXIN1_AXIN1, and Beta-NGF_NGF) as instrumental variables to assess their causal effects on thyroid cancer risk. Multiple MR methods (MR Egger, weighted median, inverse variance weighted, simple mode, and weighted mode) were used to enhance robustness. For the cellular characterization, scRNA-seq was performed on thyroid cancer samples, followed by dimensionality reduction, clustering analysis, cell type annotation, and pseudotime trajectory inference. Results MR analyses revealed a significant positive causal association between AXIN1_AXIN1 expression and thyroid cancer risk (weighted median: OR = 1.396, p < 0.05; inverse variance weighted: OR = 1.291, p < 0.05), while ADA_ADA showed protective effects (simple mode: OR = 0.731, p < 0.05). The scRNA-seq analysis identified six major cell populations within the thyroid cancer microenvironment: epithelial cells, T cells, natural killer cells, fibroblasts, stromal cells, and macrophages. Pseudotime analysis revealed distinct differentiation trajectories with natural killer cells and macrophages appearing in early pseudotime, while epithelial cells and fibroblasts demonstrated multiple developmental states. Gene expression profiling identified four distinct cellular states with unique molecular signatures, including immune/inflammatory, stromal, and vascular components. Conclusion Our findings suggest that inflammatory protein genes, particularly AXIN1, have causal effects on thyroid cancer risk, providing potential targets for risk prediction and intervention.https://doi.org/10.1007/s12672-025-03348-8Thyroid cancerMendelian randomizationAXIN1Single-cell RNA sequencingTumor microenvironment
spellingShingle Yichun Qian
Wei Chen
Xinyuan Chen
Shuai Cheng
Fangzhou Liu
The role of cellular heterogeneity in thyroid cancer: a multi-modal analysis approach with Mendelian randomization
Discover Oncology
Thyroid cancer
Mendelian randomization
AXIN1
Single-cell RNA sequencing
Tumor microenvironment
title The role of cellular heterogeneity in thyroid cancer: a multi-modal analysis approach with Mendelian randomization
title_full The role of cellular heterogeneity in thyroid cancer: a multi-modal analysis approach with Mendelian randomization
title_fullStr The role of cellular heterogeneity in thyroid cancer: a multi-modal analysis approach with Mendelian randomization
title_full_unstemmed The role of cellular heterogeneity in thyroid cancer: a multi-modal analysis approach with Mendelian randomization
title_short The role of cellular heterogeneity in thyroid cancer: a multi-modal analysis approach with Mendelian randomization
title_sort role of cellular heterogeneity in thyroid cancer a multi modal analysis approach with mendelian randomization
topic Thyroid cancer
Mendelian randomization
AXIN1
Single-cell RNA sequencing
Tumor microenvironment
url https://doi.org/10.1007/s12672-025-03348-8
work_keys_str_mv AT yichunqian theroleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT weichen theroleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT xinyuanchen theroleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT shuaicheng theroleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT fangzhouliu theroleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT yichunqian roleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT weichen roleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT xinyuanchen roleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT shuaicheng roleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization
AT fangzhouliu roleofcellularheterogeneityinthyroidcanceramultimodalanalysisapproachwithmendelianrandomization