Zhao Yu, Zhang Juan, Yao Yong, Yu Linna
People's Hospital of Qianxinan Prefecture, Xingyi, Guizhou, China.
Department of Pharmaceutics, Key Laboratory of Drug Quality Control and Pharmacovigilance (Ministry of Education), State Key Laboratory of Natural Medicines, NMPA Key Laboratory for Research and Evaluation of Pharmaceutical Preparations and Excipients, China Pharmaceutical University, Nanjing, China.
Front Cell Dev Biol. 2025 Aug 29;13:1638348. doi: 10.3389/fcell.2025.1638348. eCollection 2025.
Papillary thyroid carcinoma (PTC) is the most prevalent form of thyroid cancer, yet its cellular heterogeneity and prognostic determinants remain poorly defined. Here, we integrate two single-cell RNA sequencing datasets comprising 20 human thyroid samples to construct a high-resolution cellular atlas of PTC. We identify 29 distinct cellular subpopulations and delineate their composition, dynamics, and interactions in healthy versus tumor tissues. Notably, epithelial and monocyte populations were markedly expanded in PTC, whereas adaptive immune subsets such as B and T cells were diminished. Cell-cell communication analysis revealed enhanced intercellular signaling in the tumor microenvironment, with epithelial and endothelial cells receiving the strongest inputs. Among monocyte-specific transcriptional signatures, we identified 65 prognostic genes via univariate Cox analysis. A LASSO-derived 14-gene risk score robustly stratified patient outcomes, with CSGALNACT1 emerging as a key epithelial-specific, independent prognostic gene. Pseudotime analysis further supported its role in epithelial cell differentiation. Functional validation demonstrated that CSGALNACT1 promotes proliferation in PTC cell lines, suggesting a potential oncogenic function. Immune deconvolution across risk groups revealed substantial divergence in innate and adaptive immune infiltration, indicating a close interplay between tumor-intrinsic transcriptional programs and immune microenvironment remodeling. Collectively, our study provides a comprehensive single-cell framework for PTC, identifies a clinically relevant risk model, and highlights CSGALNACT1 as a potential therapeutic target.
甲状腺乳头状癌（PTC）是甲状腺癌最常见的形式，但其细胞异质性和预后决定因素仍不清楚。在这里，我们整合了两个包含20个人类甲状腺样本的单细胞RNA测序数据集，以构建PTC的高分辨率细胞图谱。我们识别出29个不同的细胞亚群，并描绘了它们在健康组织与肿瘤组织中的组成、动态变化及相互作用。值得注意的是，上皮细胞和单核细胞群体在PTC中显著扩增，而B细胞和T细胞等适应性免疫亚群则减少。细胞间通讯分析显示肿瘤微环境中的细胞间信号增强，上皮细胞和内皮细胞接收的输入最强。在单核细胞特异性转录特征中，我们通过单变量Cox分析鉴定出65个预后基因。基于套索回归的14基因风险评分有力地分层了患者预后，CSGALNACT1成为关键的上皮细胞特异性独立预后基因。伪时间分析进一步支持了其在上皮细胞分化中的作用。功能验证表明CSGALNACT1促进PTC细胞系的增殖，提示其潜在的致癌功能。跨风险组的免疫反卷积揭示了先天免疫和适应性免疫浸润的显著差异，表明肿瘤内在转录程序与免疫微环境重塑之间存在密切相互作用。总体而言，我们的研究为PTC提供了一个全面的单细胞框架，识别出一个与临床相关的风险模型，并突出了CSGALNACT1作为潜在治疗靶点的作用。
