食管癌是我国特色的高发肿瘤，发病率约占世界的1/2，其中超过90%为食管鳞癌。由于早期发病隐匿，大多数患者确诊时已处于中晚期。目前食管鳞癌治疗的主要手段仍是手术切除，但对于这些中晚期患者，大多数单纯手术治疗的患者3年内即出现局部复发或远端转移，5年生存率仅约为20.64％～34.00％。针对可手术切除的局部晚期患者，新辅助放化疗和手术联合治疗已成为标准的治疗选择。这种联合方案的应用一定程度上提高了患者的总生存率，但较低的新辅助放化疗有效率和较为广泛的毒副作用也给这种治疗方式带来了新的挑战。因此，探究新辅助放化疗抵抗的机制，寻找具有临床转化潜能的疗效标志物，深入挖掘逆转治疗抵抗的潜在靶点，是目前食管鳞癌临床转化研究的重要方向。

近年来，随着表观修饰研究的深入，N⁶-甲基腺嘌呤(N⁶-methyladenosine，m⁶A)修饰作为最为常见的RNA修饰方式在肿瘤发生发展中的关键角色已逐渐被揭示。目前研究证实，m⁶A修饰广泛参与到肿瘤的转移、复发和耐药等相关生物学进程。新近研究表明，其作为关键调控因子参与了肿瘤细胞的干性维持和肿瘤免疫微环境的重塑，而干细胞样肿瘤细胞和肿瘤免疫微环境已被证实与食管鳞癌放化疗疗效密切相关。因此，我们推测m⁶A修饰的异常可能是食管鳞癌放化疗抵抗的重要机制。然而，目前关于m⁶A修饰在食管鳞癌进展和放化疗抵抗中的作用尚不清楚。

在本研究中，我们通过分析课题组前期的基因芯片结果，全面揭示了m⁶A调控因子在食管鳞癌中的表达情况和临床意义，并发现m⁶A结合蛋白HNRNPC是临床意义最为重要的调控因子。通过数据库分析、qPCR验证和免疫组化，我们利用大样本、多中心队列明确了HNRNPC在食管鳞癌中表达显著上调且与较短的生存期密切相关。此外，通过多个数据库近万例泛癌样本分析，我们发现HNRNPC在多种实体瘤中高表达且与不良预后密切相关。通过分析HNRNPC在食管鳞癌新辅助放化疗治疗前内镜样本的表达情况，我们发现其高表达可以预测新辅助放化疗抵抗和治疗后较短的生存获益。通过体外实验和动物实验，我们证实了HNRNPC表达与食管鳞癌干细胞样特性密切相关，且敲降或敲除HNRNPC可以显著抑制肿瘤细胞的成球能力、转移能力和放化疗抵抗能力。进一步对HNRNPC敲除细胞系进行转录组测序分析并联合基于HNRNPC的紫外交联免疫沉淀结合高通量测序（Crosslinking-immunoprecipitation and high-throughput sequencing, CLIP-seq）数据分析，我们发现OGT可能是HNRNPC结合的经过m⁶A修饰的关键靶mRNA。细胞实验进一步证实了OGT参与了HNRNPC介导的食管鳞癌成球能力、转移能力和放化疗抵抗能力的维持。通路分析表明，HNRNPC-OGT轴可能激活了Wnt/β-catenin信号通路促进食管癌进展。组织标本的qPCR和免疫组化分析证实了OGT在食管鳞癌中高表达且与较差的新辅助放化疗疗效密切相关。免疫组化和多色免疫荧光证实了HNRNPC与OGT在食管鳞癌中表达具有较高的一致性。此外，通过数据分析和临床标本检测，我们发现HNRNPC表达与肿瘤免疫抑制微环境密切相关，且肿瘤免疫微环境异质性与食管鳞癌进展和放化疗疗效密切相关。通过细胞迁移实验，我们证实了敲除HNRNPC可以显著提高肿瘤细胞对CD8⁺ T细胞的招募能力。免疫组化和免疫荧光结果证实，基于HNRNPC和CD8⁺ T细胞的分子分型可以有效预测食管鳞癌新辅助放化疗的生存获益。

综上所述，我们首次揭示了m⁶A结合蛋白HNRNPC在食管鳞癌进展和放化疗耐药中的关键作用。HNRNPC在维持肿瘤细胞干性和重塑肿瘤免疫微环境中的决定性作用表明，HNRNPC不仅是个体化治疗的潜在生物标志物，更有望成为食管鳞癌联合治疗的潜在靶点。

Esophageal carcinoma (EC) is one of the most common malignancies in China, which makes up almost half of the global disease burden. Esophageal squamous cell carcinoma (ESCC) is the predominant histological type in China, which accounts for more than 90% of ECs. Due to the early symptoms is not apparent, patients with ESCC are often identified as advanced stage at the time of diagnosis. The surgery is the main treatment method for ESCCs. However, most of the patients with advanced stage will suffer from metastasis and recurrence within 3 years, and the 5-year overall survival (OS) rates is only about 20.64%-34%. In patients with resectable local advanced ESCC, the combination of neoadjuvant chemoradiotherapy (nCRT) and esophagectomy has become the standard treatment option. Compared with surgery alone, the application of this combined approach improves the OS of local advanced ESCC to a certain extent. However, most of the patients could not benefit from this combined therapy and the toxic and side effect widespread, which bring new challenges for the researchers. Therefore, exploring the mechanism of neoadjuvant chemoradiotherapy resistance, seeking for predictive markers with clinical translational value, and investigating the potential therapeutic targets for reversing treatment resistance are of great interest to the clinical decision-making process for ESCC.

In recent years, as the research about epigenetic modification moves along, the crucial role of N⁶-methyladenosine (m⁶A) modification in tumorigenesis is being revealed gradually. Studies have confirmed that m⁶A modification has been widely participated in the process of tumor metastasis, recurrence and therapeutic resistance. The latest research shows that m⁶A modification is a novel decisive factor in the maintenance stemness and remolding of the tumor immune microenvironment (TIME). Given the fact that stem cell-like tumor cells and TIME are important determinants for chemoradiotherapy, we speculate that m⁶A modification may play a crucial role in the progress of chemoradiotherapy resistance. However, little is known about the landscape of m⁶A modification in ESCC.

In this study, using our previously published microarray data, we conducted the first and most comprehensive study to reveal the expression profile and clinical relevance of m⁶A regulators in ESCC. We found that, of all the m⁶A regulators, HNRNPC—a m⁶A binding protein—showed the most significant clinical relevance in ESCC. The clinical significance of HNRNPC in ESCC was then validated in cases from multiple centers with large sample size by bioinformatics, qPCR, and immunohistochemistry (IHC). In addition, we found that HNRNPC was highly expressed in multiple solid tumors and closely associated with poor prognosis through analysis of nearly 10,000 pan-cancer samples from multiple databases. The expression detail of HNRNPC in the endoscopic cancer biopsies obtained from ESCCs before nCRT was also detected. We found that high HNRNPC expression was related to treatment resistance and poor survivals after nCRT. Through in vitro and in vivo experiments, we found that HNRNPC expression was closely related to the stem cell-like phenotype, and knocking down or knocking out HNRNPC can dramatically decreased the capacity of sphere formation, metastasis, and chemoradiotherapy resistance. Subsequently, we combined our RNA-seq data from HNRNPC knock out cells with the HNRNPC specific CLIP-seq data from GEO, and found that OGT was a potential m⁶A modified target mRNA of HNRNPC in ESCC. Then, in vitro experiments demonstrated that OGT played a crucial role in the HNRNPC-mediated maintenance of sphere formation, metastasis, and chemoradiotherapy resistance. Pathway analysis indicated that the HNRNPC-OGT axis may activate the Wnt/β-catenin signaling pathway to promote the progression of ESCC. Results of qPCR and IHC suggested that OGT was significantly upregulated in ESCCs and high expression of OGT was closely related to poor treatment response after nCRT. Using IHC and immunofluorescence (IF), we found that there was high consistency between the expression of HNRNPC and OGT in ESCCs. Moreover, through bioinformatics analysis and clinical sample testing, we found high HNRNPC expression was closely associated with immunosuppressive microenvironment. Simultaneously, we found the heterogeneity of TIME was tightly related to the progression and nCRT response. CD8⁺ T cell migration assays confirmed that HNRNPC knock out significantly enhanced the recruitment ability of tumor cells for CD8⁺ T cells. Finally, by IHC and IF, we built a novel HNRNPC and CD8-based molecular classification system, which showed robust outcome prediction ability for ESCCs with nCRT.

In summary, for the first time, we revealed the crucial role of m⁶A binding protein HNRNPC in tumor progression and chemoradiotherapy resistance of ESCC. The decisive role of HNRNPC in the maintenance stemness and remolding of the TIME indicates that, HNRNPC will not only act as a promising biomarker for the individualized therapy, but also will be more likely to be a potential target for combined therapy of ESCC.