Esophageal cancer is one of the most aggressive malignancies globally, with a poor clinical prognosis. According to the latest global cancer statistics, there are approximately 511,000 new cases and 445,000 deaths annually, ranking it 11th and 7th, respectively, among all cancers [1]. Esophageal squamous cell carcinoma (ESCC) is the predominant histological subtype of esophageal cancer [2]. More concerningly, due to the lack of specific symptoms in the early stages, over 80 % of patients are diagnosed at an advanced stage, with a five-year survival rate of only 15 %–20 % [3,4]. Despite recent breakthroughs in the treatment of ESCC with immune checkpoint inhibitors (ICIs), the overall response rate remains under 30 % [[5], [6], [7]], primarily due to the high heterogeneity of ESCC and the complexity of its tumor microenvironment (TME) [[8], [9], [10], [11], [12]]. Therefore, a deeper understanding of the key regulatory mechanisms within the ESCC microenvironment is of significant clinical value for the development of novel therapeutic strategies.

In recent years, CAFs, the most prominent stromal cell population in the TME of ESCC, have been shown to play critical roles in tumor progression. They promote cancer development through direct interactions with tumor cells or other stromal components, as well as via the secretion of cytokines, exosomes, proteins, and by remodeling the extracellular matrix (ECM) [[13], [14], [15], [16]]. CAFs are involved in various malignant processes, including tumor growth [17], metastasis [18], and therapeutic resistance [19]. Studies have revealed significant heterogeneity among CAF populations. Notably, the myCAFs subtype, characterized by high expression of α-smooth muscle actin (α-SMA/ACTA2) and fibroblast activation protein (FAP), exhibits strong profibrotic activity and is believed to play a pivotal role in tumor progression. However, the specific markers defining functionally distinct myCAF subpopulations in ESCC, and their precise mechanisms in orchestrating immunosuppression, particularly through interaction with macrophages, remain largely unexplored.

G protein-coupled receptor class C group 5 member A (GPRC5A), a class C GPCR, has been reported to exhibit context-dependent “dual roles” in various cancers. For instance, in non-small cell lung cancer, it acts as a tumor suppressor by inhibiting the NF-κB signaling pathway [20], whereas in breast cancer, it promotes liver metastasis and docetaxel resistance through activation of the mTOR pathway [21]. Additionally, GPRC5A has been shown to be highly expressed in ESCC [22] and gallbladder cancer (GBC) [23], where it facilitates pulmonary metastasis. Our previous work demonstrated that GPRC5A is aberrantly overexpressed in ESCC tissues and is associated with poor prognosis; however, its expression and function within specific CAFs subpopulations have not been investigated.

In this study, we leverage scRNA-seq to deconvolute the CAFs heterogeneity in ESCC. We identify and characterize a novel, terminally differentiated myCAF subset marked by high GPRC5A expression, which is associated with poor patient outcome. Beyond merely identifying this subset, our work unveils a novel dual mechanism by which GPRC5A+ myCAFs drive tumor progression: (1) intrinsically, through TGF-β/SMAD2/3 pathway activation to enforce fibroblast activation, and (2) extrinsically, via ANXA1 secretion to reprogram macrophages towards a pro-tumoral M2 phenotype. This GPRC5A/TGF-β/ANXA1 axis represents a previously unrecognized signaling cascade that links CAF heterogeneity to immune evasion in ESCC. Our findings not only provide a specific biomarker for a high-risk CAF subset but also reveal a promising therapeutic target to disrupt the stroma-immune crosstalk in ESCC.