Esophageal squamous cell carcinoma (ESCC) is a leading cause of cancer-associated mortality worldwide. In 2022, there were ~510,000 new cases of esophageal cancer (EC) and 440,000 mortalities worldwide, with 75% of the cases and mortalities occurring in Asia, this may be related to Asians' preference for hot food (1,2). The risk factors for EC include smoking, drinking (3), eating salted vegetables (4) and low fruit intake (5). Although conventional endoscopy is still the gold standard for diagnosing ESCC, with a pooled sensitivity of 94% and specificity of 92% (6), its high cost and the need for specially trained personnel are significant barriers to widespread implementation. Novel EC markers include tissue biomarkers (7), exhaled gas markers (8), saliva markers (9) and circulating biomarkers (10). Previous studies have highlighted the key role of long non-coding RNAs (lncRNAs) as master regulators of ESCC pathogenesis, orchestrating oncogenic processes through intricate molecular networks (11,12). Clinically, lncRNAs have been associated with various aspects of ESCC, including lymph node metastasis, tumorigenesis (13), tumor nodes metastasis stage (14) and prognosis (15). In ESCC cell lines, lncRNAs have been implicated in processes such as cell proliferation (16), migration (17), invasion (18) and epithelial-to-mesenchymal transition (19). In both cell-derived xenograft models and patient-derived tumor xenografts, lncRNAs have been linked to tumor growth (20), lymph node metastasis (21), liver metastasis (19) and bone metastasis (16). These lncRNAs transcripts modulate proliferation, invasion and metastasis by targeting key signaling pathways. For example, The oncogenic effect of lncRNA BBOX1-AS1 in esophageal squamous cell carcinoma was found to activate the Hedgehog signaling pathway through the miR-506-5p/EIF5A/PTCH1 axis, promoting cancer cell proliferation and stemness (22), while cleavage and polyadenylation specific factor 3 (CPSF3) disrupts miR-125a-5p-mediated suppression of cornichon family AMPA receptor auxiliary protein 2 (CNIH2) to enhance proliferation and migration (23). Conversely, membrane associated guanylate kinase, WW and PDZ domain containing 2 (MAGI2)-AS3 acts as a tumor suppressor by binding enhancer of zeste homolog 2 (EZH2) to downregulate homeobox B7 (HOXB7), thereby inhibiting proliferation and improving radiosensitivity (24). Such dual regulatory roles underscore the versatility of lncRNAs in ESCC progression, mirroring the competing endogenous RNA (ceRNA) network complexity observed in nasopharyngeal carcinoma where lncRNA/circular RNA-miRNA-mRNA interactions govern therapeutic resistance (25).

Epithelial-mesenchymal transition (EMT), a cornerstone of metastasis, is controlled by lncRNAs. While hepatocyte nuclear factor 1 homeobox A (HNF1A)-AS1 suppresses EMT by sponging miR-298 to inhibit transcription factor 4 (TCF4) and mesenchymal markers (26), kinectin 1 (KTN1)-AS1 promotes EMT via the Hippo/YAP1 pathway (27). Notably, LINC00858 exacerbates invasion by inducing EMT markers (such as N-cadherin) through miR-425-5p/ABL proto-oncogene 2, non-receptor tyrosine kinase (ABL2) signaling (28), whereas small nucleolar RNA host gene 12 (SNHG12) dual-regulates EMT and stemness via β-catenin stabilization (29), revealing the complexity of lncRNA-mediated EMT networks. This regulatory intricacy parallels findings in other types of cancer, such as colorectal cancer (30) and breast cancer (31), where miRNA-mediated EMT modulation (such as CD44-STAT3 axis inhibition) enhances radiosensitivity (32), suggesting potential cross-cancer therapeutic strategies.

Cell cycle dysregulation, a hallmark of ESCC, is governed by lncRNAs such as cancer susceptibility 9 (CASC9), which recruits EZH2 to suppress programmed cell death 4 (PDCD4) and accelerate G₁/S transition (33) and LINC00022, which promotes p21 degradation via m6A demethylation (34). Conversely, LINC01980 induces G₁ arrest through growth arrest and DNA damage inducible a (GADD45A) upregulation (35), illustrating their context-dependent roles in cell cycle control.

Therapeutic resistance, a notable barrier in ESCC management, is associated with lncRNAs. Colon cancer associated transcript 2 (CCAT2) enhances radioresistance by inhibiting miR-145/p53 pathways (36), while CASC8 stabilizes heterogeneous nuclear ribonucleoprotein L (hnRNPL) to activate anti-apoptotic Bcl2/caspase3 signaling (37). Hypoxia-driven LINC01116 and exosomal myocardial infarction associated transcript (MIAT) mediate resistance via hypoxia inducible factor 1α (HIF-1α)/miR-3612 and TATA-box binding protein associated factor 1 (TAF1)/sterol regulatory element binding transcription factor 1 (SREBF1) axes (38,39). These findings align with studies demonstrating that lncRNAs such as ectodermal-neural cortex 1 regulate tumor microenvironments and immune evasion in cancers (39,40), highlighting their systemic role in therapy resistance. Cancer stemness, a driver of recurrence, is amplified by BBOX1-AS1 via Hedgehog activation (22) and LINC00941 via SOX2/OCT4/NANOG feedback loops (41). Apoptosis evasion, which is key for tumor survival, is regulated by deleted in lymphocytic leukemia 1 (DLEU1; via BCL2 stabilization) (42) and reversed by interferon regulatory factor 1 (IRF1)-AS through interferon signaling (43), highlighting lncRNAs as dual modulators of cell death.

Collectively, lncRNAs are central players in ESCC malignancy, governing proliferation, metastasis, therapy resistance and stemness through multitarget mechanisms (Fig. 1). Their interactions with super-enhancers (a type of DNA regulatory element with strong transcriptional regulation ability), revealed by tools such as SEanalysis 2.0 (44), may further elucidate context-specific transcriptional networks. Their regulatory complexity positions them as promising diagnostic biomarkers and therapeutic targets. Integrating multi-omics approaches, such as ceRNA networks (25) and phosphorylation proteomics (32), may accelerate biomarker discovery and combination therapy design. The present review aimed to summarize the involvement of lncRNAs in ESCC, providing insight into the clinical diagnosis, treatment and prevention of ESCC.

A class of RNA molecule >200 nucleotides, lncRNAs were first discovered in the 1970s. Scientists mainly concentrated on mRNA that encodes proteins, while ncRNAs were regarded as ‘noise’ or ‘by-products’. Nevertheless, the advancement of technologies such as high-throughput sequencing (45) for detecting non-coding RNA and research (46) demonstrated that ncRNAs serve vital roles in gene regulation, epigenetics and the occurrence of disease. In 2003, researchers first found an lncRNA associated with gene silencing on the X chromosome (47). Guttman et al (48) discovered the lncRNA-HOTAIR, which is notable in gene locus regulation. In 2007, Rinn et al (49) identified an lncRNA located in the HOX gene cluster and found its crucial participation in gene locus regulation. lncRNAs serve important roles in embryonic development (50). Studies have also demonstrated the role of lncRNA in the initiation and progression of tumors, which has triggered research into the role of lncRNA in cancer (51–53).

According to genomic annotations in Ensembl Release 96 (April 2019) (54), human lncRNAs are categorized into distinct functional groups. These include 3′overlapping nc transcripts, antisense-oriented lncRNAs, genomic-spanning interspersed RNAs, intronic sequence-retained transcripts, intra-exonic sense RNAs, coding sequence-overlapping molecules and macro-scale regulatory RNAs. Intronic lncRNAs reside within the nc segments (introns) of protein-coding genetic loci. Although transcribed from these intronic regions, they are functionally independent from protein synthesis pathways (55). Antisense lncRNAs exhibit sequence complementarity to protein-coding gene strands, potentially modulating gene activity through RNA duplex formation via Watson-Crick base pairing mechanisms (56,57). Positioned between coding genetic elements, intergenic lncRNAs may exert cis-regulatory effects on neighboring genes (48). Sense strand-aligned lncRNAs demonstrate exonic overlap with protein-coding transcripts through co-directional transcription (58). Messenger-like lncRNAs serve as epigenetic modulators capable of fine-tuning target gene expression (50). Structure-encoded lncRNAs contribute to cell organization by maintaining nuclear compartmentalization and chromatin topology (59).

lncRNAs are found in the cytoplasm (60), nucleus (61), nucleolus (62), as well as other subcellular regions and vesicles (such as nucleolus and exosomes). Their localization is associated with molecular functions, including nuclear lncRNA target-specific genomic regions by binding to chromatin modification complexes, while cytoplasmic lncRNAs act by binding to mRNA or miRNA (60,63). Specific sequence motifs in their primary sequences are associated with subcellular localization (64). Investigating the localization of lncRNAs is key for understanding their roles in gene regulation, disease development and cell function. The techniques used to study the localization of lncRNAs include in situ hybridization (65), RNA immunoprecipitation (66), RNA (67) and single-cell RNA sequencing (68) and fluorescence in situ hybridization-flow cytometry (69).

While lncRNAs serve key biological roles, the majority of these molecules demonstrate poor evolutionary preservation in nucleotide alignment across organisms. This sequence divergence implies potential challenges in cross-species identification through conventional homology searches. This low degree of conservation is considered to reflect the diversity and specificity of lncRNA functions, as well as their rapid evolution (70). Notwithstanding this sequence variability, certain lncRNAs exhibit conserved architectural features or regulatory capacity in divergent species. These molecules may retain comparable spatial configurations or participate in equivalent gene expression networks despite sequence dissimilarity (59,71). Numerous lncRNAs display pronounced phyletic restriction, being uniquely present in particular evolutionary lineages. This taxonomic exclusivity implies specialized biological contributions during species-specific developmental or adaptive mechanisms (70,72). Notably, the regulatory regions controlling lncRNA transcription show evolutionary stability similar to those governing protein-coding genes, suggesting shared regulatory principles (73,74).

For ESCC diagnostic biomarkers, ease of acquisition and detectability are key. In early-stage cases, invasive methods such as endoscopic biopsy are often declined by patients due to discomfort and complexity, making blood or saliva-based biomarkers key for improving early screening compliance (75–85). LncRAs show promise as non-invasive tools, given their stability in the circulation (86). Emerging evidence highlights their potential to address the need for accessible screening in high-prevalence regions. LncRNAs can circulate in bodily fluids such as blood and urine due to their ability to traverse cell membranes (87). This enables their identification through non-invasive diagnostic approaches (88). LncRNAs in bodily fluids directly reflect the expression patterns of specific genes and distinguish patients with cancer from healthy individuals (89). Furthermore, circulating lncRNAs resist enzymatic degradation by RNases (86,90). Apoptotic bodies, microvesicles and exosomes represent phospholipid bilayer-enclosed vesicles carrying DNA, RNA, lipid, proteins, polysaccharides and metabolites. These structures are released into the circulatory system for systemic distribution, mediating intercellular material transfer (91–93). Reverse transcription-quantitative PCR (RT-qPCR) is utilized for detecting circulating lncRNAs due to its high sensitivity and specificity (94). Multiple lncRNAs are stably present in the blood of patients with ESCC through plasma, serum, whole blood or exosomes, demonstrating non-invasive diagnostic potential; for example, upregulated lncRNAs such as Linc00152 (plasma, AUC=0.698) (95) and MALAT1 (serum exosomes, AUC=0.755) (96), and downregulated lncRNAs, such as LINC-PINT (plasma, AUC=0.862) (97), were detected in body fluids to distinguish tumors from healthy controls. And as detected in the whole blood, LINC00324 (AUC=0.627) (98), EWSAT1 (AUC=0.717) (99) and NEF (AUC=0.904) (100) were also of a certain diagnostic value. It is worth noting that the high diagnostic efficacy of 5 plasma exosome lncRNA combinations (AUC=0.999) (101) and 4 serum exosome lncRNA panels (AUC=0.853) (102) confirms a synergistic detection advantage of circulating lncRNAs. In this chapter, the research on lncRNAs as diagnostic or prognostic biomarkers for ESCC in blood is discussed (Table I).

lncRNAs serve a key role in the onset and development of ESCC. lncRNAs serve multifaceted roles in ESCC, influencing biological processes including cell cycle control, cell proliferation, EMT, migration, invasion, drug resistance, apoptosis and cell stemness. lncRNAs within ESCC contribute to both the development and progression of the disease (Table II). Potential lncRNA candidate markers in EC are presented in Fig. 2.

lncRNAs are pivotal orchestrators of ESCC pathogenesis, efficiently regulating hallmark oncogenic processes, including unrestrained proliferation, metastatic dissemination, treatment resistance and cancer stemness, through their ability to engage in multifaceted regulatory networks (22,26,37). Their capacity to modulate epigenetic reprogramming, post-transcriptional modulation and intercellular communication underscores their dual role as molecular linchpins in tumor evolution and promising candidates for precision oncology. Crosstalk between lncRNAs and key signaling cascades (such as Hedgehog, PI3K/AKT/mTOR and JAK/STAT) (104,117) not only demonstrates their mechanistic versatility but also positions them as novel diagnostic biomarkers and druggable targets for overcoming therapeutic bottlenecks in ESCC management.

LncRNAs have diverse modes of action, complex mechanisms and unclear functions. lncRNAs affect gene expression through various mechanisms such as epigenetic modifications (161), RNA-RNA (162) and RNA-protein (163) interactions and regulation of miRNA activity (164); however, the specific mechanisms are often difficult to clarify. The function of lncRNA is associated with its location within the cell (such as the nucleus or cytoplasm), but the technology for dynamically tracking its localization changes is still immature (61,64). Technical bottlenecks limit research; current detection methods have low sensitivity and lncRNA expression levels are typically low, and traditional sequencing techniques may miss key molecules (165). Evolutionary conservation is low, model construction is difficult and species specificity is strong; >80% of lncRNAs only exist in humans or primates, and model animals such as mice cannot fully simulate their function. In addition their function is difficult to extrapolate. The lncRNA mechanisms discovered in animal experiments may not be directly applicable to the treatment of human disease. Clinical applications face multiple obstacles such as low biomarker specificity as numerous lncRNAs (such as HOTAIR) are abnormally expressed in various types of cancer, making it difficult to serve as a single diagnostic biomarker. There are difficulties in developing therapeutic targets as there are high off-target risks; targeted intervention may affect other non-target RNAs or genes. Finally, the detection methods for circulating lncRNA, such as exosome extraction and RT-qPCR primer design, lack a unified standard, which affects the comparability of results. Although this review comprehensively summarizes the latest research on lncRNAs in EC, it does not delve into the complex ceRNA network formed by lncRNAs, circRNAs, miRNAs and other factors; the application of cutting-edge RNA detection technology in EC is also rarely mentioned, and these contents will be supplemented in future research.

Future investigations should deconstruct the spatiotemporal dynamics of lncRNA interactions within tumor microenvironments, identifying their isoform-specific functions and translating these insights into clinically actionable strategies, ranging from liquid biopsy-based early detection platforms to RNA-targeted therapeutics. Meanwhile, future research should also clarify their dynamic interactions and transformation potential to overcome clinical challenges in ESCC. By summarizing lncRNA-mediated oncogenic pathways, the present review provides a multidimensional framework for understanding ESCC biology, while proposing innovative avenues to reimagine clinical paradigms in diagnosis (such as leveraging exosomal lncRNA signatures) and therapeutic intervention (such as the mechanism of drug resistance regulated by lncRNA). Such efforts hold transformative potential to mitigate the global burden of ESCC, a malignancy notable for its late-stage detection and poor survival outcomes.