Head and neck cancer ranks as the sixth most prevalent cancer globally, with about 950,000 new cases and 480,000 deaths reported in 2024(Bray et al., 2024). Head and neck squamous cell carcinoma (HNSCC) makes up 90 % of all head and neck cancers, boasting a 5-year survival rate of roughly 66 %(Johnson et al., 2020, Johnson et al., 2023). OSCC constitutes around 90 % of all oral malignancies and stands as one of the most common head and neck cancers worldwide, with over 130,000 new cases and more than 60,000 deaths each year(Ghanem et al., 2024; Liu et al., 2023). Therefore, OSCC has become a social problem that threatens public health for a long time. Despite notable advancements in surgery, radiotherapy, and targeted therapy, the 5-year survival rate for OSCC has long remained at approximately 50 %-60 %, primarily attributed to low rates of early diagnosis and high incidences of recurrence and metastasis(Jagadeesan et al., 2024; Liu et al., 2022; Radaic et al., 2024). Thus, there is an pressing requirement to investigate molecular markers with high sensitivity and specificity for early screening and prognostic evaluation, as well as to develop precise therapeutic approaches that target the key pathways involved in carcinogenesis.

Glycosylation is pivotal to the development and progression of cancer. Abnormal glycosylation events in cancer—such as sialylation, fucosylation, O-glycan truncation, and N-glycan branching—promote tumor progression via mechanisms that encompass the regulation of cell adhesion, signal transduction, immune evasion, and metabolic reprogramming(Arora et al., 2024; Guo et al., 2022; Pinho & Reis, 2015). Deciphering the role of glycosylation in cancer not only enables glycans to serve as important sources for developing novel clinical biomarkers but also paves the way for identifying promising therapeutic targets in cancer treatment by intervening in key steps of tumor progression involving glycans or targeting core mechanisms in glycan biosynthesis pathways(Thomas et al., 2021). Studies have shown altered protein glycosylation in OSCC. Among these, N-acetylgalactosaminyltransferase 2 (GALNT2) is overexpressed in OSCC, especially in cancer cells located at the tumor invasion front. The overexpression of GALNT2 markedly boosts the invasive capacity of OSCC cells through the regulation of O-glycosylation modifications and the activity of the epidermal growth factor receptor (EGFR)(Lin et al., 2014). Additionally, an integrated glycomic and glycoproteomic study analyzing N-glycosylation in 31 primary OSCC tissues with or without lymph node metastasis revealed that although the overall N-glycome remained relatively stable during tumor progression, six sialylated N-glycans were associated with lymph node metastasis. Specifically, elevated levels of core-fucosylated and sialylated N-glycans, together with fibronectin N-glycopeptides, showed a correlation with poor survival outcomes in patients. In contrast, reduced levels of afamin and CD59 N-glycopeptides were also found to be indicative of an adverse prognosis(Carnielli et al., 2023). While a wealth of studies have documented the involvement of glycosylation in tumorigenesis and progression, the ways in which glycosylation affects OSCC are still not fully elucidated(Hakomori, 2002; He et al., 2024). Recent studies focusing on the crosstalk between glycosylation and signaling pathways such as mTOR and PI3K-AKT have provided new insights into deciphering glycosylation-driven tumor evolution(Frappaolo et al., 2025; Vásquez Martínez et al., 2024; Very et al., 2018).

As a key mannosyltransferase, DPMS occupies a central position in post-translational modification processes—including the biosynthesis of glycosylphosphatidylinositol (GPI) anchors, as well as protein O-mannosylation and C-mannosylation—in both yeast and mammalian cells(Banerjee et al., 2017). DPMS assembles into a functional complex that consists of three subunits: DPM1, DPM2, and DPM3. Among them, DPM1 serves as the main catalytic component of DPMS, while DPM2 and DPM3 act as regulatory subunits, assisting DPM1 in localizing to the endoplasmic reticulum membrane and enabling its catalytic activity(Ashida et al., 2006; Rathod et al., 2024). In the field of cancer research, continuous progress has been made in studies related to DPMS. Alterations in DPMS activity in tumor cells lead to abnormal glycosylation modifications, affecting the structure and function of cell surface glycoproteins and glycolipids. This, in turn, influences cell-cell recognition, adhesion, and signal transduction, creating conditions for tumor invasion and metastasis. Existing studies have confirmed that DPMS exhibits increased enzymatic activity in tumors such as breast cancer and liver cancer, and it is positively correlated with cell proliferation and angiogenesis(Baksi et al., 2008; Baksi et al., 2009; Li et al., 2020; Palmer et al., 2019). Therefore, the genes that encode DPMS and their corresponding protein activities are likely to be positively linked to tumor progression. Nevertheless, the expression patterns, functional mechanisms, and specific roles of DPMS in the occurrence and development of OSCC are yet to be clarified. In this study, we aim to elucidate the potential role of DPMS in OSCC progression by analyzing the expression patterns of the three DPMS subunits (DPM1/2/3) and their correlation with clinical parameters of OSCC patients, as well as by deeply exploring the predicted functions and signaling pathways of DPMS and its homologous genes.