Developing a prognostic stratification model based on glutathione metabolism in thyroid cancer and validating RRM2's tumor-promoting role.

INTRODUCTION: Glutathione (GSH), the most abundant antioxidant in cells, acts as free radical scavenger and detoxifying agent. Elevation of GSH metabolism protects tumor from damage of oxidant and even promotes tumor progression. However, the clinical value of GSH metabolism in thyroid cancer (THCA) remained largely unknown. METHODS: The expression and prognostic value of GSH metabolism-related enzymes were first investigated using a large The Cancer Genome Atlas (TCGA) cohort of 510 THCA patients. To expand the prognostic application, a risk stratification model based on these enzymes was developed using the LASSO Cox regression algorithm. Patients were categorized into high- and low-risk groups based on the median risk score, and the model's predictive performance for disease-freesurvival (DFS) was validated. Further correlation analysis, pan-cancer analysis (using TCGA and GTEx data), and detailed analysis across pathological types and TNM stages were performed to identify and characterize key molecules, such as RRM2. Finally, the biological role of RRM2 was validated in vitro (CCK-8 and colony-formation assays) and in vivo (subcutaneous tumor formation in nude mice). Furthermore, the molecular mechanism underlying RRM2's tumor-promoting function was preliminarily investigated through mRNA sequencing and subsequent experiments. RESULTS: The majority of GSH metabolism-related enzymes were significantly upregulated in THCA tumor tissues and their expression was negatively associated with DFS. The LASSO Cox model stratified patients into high-risk and low-risk groups with significantly different DFS. High-risk status was also positively correlated with increased infiltration of naïve B cells, activated memory CD4+ T cells, helper T cells and regulatory T cells. RRM2, screened as a key molecule, exhibited high expression in THCA tissues, especially in more aggressive subtypes (classic and tall-cell variants of papillary THCA) and N stages. Paired-sample IHC confirmed higher RRM2 in PTC versus adjacent tissue. High RRM2 expression was significantly and negatively correlated with DFS. Functionally, RRM2 overexpression promoted TPC-1 cell proliferation and colony formation (CCK-8 and colony assays) while knockdown suppressed growth. Subcutaneous tumor formation experiments recapitulated these findings. Mechanistically, RRM2's oncogenic effects may be mediated through cell cycle regulation and activation of the PI3K/Akt signaling pathway. DISCUSSION: GSH metabolism-related enzymes are upregulated in THCA and associate with a worse prognosis and an immune landscape suggestive of antigenic stimulation coupled with immunosuppression. RRM2 is a tumor-promoting gene that correlates with aggressive clinicopathologic features and functionally drives thyroid tumor growth in vitro and in vivo. These data support further investigation of GSH metabolism and RRM2 as prognostic biomarkers and potential therapeutic targets in thyroid cancer.