3552 - Tumor Tissue O-GlcNAc Level as a Dual Biomarker for Radiation-Induced Lung Injury and Radiotherapy Prognosis in NSCLC
Presenter(s)
Y. Ma1, X. Wang1, J. Yuan2, B. Tian3, F. Wang2, Y. Xu2, J. Yu2, and D. Chen2; 1Shandong university, Department of Radiation Oncology and Shandong Provincial Key Laboratory of Precision Oncology, Jinan, Shandong, China, 2Shandong Provincial Key Laboratory of Precision Oncology, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, Shandong, China, 3Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, Shandong, China
Purpose/Objective(s): Radiation-induced lung injury (RILI) remains a critical dose-limiting toxicity in thoracic radiotherapy for non-small cell lung cancer (NSCLC), and current dosimetric models inadequately predict risk due to inter-patient biological heterogeneity. This heterogeneity is likely driven by the tumor's intrinsic metabolic phenotype, meaning its nutrient-sensing pathways could be key to identifying susceptible individuals. Accordingly, this study aimed to determine if O-GlcNAcylation, a central mediator of nutrient sensing, serves as an independent predictor for RILI. We further sought to contextualize its role by exploring its relationship with systemic glucose and macroscopic tumor glucose uptake (PET-CT SUVmax), with the overarching goal of validating a biomarker capable of enabling precise patient risk stratification and guiding targeted radioprotective strategies.
Materials/Methods: This study was based on a retrospective analysis of a cohort comprising 263 NSCLC patients who underwent radiotherapy between 2014 and 2025. For each patient, pre-treatment tumor O-GlcNAc levels were quantified via immunohistochemistry, and corresponding macroscopic glucose uptake was assessed by SUVmax derived from 18F-FDG PET-CT scans. Systemic glycemic status was tracked by serially measuring fasting blood glucose at baseline, and again at weeks 2 and 4 during the course of radiotherapy. Multivariate logistic regression models were employed to identify independent predictors of RILI, while survival analyses were utilized to explore potential prognostic associations.
Results: Patients with high O-GlcNAc expression (n = 170) had a significantly higher RILI incidence than the low-expression group (n = 93) (45.9% vs. 31.2%, P = 0.02). In multivariate analysis, high O-GlcNAc was confirmed as an independent predictor for RILI occurrence (OR = 0.52; 95% CI: 0.30–0.89; P = 0.02), regardless of TNM stage. Notably, we found no correlation between tumor O-GlcNAc levels and SUVmax, radiation dose, or systemic blood glucose at any time point. Paradoxically, high O-GlcNAc expression was also associated with improved progression-free survival (PFS, P = 0.007) and overall survival (OS, P < 0.001). While elevated week 4 blood glucose correlated with RILI, it showed no association with tumor O-GlcNAcylation.
Conclusion: High tumor O-GlcNAc expression is an independent biomarker for predicting RILI risk in NSCLC, reflecting an intrinsic tumor metabolic phenotype independent of systemic hyperglycemia or macroscopic glucose uptake. We propose hyper-glycosylated tumors may increase normal tissue radiosensitivity via metabolic-immune interactions. This offers a novel target for precise RILI prevention without compromising tumor control, holding significant implications for developing new radioprotective strategies and future clinical practice.