Abstract
BACKGROUND: Due to high sensitivity of the skin on the head and neck to radiation, patients with head and neck cancer (HNC) are more susceptible to developing radiation dermatitis (RD). It is important to predict the severity of RD in patients with HNC after radiotherapy. The aim of this research is to develop a nomogram model to predict the RD with a subjective and objective combined scale. METHODS: During the period from May 24, 2020 to December 13, 2023, HNC patients with intensity-modulated radiotherapy (IMRT) combined concurrent chemotherapy were recruited. The main outcomes considered were scores on the Late Effects Normal Tissue Task Force-Subjective, Objective, Management, Analytic (LENT-SOMA). And the correlation among early LENT-SOMA score and scores of other scales for acute RD was assessed. A total of 257 patients were divided into the training and validation set (7: 3). Independent risk factors were identified using both single-variable and multiple-variable logistic regression analyses. A nomogram was subsequently developed using the factors that were significantly different in the multivariate logistic regression model (P<0.05). The nomogram’s predictive perform and clinical practicality was evaluated by receiver operating characteristics (ROC) curves, calibration curves and decision curve analysis (DCA). RESULTS: Correlation analysis indicated that the LENT-SOMA early score was correlated with Radiation Therapy Oncology Group (RTOG), Common Terminology Criteria for Adverse Events (CTCAE), World Health Organization (WHO), and Oncology Nursing Society (ONS) criteria. Multivariate logistic regression indicated that higher lymphocyte (Lym), blood glucose (BG), lower fibrinogen (FB), and albumin (ALB) before radiotherapy were independent prognosticators of LENT-SOMA score. A visible nomogram was built by integrating the Lym, BG, FB, and ALB before radiotherapy. The area under the curve (AUC) values of the nomogram in training cohort and validation cohort showed good power for the discrimination of 2 + grade RD. Using calibration curves and DCA, the results were confirmed. CONCLUSION: The LENT-SOMA scale exhibits potential for assessing acute RD. Higher Lym, BG, lower FB, and ALB before radiotherapy were independent prognosticators of LENT-SOMA score. We built a visible nomogram that showed good power for the discrimination of 2 + grade RD.