S676
Clinical – Head & neck
ESTRO 2026
compared with their dynamic (delta) counterparts. Conclusion: Radiotherapy-induced changes in blood cell counts may reflect systemic treatment response and offer additional predictive information beyond baseline values. Overall, delta biomarkers showed improved prognostic performance. Validation in larger, independent cohorts is warranted. Keywords: delta hematologic biomarkers, radiotherapy
Digital Poster 5208
Prognostic Value of Delta Hematologic Biomarkers Pre- and Post-Radiotherapy for Head and Neck Squamous Cell Carcinoma Patients (HNSCC) YongXin Guo 1,2 , Dorothy M. Gujral 1 1 Department of Radiotherapy, Imperial College Healthcare NHS Trust, London, United Kingdom. 2 Department of Surgery and Cancer, Imperial College London, London, United Kingdom Purpose/Objective: Baseline hematologic biomarkers such as NLR, PLR, LMR, and SII have been demonstrated in multiple meta-analyses to predict prognosis in HNSCC. During radiotherapy, treatment-induced toxicity and inflammatory responses can lead to substantial changes in blood cell counts. This study aims to compare the prognostic performance of baseline biomarkers with the delta biomarkers ( , before and after radiotherapy) to evaluate whether dynamic hematologic alterations provide additional predictive value. Material/Methods: This retrospective study included patients diagnosed with HNSCC who received radical or adjuvant radiotherapy between 2015 and 2022. Three types of biomarker changes were calculated: 1. Absolute change = post – pre; 2. Relative change = (post − pre) / pre; 3. Log - ratio change = ln(post) − ln(pre). As the hazard associated with biomarker changes is time- dependent, Flexible Parametric Survival Model (FPM) was applied to predict overall survival (OS). Locoregional relapse-free survival (LRRFS) was assessed using the Fine–Gray competing risks model. Hazard ratio (HR), subdistribution hazard ratio (SHR), 95% confidence intervals (95% CI), and Akaike Information Criterion (AIC) were calculated accordingly. A two-sided p-value < 0.05 was considered statistically significant. Results: A total of 455 patients was included in the analysis. The median follow-up time was 81.3 months (95% CI: 76.7–86.4). 316 patients received radical radiotherapy and 139 received adjuvant radiotherapy. In FPM analysis, pre-treatment SII was significantly associated with OS (p = 0.016), indicating that higher systemic immune–inflammatory burden before radiotherapy predicts poorer prognosis. Among the dynamic biomarkers, Δ LMR_R and Δ PLR_A demonstrated better model fit than baseline (AIC 2104.4 vs 2120.6, 2103.5 vs 2105.3), p = 0.028 and 0.025, respectively. Δ NLR_A has better trend but exhibited non-significant but directionally consistent trends toward poorer survival. For predicting LRRFS, Δ LMR_R showed a trend toward association with relapse risk, whereas baseline NLR and SII demonstrated stronger predictive performance
Made with FlippingBook - Share PDF online