BLUE RIBBON POSTER THERAPY: Time-Aware Dosimetric Assessment of Peg-Defined Swallowing Toxicity In Head and Neck Re-Irradiation
Abstract
Purpose
PEG dependence is widely considered a consequence of head and neck re-irradiation (HN re-RT), yet PEG persistence around treatment is frequently misinterpreted as durable swallowing toxicity. We introduce a time-anchored definition of persistent PEG dependence through 3 months post re-RT and evaluate whether integrating baseline nutritional reserve materially improves dosimetric discrimination of functional swallowing failure.
Methods
We retrospectively analyzed 62 patients treated with curative-intent HN re-RT. The primary endpoint was persistent PEG dependence, defined as PEG present during and within 3 months after reRT completion. Nutritional reserve was assessed using baseline serum albumin and percent weight change from baseline to ~30 days post-reRT. Cumulative dose to swallowing-related organs at risk (OARs), including the pharyngeal constrictors and left-submandibular glands, was extracted using dose summation. Logistic regression was used to evaluate associations between nutritional factors and persistent PEG dependence, and model discrimination was assessed using the area under the ROC curve (AUC).
Results
Persistent PEG dependence occurred in 11/62 patients (18%) and was associated with higher cumulative swallowing-OAR dose (pharyngeal constrictors 86 vs 54 Gy, p=0.02; left submandibular gland 98 vs 58 Gy, p=0.008) and poorer nutritional reserve/trajectory (baseline albumin 3.6 vs 4.0 g/dL, p=0.07; late % weight change −0.3% vs −2%, p=0.4). In logistic regression, post-reRT weight loss alone showed poor discrimination (AUC 0.53), improved with the addition of baseline albumin (AUC 0.68), and was excellent with further inclusion of cumulative dose (AUC 0.82); baseline albumin (OR 0.20 per 1 g/dL, p=0.02) and cumulative dose (OR 1.03 per Gy, p=0.009) were independent predictors.
Conclusion
Persistent PEG dependence at ≥3 months post-reRT better reflects durable swallowing toxicity than peri-treatment PEG use. Adding cumulative swallowing-OAR dose and baseline nutritional reserve improves prediction beyond weight-based surrogates. This supports a practical, time-aware dose–reserve framework for re-irradiation planning and early supportive intervention.