Radiation-Related Dysphagia Development Prediction Using a Two-Step Ultrasonographic Model (R-2D-2)

Radiotherapy for head and neck cancers (H\&NC) heightens the risk of swallowing disorders (SD), impacting nutrition, quality of life, and overall health, leading to increased hospitalization and mortality. Dietary plans hinge on patients' nutritional status, swallowing ability, and prognosis. Early interventions are crucial, emphasizing the need for precise assessments guiding prognosis, specifying structures for intervention, and facilitating targeted rehabilitation. Clinical examinations lack precision, while existing complementary methods like videofluoroscopy or Fiberoptic Endoscopic Evaluation of Swallowing are invasive, irradiating, resource-intensive, and challenging to access, with uncertain prognostic values. Ultrasound imaging emerges as a non-invasive alternative, offering morphological and dynamic evaluation of swallowing-related structures. It enables qualitative and quantitative analyses, improving precision in targeting structures for rehabilitation. Researchers propose an ultrasound predictive model to anticipate SD risk during H\&NC radiotherapy, assessing its reliability and accuracy. Over eighteen months, 124 outpatients beginning H\&NC radiotherapy at Forcilles's Hospital will undergo weekly clinical and water-swallow tests by a speech language therapist, with videofluoroscopy when SD is suspected. Ultrasound evaluations pre-treatment and at seven and fourteen days will be conducted by blinded ultrasonographers. Cox models will test ultrasound measurement thresholds for SD prediction, estimating sensitivity, specificity, and prediction values. A global ultrasound predictive model will be developed via logistic multivariable regression. The study aims to establish an association between ultrasound markers and SD, improving early detection for tailored management. This non-invasive alternative to videofluoroscopy offers potential for enhancing patient outcomes in H\&NC radiotherapy.