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

Radiotherapy for head and neck cancers increases the risk of Swallowing Disorders (SD), which can have serious consequences for nutrition, quality of life, and overall health.

Early intervention strategies could greatly benefit, highlighting the importance of assessments offering insights into patients' prognosis, identifying which specific structures require intervention and enabling a targeted rehabilitation approach.

However, the clinical examination lacks precision. Complementary examinations, such as videofluoroscopy or Fiberoptic Endoscopic Evaluation of Swallowing (FEES), are irradiating or invasive, consume numerous resources and are difficult to access. What's more, their prognostic values are not known.

In this context, ultrasound imaging has garnered increased interest for assessing swallowing disorders. Recognised for its non-invasive nature and its ability to overcome conventional constraints, ultrasound holds promise in integrating morphological and dynamically swallowing-related structures evaluation. This approach, enabling both qualitative and quantitative analysis of multiple structures, would enhance the precision in targeting structures for rehabilitation.

The researchers intend to create an ultrasound predictive model to anticipate the risk of developing swallowing disorders during head and neck cancer radiotherapy.

Hypothesis : The investigators hypothesize that one or more ultrasound variables are independently associated with swallowing disorder development and worsening nutritional status during radiotherapy treatment for head and neck cancer.

Objectives Main objective : The main objective of this study is to evaluate the predictive capacity of different ultrasound markers of the VADS on the incidence of TD during radiotherapy treatment for CTC

Secondary objectives :

In a population of patients treated with radiotherapy for CTC :

• Assessing the reliability of ultrasound measurements ;
• Evaluate the feasibility of ultrasound measurements;
• To evaluate the predictive ability of the change in ultrasound markers after 1 and 2 weeks of radiotherapy on the incidence of TD ;
• To determine whether ultrasound markers associated with the onset of TD improve a predictive model including known predictors ;
• Assessing the association between ultrasound markers and the incidence of undernutrition;
• To evaluate the predictive ability of the change in ultrasound markers after 1 and 2 weeks of radiotherapy on the incidence of undernutrition ;
• Evaluate the association between ultrasound markers and quality of life in relation to TD.

Study design :

Prospective, observational single-centre study will be performed per the ethical standards of the Declaration of Helsinki and will be reported following the standards for the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis guidelines (TRIPOD)

Population :

One hundred and twenty-four outpatients starting a radiotherapy treatment for head and neck cancer will be enrolled at Radiotherapy Services at Forcilles's Hospital, during a eighteen-months period. All consecutive patients will be screened at admission, and will be included if they satisfy the eligibility criteria. All patients will provide oral consent to participate, by the clinical research French laws. The patient's participation will be equivalent to the duration of their radiotherapy treatment. The maximum additional time needed for research interventions is 40 minutes.

Data collection : Data will be recorded using an online case report form (Datacapt). Demographic data, comorbidities, medical and surgical history will be recorded at inclusion. Clinical, videoflyoroscopy and ultrasound measurements will also be recorded. On-site monitoring is planned every month.

The ultrasound examination will be conducted initially, on the seventh and fourteenth days of radiotherapy, always preceding other swallowing evaluations.

Ultrasound examination :

Ultrasonography will be performed using a Sonosite E2 Expert ultrasound device (Digital Color Doppler Ultrasound System ; Sonoscape medical corp. China) with a 7-10 MHz linear and a 3-5 MHz probes. Detailed information for ultrasound procedure is described in the research protocol. Ultrasonography will be performed with the patients in near-to-supine position (30°).

To assess the value of variation in these examinations during radiotherapy as a predictive factor, a total of three ultrasound examinations will be performed prior to the first radiotherapy treatment and after seven and fourteen days. Ultrasonographers will be blinded from all patients' status and previous examination.

Ultrasound evaluation will focus on the movements or morphology of structures such as the tongue, laryngeal movement, or suprahyoid muscles.

In order to assess inter-examiner reliability, and only during the first ultrasound evaluation, the tests will be carried out by two different operators on a consecutive sample of patients. To limit the examination time for each patient, the investigators will use another consecutive sample of patients for intra-examiner reliability. For the patients making up the samples used for reliability assessment, only 10' extra will be added.

Clinical evaluation All patients will undergo a standardized weekly clinical assessment conducted by a speech language therapist. This assessment will involve an analytical evaluation of various structures (tongue movement and strength, laryngeal movement, etc.) and swallowing tests following the volume-viscosity swallow test.

Videofluoroscopy :

VFSS is a radiographic procedure offering a dynamic view of oral, pharyngeal, and upper esophageal function during swallowing. The radiologist and speech-language therapist collaborate to realise these evaluations. Barium-mixed food and fluid aid visibility via videofluoroscopy. The examiner must evaluate the broadest range of the patient's feeding options, starting with the safest quantity and texture. VFSS assesses swallow characteristics, bolus movement patterns, initiation, nasopharyngeal reflux, pharyngeal clearance, laryngeal penetration, and aspiration. At any time, examiners may halt the evaluation if deemed harmful, with reasons recorded. In instances of penetration or aspiration, the speech-language therapist employs the 'Penetration-Aspiration Scale.'

Statistical plan :

The study aims to assess the prognostic value of swallowing ultrasound in predicting the occurrence of swallowing disorders (TDs) during radiotherapy for CTC treatment. The event under consideration is the development of TDs between patient inclusion and the completion of radiotherapy. The analysis involves investigating the relationship between variables and TD occurrence. Quantitative variables in multivariate models will be studied using restricted cubic splines and fractional polynomial methods for accurate assessment.

Variable selection for the TD predictive model includes testing the prognostic values of ultrasound variables and adjustment variables through univariate Cox models. Significant variables (p ≤ 0.10) or those with known prognostic impact will be included in a multivariate Cox model. Model calibration, discrimination, and internal validation will be evaluated using the c-index, graphical calibration at 30 days, and the bootstrap method.

The estimation of TD onset times will utilize the Kaplan-Meier method, comparing survival curves between groups defined by ultrasound variable thresholds with the log-rank test. A prognostic score will be constructed from multivariate model results, categorizing patients into low-risk, intermediate-risk, and high-risk groups. Thresholds for categorization will be determined using Cox's method.

Handling missing data involves either deleting observations with missing data or using data imputation methods depending on the origin and type of missing data. Reproducibility assessment of ultrasound measurements includes inter-operator and intra-operator evaluations. To assess inter-operator reproducibility, measurements by two different operators will be compared, while intra-operator reproducibility will be evaluated from measurements performed by the same operator 30 minutes apart. A sample size of 40 patients is deemed necessary for reproducibility assessment based on Walter et al.'s methodology.

The tests will be performed at a significance level of 5%. 95% confidence intervals will be provided for each estimate.

Calculations will be performed using SPSS v21 IBM and R software (version 3.6.1, www.R-project.org).

Sample size calculation :

According to Harrell's recommendations on the construction of multivariate models, a maximum of 1 variable per 10 events should be included. The investigators estimates the incidence of swallowing disorders in our patient population at 45% according to the literature. The investigators aims to include a maximum of 5 variables in the multivariate model, i.e. 112 patients required.

Among the eligible patient population, taking into account a 10% loss-of-sight rate, it is necessary to include 124 patients