Personalized Volume-deescalated Elective Nodal Irradiation in Oropharyngeal Head and Neck Squamous Cell Carcinoma

Local treatment of squamous cell carcinoma (SCC) of the oropharynx can consist of surgery, radiotherapy, or a combination of both. When treated with radiation, the target volume contains not only the primary tumor and clinically detected lymph node metastases. In addition, a large part of the lymph drainage system of the neck which is at risk of harboring occult metastases is irradiated, the so called "elective clinical target volume (CTV)". This elective CTV is currently based on clinical recommendations, but there is limited data and evidence on (occult) lymphatic spread and the required size of the elective CTV. This standard radiotherapy approach is associated with early and late toxicity. Toxicities such as pain, dermatitis, mucositis, but also long-term sequela like swallowing dysfunction, lymphedema and dysgeusia are commonly described, which can even lead to hospitalization or long-term symptoms with subsequent life-quality impairment.

A de-escalation of the treatment could result in less toxicity. Multiple studies have evaluated potential ways to de-escalate treatment and reduce toxicity, such as dose reduction or change of chemotherapeutic agent. Another possible de-escalation strategy, which is pursued here, is to reduce the elective clinical target volume.

A multi-institutional dataset of 598 oropharyngeal SCC patients in whom the detailed patterns of lymph node involvement are reported was collected. The publicly available online platform www.LyProX.org was developed to share and visualize the data. Based on this data, a statistical model of lymphatic tumor progression to perform a statistical analysis to estimate the probability of occult metastases in the clinically negative lymph node levels was developed. The patient's state of metastatic lymphatic progression is described via a hidden Markov model. The state of tumor progression is described by a collection of hidden binary random variables that indicate the involvement of lymph node levels. The model parameters are the probabilities for the tumor to spread to and between lymph node levels and are learned from the dataset. Supporting clinical experience, these statistical calculations can subsequently be used as a basis, to personalize the risk estimation of occult lymph node metastases in newly diagnosed patients based on their distribution of macroscopic metastases, T-stage, and lateralization of the primary tumor. A table with the possible different combinations of clinically observed lymph node involvement and the associated risk of occult lymph node involvement in the remaining, clinically negative lymph node levels (LNL) was created. By interpreting the results from both the statistical analysis and clinical experience, the elective clinical target volume (CTV) was personalized based on a patient's individualized risk profile. As a final measure of quality assurance, the elective CTVs (CTV-3s) constructed in this way have been discussed individually by the investigators, to ensure consistency with data and clinical judgement and experience. This leads to a reduction of irradiated volume and, potentially, to a reduction in early and late toxicity.

The aim of this clinical trial is to determine the safety of the use of a personalized de-escalated elective nodal CTV in oropharynx SCC patients treated with primary (chemo)radiotherapy.