Prediction models for tube feeding dependence in head and neck radiotherapy

Aim of the research presented in this thesis was the development of prediction models that aid in the selection of patients with cancer of the head and neck at high risk for tube feeding dependence after radiotherapy.
In head and neck cancer (HNC) the tumour is surrounded by critical structures like the spinal cord, the salivary glands and structures important in swallowing. Radiotherapy in the head and neck region often results in side-effects; examples are xerostomia (dry mouth) and severe swallowing problems. Severe swallowing problems often result in long-term use of tube feedings. Research has shown that this decreases quality of life.
Radiotherapy techniques have improved significantly during the last decades. With these improved techniques salivary glands and structures important in swallowing can be spared to a certain degree. Unfortunately, not all structures can be spared since this would result in an unacceptable treatment plan with insufficient radiation dose to the tumour. It is important to know which structures should be spared the most. In previous research structures important for the development of xerostomia have been identified. It is, however, mostly unknown which structures, patient-, tumor- and treatment characteristics are most important for the development of tube feeding dependence.
In this thesis prediction models are presented that aid in the selection of patients who are likely to become tube feeding dependent, for preventive measures or new radiation techniques like proton therapy. This thesis also describes how the models can be used in optimizing radiotherapy treatment plans.