An Evaluation of Two Approaches to Skin Bolus Design for Patients Receiving Radiotherapy for Head and Neck Cancers

PURPOSE: This radiation treatment planning study compares two approaches to designing a bolus for patients with head and neck cancer. Our current approach, based on clinical examination, is compared with an alternative approach, based on the patient's computed tomographic image data set, to investigate potential improvements in delivering the prescribed dose to the superficial regions of the clinical target volume (CTV) while limiting the dose to normal skin. METHODS: Twelve consecutive head and neck radiotherapy plans requiring a bolus were selected. A clinically placed bolus was designed by a radiation oncologist through physical examination of the patient. A virtual bolus was designed using an algorithm that configured it to overlay only the superficial CTV delineated on the patient's CT data set. These two approaches were compared on the basis of dose-volume histograms of normal skin and the superficial CTV, as well as the total bolus area. RESULTS: Of 12 patients, the virtual bolus plan resulted in a decrease in the bolus area of at least 4 cm² for nine patients, an increase in the bolus area of at least 30 cm² for three patients, and an improvement in the minimum dose to the superficial CTV in six patients. Of these six patients, half had a reduction in the bolus area with a corresponding modest 2% improvement in the minimum dose to the superficial CTV, whereas the other half had an increase in the bolus area with a corresponding dramatic 10%-57% improvement in the minimum dose to the superficial CTV. CONCLUSIONS: Basing bolus design on computed tomography image data rather than on clinical examination reduced the area of normal skin under the bolus in 9 patients (75%) and improved dose coverage of the superficial CTV in 3 patients (25%). All plans benefited from the virtual bolus approach because it has been shown to be more appropriate for balancing skin sparing with target coverage.