Contralateral parotid volume as an objective marker for adaptive radiotherapy in head and neck cancer.

Abstract

5596^ Background: Deformable image registration simplifies the process of adaptive radiotherapy and allows for sophisticated dosimetric analyses of anatomic changes. Lacking is an objective measurement that could reliably predict for undesirable dosimetric changes during a course of radiotherapy. We analyzed daily changes in contralateral parotid volume and correlated these changes to mean dose delivered to the parotid gland over a course of radiotherapy in four patients. Methods: Daily cone-beam CT (CBCT) scans of four patients were imported into the MIM Maestro software. The simulation CT scan was registered to 35 daily CBCT’s per patient using a sequence of rigid and deformable registrations. Parotid contours were resampled and a deformed reconstruction of the planning CT was exported to the Eclipse treatment planning software. The CT reconstructions were extended superiorly and inferiorly and the original fluence map was re-computed on the reconstructions with the isocenter at daily treatment position. Mean dose percent change and mean volume percent change were calculated for all CBCT’s and averaged across all patients. Average mean dose percent change was plotted against average mean volume change and evaluated with simple regression analysis. Results: A mean of 31 CBCT’s per patient were evaluated. Contralateral parotid volume percent changes for all patients were -31%, -40%, -40%, and -22% with a mean percent change of -33%. Mean dose percent changes were 25%, 35%, 19%, and 25% with a mean percent change of 26% and a mean absolute dose increase to the parotids of 6.8 Gy. Simple regression analysis identified a coefficient of determination (R2) of 0.23 where the independent variable is volume percent change and the dependent variable is mean dose percent change. Conclusions: Daily or weekly volume assessment of critical volumes is feasible utilizing deformable image registration and CBCT. Our data suggests there may be a correlation between parotid volume changes and increases in mean parotid dose though no causality is inferred. This relationship could be useful as a simple objective volumetric threshold measurement to prompt adaptive re-planning and warrants further investigation.