Physical radiotherapy treatment planning based on functional PET/CT data.

Positron emission tomography (PET) provides molecular information about the tumor microenvironment in addition to anatomical imaging. Hence, it seems to be highly beneficial to integrate PET data into radiotherapy (RT) treatment planning. Functional PET images can be used in RT planning following different strategies, with different orders of complexity. In a first instance, PET imaging data can be used for better target volume delineation. A second strategy, dose painting by contours (DPBC), consists of creating an additional PET-based target volume which will then be treated with a higher dose level. In contrast, dose painting by numbers (DPBN) aims for a locally varying dose prescription according to the variation of the PET signal. For both dose painting approaches, isotoxicity planning strategies should be applied in order not to compromise organs at risk compared to conventional modern RT treatment. In terms of physical dose painting treatment planning, several factors that may introduce limitations and uncertainties are of major importance. These are the PET voxel size, uncertainties due to image acquisition and reconstruction, a reproducible image registration, inherent biological uncertainties due to biological and chemical tracer characteristics, accurate dose calculation algorithms and radiation delivery techniques able to apply highly modulated dose distributions. Further research is necessary in order to investigate these factors and their influence on dose painting treatment planning and delivery thoroughly. To date, dose painting remains a theoretical concept which needs further validation. Nevertheless, molecular imaging has the potential to significantly improve target volume delineation and might also serve as a basis for treatment alteration in the future. Copyright 2010 European Society for Therapeutic Radiology and Oncology and European Association of Nuclear Medicine. Published by Elsevier Ireland Ltd.. All rights reserved.