Premium
SU‐GG‐T‐499: TCP‐Based Dose‐Painting by Numbers at the Example of a Prostate Case with 18F‐Choline PET
Author(s) -
Dirscherl T,
Rickhey M,
Bogner L
Publication year - 2010
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.3468895
Subject(s) - voxel , nuclear medicine , radiation treatment planning , dosimetry , radiation therapy , medicine , histogram , medical imaging , computer science , radiology , artificial intelligence , image (mathematics)
Purpose : A biologically adaptive radiation treatment method to maximize the TCP is shown. Functional imaging is used to acquire a heterogeneous dose prescription in terms of dose‐painting by numbers and to create patient‐specific IMRT plans. Method and Materials : Adapted from a method for selective dose escalation under the guidance of spatial biology distribution, a model, which translates heterogeneously distributed radiobiological parameters into voxelwise dose prescriptions, was developed. At the example of a prostate case with 18F‐choline PET imaging, different sets of reported values for the parameters were examined concerning their resulting range of dose values. Furthermore, the influence of each parameter of the linear‐quadratic model was investigated. A correlation between PET signal and proliferation as well as cell density was assumed. Using our in‐house treatment planning software Direct Monte Carlo Optimization, a treatment plan based on the obtained dose prescription was generated. Gafchromic EBT films were irradiated for evaluation. Results : When a TCP of 99% was aimed for, the maximal dose in a voxel of the prescription exceeded 100 Gy for every considered set of parameters. With a less ambitious aim for the TCP, one of the parameter sets resulted in a dose range of 87.1 Gy to 99.3 Gy, still yielding a TCP of 94.7%. The TCP of the plan decreased to 73.5% after optimization based on that prescription. The dose difference histogram of optimized and prescribed dose revealed a mean of −1.64 Gy and a standard deviation of 4.02 Gy. Film verification showed a reasonable agreement of planned and delivered dose. Conclusion : If the distribution of radiobiological parameters within a tumor is known, this model can be used to create a dose‐painting by numbers plan which maximizes the TCP. It could be shown, that such a heterogeneous dose distribution can be planned and applied.