2 Brachytherapy Physics-Sources and Dosimetry

Brachytherapy Physics: Sources and Dosimetry

20

THE GEC ESTRO HANDBOOK OF BRACHYTHERAPY | Part I: The basics of Brachytherapy Version 1 - 01/12/2014

using the cavity theory. The choice for D w,m is the situation by de- fault in external beam therapy. Concerns relating to the conver- sion method may arise at low BT energies (i.e. < 50 keV) where cellular components have dimensions similar to or smaller than ranges of secondary electrons and where calculated doses are sensitive to small differences in atomic composition. For high energy brachytherapy sources the problem is much less, with a situation close to that in encountered in external beam radio- therapy. There is, at the time of writing this chapter, a lack of published investigations on the topic at brachytherapy energies and the aim of this section is to provide the current status while the problem is still under discussion. At this point, the authors of the present chapter have chosen to refrain from speculation on the future and not to add personal conclusions but just to present a number of key paragraphs from original work. First, a crucial document for further reading is the AAPM TG- 186, a report approved and endorsed by the AAPM, ESTRO, American Brachytherapy Society/ABS, and the Australasian Brachytherapy Group: ‘Report of the Task Group 186 on mod- el-based dose calculation methods in brachytherapy beyond the TG-43 formalism: Current status and recommendations for clinical implementation’ (4). From this report the text of the Conclusions is cited: “This report recognizes the benefit of the current stand- ard and has built upon the developed infrastructure with specific recommendations to help move the brachytherapy field towards greater adoption of more accurate, advanced model-based dose calculation algorithms. Application of TG-186 guidance should help to ensure that the shared worldwide experience in using the TG-43 formalism is not lost in the process of seeking better brachytherapy dose calculation tools. The report also underlines specific research areas that brachytherapy physicists and physi- cians should tackle expeditiously, perhaps through consortium efforts to provide data required to move forward with MBDCAs. Furthermore, it is important to acknowledge that most of clinical brachytherapy dosimetry experience is with radiation transport and energy deposition in water, with dose to water reported (as calculated under the TG-43 approach). Hence, for the time be- ing, the recommendation that TG-43 calculations be performed in parallel with model-based dose calculations is crucial. Only in this way will the radiation therapy community become famil- iar with dose differences, including the impact on prescription dose and doses to points, organs, and regions of interest. This will enable assessment of the implications of adopting MBDCAs for treatment planning, and help the brachytherapy community to understand new study results and place them in the context of previous dose calculations, treatment planning, and research.” And from the Conclusions section of that report: “Until suffi- cient clinical data become available to issue specific societal rec- ommendations on dose prescriptions, the AAPM, ESTRO, ABS, and ABG recommend that prescriptions based on the TG-43 dose calculation formalism remain in effect. Similarly, for the time being MBDCAs are generally impractical (from a compu- tational time perspective) for dose optimization and the TG-43 formalism remains the standard of practice for this task.” If the implementation of MBDCAs is slow and unclear, the QA recommendations for MBDCAs would be even more so. From a another recent report by Rivard et al ., ‘Enhancements to com- missioning techniques and quality assurance of brachytherapy treatment planning systems that use model-based dose calcu- lation algorithms, the clinical recommendations are cited’ (47): “While there is substantial literature on the influence of inter- seed attenuation for permanent prostate seed implants, there are

D w-­‐TG43 OUTPUT

INPUT

Superposi-on of data from source characteriza-on CALCULATION

Source characteriza-on

TG-­‐43

OUTPUT

INPUT

CALCULATION

Source model characteriza-on Tissue/applicator informa-on

D m,m D w,m

MBDCA

Monte Carlo GBBS Collapsed Cone

Fig. 2.18 Main steps of the dose calculation process for treatment planning using the TG-43 method (top) and model-based dose calculation algorithms (bottom). The output in terms of calculated doses relies on the calculation algorithms and the type of input includes the radiation source (source model/characterization) and the geometry in which calculations are performed (i.e., tissue and applicator information for the actual patient). (Courtesy: Å. Carlsson Tedgren)

during the treatment planning process. If a TG-43 approach is used for the optimization step, MC can calculate the chosen con- figuration with a higher level of accuracy than TG-43. The principles of analytical solvers to the so-called linear Boltz- mann transport equation , also referred to as Grid Based Boltz- mann Solvers (GBBS), have been described in detail in several other publications and will not be covered here. Analytical solv- ers have been tested successfully for brachytherapy (12, 13, 20, 55). An important rationale for their use in treatment planning is that they can be faster than MC simulations. A GBBS named Acuros developed by Transpire, Inc. (Gig Harbor, WA, USA) was recently implemented in the Varian Medical Systems Inc. (Palo Alto, CA, USA) treatment planning system BrachyVision for 192 Ir dose calculations and has been benchmarked against MC and measurements (42,56,57). The collapsed cone algorithm (CC) is a fast method for kernel superposition designed for treatment planning applications with the calculation speed stemming from an efficient ray-tracing kernel provided through “collapsing” the transport of energy released onto cone axes (angular discretization) in combination with suitable lattices of transport lines (computational grid) over the calculation volume. CC algorithms are in clinical use for ex- ternal beam radiotherapy and they compare well with full MC simulations. A brachytherapy version of CC has been developed (8). It calculates the primary dose analytically under the charged particle equilibrium (CPE) assumption and the dose from scat- tered photons in two subsequent steps using a “successive scat- tering” approach designed to minimize systematic errors due to kernel discretization and approximations in multiple scatter dose. The CC method has been used to calculate dose for sources in the range from 30 keV to 662 keV. The brachytherapy col- lapsed cone algorithm is currently being integrated into the On- centra Brachy TPS from Elekta AB (Sweden) (54). First principles MBDCAs are used for transport calculations in the local medium. The calculation requires knowledge of the composition of the calculation volume which is a separate prob- lem. Recent work done by Landry et al . showed results obtained using dual energy CT scanning (29-32). From Fig. 2.18 one can see that MBDCAs allow the user to define the outcome of the treatment planning calculation in terms of a Dose-to-medium in a medium surrounding ( D m,m ), or as a Dose-to-water in a me- dium surrounding ( D w,m ) (9). D w,m and D m,m are closely related. MBDCA radiation transport provides D m,m but for obtaining D w,m the dose to bulk medium is converted into dose-to-water