2 Brachytherapy Physics-Sources and Dosimetry

SECOND EDITION

The GEC ESTRO Handbook of Brachytherapy

PART I: THE BASICS OF BRACHYTHERAPY 2 Brachytherapy Physics: Sources and Dosimetry Jack Venselaar, Dimos Baltas

Editors Erik Van Limbergen Richard Pötter

Peter Hoskin Dimos Baltas

Brachytherapy Physics: Sources and Dosimetry

3

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

2 Brachytherapy Physics: Sources and Dosimetry Jack Venselaar, Dimos Baltas

1. Summary 2. Introduction

3 3 4 8

5. Dose calculation in brachytherapy

11 21 22

6. Key messages 7. References

3. Radioactive sources 4. Source dosimetry

1. SUMMARY

should be available with traceability to a primary standard.

to know the strength of the sources. Dose calculation algorithms were developed for clinical use from very simple calculation meth- ods, via parameter based approaches, to complex, so-called mod- el-based dose calculation algorithms (MBDCA’s, including Monte Carlo techniques). To be able to perform such calculations for real clinical cases, the target definition needs to be determined. So, im- aging techniques are employed to reconstruct the source positions, target and treatment volumes, organs at risk, and prescription points and volumes using different types of imaging modalities. There must be a clear understanding of the volume definitions so that the strength of brachytherapy (a high dose fall-off due to the inverse square law, and therefore high dose conformity to the tar- get volume) can be used to optimize the individual treatment plan in terms of highly conformal coverage of the target and limitation of surrounding tissues. Treatment delivery systems are commer- cially available to treat our patients, but these systems have to be commissioned, sources need to be replaced on a regular basis, and servicing must be organized. This part falls within the scope of a departmental quality management system for which the medical physicist is usually responsible. Radioprotection is an important Calculation of the absorbed dose to the target is a key ele- ment in the process of clinical treatment preparation. In this chapter, two methods for the calculation of dose to a point in water at a given distance from a source are explained: the conventional approach and the methodology described in the AAPM Task Group 43. The latter has become a standard of practice in all present-day brachytherapy treatment plan- ning systems. References to the open literature and practical webbased databases are provided so that the reader can find reliable data for the validation of the TG-43 calculated dose of a brachytherapy treatment planning system. Some examples are shown of one- and two-dimensional dose distributions to demonstrate typical dependence of the dose deposition in relation to the energy of the emitted photons and absorption phenomena in the source walls, such as anisotropy. However, these two methodologies of dose calculation have their limitations such as their inability to account for shielding and inhomogeneities with, for example, lack of scattered radi- ation. New solutions are being developed to avoid such draw- backs. The last section of the chapter therefore discusses the present status of so-called Model Based Dose Calculation Al- gorithms (MBDCAs) and their possible impact on dosimetry.

2. INTRODUCTION The physics of radiotherapy, and specifically the physics of brachytherapy covers a large number of topics that are all strongly interrelated. It deals with the understanding of the phenomenon of radioactivity and radioactive decay. Interaction of the emitted ionizing radiation with matter -with the different processes of ab- sorption and scatter effects: the photoelectric effect, compton ef- fect, and pair production at high energies- is essentially the same in brachytherapy as in external beam therapy and will therefore only be touched on in this chapter. Factors that influence the tissue re- sponse to the absorption of radiation, with variations in dose dep- osition over time, fraction size and interfraction intervals, is the realm of radiobiology. What remain as important topics to discuss here start with the characterization of the radioactive material and the physical sources used in brachytherapy: the different source types and the influence of the encapsulation of the sealed sources. Then, for accurate dosimetry, i.e. predictive dose planning using dedicated brachytherapy treatment planning computers, we need This chapter deals with the first steps in understanding of brachytherapy physics in this rapidly evolving field of treat- ment with ionizing radiation. The chapter starts with an ex- planation of the physical phenomena of radioactive decay and the properties of radionuclides that are (or have been) widely used for the superficial, interstitial, intracavitary, and endolu- minal treatment of tumours. The properties of these nuclides are discussed in relation to their relevance for certain types of clinical application. Drawings of real sources are shown to demonstrate the wide variation of commercial source designs that have been developed as solutions for clinical demands. Treatment duration is a critical factor for the comfort of the patient and the potential workload of an afterloader. It is re- lated to the prescription dose to the target. It needs to be de- termined with high accuracy as the accuracy of dose delivery is directly dependent on knowledge of the time of treatment. The quantity source strength therefore needs to be clearly defined according to national or international recommenda- tions. Methods for determining this strength and performing in-house calibration or validation of newly acquired sources is discussed. Awell-type chamber is usually the preferred type of equipment for this purpose, for which a calibration factor

Brachytherapy Physics: Sources and Dosimetry

4

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

part of the work of the radiation safety officer (RSO), a job often assigned to the medical physics team. Therefore the content of this chapter is confined to a descrip- tion of sources, source types, and their characterization. The ba- sic theory of dose calculations is explained. It does not address the general issues of radiotherapy physics which are already covered in many other textbooks. However, the reader will find separate chapters in this Handbook on Radiation Protection in Brachytherapy (chapter 3), Brachytherapy Equipment and Qual- ity Assurance (chapter 4), Radiobiology (chapter 5), Modern Im- aging in Brachytherapy (chapter 6), Principles of Brachytherapy Planning Systems (chapter 7), Treatment Planning and Evalu- ation (chapter 8), and Reporting in Brachytherapy (chapter 9). Today, gamma-ray sources are the most commonly used sources in the treatment of malignancies. Beta radiation is used only in very specific types of treatments, and neutron emitting sources are rarely used. The most widely used radionuclide in brachyther- apy is iridium-192, in the form of the miniaturised pulsed-dose rate (PDR) and high-dose rate (HDR) stepping sources used in dedicated afterloading equipment. Iridium-192 has also been applied in many manual afterloading techniques in the form of thin wires which are cut to the desired lengths, and in ribbons containing small iridium seeds. Other radionuclides common- ly used in brachytherapy are iodine-125, palladium-103, and cesium-137. There are other gamma-ray emitting radionuclides which are discussed in this chapter for their specific properties, such as cobalt-60, ytterbium-169 and thulium-170. Eye lesions are sometimes treated with plaques that are covered with a layer of the β-emitter ruthenium-106. Specific sources containing the β-emitting radionuclides strontium-90 and yttrium-90 have been developed for use in endovascular brachytherapy, but in practice this technique was more or less abandoned several years ago (11, 37). The focus of this chapter will be on photon emitting sources. For further information on the production methods and phys- ical properties of sources that are currently in use or were used more extensively in the past, such as radium-226, gold-198, or on developments of other nuclides like americium-241 and samar- ium-145, the reader is referred to other publications, e.g. (2, 3). Note: in this physics and dosimetry chapter the radionuclides will be referred to by their name and atomic weight. In the clinical chapters of this Handbook, the atomic weight is usually omitted for reasons of brevity. A clarification will be given on the methods of calculation of dose deposition to a point in tissue at a given distance from a source. This chapter shows the mathematical background of the algorithms used in treatment planning systems. In addition, some aspects of 2-D and 3-D dose calculation other than dis- tance, such as oblique filtration and source anisotropy, are briefly discussed. Dose characteristics of a single point-source and of line-sources are shown. Brachytherapy is team work. Radiation oncologists, medical phys- icists and radiation technologists work jointly to obtain the best results for their patients. Education and training are crucial parts of each successful team, a statement that cannot be repeated too often as shown in several reports on incidents and accidents (22, 25). The clinical parts of this Handbook will probably be read and used most frequently by the radiation oncologists. As medical specialists, they are responsible for the treatment of their patients

and specifically for the correct geometric localisation of the appli- cator in order to treat the defined target volume adequately. Con- sistent applicator placement is crucially dependent on the skills of the radiation oncologist. The radiation oncologist is responsible overall for the whole procedure, in which it is primarily the phys- icist’s task to ensure that the treatment is delivered accurately and safely during all steps and in accordance with the radiation oncol- ogist’s prescription. The physicist must ensure that sources of cor- rect strength and type are accurately positioned in the applicators, and that the source positions are accurately reconstructed either from the 2D or –preferably- 3D reconstruction images together with the organ and tumour structures. An accurate and consistent treatment planning procedure is key. High-quality performance of the afterloading systemmust be guaranteed. The radiation tech- nologist is a co-ordinating specialist in the brachytherapy team, supporting the work of oncologists and physicists, with dedicated tasks that may differ from institute to institute. 3.1 Radioactivity Radioactivity is a phenomenon, discovered more than 100 years ago, in which ionising radiation is emitted by the nuclei of ra- dionuclides. This radiation can be in the form of particles, or electromagnetic radiation, or both. The process of radioactive decay or disintegration is a statistical phenomenon. The number of atoms that disintegrate per unit of time is proportional to the number of atoms in a given amount of source material. This rate of decay of a radioactive source is referred to as its activity. The SI (Système International) unit for activity is the bequerel (Bq), which equals to 1 disintegration per second. Another unit, the curie (Ci) has been in use for many decades and was defined by the activity contained in 1 g of the radionuclide radium-226. It is equal to 3.7 10 10 Bq. Consequently, 1 Bq equals 2.70 10 -11 Ci. Note: in practice, units as those discussed in these chapters can be altered by various factors of 10 through the use of appropriate prefixes. For instance, 10 -12 = pico (p), 10 9 = giga (G). So one can write: 1 Ci = 37 GBq or 1 MBq = 27 µCi. Radioactive decay occurs spontaneously, and there is no way to predict when each individual atom will disintegrate. However, on the average, one can state that in a given time frame, called the half-life , T 1/2 , half of the atoms will disintegrate. In the next half-life, one-half of the remaining atoms will decay, etc. Conse- quently, the number of atoms and the activity, A , of any radionu- clide vary exponentially over time as given in the formula: 3. RADIOACTIVE SOURCES

period) (time -e n)

A

A

(actual)

=

calibratio (at

•

(2.1)

1/2 0.693 T

=

In this equation the decay constant of the radionuclide λ is used (in units s -1 ). The decay formula can also be written as follows:

Brachytherapy Physics: Sources and Dosimetry

5

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

Table 2.1 Some properties of γ-ray-emitting sources discussed in this chapter. Only penetrating energies above 10 keV have been considered. The maximum specific activity is calculated for ideal pure point sources. Data were taken from the Table 5.2 of reference (3).

Photon energy range (MeV)

Effective energy, E eff (MeV)

Maximum specific activity (GBq mg -1 )

Γ δ . (μGy h -1 MBq -1 m 2 )

HVL (mm Pb)

Name

Symbol

Half-life, T ½

Z

A

Cesium Cesium Cobalt

Cs Cs Co Au

55 55 27 79 53 77 47 88 69 70

137 131

30.07 a 0.032-0.662 9.689 d 0.029-0.034

0.652 0.030 1.257 0.417 0.028 0.398 0.021

7.0

3.202

0.0771 0.0151 0.3059 0.0545 0.0348 0.1091 0.0361

0.030

3,809.13

60

5.27 a

0.347-2.159

12.0

41.91

Gold

198 125 192 103 226 170 169

2.695 d 0.069-1.088 59.49 d 0.027-0.035 73.81 d 0.061-1.378 16.991 d 0.020-0.497

2.8

9,055.12

Iodine Iridium

I

0.025

650.15 340.98

Ir

3.0

Palladium Radium Thulium Ytterbium

Pd Ra

0.008

2,763.13

1600 y

0.047-2.45

-

13.0 0.17 0.23

0.0366 221.07 893.29

-

Tm Yb

128.6 d 0.048-0.084 32.015 d 0.021-0.773

0.067 0.131

0.00053 0.0431

Note to the table: these data were taken from NIST Physical Reference Data and NUDAT vs. 2.0, both accessed in 2004 for the purpose of a book publication (3). Furthermore, data from ICRP 21 Figs. 50 and 51 were used (24). See Baltas et al. for the notes to the original table, the references, and further details (3).

Table 2.2 Some properties of the ß-ray emitting sources that contain yttrium-90, stron- tium-90 or ruthenium-106 radionuclides. Data were taken from Dutreix et al . (16) and Soares (50).

T period time

(

)

Maximum electron

2 1

Mass for 100 MBq (μg)

1/2

λ (s -1 )

Radio Nuclide

Half-life T ½

A

A

=

•

(actual)

calibratio (at

n)

(2.2)

Energy (MeV)

90 Y

64.1 h 3.01 10 -6

0.005

2.28

90 Sr / 90 Y *

28.2 y

7.60 10 -10

19.2

0.54, 2.28

In these formulae time period = (actual time – time at calibra- tion) . If the half-lives from the following Table 2.1 and Table 2.2 are converted and expressed in s, then λ values can easily be found using the equation (2.1) given above. Table 2.1 shows the γ-ray and Table 2.2 the ß-ray emitting radio- nuclides that are mentioned in the following chapters, and some of their properties. To correct for the decay of a radioactive source to obtain an ac- tual dose rate value at the time of application (or: the activity of that source), the initial dose rate at the time of calibration must be multiplied by a decay factor. For instance, for a time period of 24 days between the time of calibration of the source and the time of its clinical application, the decay factor for an iridium-192 source with a half-life of 74 days is 0.80 (see Table 2.3). For iridium-192 sources, the correction to the dose rate for decay is very close to 1% per day. For iodine-125 with a half-life of about 60 days, one needs to correct for the decay only slightly more than for iridium-192 (see Table 2.4). For a treatment duration in brachytherapy which is much short- er than the half-life of the given radionuclide, the calculation of treatment time can be performed assuming that the activity or the source strength remains constant. For sources such as ce- sium-137 with very long lives this will be the case. Due to the long half-life of 30 years, the activity of cesium-137 sources can adequately be corrected for decay twice a year only during its clinical use. For cobalt-60 with a T 1/2 of about 5 years, a monthly correction of about 1% is needed. When source decay is short compared to the duration of the application, it may be necessary to take into consideration the

106 Ru / 106 Rh*

372 d 2.19 10 -8

-

3.55

* the two nuclides are in radioactive equilibrium. λ = ln2 / T 1/2 , see text for a definition.

Table 2.3 Decay factor for iridium-192.

Iridium-192

Days

0

2

4

6

8

0.94 0.86 0.78 0.71 0.65 0.59

0.93 0.84 0.77 0.70 0.64 0.58

0

1.00 0.91 0.83 0.75 0.69 0.63

0.98 0.89 0.81 0.74 0.67 0.61

0.96 0.88 0.80 0.73 0.66 0.60

10 20 30 40 50

Table 2.4 Decay factor for iodine-125.

Iodine-125

Days

0

2

4

6

8

0.93 0.83 0.74 0.66 0.59 0.52

0.91 0.81 0.72 0.64 0.57 0.51

0

1.00 0.89 0.79 0.71 0.63 0.56

0.98 0.87 0.77 0.69 0.61 0.55

0.95 0.85 0.76 0.67 0.60 0.53

10 20 30 40 50

Brachytherapy Physics: Sources and Dosimetry

6

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

decrease of strength during treatment time. A simple approxi- mation to the calculation of total dose of an implant is to use the dose rate value half-way through the planned application. Note: the activity of all sources at the end of 7 half-lives is less than 1%. Ten half-lives will reduce the source activity by about 10 -3 . This is a simple rule of thumb which is often used in calculations of remaining activity, e.g. for dealing with radioactive waste disposal. Sometimes sources are permanently implanted into patients, such as iodine-125 seeds in prostate treatment. It is then nec- essary to know the total number of disintegrations that occur in an infinite period of time. A radioactive source that decays expo- nentially over an infinite time with a half-life T 1/2 is equivalent to a virtual source that radiates at a constant rate (thus, equal to its initial rate) for a given time period, T a . T a is called the “average life” of the radionuclide and equals. The factor 1.44 (or: 1 / 0.693) derives from a simple mathematical operation ( integration over time ), and is universally applicable. As a result, the average life T a of iodine-125, commonly used for permanent implants of the prostate, is about 86 days (59.49 / 0.693 = 85.8 d). The total dose of an implant is equal to the prod- uct of the initial dose rate and T a . Many radionuclides undergo successive transformations in which the original parent nuclide gives rise to a daughter radio- nuclide (3, 16). If the half-life of the parent is longer than that of the daughter, then after a certain period of time a condition of equilibrium will be achieved. This means that the ratio of daugh- ter activity to parent activity will become constant. The apparent decay rate of the daughter nuclide is then governed by the half- life of the parent. Two examples were given in Table 2.2, parent strontium-90 together with daughter yttrium-90 and parent ru- thenium-106 together with daughter rhodium-106. 3.2 Decay schemes of radionuclides The choice of a radionuclide to be applied in brachytherapy sys- tems depends on a number of considerations and on the physical properties. The radiation emitted by the radionuclide determines how a source can or should be constructed and is therefore im- portant for the possible source geometry and structure. The half- life, T ½ , determines if a source can be used in permanent or tem- porary implants or in both types. The specific activity defines the possible source size and dose rate. The energy of emitted radia- tion influences the dose distribution within tissue, and also im- poses the measures to be taken for radiation protection. Density and atomic number of the radionuclide are important for radio- graphic visibility and source localisation, and influence the (an) isotropy of the resulting dose distribution around the source. The radiation emitted by a radionuclide can be clarified with a graphical representation of the energy levels of the moth- er-daughter nuclides. Fig. 2.1 and Fig. 2.2 present two such de- cay schemes of commonly used radionuclides, iodine-125 and cesium-137. The schemes of these two nuclides are relatively simple, compared to a very complex decay scheme of the irid- T a = T 1/2 0.6 93 = 1.44 T 1/2 (2.3)

Fig. 2.1 Schematic representation of electron capture decay of iodine-125 to the first excited state of tellurium-125. The disintegration energy for the decay is Q EC = 0.1858 MeV. There is a single γ-ray of 0.035 MeV with an intensity of 6.68% emitted, whereas there are several characteristic x-rays emitted in the range 0.027 to 0.032 MeV as a result of internal conversion processes. The average number of photons emitted per disintegration with energy above 10 keV is 1.4. The half- life for iodine-125 decay is 59.49 days (3). (Courtesy: D. Baltas)

Fig. 2.2 Schematic representation of β- decay of cesium-137 which decays mainly (94.4%) to the second excited state of barium-137. The disintegration energy for the decay is Q β - = 1.1756 MeV. In practice a single γ-ray of 0.6617 MeV is emitted with an absolute intensity of 85.1%. The aver- age number of photons emitted per disintegration with energy above 10 keV is 0.9. The half-life for cesium-137 decay is 30.07 years (3). (Courtesy: D. Baltas)

ium-192 nuclide which has a multitude of photon energies in its spectrum. The iridium-192 scheme is not presented in this publication. Decay schemes, emission spectra, production methods and in- formation on source construction of other widely used radio- nuclides and sources can be found in other textbooks, such as in chapter 5 of Baltas et al. (3) or Ballester et al. (2). 3.3 Types of sources In brachytherapy different types of sealed or encapsulated sourc- es are used. They all contain a certain amount of a radionuclide that is encapsulated in layers of a metal such as platinum, titani- um, or stainless steel, or in some kind of thin foil in the case of ß-emitters. Various types of sources such as tubes, needles, wires, pellets, seeds, and a single stepping source connected to a cable, are available (see Fig. 2.3 and Fig. 2.4). The definition of the length of a source should not lead to confu- sion among the members of the brachytherapy team. In order to

Brachytherapy Physics: Sources and Dosimetry

7

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

3.4 Clinical application in brachytherapy Although the radionuclide radium-226 is very well-known, it was abandoned for clinical use many years ago because of its long half life and the associated environmental risks. Another reason is that the daughter product radon-222 in its gaseous form can escape from damaged sources, leading to severe risks of radioactive contamination. In the first instance, the radionu- clide cesium-137 was used as a replacement for radium-226, and cesium-137-containing sources were constructed in such a way that a very similar dose distribution was obtained around the sources. These sources were commonly available in the form of needles and tubes. Cesium-137 material is readily available from nuclear power reactor waste products. Both radium-226 and cesium-137 decay with relatively high energy photons, as does cobalt-60, which is apparent from the higher values of the half value layer ( HVL , the thickness of a radiation protection shield or barrier that reduces the exposure with a factor 2; values for lead are shown in Table 2.1) of these radionuclides. Therefore, protective walls and -devices need to be thicker for these sources and are more expensive than with most other radionuclides. Iridium-192 is produced in nuclear reactors and has been the most popular radionuclide in brachytherapy since the nine- teen-sixties. It combines a relatively high mean photon energy with an HVL which is considerably smaller than, e.g., with cesi- um-137 or cobalt-60. Therefore, less shielding thickness in lead barriers or concrete walls is required, thus reducing facility costs. The main advantage of the iridium-192 in radioactive sources, however, is the very high specific activity . The specific activity is defined as the maximum activity of a radionuclide that can be contained in 1 g (or 1 mg) of the material. These values are shown for the sources listed in Table 2.1. A very small amount of pure iridium material can still have a very high yield of photons, which means that the source can be small. This property allows the construction of thin iridium wires of only 0.3 mm diameter for manual brachytherapy applications, or miniaturized highly activated sources for use in HDR or PDR afterloaders. Source guiding tubes or needles can have an outer dimension of 1.9 mm or less. The half-life of iridium-192 is close to 74 days, requiring sources to be exchanged at intervals of 3-4 months.

speak the same language, three different lengths of a source are defined: the physical length, PL , the active length, AL , and the equivalent active length, EL , as shown in the Fig. 2.3. These sources can usually be delivered in a wide variety of source strengths. In addition, designs of four seed type sources developed specif- ically for use in prostate permanent implant brachytherapy are shown here with their construction details in Fig. 2.4. These four examples form a small selection of the many seeds, each with their own details that, have been proposed to the market by a multitude of vendors. See for example in the ESTRO Web pages where dosimetry data useful for brachytherapy treatment plan- ning have been collected: http://www.estro.org/about/govern- ance-organisation/committees-activities/tg43 (18).

Fig. 2.3 Different types of sources used in brachytherapy. AL : active length, PL : physical length, EL : equivalent active length, s : spacing between the centres of seeds.

Fig. 2.4 In this graph iodine-125 seed sources for use in prostate implants from 4 different vendors are shown: the Oncura/GE-Healthcare (previously Amersham Health) model 6711, the Best model 2301, the Bebig/Theragenics model I25.S06, Elekta (previously Nucletron) I-125 SelectSeed. The seeds have a typical outer diameter of 0.8 mm and a length of 4.5 - 5.0 mm. Figures taken from (18).

Brachytherapy Physics: Sources and Dosimetry

8

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

The data in Table 2.1 shows that the specific activity of radionu- clides is more or less inversely related to theier half-life. The val- ues show that, for example, the mass needed for a certain activity of iridium-192 with a half-life of 74 days is much smaller (by a factor of about 100) than for the same activity of cesium-137 with a half-life of 30 years. It is therefore not feasible to use ce- sium-137 as a similarly miniaturised source for brachytherapy applications. Note that, in order to be suitable for use in HDR brachytherapy, the specific activity is not the only factor that determines the ap- plicability of a given radionuclide. The yield , i.e. the number of useful photons per disintegration, the presence of beta emission, the energy of the emitted photons, and the radiochemical puri- ty also influence the suitability. Table 2.1 shows a significantly lower value of the specific activity of cobalt-60 vs. iridium-192 (41.91 vs. 340.98 GBq mg -1 ), but this is again compensated for by a factor of 3 because of the higher value of the gamma ray constant , Γ δ , of cobalt-60 (0.3059 vs. 0.1091 μGy h -1 MBq -1 m 2 ). As can be seen from the units, the gamma ray constant Γ δ is a measure of the emitted energy at a given reference distance from the source per unit of activity. Iodine-125 is mostly used in the form of seeds for permanent implants. The low photon energy is easily absorbed in layers of tissue, so that with deep seated implants (prostate) the risk of exposure to other individuals (caregivers, family members) is minimal. The photons emitted by palladium-103 have an even lower energy and this radionuclide has a shorter half-life. Appli- cations with palladium-103 therefore have a somewhat different radiobiological effect. The same holds for application of the ra- dionuclide cesium-131, which is not discussed here any further. Research of vendors is continuously directed towards new prod- ucts and new isotopes. Both ytterbium-169 and thulium-170 have elicited increasing interest. From Table 2.1 it is clear that the relatively low effective energy of the emitted photons is inter- esting from the point of view of radioprotection. Ytterbium-169 has a relatively short half-life which has implications for frequent source exchange, while thulium-170 has a relatively low value of the specific activity. The future will show if the application of these radionuclides in temporary applications (e.g., in HDR or PDR systems) is successful or not. Considering the overall numbers of patients treated with brachytherapy, the γ-ray-emitting sources are the most fre- quently used radionuclides. Tumor sites where brachytherapy using iridium-192 or cesium-137 is the treatment of choice or is part of this treatment are: uterine cervix, endometrium, vagina, urethra, oral cavity, oropharynx, bladder, prostate, oesophagus, bronchus, breast, lip and skin. For treatment of the prostate, large numbers of implanted iodine-125 or palladium-103 seeds are used, typically 60 to 80. For temporary applications for shallow lesions of the skin or the eye, a ß-emitter is sometimes used. ß-particles have a limited range when interacting with matter, depending on their energy. Their penetration in tissue is therefore limited in depth. Appli- cators employing the ß-emitters strontium-90 or ruthenium-106 are predominantly used to treat such shallow lesions, because the 50% isodose curve is at about 3 mm while these radionuclides are not used for lesions with extension over 5 mm in depth. Sources with a ß-emitter have been used for endovascular inter- vention techniques for prevention of vascular restenosis before chemically coated stents were introduced. Iodine-125 is intensively used in the form of seeds as well as to construct individual eye plaques. For distances from 1 up to

Table 2.5 Physical properties of radionuclides and their relevance for application in brachytherapy. From Table 5.3 in reference (3).

Physical property Radiation emitted

Relevance in brachytherapy

Source geometry and structure Determines if permanent or temporary implant or both are practical

Half-life, T

1/2

Specific activity Source size, dose rate Energy of emitted radiation Dose distribution within tissue, radiation protection requirements

Radiographic visibility/ localization, isotropy/anisotropy of dose distribution

Density and atomic number

20 mm, the dose distribution for a point source of iodine-125 is within 10% of the dose distribution from iridium-192 or ce- sium-137 sources. The low energy (<35 keV) allows gold or stainless steel foil to be used at the plaque edge to shield adja- cent structures (retina). The relative biologic effectiveness of io- dine-125, possibly from 1.2 to 1.4 compared with 250-kV X-rays, may be important for use with a radio-resistant tumor like mel- anoma. A summary of the physical properties discussed in this section of radionuclides and their relevance in brachytherapy is shown in Table 2.5.

4. SOURCE DOSIMETRY

4.1 Specification of source strength Several international organisations such as the ICRU (26,27), the AAPM in its report 32 (1), and some national organisations such as the BCRU (5), SFPH (51), NCS (38-40), DIN (14,15) have recommended the specification of the strength of γ-ray sources in terms of the quantity kerma rate to air at the point along the transverse axis of the source. The ICRU (27), ESTRO (17), IAEA (23) recommend the use of the Reference Air Kerma Rate for this purpose. In this chapter the acronym RAKR will be used occa- sionally for clarity. The RAKR , written as the symbol K . ref , is the kerma rate to air in air at a reference distance of 1 meter from the centre of the source, after correction for air attenuation and scattering. The quantity kerma , from kinetic energy released to matter , refers to the kinetic energy of charged particles, for example electrons and positrons, that have been liberated by uncharged particles such as by photons emitted by the brachytherapy source. Kerma does not include the energy that has been expended against the binding energies of these charged particles, even if this is usually a relatively small component. The name of the SI unit for kerma is gray (Gy), with 1 Gy = 1 J/kg. The quantity air kerma rate is the kerma per unit of time, K . ref , and is expressed in Gy s -1 or a multiple of this unit. The reference air kerma rate or RAKR is the air kerma rate at a defined reference distance, which is taken as 1 m in the reports cited above. For clarity, it is noted here that in AAPM task group reports, a slightly different approach was followed to specify the strength of brachytherapy sources, i.e. the air kerma strength , S K . S K was

Brachytherapy Physics: Sources and Dosimetry

9

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

introduced and defined as the product of the air kerma rate in free space at a measurement distance r from the source centre and the square of the distance r . The reference air kerma rate K . ref and air kerma strength S K are thus mutually related through the inverse square law to the reference distance r 0 as

ufacturer. The medical physicist is, however, always responsible for the validation of the source strength. The measured value should be traceable to a primary standards dosimetry laboratory (PSDL) such as PTB, NPL, or the National Institute of Standards and Technology (NIST) (35). In sections 2.3 and 2.4 on source dosimetry and dose calculation in brachytherapy, the two quan- tities, reference air kerma rate K . ref and air kerma strength S K , will be used in more detail in the given formalisms. For ß-emitters, the AAPM report (37) recommends that the source strength should be expressed in terms of dose rate in wa- ter at a reference distance of 2 mm. In a few reports, the most recent scientific views and recommendations for brachytherapy beta source dosimetry are discussed (28, 41). A recent project, initiated in the joint group of European Associ- ation of National Metrology Institutes (EURAMET, project TP2. JRP6: Increasing cancer treatment efficacy using 3D brachythera- py ; see http://brachytherapy.casaccia.enea.it/), aimed to develop primary standards for brachytherapy source strength specifica- tion in terms of dose-to-water at 1cm from the transversal axis of a source. Clinical introduction of this new concept has not yet been established by the professional societies. The introduction of such standards will need a reconsideration of the procedures for in-house calibration and for implementation in dose calcu- lation algorithms. Research groups working with or in contact with vendors and the standards laboratories must be included in the working groups and task groups of the professional societies to avoid any disparities occuring either from the dose-to-water concept for calibration or from new consensus algorithm types, that might otherwise distract and confuse individual users. In practice the new procedure would mean that for a given source, the reference air kerma rate calibration and the conversion factor of air kerma rate to dose rate will be replaced by a direct dose-to- water calibration. Coordination for this transition from reference air-kerma to dose to water across the entire field of brachythera- py is key. A distinct requirement from the clinical user’s point of view is that such a new approach should at least give the same, but preferably, lower source calibration uncertainties. 4.2 Calibration of sources The manufacturers of sources issue some form of calibration test certificate. This certificate is usually not directly traceable to a Standards Laboratory. However, all sources should have cali- brations traceable to a national or international standard before clinical use (27). As indicated above, and in accordance with the

2

S K =

r 0

(2.4)

K

ref

The recommended unit for air kerma strength is μGy m 2 h -1 . It has been denoted by the symbol U in such a way that

(2.5)

1 U = 1 µGy m 2 h -1 = 1 cGy cm 2 h -1

For elongated rigid sources, the direction from the source centre to the reference point at 1 m is defined in the transversal plane of the source, i.e. at right angles to the long axis of the source (Fig. 2.5). For low dose rate (LDR) brachytherapy applications it is convenient to use the unit μGy h -1 . The current and world-wide adopted approach for brachythera- py dose calculation is based on the AAPM TG-43 dosimetry for- malism (35, 44), which relies on superposition of single-source dose distributions obtained in a liquid water phantom with a fixed volume for radiation scattering. Note that, although the air kerma strength S K and the reference air kerma rate K . ref are dimensionally different, the numerical values are equal. A useful overview of conversion factors for reference air kerma rate and air kerma strength is presented in Table 2.6. The reference air kerma rate is determined for specific sources either by the clinical medical physicist or provided by the man-

Fig. 2.5 Schematic representation of the geometry as defined in AAPM Report 21 of the AAPM task group 32 (1) for defining the brachytherapy source strength. A cylindrical source is assumed, with L s being the length of the active part of the sealed source (active core). The radial distance r used for the measurement is taken in the transversal plane of the source, determining the air kerma rate in free space, K . a (r) . Figure from (3). (Courtesy: D. Baltas)

Table 2.6 Units and unit conversion factors for reference air kerma rate and air kerma strength (3).

Source Strength Quantity

Symbol

SI

Current unit

Conversion factor

μGy h -1

(1/3600) x 10 -6 Gy s -1 = 2.778 x 10 -10 Gy s -1

K .

Gy s -1

Reference air kerma rate

mGy h -1 mGy h -1

(1/3600) x 10 -3 Gy s -1 = 2.778 x 10 -7 Gy s -1

ref

1.0 x 10 3 μGy h -1

U, 1U = 1 μGy m 2 h -1 = 1 cGy cm 2 h -1

(1/3600) x 10 -6 Gy m 2 s -1 = 2.778 x 10 -10 Gy m 2 s -1

S

Air kerma strength

Gy m 2 s -1

K

(1/3600) x 10 -12 Gy s -1 Bq -1 m 2 = 2.778 x 10 -16 Gy s -1 Bq -1 m 2

μGy h -1 MBq -1 m 2

J kg -1 or Gy s -1 Bq -1 m 2

Γ

Air kerma rate constant

δ

μGy h -1 MBq -1 m 2

1.0 U MBq -1

Brachytherapy Physics: Sources and Dosimetry

10

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

viewpoints of the professional societies for medical physics , the medical physicist is always responsible for such a calibration. The procedure should rely on a properly calibrated standard ra- dioactive source or he/she should perform a calibration based upon a calibrated measuring device. Recommendations for cali- bration procedures have been published by several organisations including the ICRU (27), the IAEA (23), the NCS (38, 39) and the AAPM (36). The aim of the calibration is two-fold: to ensure that the value entered into the treatment planning system agrees with the source calibration certificate to within a predetermined limit and to ensure traceability to (inter)national standards. The traceability is important as it simplifies national and internation- al comparison of treatment results and improves consistency in clinical outcome. As stated before, the quantities recommended to specify the source strength are the reference air-kerma rate K . ref and air-ker- ma strength S K . Other quantities for specifying source strength, such as the apparent activity (Bq or Ci), are now considered ob- solete. Note that a source specification based on contents of activity may be obligatory in the realm of radiation protection, where the RSO should comply with the legal requirements of the license. Generally, the presence of all radioactive material in the institute needs to be described in terms of activity. The calibration chain proposed by the ICRU (27) to guarantee traceability is as follows: • In the primary laboratory , the RAKR of a source of a given radi- onuclide is measured with a primary ionisation chamber. The source so calibrated is a primary standard source . • The secondary laboratory receives a primary standard source to calibrate the secondary ionisation chamber and calibrates secondary standard sources . The only requirements are that the sources have similar shape and size, same filtration and same radionuclide. • The user may order a calibrated source from the secondary standard laboratory in order to calibrate his own (well-type) ionisation chamber and is therefore able to calibrate any source of the same radionuclide, with the same filtration and similar shape and size. A long half-life source can be used to check the reproducibility of the measuring device in the department. High intensity iridium-192 sources used in afterloading devices require special considerations. In the absence of a suitable pri- mary standard in many countries, the interpolative free-air sec- ondary standard method has become the interim standard for measurement of high intensity iridium-192 source strength (17, 21, 23). Briefly this approach consists of measuring air kerma rate on the transverse axis of an HDR source at a distance of 10- 100 cm in a free air geometry using a thimble-ionisation cham- ber with a build-up cap thick enough to establish secondary electron equilibrium at the highest photon energy encountered. The method can be used as an in-house procedure to calibrate a more simple measurement set-up such as a well-type chamber or a solid phantom. Note that some laboratories provide a ser- vice for direct calibration of a suitable HDR well-type chamber instrument for HDR iridium-192 sources. Total reference air kerma rate of the radioactive sources can now be verified using a well-type ionisation chamber, previously cal- ibrated with at least one source provided by a National Standard Laboratory. At the time of calibration, the well-type ionisation

chamber must be corrected to take into account the energy, the dimensions, and the position in the cavity of the radionuclide, if these are different from the calibration source. The methods used for calibration are either based on a so-called in-air measurement technique (e.g., Fig. 2.6), on the use of a well-type ionization chamber (Fig. 2.7), or on a solid phantom dedicated for calibration purposes. In principle, any source can be calibrated with these methods, but there are some practical limitations. With in-air and solid phantom calibrations the sig- nals, typically obtained when using low strength sources (i.e., LDR seed sources) are small, and the final uncertainty in the air-kerma strength or reference air kerma rate may be unneces- sarily high. For HDR sources, however, all calibration methods discussed here may be considered. Most of the present recommendations from national or inter- national organizations rely on the use of a dedicated well-type ionization chamber. Even though such chambers provide an easy, fast and reliable method for source calibration, it must be kept in mind that in-air calibration is a more fundamental meth- od. However, well-type chambers offer the best in practice for a brachytherapy institution: it is a reliable, reproducible and easy to use method. Several primary or secondary standards labo- ratories, or (in USA) ADCLs, provide users all over the world with a calibration factor for their instrument which must be used with its accompanying electrometer and inserts. The procedure for in-house calibration is then simple and can be repeated at a few intervals in the lifetime of the local HDR afterloader source. A general requirement is to use the available dosimetry equip- ment at each source exchange before any clinical treatments take place. In many institutions, the procedure is repeated at the end of the life time of the source. In this way, the old and the new sources are checked in a short period of time, the stability of the measurement system is demonstrated and confidence in the use of the correct decay factor is established. In most commercial well type chambers, a guide tube is provid- ed to hold the source catheter along the axis of the cylindrical well. The sensitivity of the chamber versus the source position along the guide tube must be checked: the determination of the so-called “sweet spot”. This can be done by varying the position of a small source along the length of the guide tube. Usually, the signal is within about 1% over a trajectory of several mm of the guide tube, indicating that it is very easy to have highly repro- ducible readings. It should be noted that well type chambers with thick internal walls may show energy dependence which is particularly em- phasised when calibrating low energy photon sources, such as iodine-125 and palladium-103. For instance, the filtration of low energy photons depends on the thickness of the wall of the source holder. It is important to understand that a well type chamber in general exhibits a larger dependence on the source design compared to Farmer type chambers. The well type cham- ber’s calibration coefficient is valid only for the type of source for which it has been calibrated. This is specifically true for the low energy photon emitters. The instrument (the ion chamber, electrometer, and cables) itself should be checked for linearity, leakage, and for consistency of the readings, with a measurement at regular intervals with a long lived source, e.g. a cesium-137 source, once a year. Any deviation for the source under consideration larger than 3% must be inspected by repetition of the measurement and/or by independent means.

Brachytherapy Physics: Sources and Dosimetry

11

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

age from the vendor value by 0.3% (10). Measurements with the well chambers were 0.5% ± 0.1% higher than the vendor-pro- vided source strength. Measurements with the Farmer chamber were 0.7% lower than the average well chamber results and 0.2% lower than the vendor-provided source strength. The compar- isons were made using equipment with traceability to different standards from the laboratories. 4.3 Quality control of sources Quality management is a responsibility of the medical physicist and/or the RSO in the radiation therapy department. A detailed description of radiation safety issues and a general overview of quality assurance and quality control procedures regard- ing sources is given in chapter 3 on Radiation Protection in Brachytherapy. A very comprehensive overview of quality man- agement in brachytherapy is given in the book ‘Achieving Quali- ty in Brachytherapy’ by Thomadsen (53). With regard to brachytherapy sources, in this paragraph only the following important topics that must be addressed are listed: • The necessity to have an accurate logbook of all available sourc- es, their identification numbers, their strengths, and storage position. • The necessity to perform leakage testing using a wipe test pro- cedure, according to the requirements of the licence of the de- partment with regard to the minimum frequency. • The necessity to perform inspections on a regular basis of safe- ty measures, the loggings, the vaults, and the (written) proce- dures. • The necessity to perform a source strength verification prior to clinical use of a brachytherapy source, as described in the previous section. • The necessity to perform a check of the uniformity of the ac- tivity distribution, e.g. the uniformity of linear sources, source needles, ribbons, or the uniformity of source strength in a batch of sources. • The necessity to –if possible- inspect the source integrity before clinical use. • The necessity to validate the dosimetric parameters of the sources that are used in the dose calculations. It is noted that high-dose rate sources constitute a high risk of ra- diation exposure. Therefore direct wipe testing and visual inspec- tion are not recommended for these sources. Often checks are per- formed on the applicators or appliances such as transfer tubes to inspect if any radioactive contamination from a leaking source is present on these materials.

Fig. 2.6 A typical in-air calibration jig suitable for calibrating an HDR or PDR 192 Ir source to be positioned 10 cm left and right from a centrally placed Farmer type ionisation chamber with a build-up cap. A small plastic tube is used in this jig to keep the catheters exactly 20 cm apart. Readings with the source in the left and right catheter are averaged to correct for positional inaccuracies of the tubes. (Courtesy: J. Venselaar)

Fig. 2.7 Example of a well-type chamber with electrometer, a system nowadays generally recom- mended in societal reports for use in the clinic. (Courtesy: J.Venselaar)

A source calibration accuracy of ≈3% relative to existing air ker- ma standards is recommended. If the institution’s verification of source strength disagrees with the manufacturer’s data by more than 3%, the cause of the disagreement should be investigated. It should be noted that the recommended 3% tolerance between manufacturer and institution calibrations discussed above ap- plies to the mean of a batch of sources . Since individual sources may differ from the mean by a greater amount, a maximum de- viation from the mean of 5% is acceptable for individual sources in such a batch (6). Although traceability to primary standards for HDR iridium-192 sources is still not readily available for all users world-wide, those standards that have been developed seem to be at an acceptable level. In a paper on a comparison of HDR source strength meas- urements for these sources using equipment with traceability to different standards, Carlsson Tedgren et al . showed that the measured source strengths varied by 0.8% and differed on aver-

5. DOSE CALCULATION IN BRACHYTHERAPY

A full description of what is called here the conventional meth- odology for the calculation of the absorbed dose around a brachytherapy source can be found in many other text books. In present-day practice, however, this conventional method has been abandoned in modern treatment planning systems as it has been replaced by another formalism. This new formalism was published in a report of AAPM Task Group 43 , generally indicat- ed as the TG-43 formalism (35). This report was updated in 2004 (44), supplemented in 2007 (45), and has become a standard practice in computerised dose planning systems. Nevertheless,