Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The Use of Electrons for External Beam Radiotherapy
15.1
Introduction
Electron beams have been used in radiotherapy since the 1940s but did not gain widespread use until the 1970s with the commercial development of linear accelerators (linacs). Electrons lose energy as they traverse a medium through various elastic and inelastic collisions with atomic electrons and the nucleus. Inelastic collisions with the nucleus result in a radiative loss of a photon, called bremsstrahlung production. The bremsstrahlung radiation created by electron beam interactions with high atomic number scattering foils and other materials in the beam path is called photon contamination. Electrons are a charged particle with a finite range and are therefore suitable for treatment of superficial or shallow targets.
As charged particles, megavoltage electrons undergo more scattering than megavoltage photons, particularly in air. It is because of this that electron beams use cones with beam trimmers that are close to the patient surface to minimize scatter outside the treatment field. To further customize the shape of the beam, cerrobend cutouts are placed on top of the trimmers.
The electron scatter changes with initial electron energy and depth in the medium. This makes matching of electron fields with other electron fields or photon fields difficult. The many sources of scatter in the electron beam path (scattering foils, linac dose chamber, electron cone) also affect the distance dependence of the beam output. The concept of an effective source position is used to describe this.
Three factors that greatly affect the dose distribution of electron beams are beam obliquity, surface irregularity, and tissue inhomogeneity. Traditional pencil beam algorithms do not accurately calculate dose under these conditions. Monte Carlo calculations for electrons are becoming increasingly common and perform much better in these conditions.
American Association of Physicists in Medicing (AAPM) Task Group (TG)-25, Clinical Electron Beam Dosimetry, provides a good overview of electron beam characteristics. Chapter 8 in the IAEA Radiation Oncology Physics Handbook also gives a nice review. Hogstrom and Almond published a review of electron beam therapy physics. AAPM TG-70, Recommendations for Clinical Electron Beam Dosimetry, produced a supplement to AAPM TG-25 that updated the absolute dosimetry sections to reflect the change to absorbed dose in water standards. The discussions in AAPM TG-25 on obliquity, surface dose, effective source position, air gap corrections, and use of diodes remain valid. The AAPM TG-70 report gives a detailed discussion on the use of electrons in many clinical situations. What follows in this chapter is a brief discussion on electron beam characteristics and the impact of the three factors mentioned above, beam obliquity, surface irregularity, and tissue inhomogeneity. This is followed by a discussion of three common clinical situations: bolus, surface shielding, and compensators. Total skin electron therapy is described in Chapter 24 , and electron intraoperative therapy is described in Chapter 23 .
15.2
Beam Characteristics
Depth Dose
Typical electron beam depth dose curves are shown in Figure 15.1 . They show a relatively high surface dose compared to photon beams of the same nominal energy, a build-up to maximum dose, a rapid fall-off in dose, and then a small low dose component called the bremsstrahlung tail or photon contamination. The photon contamination increases with increasing energy as shown in Figure 15.1 . There are various depth dose parameters that are used to describe an electron depth dose curve: surface dose, depth of maximum dose (R100 or D max ), depth of 90% dose (R90), depth of 80% dose (R80), depth of 50% dose (R50), the practical range (Rp), and the photon contamination. Depth dose parameters do change with field size, but the change is small for field sizes larger than the practical range. In fact, the data from Hu et al. demonstrate that for energies less than or equal to 16 MeV, the changes in any of the range parameters are less than 1 mm from 10 cm × 10 cm to 25 cm × 25 cm cone sizes. For energies less than or equal to 12 MeV this extends down to a 6 × 6 cone. Sample data are shown in Table 15.1 . For higher energies there can be significant differences, as shown in Figure 15.2 . For electrons the surface dose increases with energy, which is the opposite of photons. This is also shown in Figure 15.1 .
TABLE 15.1
R90 (in mm) by Energy and Cone Size from Hu et al.
| Cone Size (cm) | |||||
|---|---|---|---|---|---|
| Energy (MeV) | 6 | 10 | 15 | 20 | 25 |
| 6 | 18.5 | 18.4 | 18.4 | 18.5 | 18.6 |
| 9 | 28.5 | 28.4 | 28.5 | 28.5 | 28.7 |
| 12 | 39.5 | 39.7 | 39.7 | 39.9 | 40.0 |
| 16 | 49.2 | 51.2 | 51.5 | 51.5 | 51.7 |
| 20 | 54.5 | 59.0 | 60.1 | 60.4 | 61.0 |
There are general rules of thumb that can be used to estimate the range parameters, and they are shown in Table 15.2 . These are intended to be quick estimates and cannot replace measured data. The values may differ slightly for linacs of different models. In clinical practice, these values should be tabulated in a convenient format similar to the table in Figure 15.1 and made readily available to the clinicians as an aid in selecting beam energy. In clinical practice the energy is often selected so that the R90 depth is at least as deep as the distal-most aspect of the planning target volume (PTV). The energy should also be limited to reduce dose to any normal tissues or organs at risk (OARs) distal to the target.
TABLE 15.2
Electron Beam Range Parameters Rules of Thumb
| Parameter | Estimation Formula (E is the nominal energy in MeV) |
|---|---|
| R90 in cm | E/(3.2 to 3.3) |
| R80 in cm | E/(2.9 to 3.0) |
| Rp in cm | E/2 |
| Surface dose (%) | E+(70 to 74) |
Output and Cutouts
The output trend of the electron cones will vary by energy. The Radiological Physics Center (RPC), now IROC Houston, has published benchmark data for many linac models. There are also other benchmark data sets as described in Chapter 5 . A sample set of data is shown in Figure 15.3 . As mentioned above, the depth dose is unchanged for field sizes greater than the practical range. Similarly, there is little change in the output of an electron beam down to cutout sizes equal to the practical range. Given that Rp can be estimated by E/2, this implies that only fields less than 3 cm for a 6 MeV beam will have significant changes. This is only an estimate, and the validity for a particular linac should be established at the time of commissioning. AAPM TG-70 describes a method to estimate the depth dose and output of rectangular cutouts of dimension X × Y via the geometric mean:
This requires that a series of cutouts be measured at the time of commissioning. Although it does work for rectangular cutouts, it is not particularly accurate for small or irregularly shaped cutouts. Khan described a method based on the lateral buildup ratio (LBR) concept that more accurately describes changes in output and depth dose. The LBR is derived from the ratio of the percent depth dose (PDD) for a 2 cm diameter cutout to that for the open cone. For irregular cutouts, a method of sector integration is used to calculate the PDD and output. This is easily implemented in computer code but is difficult to use in a manual calculation. Kehwar and Huq have proposed an improvement in this method. In practice, small or irregular cutouts should be measured to confirm the output and depth dose.
For small cutouts, the depth dose curve will shift toward the surface as shown in Figure 15.4 . Therefore both the output and depth dose should be verified because this may change the choice of energy. Figure 15.4 also shows that the effect is more pronounced at higher energies, as expected.
The shape of the isodose lines for electron beams is also energy and depth dependent. Table 15.3 shows the 90% width, proximal depth, and distal depth of 2 cm × 9 cm, 4 cm × 9 cm, and 9 cm × 9 cm cutouts in a 10 cm × 10 cm cone for various energies. The 90% width is measured at the depth of maximum dose. The data show that the margin needed for the cutout is about 1 cm. It should be noted that the 90% width narrows significantly with depth and the distal 90% depth becomes much shallower as the cutout size decreases for higher energies, so computerized treatment planning is recommended to ensure coverage (see Figure 15.5 ). Lead collimation resting on the skin is sometimes used for small fields since the penumbra is relatively wide compared to the tumor. Skin collimation can also achieve better shielding for superficial nearby structures.
TABLE 15.3
90% Coverage for Different Cutout Sizes
| 90% Width at D max Depth for the Smallest Dimension/Proximal 90% Depth/Distal 90% Depth (cm) | ||||
|---|---|---|---|---|
| Cutout Size (cm) | 6 MeV | 9 MeV | 12 MeV | 15 MeV |
| 9 × 9 | 8.0/0.5/1.8 | 8.1/0.7/2.9 | 8.1/0.4/3.8 | 8.5/0.2/4.6 |
| 4 × 9 | 2.8/0.5/1.8 | 2.9/0.6/2.8 | 3.1/0.3/3.6 | 3.3/0.2/4.2 |
| 2 × 9 | 1.1/0.3/1.7 | 1.3/0.2/2.4 | 1.4/0.1/2.8 | 1.4/0.1/3.1 |
The thickness of the cutouts should be sufficient to reduce the transmission to less than 5%. If lead is used, a good rule of thumb is the thickness required in mm is E/2; if cerrobend is used, multiply by a factor of 1.2 (TG-25).
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here
