Simulation for Radiotherapy Treatment Planning


Introduction

Prior to the early 1990s, “simulation” for an external beam radiotherapy consisted of just that: simulating the treatment with a device that mimicked a linear accelerator (linac) in all aspects except the delivery of a therapeutic beam. These devices (now called “conventional simulators”) are equipped with diagnostic x-ray tubes and imaging equipment and are still in use. They have largely given way, however, to computed tomography (CT) simulation whereby a “virtual simulation” is performed on a CT scan acquired with the patient in treatment position. This has enabled the era of 3D conformal radiotherapy (3DCRT, in which targets and normal tissues are delineated on the volumetric CT scan) and image-guided radiation therapy (IGRT) (which uses the CT scan as a reference image). For brachytherapy procedures orthogonal planar imaging is still widely used to locate the applicator and tissue structures, though this is also giving way to volumetric imaging for simulation. A 2010 survey of clinicians performing brachytherapy for gynecologic tumors, for example, showed that the majority of respondents used CT over planar films, though prescription techniques were more often based on 2D definitions.

Technical aspects of CT simulators and QA are considered further in Chapter 7 and in American Association of Physicists in Medicine (AAPM) Task Group (TG)-66, Quality Assurance for Computed-Tomography Simulators and the Computed-Tomography-Simulation Process. Some specific issues to consider here are:

  • Hounsfield units to electron density conversion: The pixel values in CT scans are in Hounsfield units (HU), and a conversion is required from HU to electron density which is what dose calculations require. Conversion is accomplished with a phantom calibration, and annual checks are required.

  • Metal artifacts: These can obscure tissue structures and result in inaccurate dose com­putations. Examples include dental artifacts in the head and neck or a hip prosthesis in the pelvis (see, e.g., AAPM TG-63, Dosimetric considerations for patients with hip prostheses undergoing pelvic irradiation ). The density of metal artifacts may need to be overridden in the treatment planning system. Algorithms for metal artifact reduction are now available on some CT scanners. Problems with CT artifacts can also be avoided by using clinical setups which are appropriate for some treatments or by using megavoltage CT (e.g., TomoTherapy).

  • Large-bore CT scanners: Because it is often necessary to accommodate patients in the treatment position (e.g., on an incline board), large diameter bore scanners are useful. Available models include the Brilliance CT Big Bore (Philips Inc., Eindhoven, Netherlands) with an 85 cm diameter bore, the Somatom (Siemens Inc., Ehrlangen, Germany) with an 80 cm diameter bore, and the Airo intraoperative CT with a 107 cm bore (Brainlab Inc., Feldkirchen, Germany). In addition to the bore size, the size of the field of view (FOV) is also important. If the scanner has a large bore but a standard FOV, the patient and devices may fit through the bore but the skin surface may not be imaged, rendering the scan useless. Some scanners can reconstruct larger FOVs but these must be checked carefully for HU and geometrical accuracy.

Magnetic resonance imaging (MRI) and positron emission tomography (PET)/CT are being increasingly employed in Radiation Oncology treatment planning. If the goal is to use these studies to define target volumes, then the patient position in the MRI or PET/CT scanner should match the treatment position as closely as possible (e.g., flat table top, same devices, and same pose). MRI-compatible immobilization devices are now available from vendors. One challenge is that MRI or PET/CT scans are often acquired for staging before the radiotherapy simulation has even been performed. The use of MRI in simulation is likely to be an important topic as MRI-enabled treatment devices come into more widespread use.

Immobilization, Setup, and Marking

The technology for patient immobilization is discussed in Chapter 6 . Overall, the goal of patient setup is to reduce motion, make daily setup more reproducible, improve comfort, and optimize the avoidance of normal tissue. Reducing motion is especially important given the fact that patients are typically on the treatment table for 15 minutes or more, and even longer for Stereotactic Radiosurgery (SRS)/Stereotactic Body Radiation Therapy (SBRT) treatments. The same immobilization accessories should be used for simulation and treatment. Some options for immobilization include specialized head rests, incline boards and arm holders for breast patients, and prone “belly boards” for some abdominal treatments (see Chapter 6 for a full description of immobilization devices). One widely used device is the thermoplastic shell, typically a perforated plastic that softens when heated. Thermoplastics are available in different thicknesses, with thicker materials offering more rigidity at the cost of decreased skin sparing. Thicker thermoplastics can double or even triple the skin dose, as described in AAPM TG-176, Dosimetric Effects Caused by Couch Tops and Immobilization Devices. For mildly claustrophobic patients, cutting away the mask above the eyes and, if necessary, above the mouth may increase patient comfort sufficiently to avoid the need for sedatives. Several manufacturers are now offering pre-cut masks for this purpose.

To promote safety, most modern immobilization devices can be “indexed”—that is, affixed to the treatment table at clearly defined locations. Since the patient's position is fixed relative to the treatment table, the X, Y, and Z coordinates of the table can be checked for consistency from day to day. Further description and figures are found in Chapter 6 .

Marking Patients

A three-point setup is commonly used to align patients. With this technique, three lasers' crosshairs are visualized on the skin: two lateral points (one on each side) and one anterior point (assuming the patient is supine). Conceptually the three-point setup determines the location of the isocenter and controls for the patient roll (pitch and yaw are somewhat less well constrained). Extra skin marks are sometimes used to aid in positioning. For example, a set of “leveling marks” placed inferiorly can be used to set up a breast cancer patient because the mobility of the breast tissue makes it less suitable for alignment. Prior to CT scanning, a plastic ball bearing (BB) is typically placed on each of the three-point setup marks so that the location can be visualized on the CT scan. Alternative markers are available such as the CT-SPOT plastic crosshair (Beekley Medical Inc., Bristol, CT). The patient is typically given a small permanent ink tattoo at each of the final setup points. In some cases, however, the final isocenter may not be determined at the time of simulation, and so a temporary mark is left on the patient. The necessary shifts are then determined in the treatment planning system, and then final tattoos are made on the patient (e.g., on first imaging day). To verify the correct position of the patient and/or treatment field, radiographs (“port films”) are often used on the treatment machine. These images are compared to digitally reconstructed radiographs (DRRs) generated from the simulation CT.

Other markers are sometimes used at the time of simulations for various purposes. For patients receiving postoperative treatment, for example, the surgical scar may be marked for later reference in designing the treatment plan (scars being a site of recurrence in some cases). A radio-opaque plastic “wire” is typically used for this purpose. Metal wire is avoided because it produces artifacts on the CT scan. In postoperative cases it is also common to mark a surgical drain site with wire or a BB. Wires are sometimes used to mark the extent of palpable breast tissue. As a final example, markers are often used to indicate laterality and help ensure that the correct treatment site is identified. An example of this is stereotactic radiosurgery for trigeminal neuralgia. A BB is placed on the patient's head on the involved side, which helps to ensure that the correct fifth cranial nerve is targeted.

There are several pitfalls related to setup and marking: (1) If the isocenter point is moved in the planning system relative to the point set at simulation, this could result in an incorrect patient setup unless the change is clearly communicated to the entire treatment team. (2) Care must be taken to identify the correct BB for the setup point on the CT images. The potential for confusion may exist if extra BBs or other fiducials are present (e.g., a BB on a drain site or plastic wire on a surgical scar). (3) If a patient receives repeat radiation to a similar area care must be taken not to confuse the old tattoo with the new intended location. Good communication is essential. (4) Moles or freckles can sometimes be confused with tattoos. Photos and documentation can help avoid this.

Bolus

Tissue-equivalent bolus may be added to photon or electron beams to increase a superficial dose. Although bolus can be added digitally in the treatment planning system, it is sometimes advantageous to place the actual bolus material at the time of simulation to accurately determine its position, shape, and dosimetric effect. An excellent review of bolus material options and effects on dosimetry can be found in Vyas et al. Tissue-equivalent bolus can take a variety of forms, including:

  • Gel sheets such as Superflab or Elastogel: These are easy to use and thickness is well controlled, but they may not conform well to curved surfaces.

  • Meltable materials such as thermoplastic pellets, bees wax, or paraffin wax (also available in pellet form): Useful for highly curved surfaces and particularly for electron treatments. For electron treatments bolus material can be formed into shapes to fill air cavities such as the nasal cavity.

  • Superstuff: Like wax, this material can be molded to shape.

  • Wet gauze or towels: These are useful for curved surfaces, but the gauze must be consistently saturated with water to achieve its intended dosimetric effect.

  • Brass metal mesh, a beam spoiler, or thick Vaseline: These are all useful for breast cancer patients.

Whatever bolus material is used, care must be taken to ensure that the material lies flat against the skin and avoid air gaps. Large air gaps produce a lateral spread of electrons, which can lead to a reduced dose on the distal edge of the cavity and even a secondary build-up region. The effect is larger for small field sizes, large air gaps, and higher energies. The effect of air gaps has been studied, but most reports have focused on the accuracy of planning algorithms to predict the dose beyond such air gaps. However, studies in head and neck treatment suggest that the effect of air cavities is on the order of 2% for large field sizes. For air gaps smaller than approximately 1 cm the impact does not appear to be clinically significant.

Shields

Radiation shields are often constructed or tested at the time of simulation. Examples include:

  • Eye shields used to protect the lens and cornea during electron treatments of the eyelid: Commercially available devices specifically designed for high energy electrons should be used. An example product is a curved tungsten shield 2 to 3 mm thick covered with acrylic to prevent backscattered electrons. A 2 to 3 mm shield results in an approximately 4% transmission.

  • Lead shields placed on the skin for electron treatments: The main purpose is to create a sharper penumbra at the field edge. The shield should be made thick enough to attenuate the beam. A typical rule of thumb is: thickness of lead (mm) = Energy (in MV)/2 + 1. A 20% thicker shield should be used if cerrobend is used.

  • Testicular shield or “clamshell”: These are typically constructed from 1.25 cm ( ) thick lead with a slit for access, designed to prevent scatter radiation from reaching the testicles. Clamshells may be useful in a variety of applications because permanent aspermia results at doses >2 Gy. Data from SWOG-8711 supports the use of such shields in treatment of testicular cancer. Useful radiation dose data can be found in Mazonakis et al.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here