Ionizing radiation is the name given to high-energy particles or electromagnetic waves that are capable of stripping electrons away from their atoms, producing physical and biochemical changes when they interact with matter. Electromagnetic ionizing radiation is subclassified based on its source: gamma rays are emitted in discrete energy peaks by naturally occurring radioactive isotopes, while x-rays are broad-spectrum radiation produced by manufactured devices such as linear accelerators. Particle radiation beams used in radiation therapy include both types of electron, proton, and neutron beams.

When a beam of ionizing radiation interacts with matter, many electrons are ejected from their atoms. These high-speed electrons are then either reabsorbed, transferring their energy back into the material, or produce further ionization downstream along their pathways.

High-energy x-rays produced by linear accelerators are by far the most commonly used beams for thoracic radiation therapy in the United States. Linear accelerators produce high-energy x-rays by bombarding a heavy-metal target (usually tungsten) with fast electrons. Low-energy x-rays, such as those produced by diagnostic x-ray units, have poor tissue penetration characteristics and deposit excessive doses in superficial tissues and bone,
making them unsuitable for the treatment of deep-seated thoracic tumors.

Cobalt teletherapy units house the isotope cobalt-60 (⁶⁰Co), which emits gamma rays. ⁶⁰Co machines have been extensively used for thoracic radiation therapy in the past but have now been largely replaced by linear accelerators. ⁶⁰Co beams have relatively low energy and poor edge definition; thus they deposit more dose in normal lung than do linear accelerator beams, likely increasing the risk of radiation pneumonitis. Thus the use of ⁶⁰Co is no longer recommended for thoracic radiation therapy.

It is often said that patients undergoing radiation therapy are having their tumors “burned out.” However, “burning” is not what is really happening. From the definition of a Gy (1 J/1,000 g) and the fact that 1 calorie (cal) = 4.184 J, it follows that it takes 4,814 Gy to deposit 1 cal of energy into 1 g of tissue. Since 1 cal is the amount of energy required to raise 1 g of water by 1°C, the amount of radiation needed to raise the temperature of the patient’s tissue by just 1°C is about 60 times a typical total therapeutic dose of 70 Gy. Thus cell death due to radiation therapy cannot be to the result of heating.

The critical target in radiation-induced cell death is in the nucleus and is probably DNA. Ionizing radiation damages DNA in at least four different ways: (a) double-helix strand breaks, (b) single-strand breaks, (c) base damage, and (d) damage to crosslinks, in which both DNA–DNA and DNA–protein cross-links are involved. Chromosomal damage is not detectable in interphase and appears only as the cells divide.

These physical and biochemical phenomena take place in fractions of a second at the time of the radiation exposure. Various nuclear and cytoplasmic molecular changes then follow, leading to loss of reproductive ability, cell cycle delays, somatic transformation, and mutations, among other effects.

The process of radiation therapy begins with an evaluation and staging of the patient, both anatomically and pathologically. Comprehensive evaluation of the physiologic and functional status of the normal tissue organs—such as lungs, heart, and spinal cord—are an essential part of this evaluation. After the regimen to be used has been decided on (e.g., conventional fractionation, hyperfractionation, radiation plus chemotherapy), the technical process of administering radiation therapy begins with simulation. In the past, this planning process was done with a single radiograph, called a simulator film, which is a plane radio- graph with an overlaid grid for measurement purposes. On obtaining the simulation film, usually in an anteroposterior or lateral projection, a portal would be outlined by the radiation on- cologist that presumably contained the area of gross tumor as well as areas of subclinical disease and outlined the area to be radi- ated. After appropriate measurements of the dimensions of this area and the thickness of the patient, calculations would be carried out and the delivery of treatment would commence. This technique is currently labeled two-dimensional (2D) radiation therapy.

In the 1990s, the advent of computer technology led to development of three-dimensional conformal radiation therapy (3D-CRT), which represented a radical change in practice. Instead of doing the treatment planning on a plane film, one starts by obtaining a CT scan of the patient’s chest in the treatment position. An isocenter for the treatment is then selected and marked on the patient; usually it is at or near the center of the tumor volume. The patient is then excused and the process of treatment planning begins.

3D radiation treatment planning begins by identifying the gross tumor volume (GTV), clinical target volume (CTV), and planning target volume (PTV) as defined by the ICRU-50 recommendations discussed above. Delineation of these volumes takes a great deal of knowledge and skill and still remains significantly operator-dependent. For example, what window setting of the CT scan is most representative of the actual pathology? In general, the GTV will not be the same if it is determined on different CT window settings.

Delineation of the CTV can be even more problematic than that of the GTV. Recall that, by definition, the CTV represents the volume of tissue at risk for tumor but one in which no tumor can be imaged. For lung, this is the microscopic extension of the primary tumor into the surrounding apparently normal tissue or clinically normal lymph nodes suspected of bearing microscopic disease. Recently it has been the trend to make the CTV as small as possible, not treating the radiographically negative mediastinum. Advocates of this point of view stress that there is no evidence of a significant recurrence rate in the untreated mediastinum. They also point to the potential for esophageal morbidity in treating the entire mediastinum. The counterargument is that the source of tumor recurrence is always difficult to determine and that it is naive to think that just because a lymph node is not enlarged on CT means that it does not contain tumor, especially if the patient is known to have positive mediastinal nodes. In addition to the “small field/big field” argument, there is significant uncertainty as to whether to change the original CTV if tumor regression takes place after induction chemotherapy. At the University of North Carolina, we have taken the position that the CTV should not just be a mathematical expansion of the GTV but should include some of the proximal uninvolved mediastinum with the radiation fields limited by normal tissue tolerance (see below). We tend to regard tumor volumes that have regressed as a result of induction chemotherapy as still needing treatment, but to a lower radiation dose.

Particular care must be taken to deal with tumor motion due to breathing. Four general approaches have been proposed. The first and simplest is to observe the patient under fluoroscopy and draw an additional volume outside the CTV (the PTV) that is an envelope of the tumor movement. This approach will, of necessity, result in an increased volume of lung to be treated but often will still lead to acceptable treatment volumes. The second approach is to have the patient hold a deep breath during treatment (deep inspirational breath hold, or DIBH). (See, for example, Stock and colleagues.⁴²) DIBH not only arrests tumor motion but also has the advantage of thinning the lung around the tumor, so that less normal lung is treated. Unfortunately, many lung cancer patients have COPD and cannot hold their breath for long. A third approach is known as gating. See, for example, Berman et al.⁴ Gating can be either active or passive—that is, with patient cooperation or not. The idea is to leave the accelerator on only during times when the lung is at a predetermined expansion. The treatment is, in effect, “strobed.” A fourth approach, more theoretical than currently practical, is to track the patient’s breathing, making adjustments in the accelerator treatment head or table movement to keep the tumor at isocenter. (See, for example, Zhang.⁴⁹) Although this method requires the least cooperation on the part of the patient, it has proved technically difficult to implement and requires a continuous source of tumor position updating, which is not yet available on conventional linear accelerators.

A significant problem in defining GTV is distinguishing between the actual tumor and postobstructive atelectasis or pneumonia. Inclusion of other imaging modalities, such as positron emission tomography (PET), may further improve tumor delineation. PET scanning has shown significant promise in both initial staging of lung cancer, as noted by Caldwell⁸ and Haberkorn²⁰ and their associates, and target volume delineation for 3D-CRT, as reported by Erdi and colleagues.¹⁴ Nodal involvement has been detected by PET scan in 80% to 100% of the studies in a meta-analysis comparing the efficacy of PET versus CT in the detection of nodal metastases reported by Dwamena and coworkers.¹² In the Leuven Lung Cancer Group experience of the evaluation of 980 lymph node stations, the accuracy of PET was 85% (visually, 90%), as compared with 64% for the CT scan. The PET scan has been shown to be of significant help in distinguishing gross tumors from atelectasis or postobstructive phenomena. Fusion of PET images into treatment-planning CT scans is becoming an increasingly routine practice.²⁹