Radiation Biology

Most radiation-induced cell death is caused by damage to nuclear DNA and is referred to as mitotic cell death. The interaction of photons or charged particles with water produces highly reactive free radicals that interact with DNA, causing breaks that can interfere with cell division. Although cells are equipped with very effective mechanisms for repair of the damage caused by free radicals, accumulated injury can lead to irrecoverable DNA breaks that prevent successful mitosis. Oxygen in the environment enhances the lethal effects of radiation by fixing free radical damage. Damaged cells that have lost their ability to reproduce indefinitely may continue to be metabolically active or even undergo several divisions before losing their integrity. For this reason, radiation-induced damage to tumors may not be expressed morphologically for days or even weeks after the radiation exposure.

Another type of radiation-induced cell death is referred to as apoptosis or programmed cell death. Apoptosis can occur before or after mitosis.⁵ The plasma membrane and nuclear DNA may be important targets for apoptotic injury. Apoptosis appears to play an important role in the radiation response of some tumors and in certain normal tissues such as salivary glands and lymphocytes.

In vitro studies of the relationship between radiation dose and cell survival demonstrate that mammalian cells differ widely in their inherent radiosensitivity. These differences contribute to the wide range of doses required to cure tumors of different cell types. Even bulky lymphomas can typically be controlled with doses of 35 to 40 Gy, whereas 2- to 3-cm squamous carcinomas usually require doses of more than 60 Gy. Melanomas and most sarcomas require even higher doses and usually cannot be controlled with tolerable radiation doses if there is more than microscopic residual disease after surgery.

A number of factors influence cellular radiosensitivity and tumor responsiveness:

Normal Tissue Effects of Radiation

The extent, nature, and likelihood of radiation-related normal tissue effects depend on the tissue, the dose, and the volume irradiated as well as patient factors. Surgery, chemotherapy, and other treatments may worsen the normal tissue effects of radiation. An understanding of the complex relationships between dose, volume, and normal tissue effects is critical in radiation therapy treatment planning.

Tissues and cells that have a rapid turnover rate (e.g., bone marrow stem cells, skin, hair follicles, gastrointestinal epithelium) tend to exhibit side effects during or soon after a course of fractionated radiation therapy and are referred to as acutely responding tissues. Examples of acute radiation reactions include diarrhea caused by pelvic irradiation, oral mucositis caused by irradiation of the head and neck, and hair loss, which can occur in any irradiated area. The renewal rate of acutely responding tissues typically limits the rate at which radiation therapy can be safely delivered to 900 to 1000 cGy/wk. Most acute side effects resolve within weeks of the completion of a course of radiation therapy.

Tissues that are more slowly proliferating are referred to as late-responding tissues and tend to manifest side effects weeks or months after radiation therapy. These effects may reflect direct damage to parenchymal cells or damage to vascular stroma, and the dose-response relationship varies according to the tissue irradiated and other factors. Table 4-1 presents some of the conclusions of a 1991 task force¹⁰ charged with summarizing relevant data concerning the effect of ionizing radiation on normal tissues. A more detailed update was subsequently published in 2010.¹¹ The duration of a course of radiation therapy has little impact on the incidence of late complications, but the dose per fraction has a major impact. In general, radiation schedules that involve fractional doses of 2 Gy or less permit maximal recovery of sublethal damage to normal tissues. For this reason and because acute side effects usually limit the weekly dose of radiation to no more than approximately 10 Gy, radiation therapy is most commonly delivered with a schedule of 1.8 to 2 Gy per fraction, five times per week. Most tumors repair cellular damage less effectively than late-responding normal tissues; as a result, the differential effect on tumor versus normal tissues is increased when a dose of radiation is fractionated. This is referred to as the fractionation effect.

Under certain circumstances, alternate fractionation schedules may be used to reduce the overall duration of a course of radiation therapy. Hypofractionation, the use of daily fractional doses of more than 2 Gy per fraction, is routinely used for palliative radiation therapy to optimize convenience, cost, and the rapidity of symptom relief. Common schedules used for palliation include 30 Gy in 10 fractions, 20 Gy in 5 fractions, and in some cases, 8 to 10 Gy in a single dose of radiation. Recently, the development of highly conforming radiation technique has led investigators to explore the value of hypofractionated radiation therapy for curative radiation therapy in certain situations. This approach is most effective if adjacent critical structures receive a significantly lower dose and dose per fraction than the target. Accurate target delineation, precise patient positioning, and clear understanding of internal target motion are critical to successful treatment. Stereotactic body radiation therapy is a form of ultra-hypofractionated radiation therapy in which very large daily doses of radiation are delivered with great precision under image guidance. In contrast, hyperfractionation is the delivery of small doses of radiation two or more times daily (usually with a minimum interfraction interval of 5-6 hr). This approach is most useful when repopulation is considered to be an important factor in tumor curability but proximate late-reacting normal tissues prohibit the use of hypofractionation.

Normal tissues can be categorized as “serial” or “parallel” according to the influence of their structure on radiation tolerance. Serial structures, such as spinal cord, small bowel, and ureter, may fail when even a small portion of the organ is irradiated to a high dose. In contrast, parallel structures, such as liver, kidney, and lung, can sustain very high doses to partial volumes but are less tolerant of moderate whole organ doses.

Therapeutic Gain

The goal of radiation therapy is to sterilize tumor with the fewest possible side effects. The difference between the rate of tumor control and the rate of normal tissue complications is referred to as the therapeutic gain or therapeutic ratio (Figure 4-1). The probabilities of tumor control and late normal tissue effects can generally be described by two sigmoid dose-response relationships. Below a threshold dose, the probability of tumor control is very low; it then rises steeply to a dose above which little additional benefit can be achieved by further increases in dose. The shape and slope of the curve are related to the type and size of the tumor and other factors including the use of concurrent systemic treatments. The likelihood of complication-free tumor control is determined by the separation between the tumor control and the late effects dose-response curves. The most successful treatment strategies are those that maximize the separation between these curves. Strategies that combine surgery or chemotherapy with radiation in a way that shifts the tumor control dose-response curve to the left without a commensurate shift in the complication curve probability increase the likelihood of a good result. Conversely, multidisciplinary treatments that increase the risk of complications without significant improvement in the probability of tumor control should be avoided.

Figure 4-1 The difference between the rate of tumor control and the rate of normal tissue complications is referred to as the therapeutic gain or therapeutic ratio.

Surgery and Radiation Therapy

In many cases, the close proximity of critical structures prohibits delivery of a dose of radiation sufficient to eradicate gross disease. In other cases, surgical resection leaves microscopic disease that could lead to future recurrence. In cases such as these, judicious combinations of surgical resection and radiation therapy may significantly improve local control and survival and preserve organ function. Postoperative radiation therapy is often used to prevent local recurrence after gross total resection.^12,13 In some cases, preoperative radiation therapy is used to “downstage” tumor, improve local control, or enable the surgeon to use organ-sparing operations. This approach has been particularly effective in the treatment of rectal cancers.¹⁴ However, unnecessary multimodality treatment can increase complications, and poorly chosen surgical procedures can, in some cases, compromise the ability to deliver curative radiation therapy.

The information gained from surgery can also help guide planning of radiation therapy. Operative findings frequently provide critical information about local and regional disease extent that can guide the radiation oncologist in target volume definition. However, optimal combined-modality treatment requires careful communication between surgeon and radiation oncologist, preferably before any treatment has been initiated.

Chemotherapy and Radiation Therapy

During the past several decades, randomized trials have demonstrated the benefit of combining radiation therapy and chemotherapy in the curative treatment of many cancers. A number of cytotoxic agents have been demonstrated to potentiate the toxic effects of radiation when given concurrently with a course of radiation therapy. Drugs that have proved to be particularly effective radiation sensitizers include cisplatin, 5-fluorouracil, and mitomycin-C. Concurrent chemoradiation schedules are most effective if the dose-limiting toxic effects of the drugs differ from those of radiation and if the sensitizing effect on the tumor is greater than that on normal tissues.

Concurrent or sequential combinations of drugs and radiation may also improve cure through spatial cooperation. For example, chemotherapy may be used to sterilize minimal microscopic disease in distant sites while radiation is used to treat areas of gross or high-risk microscopic local and regional disease. The use of neoadjuvant chemotherapy before radiation therapy has been explored in a number of settings. Unfortunately, when neoadjuvant chemotherapy followed by radiation therapy has been tested in clinical trials, impressive chemotherapy responses have rarely translated into significant improvements in survival. Large meta-analyses of outcome in patients with head and neck or cervical cancers suggest much smaller benefits with this approach than with concurrent chemoradiation.^2,15 However, neoadjuvant chemotherapy has been used effectively in patients with breast cancer and continues to be explored in other settings.¹⁶

Adjuvant chemotherapy is also used after local treatment to control metastatic disease in a number of disease sites.

Radiation Techniques

External Beam Radiation Therapy

Photons

Most modern external beam radiation treatments are delivered using linear accelerators that generate high-energy (6-20 MV) photon beams by bombarding a target (usually tungsten) with accelerated electrons. Photons in this energy range interact with tissues primarily through the mechanism referred to as Compton scatter; absorption is independent of atomic number but is related to the density of the absorbing material. The number of scattered electrons and ionizations increases as the photons penetrate the surface, resulting in a relative sparing of superficial tissues referred to as skin sparing.

The photon source is located in the head of a gantry that rotates around the treatment table (Figure 4-2). Collimators in the treatment head govern the radiation field size and rotation; a secondary, electronically controlled, multileaved collimator or blocks inserted between the internal collimator and the patient are used to shape the field to the irregular contours of a treatment target volume. The energy of the beam, the distance from the source to the patient, the depth of the target, and the density of intervening tissue also determine the doses delivered to the targets and normal tissues. The depth of penetration and the amount of skin sparing are correlated with the energy of the photons.

Figure 4-2 The photon source is located in the head of a gantry that rotates around the treatment table.

Electrons and Protons

Several types of particle beams have also been used in radiation therapy. By far the most common are electron beams. Most modern linear accelerators are equipped to produce electron beams of several energies in addition to photon beams. The absorbed dose from an electron beam is relatively homogeneous to a certain depth and then falls rapidly to nearly zero; the depth of penetration is related to the energy of the electrons and typically ranges from approximately 2 to 6 cm for electron energies between 6 MV and 18 MV, respectively. Electrons are used primarily to treat targets on or just below the skin surface.

Protons are positively charged particles that deposit most of their energy at a tissue depth determined by the energy of the protons. The rapid deposition of energy is referred to as a Bragg peak. For most clinical applications, the proton energy is modulated to spread out the peak, creating a dose distribution characterized by a relatively low entrance dose, homogeneous dose within the target, and very little exit dose. Although proton beams have been studied in a small number of facilities for several decades, there has been a dramatic increase in interest in proton therapy and in the number of facilities providing this therapy as the cost of proton accelerators has declined somewhat during the past 5 or 6 years. The tumors that have so far been of greatest interest are prostate, skull base, ocular, and pediatric tumors.¹⁷ However, there are as yet no randomized trials comparing proton therapy with more conventional treatments. Proton therapy continues to be a very expensive, highly complex technology that requires highly skilled physics and technical support. The best applications for proton beam therapy are still to be determined.¹⁸

Other Particles

Several other types of particle beams, including neutrons, carbon ions, and pi-mesons, have been explored for their clinical potential but currently are not in common use.¹⁹

Treatment Planning

Before the early to mid 2000s, nearly all external beam radiation therapy was “forward-planned.” In other words, the radiation oncologist drew shaped fields on orthogonal plain x-rays, using bony landmarks as a guide (Figure 4-3). These x-ray films were produced using a “simulator” that mimicked the specifications of a treatment accelerator but had a diagnostic-energy beam in the rotating head. Using the radiation oncologist’s fields and a template of the patient’s external contour, the radiation dosimetrist calculated the length of time the machine needed to be on to deliver the appropriate dose. Techniques and field designs were tailored on the basis of findings on diagnostic imaging studies but were often relatively standard, based on years of feedback from studies of patterns of disease recurrence. Whenever possible, multiple-field techniques were used to minimize the volume of uninvolved tissue treated to a high dose (Figure 4-4). However, treatment fields that were designed without direct visualization of soft tissue structures were necessarily relatively simple and collateral treatment of uninvolved structures was often substantial.

Figure 4-3 Radiation treatment fields may be drawn on x-ray films of the treatment area or on digitally reconstructed radiographs generated from a planning CT scan (A). Port films taken during treatment using the treatment beam are evaluated periodically to confirm accurate placement of the field (B).

Figure 4-4 Top panels: Multiple-field techniques can be used to minimize the volume of uninvolved tissue treated to a high dose. Bottom panels: Intensity modulated radiation therapy can generate plans that conform even more closely to the target volume, sparing adjacent critical structures.

Since the mid 1990s, computed tomography (CT)–based treatment planning has gradually become standard. CT-based planning makes it easier for the radiation oncologist to design treatment fields that conform more closely to the soft tissue structures that form most clinical target volumes.

Initially, with CT-based treatment planning, treatments were still forward-planned, although the radiation dose distributions were increasingly calculated in three dimensions with corrections for the tissue heterogeneities revealed in planning CTs.

More recently, modern technology has made it possible to take treatment planning one step further with the increasing use of “inverse planning” to design radiation treatment plans. With this approach, the radiation oncologist does not define treatment fields per se but rather designates target volumes and structures that are carefully contoured on a planning CT scan. Complex computer algorithms then design treatment plans that deliver the desired dose to the target volumes while minimizing the collateral dose to critical structures. The resulting plans are usually far more complex than could be conceived with a forward-planned approach, typically consisting of six to nine modulated treatment fields (e.g., intensity-modulated radiation therapy) or modulated rotational treatments (e.g., tomotherapy). Fields are modulated through the dynamic use of multileaf collimators.

These modern approaches can produce highly conforming treatment plans that often permit greater dose to targets or greater protection of normal structures than is possible with simpler techniques. However, the opportunity for error is also greater with these very complex treatments. The demands upon the radiation oncologist to understand normal anatomy and to correctly transmit the findings of diagnostic studies to contoured target volumes are much greater than with forward-planned techniques. Incorrectly defined target volumes lead to underdosage of the target, risking unnecessary tumor recurrences. Patient positioning and an understanding of internal organ motion also become more important when these very highly conforming treatments are used. In addition, because treatments are entirely dependent on accurate computer control of the treatment machine, meticulous quality assurance methods are required to ensure that treatments are being delivered as prescribed.

Role of Imaging in Radiation Therapy

The effectiveness of radiation therapy is heavily dependent on the quality, accuracy, and interpretation of the images that form the basis of most treatment plans and that document the accuracy of radiation delivery. Nearly all radiation treatments are planned directly using CT scans that are obtained in the treatment position. To improve the accuracy of transferred information, planning CT scans may be digitally fused with magnetic resonance imaging, positron-emission tomography/CT, or other tomographic imaging studies obtained in the course of the patient’s diagnostic evaluation. Once a plan has been finalized, the actual treatment delivery is also guided by periodic imaging. At a minimum, the accuracy of treatment is documented with weekly portal images taken with the treatment beam. However, daily image guidance with kilovoltage imaging, ultrasonography, or even CT is increasingly being used to improve the accuracy of daily setup for intensity-modulated radiation therapy, proton therapy, and other tightly conforming treatments.

As a result, close, effective communication between radiation oncologist and diagnostic imager has never been more critical than it is today. Even small misunderstandings about the location or significance of radiographic abnormalities can result in serious errors in treatment design. Diagnostic imagers greatly assist their radiation oncology colleagues by providing specific information about the location, size, and relevance of abnormal findings. Communication can be improved by specifying the series and slice number corresponding to the best views of each finding. Additional anatomic information about the laterality, vertebral level, and proximity to easily identifiable structures often helps the radiation oncologist to accurately transfer diagnostic imaging information to treatment planning studies. Radiation oncologists also often depend on their imaging colleagues’ suggestions for ways of obtaining the most accurate depiction of disease. However, in very difficult cases, no amount of written communication can replace direct discussion between the imager and the radiation oncologist through tumor boards, face-to-face reviews, or telephone discussions while both the radiation oncologist and the diagnostic imager review the patient’s images. In this context, radiation oncologists have a responsibility to ask for help and to confirm that their understanding of images and reports is accurate.

Radiation oncologists can also provide their imaging colleagues with important information to assist in accurate posttreatment diagnoses. In particular, the differential diagnosis of posttreatment abnormalities in bowel, lung, bone, and other structures can often be narrowed if the imager understands whether these structures were included in previous radiation fields. Close communication can also help the imager, radiation oncologist, and other members of the multidisciplinary team determine whether recurrences are inside, outside, or marginal to radiation therapy treatment fields. This has important implications with respect to whether the patient can be treated with additional radiation therapy and also provides radiation oncologists with important information about the reasons for treatment failure, helping to improve treatment for future patients.