NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Kufe DW, Pollock RE, Weichselbaum RR, et al., editors. Holland-Frei Cancer Medicine. 6th edition. Hamilton (ON): BC Decker; 2003.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of Holland-Frei Cancer Medicine

Holland-Frei Cancer Medicine. 6th edition.

Show details

Biologic Basis of Radiation Therapy

, MD, , PhD, , MD, PhD, and , MD.

Cellular Response to Radiation

Radiation randomly interacts with molecules within the cell. Although the critical target for cell killing is the deoxyribonucleic acid (DNA),72 damage to the cellular and nuclear membranes and other organelles may also be important. Radiation deposition results in DNA damage manifested by single- and double-strand breaks (DSBs) in the sugar-phosphate backbone of the DNA molecule. Cross-links between DNA strands and chromosomal proteins also occur. The mechanism of DNA damage differs among the various radiation types. For example, electromagnetic radiation is indirectly ionizing via short-lived, hydroxyl free radicals produced primarily by the ionization of cellular H2O.73 Protons and other heavy particles are directly ionizing and damage DNA directly.74 Consequently, different types of radiation have varying relative biologic effectiveness (RBE). Directly ionizing radiation (eg, neutrons) has a greater RBE than indirectly ionizing radiation (photons or electrons).75, 76

Radiation damage is primarily manifested by the loss of cellular reproductive capacity. Lethally irradiated cells are, thus, said to undergo a reproductive death. Consequently, most cell types do not show morphologic evidence of radiation damage until they attempt to divide. Alternatively, some cell types are killed via the induction of apoptosis.77 A cell that has sustained lethal damage following radiation exposure may undergo one or more divisions prior to metabolic death and loss from the tumor population.

The concept that cell death following radiation exposure may not be manifested until several cell divisions later has clinical relevance, such that tumors associated with very slowly proliferating cancers may persist for months and appear histologically viable. The histologic appearance may clear only after tumor cells have had the opportunity to attempt to divide. Prostate carcinoma, for example, may require up to 24 months after RT prior to obtaining a normal biopsy.78

Radiation Survival Analysis

Radiation survival can be studied both in vitro and in vivo. In vitro experiments usually involve irradiating exponentially growing cells to known doses of radiation. Cells are plated, and after 2 to 3 weeks, colonies are stained. The surviving fraction is then calculated by dividing the number of colonies by the plating efficiency of unirradiated cells. A survival curve is then generated by graphing the log of the surviving fraction versus the absorbed dose (Figure 39-13). Typically, survival curves are characterized by an initial shoulder, followed by exponential decrease in the fraction of surviving cells at higher doses.

Figure 39-13. Models for survival curve analysis.

Figure 39-13

Models for survival curve analysis. Experimental data are typically shown as the fraction of cells surviving a dose of radiation plotted on a logarithmic scale, whereas the dosage of radiation is plotted on a linear scale A. When using the multitarget (more...)

In vivo models examine both the inherent radiosensitivity of tumor cells and environmental influences, such as hypoxia and host immunity. A model commonly used to study radiation effects is the growth delay assay, which measures the time interval required for a tumor exposed to radiation to regrow to a specified volume (Figure 39-14).79 An assay that analyzes the dose required to control 50% of tumors is the TCD50 assay, which has been widely employed to study tumors in a variety of experimental systems.80 The radiobiologic use of transplantable solid tumor systems in experimental animals has been reviewed by Hall.75 Radiation survival parameters can be assayed for normal tissues in vivo as well as for tumors. For acutely responding tissues, in vivo survival is measured by studying clones of normal tissues regrowing in situ (eg, skin, jejunal crypt cells) or cells transplanted to another site (bone marrow stem cells). To study radiation effects in late-responding tissues, such as the nervous system, functional assays, such as paralysis and death, may be employed.

Figure 39-14. Data points represent volume changes observed in tumors in animal models after irradiation.

Figure 39-14

Data points represent volume changes observed in tumors in animal models after irradiation. After an initial decrease in the volume size, tumors grow back to the original volume over a time interval referred to as the growth delay. Curve 1 is the growth (more...)

Models of Radiation Survival Curve Analysis

Two empirically derived mathematical models have been used to analyze radiation survival data (see Figure 39-13). In the multi-target model, the reciprocal of the slope of the survival curve is defined as D0, the radiosensitivity of the cell population or tissue under investigation. D0 is the dose required to reduce the surviving fraction to 37% in the exponential portion of the survival curve. The width of the shoulder region is represented by the quantities n or Dq. Dq is the quasi-threshold dose, or the point at which killing becomes exponential.

The linear quadratic model (surviving fraction = eαD-βD2) fits radiation survival data to a continuously bending curve, where D is dose and α and β are constants. The linear component, a measure of the initial slope, termed alpha, represents single-hit killing kinetics and dominates the radiation response at low doses. The quadratic component of cell killing, termed beta, represents multiple-hit killing and causes the curve to bend at higher doses. The ratio of alpha to beta is the dose at which the linear and quadratic components of cell killing are equal. The more linear the response to killing of cells at low radiation dose, the higher is the value of alpha, and the greater is the radiosensitivity of the cells.81 Neither model has a firmly established biologic basis, and, therefore, both should be viewed as mathematical tools to describe the cellular radiosensitivity.

Molecular Events Following Cellular Exposure to Ionizing Radiation

The ultimate fate of the irradiated cell is not only a function of the radiation dose but also influenced by the cell’s natural defenses, including the ability to detect the DNA lesions and to repair them with high fidelity. Furthermore, recent advances have revealed DNA is not the only cellular target that influences the radiation response. IR also interacts directly with lipid and protein signaling pathways and modulates gene expression through a variety of mechanisms including the direct activation of transcription factors. IR-induced activation of these signaling pathways can affect critical processes such as cell cycle regulation, DNA repair, apoptosis and tissue repopulation. Figure 39-15 illustrates the current perspective on the interactions between IR-induced DNA damage and cell signaling.

Figure 39-15. The presence of double-strand breaks (DSBs) leads to activation of the ataxia-telangiectasia-mutated (ATM) protein kinase Targets of ATM phosphorylation include P53 and NBS1 and probably the phosphorylated histone variant H2AX.

Figure 39-15

The presence of double-strand breaks (DSBs) leads to activation of the ataxia-telangiectasia-mutated (ATM) protein kinase Targets of ATM phosphorylation include P53 and NBS1 and probably the phosphorylated histone variant H2AX. In this way, ATM influences (more...)

Molecular Sensors and Effectors of Radiation-Induced DNA Damage

As shown in Figure 39-15, components of the DNA-dependent protein kinase (DNA-PK) complex are involved in sensing DSBs as well as the repair of these breaks. DNA-PK consists of the 470 kDa catalytic serine/threonine kinase (DNA-PKCs) and the 70 and 80 kDa Ku heterodimer.82-84 Ku binds directly to sites of DSBs,85, 86 and DNA-PKCs is recruited to sites of Ku/DNA complexes where it is activated.84 The Ataxia Telangiectasia Mutated (ATM) protein is also involved in sensing DNA damage and is related to DNA-PKCs and other members of the phosphatidylinositol (PI) 3-kinase (PIK) family. ATM proteins are involved in cell cycle regulation, recombination, telomere length control and the multiple steps in the cellular response to DNA damage.87-89 AT cells, like DNA-PKCs/Ku-deficient cells, are hypersensitive to IR.90-92 AT cells that are deficient in ATM or express a nonfunctional mutant protein also exhibit defects in IR-induced growth arrest,93-95 radioresistant DNA synthesis and chromosome instability.96–97

Because ATM is one of most widely studied components in the cellular response to DNA damage, it has been placed in the center of this figure. However, numerous other factors, such as the AT and Rad3-related (ATR) protein kinase, are essential for cell-cycle modulation after the induction of other types of DNA damage.98 The red bars indicate three main cell-cycle checkpoints in late G1, late G2 and S; dashed lines show that the effect is of DSBs;85, 86 DNA-PKCs is recruited to sites of Ku/DNA complexes; and the question mark denotes links between G2 arrest and P53 mediated apoptosis currently under investigation.

The P53 tumor suppressor gene plays a pivotal role in the cellular response to IR. Through its C-terminal domain, the P53 protein can bind directly to radiation-damaged DNA and to single-stranded DNA ends, suggesting that P53 can act as a sensor of DNA damage. Exposing cells to DNA damaging agents or to restriction endonucleases that induce DNA breaks results in the stabilization and accumulation of P53 in the nucleus.93, 99 This increase in P53 is associated with transcription of P21 and BAX, leading to the induction of G1 arrest or apoptosis. ATM and CHK2, a mammalian homolog of Cds1/Rad53 that is activated by ATM, phosphorylates P53 and thereby prevents its degradation. DNA-PK interacts with the N-terminal region of P53 and contributes to its sequence-specific DNA binding, transactivation function, and stability.100-102 Thus, P53 is an effector as well as sensor of DNA damage. c-Abl, a non-receptor tyrosine kinase, associates with P53103, 104 to stabilize and induce its pro-apoptotic functions through a kinase-independent mechanism.104 c-Abl also interacts with the P53-related protein P73 in the response to IR and other genotoxic agents.105-107 c-Abl has been implicated in inhibiting the Mdm2 oncoprotein as well.108 Mdm2 binds the transcriptional activation domain of P53 and blocks its ability to regulate target genes and to exert antiproliferative effects. Mdm2 also promotes the degradation of P53 via the ubiquitination/proteasome pathway. The role of P53 in cell cycle regulation is discussed below.

The Role of Growth and Stress-Signaling Pathways in the Radiation Response

IR-induced activation of signaling pathways can have a variety of consequences including increased or decreased radiosensitivity, inducing proliferation, differentiation or apoptosis or affecting biologic behavior, eg, invasiveness or metastases. Several of the key pathways that have been explored are summarized.

Ras, Raf and MAPK Signaling

The Ras/Raf signaling pathway mediates growth signals from receptors such as the epidermal growth factor receptor (EGFR), which is a member of the ErbB family. Overexpression of EGFR has been associated with uncontrolled proliferation, anchorage-independent growth, autocrine growth regulation and increased radioresistance. IR activates the small G-protein Ras, which in turn activates c-Raf, a serine/threonine kinase that interacts with Ras.109 Activating mutations in the Ras oncogene has been associated with radioresistance in certain cell lines.110 Activation of c-Raf-1 has also been implicated in the development of radioresistance.111, 112 Ras and Raf thus represent potential targets for increasing radiosensitization. For example, farnesyltransferase inhibitors, which block cell membrane attachment and thereby activation of Ras, enhance IR-induced cell death.113 In addition, down-regulation of Raf expression with anti-sense oligonucleotides sensitizes tumor cells to the cytotoxic effects of radiation.114 IR-induced Ras activation leads to upregulation of mitogen-activated protein kinase (MAPK).115, 116

PKC signaling

Protein kinase C (PKC) was the first cytoplasmic serine/threonine kinase found to be activated during the cellular response to IR.117 IR-induced activation of PKC has been attributed to the generation of phospholipase A2-mediated oxidation products.118 c-Abl also phosphorylates and activates cytoplasmic PKCδ in irradiated cells.119 IR-induced activation of PKCδ is associated with translocation of PKCδ to the nucleus.119 PKC inhibitors block IR-induced activation of the early growth response 1 gene (EGR-1) and c-Jun.117 In addition, PKC inhibitors such as calphostin C, PKC 412, and chelyrithine chloride, exhibit greater than additive anti-tumor effects in animal models when used in combination with IR.120-122 Radioprotection of b-FGF-treated endothelial cells is also mediated by PKC activation.123 Selective modulation of PKC could thus represent a strategy to enhance tumor cell killing or normal tissue survival. Inhibitors of the EGFR block IR-induced activation of MAPK,124 indicating that EGFR, Ras, Raf, and MAPK all lie in the same IR response pathway. A recent Phase II trial of a blocking antibody to EGFR has shown substantial improvement in the responses of head and neck tumors to radiotherapy.125 Agents that target the intracellular tyrosine kinase region include small molecule tyrosine kinase inhibitors (TKIs), which act by interfering with ATP binding to the receptor, and various other compounds that act at substrate-binding regions or downstream components of the signaling pathway.124 This group of compounds offers several advantages in cancer chemotherapy, including the possibility of inhibiting specific deregulated pathways in cancer cells while having minimal effects on normal cell function. They also have favorable pharmacokinetic and pharmacodynamic properties and low toxicity. Some TKIs, such as the reversible synthetic anilinoquinazoline inhibitor ZD1839 (Iressa), are now undergoing phase II to III clinical trials. A number of preclinical studies confirm Iressa’s activity in a broad spectrum of tumor types, both as monotherapy and in combination with commonly used cytotoxic agents. A recent report suggests that the relative expression of ErbB isotypes may alter the potency of these agents. For example, SU11925, a small molecule inhibitor of the tyrosine kinase activity of both EGFR and human EGF-receptor 2 (HER-2), exhibits similar potency against EGFR and HER-2 in vitro; however, higher plasma concentrations of SU11925 are required to inhibit EGFR phosphorylation in vivo in tumors that also express high levels of HER-2 than in tumors that express EGFR alone.125 This observation suggests that it is more difficult to inhibit EGFR phosphorylation in cells that express high levels of HER-2.


Lyn, a member of the c-Src family of non-receptor protein tyrosine kinases (PTKs), is activated in IR-treated cells126, 127 and is required for induction of the stress-activated protein kinase (SAPK), a member of the mitogen-activated protein kinase (MAPK) family, which is involved in the apoptotic response to genotoxic stress. Lyn inhibits the cyclin-dependent kinase Cdc2 in irradiated cells which is necessary for progression through the premitotic checkpoint127 and inhibits DNA-PKCs activity.128 c-Abl activates MEKK-1, and thereby confers activation of the SEK1->SAPK pathway. In this context, SAPK induces the activation of transcription factors including c-Jun, Elk-1 and ATF-2.129 In turn, binding of activated c-Jun to the AP-1 site in the c-Jun gene promoter contributes to the activation of c-Jun transcription in a positive feedback loop. IR-induced activation of the c-Jun gene is blocked by the antioxidant N-acetyl-L-cysteine (NAC).130 IR also induces expression of the c-fos and jun-B genes, which encode proteins that associate with c-Jun.

Early response genes

IR induces expression of the EGR-1 gene.131 Transcription of it is conferred by IR-induced activation of serum response (CARG) elements in the EGR-1 promoter.131 As found for c-Jun, treatment with NAC blocks induction of the EGR-1 gene in the IR response.132 These findings showed that IR activates the c-Jun and EGR-1 genes by ROS-mediated signaling mechanisms. The finding that IR activates gene transcription provided the experimental basis for the design of gene therapy strategies in which the spatial and temporal control of gene expression are regulated by high energy x-rays. In this approach, a radio-inducible promoter, such as that from the EGR-1 gene, is inserted upstream to sequences encoding a therapeutic protein. The CARG elements in the EGR-1 promoter have been used to activate IR-induced transcription of tumor necrosis factor (TNF) or the HSV-1 thymidine kinase (TK).133-135 The TNF protein functions as a radiosensitizer and, in high local TNF concentrations, selectively increases the sensitivity of the tumor to IR treatment in the absence of systemic toxicity.136 Several clinical trials are underway in which the EGR-1/CARG promoter–TNF construct is delivered to tumors in an adenoviral vector and activated by local radiotherapy.

Classic Aspects of DNA Repair in Radiotherapy

Irradiated cells that are not lethally damaged may undergo repair of the damage to their DNA. Sublethal damage repair (SLDR) is operationally defined as the enhancement in survival when a dose of radiation is separated over a period of time. SLDR may be represented by the extrapolation number (n) of the radiation survival curve when a multi-target survival model is employed.137-141

Sublethal damage has been studied in vitro and in vivo. In general, SLDR experiments divide a single dose into two relatively equal doses spaced at variable time intervals. Elkind and colleagues investigated this phenomenon in great detail.137, 142 Figure 39-16 shows results representative of split-dose experiments. An enhancement in survival, following two doses separated in time, is observed in exponentially growing Chinese hamster cells at 2 hours. This enhancement in survival is due to the rapid repair of SLD and is followed by a subsequent decline in survival at 5 hours and then another increase in survival at 8 hours. This variability in survival is caused by synchronization of the exponentially growing cell populations by the first radiation dose and subsequent treatment with a second dose during the radioresistant S-phase (t = 2 hours) or radiosensitive G2-M phase of the cell cycle (t = 6 hours).138, 140, 141

Figure 39-16. The surviving fractions of Chinese hamster cells exposed to two doses of x-rays separated by various time intervals are shown.

Figure 39-16

The surviving fractions of Chinese hamster cells exposed to two doses of x-rays separated by various time intervals are shown. When the two doses are given together (time between doses = 0 hours), the surviving fraction is equal to that observed after (more...)

The concept of SLDR is important during a course of fractionated RT because the shoulder region of the survival curve is recapitulated due to SLDR.139 Fractionation magnifies the surviving fraction after each treatment to an exponent equal to the number of treatments. Therefore, small differences in survival after each dose may have a great impact on treatment outcome. Most human tumor cell lines studied in vitro have relatively small shoulders;143-148 however, a large capacity for SLDR has been reported for some human tumor cell lines.79, 144

The ability of tissues to undergo SLDR has been demonstrated using a variety of normal tissue clonogenic or functional assays.149-151 The capacity of different cell populations to repair SLD is reflected by the width of the shoulder (or initial slope) of their survival curve. An increase in the total dose required to give the same biologic damage when a single dose (D1) is split into two doses (total dose D2), with a time interval between doses to obtain a single biologic end point, is the capacity of a normal tissue to repair SLD. The difference in the two doses, D2 - D1, is the measure of SLDR by the tissue, provided that the two doses are larger than those that generate the shoulder region of the survival curve.81, 150, 152 (Figure 39-17), D2 - D1 = Dq. If the D0 is known, then n can be calculated from the equation:

Figure 39-17. Single-dose and two-dose survival curves for epithelial cells.

Figure 39-17

Single-dose and two-dose survival curves for epithelial cells. The D0 is 135 cGy. The ordinate is not the surviving fraction, as in survival curves for cells cultured in vitro, but is the number of surviving cells per square centimeter of skin (plating (more...)

Image ch39e2.jpg

Varying environmental conditions can influence cell survival after a dose of x-rays. Thus, damage that is potentially lethal under a given set of conditions may not be lethal if postirradiation conditions are altered.153, 154 The enhancement in survival seen following manipulation of postirradiation conditions is referred to as the repair of potentially lethal damage (PLD). PLD repair (PLDR) has been demonstrated in vitro155, 156 (Figure 39-18). PLDR has also been shown to occur in vivo.155, 157, 158 PLDR is reported to be more pronounced in large tumors, presumably because a large proportion of cells are in G1 or G0. PLDR has been described to occur principally in the G1 phase of the cell cycle. Efficient PLDR occurs in a variety of human tumor cell lines in vitro.159-162 Weichselbaum and colleagues,161, 163, 164 and Guichard and colleagues165 have suggested that PLDR contributes to radiotherapy failure under certain circumstances.

Figure 39-18. Maximum recovery potential (MRP) for radioresistant and radiosensitive cells.

Figure 39-18

Maximum recovery potential (MRP) for radioresistant and radiosensitive cells. Confluent cell lines (noncycling) were irradiated and immediately subcultured, which resulted in an initial surviving fraction (0), which is generally equal or slightly less (more...)

PLDR and/or SLDR may not be expressed under all conditions in vivo. For example, cells must be genetically competent to repair these types of damages, and the tumor environment may affect the proliferative status of tumor cells.154, 164, 166 Also, radiation (or chemotherapy) may induce tumor proliferation, which allows fixation of radiation damage before PLDR or SLDR is complete.154, 166 Therefore, PLDR is likely to be most important in tumor cells of intermediate or high radiosensitivity when cells are quiescent between fractions. The 24-hour PLDR-surviving fraction, following treatment of human tumor cells in plateau-phase culture with a similar dose, is referred to as the maximum recovery potential (MRP).80 Figure 39-18 shows that although two cell lines have different amounts of initial lethal damage induced by a constant radiation dose (a function of D0 and n), the surviving fraction after a 24-hour delay in subculture (a function of n, D0, PLDR) may be similar.

Molecular aspects of DNA repair in the cellular response to ionizing radiation

All eukaryotes have evolved several mechanisms to repair DSBs, which indicate the importance and difficulty of repairing this type of DNA injury. The two main pathways are homologous combination (HR) and non-homologous end-joining (NHEJ). These two repair modes differ in their requirement for a homologous template DNA and in the fidelity of DSB repair. Whereas HR ensures accurate DSB repair, NHEJ does not. The contributions of these two DSBrepair pathways are likely to differ depending on the stage of the cell cycle.98 However, the pathways are not mutually exclusive because repair events that involve both pathways can be detected. HR is most efficient in the S and G2 phases of the cell cycle because of the availability of sister chromatids as repair templates.167 In the absence of a sister chromatid, DSB repair in G1 phase could still efficiently occur through NHEJ.

HR can repair DSBs by using the undamaged chromatid as a template, which results in the accurate repair of the DSB. HR is mediated through the RAD52 family of proteins.168 These proteins include RAD50, RAD51, RAD52, RAD54 and meiotic recombination 11 (MRE11). The initial cellular response to DSBs is mediated through ATM and Nijmegen breakage syndrome 1 (NBS1).169 Subsequent steps of DSB repair through HR include DNA-end recognition, possibly by RAD52, and nucleolytic processing of the broken ends of DNA into 3'-end single-stranded DNA. This single-stranded DNA is bound by the RAD51 protein which mediates crucial steps in the reaction—the search for a homologous duplex template DNA and the formation of joint molecules between the broken DNA ends and the repair template. Other proteins, including replication protein A (RPA), RAD52, RAD5430 and several RAD51-related proteins (eg, RAD51B, RAD51C, RAD51D, XRCC2, XRCC3, and DMC1),170 are thought to function as accessory proteins for RAD51 at various stages of HR. In mammalian cells, RAD51 forms nuclear foci in response to IR and decreased expression of RAD51 confers sensitivity to IR-induced DNA lesions.171 The later steps of the process include polymerization of nucleotides to restore degraded DNA strands and resolution of the recombination intermediates. Mice with targeted disruption of the RAD51 gene exhibit an embryonic lethal phenotype.172 Moreover, the failure to generate RAD51-/- stem cells has indicated that RAD51 is essential for cell viability. RAD51 interacts with P53173 and BRCA1.174 Although the breast cancer susceptibility proteins BRCA1 and BRCA2 are clearly implicated in HR , their roles are not well understood. Other studies have demonstrated that RAD51 is phosphorylated by c-Abl in IR-treated cells and that this response contributes to the down-regulation of RAD51 activity in ATP-dependent DNA strand exchange reactions.171 Treatment of cells with IR is also associated with inactivation of RAD51 by caspase-3-mediated proteolytic cleavage.175

In contrast to HR, NHEJ uses little or no homology to couple DNA ends. This pathway is not only used to repair DSBs generated by IR or other exogenous DNA-damaging agents, but is also required to process the DSB intermediates that are generated during V(D)J recombination.176 Several proteins that are involved in NHEJ have been identified. The Ku heterodimer, which consists of Ku70 and Ku80, has a high affinity for DNA ends, which indicates that it has an early role in the NHEJ process. Ku bound to a DNA end attracts the catalytic subunit of the DNA-dependent protein kinase (DNA-PKCs), a 470-kDa polypeptide with a protein kinase domain near its carboxyl terminus.177 DNA-PKCs can subsequently phosphorylate several cellular target proteins, including P53, the Ku polypeptides and itself. At present, it is unclear which phosphorylation targets of DNA-PKCs are relevant in vivo. A complex that consists of DNA ligase IV and XRCC4 (X-ray-repair-cross-complementing defective repair in Chinese hamster mutant 4) accomplishes the final ligation step. Cell lines or animals that lack either of the genes encoding these proteins do not carry out V(D)J recombination and are sensitive to ionizing radiation.178 In addition to the involvement of the RAD50–MRE11-containing complex in HR, genetic experiments with yeast indicate that this complex also has a role in NHEJ.179

Although cells deficient in components of the NHEJ and HR pathways are more radiosensitive than their wild-type counterparts, few advances have been made in the development of inhibitors of IR-induced DSB repair. Initial attempts at inhibiting DNA repair in irradiated tumors were associated with increased normal tissue toxicity. A more recent strategy using antisense oligonucleotides to the RAD51 gene has resulted in enhancement of the effects of radiation on glioma cells in vitro and in animal models.180 With improvements in the selective delivery of radiotherapy, the spatial targeting of IR-induced DSBs to tumors should theoretically establish inhibition of DNA repair as an attractive therapeutic approach.

Cell Cycle Checkpoints, Integration of Signalling, and DNA Repair

It is necessary for one phase of the cell cycle to be completed before initiating events associated with the following phase. Failure to achieve accurate completion of events may lead to genetic instability and/or cell death. Cells have evolved mechanisms to monitor genomic instability associated with DNA damage. The surveillance mechanisms and resulting inhibition of cell cycle progression are referred to as checkpoint controls. For example, a DNA damage checkpoint control communicates information between a DNA lesion and the regulatory components of the cell cycle. Radiation-induced single-strand breaks (SSBs) and DSBs are associated with the initiating signal that activates checkpoint controls. In general, the cell cycle is delayed at one or the checkpoint by inhibiting cyclin-dependent kinases (CDKs) until DNA repair is complete. Following exposure to ionizing radiation, cells arrest at the check points. Cell cycle regulation in irradiated and nonirradiated cells has recently been reviewed.181

DNA damage and its effect on cell cycle progression have been intensively studied in yeast and Xenopus oocyte, as well as in mammalian cells. Studies show that the kinase activity of Cdc2-cyclin B complex is required for the G2-to-M-phase (G2/M) transition in the normal cell cycle and that tyrosine phosphorylation of Cdc2 inhibits its kinase activity.182 Both inhibition of Cdc2 kinase activity and enhanced phosphorylation of Cdc2 have been observed following DNA damage. The phosphorylation state of Cdc2 is maintained by the kinases Wee1 and Myt1 and by the phosphatase Cdc25. 119, 121, 123, 183, Cdc2 is inactivated by phosphorylation of Thr-14 and Tyr-15 in the ATP-binding domain by the Wee1 kinase.184 Phosphorylation of Tyr-15 is also mediated in part by IR-induced activation of the Lyn tyrosine kinase.126, 127 Although checkpoint regulation of both sides exists, it is thought that regulation of Cdc25 activity is an important factor in the maintenance of G2 arrest after DNA damage. In mammalian cells, two kinases, Chk1 and Cds1 (also known as Chk2), have been identified118, 185, 186 and shown to phosphorylate Cdc25C and prevent it from dephosphorylating (Ser-216) and activating Cdc2. 85, 118, 185, 186, 187, 188, Phosphorylation of Cdc25 facilitates association with the 14-3-3 protein, resulting in its export from the nucleus; however, the mechanism of Chk1 and Cds1 activation following irradiation is not clear. Studies in Chk1-deficient cells have also shown that this kinase is required for initiating G2 arrest189, 190 and is a downstream effector of ATM. The response of Cds1 to DNA damage has been shown to be dependent on the activity of ATM. The ATM-dependent DNA damage checkpoint pathway regulates both the G1-S and G2-M transitions in the response of mammalian cells to IR treatment. The IR-induced arrest at the G1-S DNA damage checkpoint is mediated predominately by P53-dependent induction of P21 expression and thereby inhibition of the Cdk2 kinase.191

P53 promotes arrest at the G0/G1 checkpoint. Following radiation exposure, P53 protein levels rise due to post-translational modifications that increase the P53 half-life. This rise is transient and is correlated with the presence of damaged DNA. DNA strand breaks are the most potent inducer of P53 following IR exposure.181 Cell cycle arrest is induced by P53 when the protein acts as a transcriptional activator. One of the most important genes induced is P21waf-cip. The interaction of P21 with the cyclin E/CDK2 complex inhibits the progression of cells into the S-phase. P21 also binds directly to proliferating cell nuclear antigen (PCNA). In addition, P53 interacts with MDM2 and GADD45. As described above, MDM2 negatively regulates P53 and likely functions during the recovery phase in the late G1-phase. GADD45 inhibits the progression of cells into the S-phase. A second mechanism of regulating the G1 checkpoint by P53 is mediated by direct binding to proteins such RPA. Bristow and colleagues192 showed that rat fibroblast clones expressing increasing levels of transfected mutant P53 showed loss of the G1/S checkpoint and increasing levels of radioresistance. The increased survival was not associated with loss of apoptosis, but consistent with observations of improved double strand break repair.193 This is not consistent with a “more time for repair” hypothesis for the G1/S checkpoint, because checkpoint abrogation would then be expected to sensitize and not protect. Interestingly, the radiosensitivity of P53 null fibroblasts remained unchanged, indicating a gain of function of mutant P53, possibly due to improved damage sensing via the carboxy terminus of the protein which is known to bind to damaged DNA in a nonsequence-dependent manner.

Pharmacological agents which override the G2/M block often sensitize the cells to IR.194 The classical inhibitor of the IR-induced G2 arrest response is caffeine. The precise mechanism for the effects of caffeine on G2 arrest is unresolved, although ATM has been proposed as the target.195 Treatment of irradiated P53-deficient cells with caffeine is associated with activation of Cdc2 and hence mitotic cell death.196 These findings are in concert with reports which demonstrate that abrogation of the G2 checkpoint results in differential radiosensitization of G1 checkpoint defective cells.197 Caffeine and its analogs, however, have not proven to be effective radiosensitizers in the clinic, in part because of the toxicities associated with levels required for inhibition of IR-induced G2 arrest. The CHK1 inhibitors, UCN01 and SB-218078, block DNA damage-induced G2 arrest in human cells.198-200 These findings and the demonstration that UCN01 does not inhibit hCHK2198, 199 have supported lack of involvement of CHK2 in the G2 arrest response. Whereas UCN01 is undergoing clinical evaluation as a radio- and chemosensitizer, other inhibitors of CHK1/2, such as debromohymenialdisine,201 are under development. It is noteworthy that the concept of caffeine-induced override of G2-M block has been challenged by investigators who suggest this cell cycle perturbation and enhancement of the induced DNA damage associated with caffeine are attributed to inhibition of DNA repair and or DNA synthesis.202-204

Both the initiation and elongation stages of DNA replication (S-phase checkpoint) are inhibited by IR. In budding yeast, the genes that regulate the S-phase checkpoint are RAD17, RAD24, RAD53, and MEC3 and MEC1. The sequence of gene interaction and function in mammalian cells is under investigation. The ATM gene product plays a role in replicon initiation and chain elongation. Both MEC1 and ATM are members of the PI3K gene family.


Early in the twentieth century, it became apparent that RT was equally efficacious but better tolerated when administered in divided doses,205 a concept known as fractionation. Fractionation spares normal tissues by allowing time for repair and repopulation of normal cells. In addition, fractionation increases tumor cell kill due to reoxygenation and reassortment of tumor cells into sensitive phases of the cell cycle.206 Conventional fractionation schemes involve a daily fraction of 1.8 to 2 Gy 5 days a week. Total treatment times depend on the total dose prescribed and thus range from 3 to 7 weeks. Various mathematical models have been proposed to equate total dose, time, and fraction size and achieve an isoeffective dose. However, clinical utility of these models remains controversial.

Clinical interest in altered fractionation has been resurrected due to recently reported improved outcomes in patients with advanced head and neck cancer. Hyperfractionation is defined as the use of reduced size fractions given twice or more per day such that a greater total dose is delivered by increasing the number of treatments (50–60 vs. 30–35) in the same total treatment time. Thus, relatively rapidly dividing cell populations may have a higher proportion of cells in the most sensitive phases of the cell cycle at each treatment. Cells in late-responding normal tissues are slowly proliferating, and therefore, after a few fractions, many surviving cells will be concentrated in the more resistant phases. This strategy has been applied to a variety of tumors, including primary brain tumors207 and head and neck cancer.208

Accelerated fractionation decreases the overall treatment time to diminish clonogenic proliferation between successive doses. Treatment is given 2 to 4 times per day employing fraction sizes of 1.5 to 2 Gy per treatment with 4 to 6 hour interfraction intervals.81, 209 The total daily dose is thus 3 to 6 Gy and the total dosage is given in 3 to 4 weeks. Although treatment interruptions are often necessary due to acute toxicity,210 the total treatment time is substantially reduced. Hypofractionation is the use of larger than standard daily fractions. This approach is typically used when palliation is the goal. Treatment may involve 4 to 10 Gy fractions delivered weekly or several times a week as opposed to daily.

Modifiers of the Radiation Response

The extent of DNA damage following radiation exposure is dependent on several factors. One of the most important is cellular oxygen.211 Hypoxic cells are considerably less radiosensitive than aerated cells. In fact, anoxic cells require 2 to 3 times the radiation dose to produce an equivalent amount of cell kill as do oxygenated cells. The difference in radiosensitivity between hypoxic and aerated cells is known as the oxygen enhancement ratio (OER). Oxygen is believed to prolong the lifetime of the short-lived free radicals produced by the interaction of x-rays and cellular H2O. Indirectly ionizing radiation is consequently less efficacious in tumors with significant areas of hypoxia and necrosis. In contrast, damage following exposure to directly ionizing radiation is independent of cellular oxygen levels.212

Thomlinson and Gray observed that tumors “outgrow” their vasculature resulting in necrotic areas and suggested that tumor cells adjacent to the anoxic region may be clonogenic but hypoxic.213, 214 As shown in Figure 39-19, the oxygen tension within the tumor falls with the distance from the capillary producing a hypoxic region. In some experimental tumor systems, reoxygenation of hypoxic tumor cells occurs within 24 hours.215 A clinical observation which supports the importance of oxygen levels in patients undergoing RT is that severe anemia is associated with worsened local control in a variety of tumors. Several studies have demonstrated improved survival in patients with locally advanced cervical cancer treated with RT when hemoglobin levels were above 11 g as compared with patients with lower hemoglobin levels.216–218 Anemia has been associated with poorer survival rates following RT in other tumors, including endometrial, bladder, and pharyngeal cancers.219–221 The importance of cellular oxygen is also demonstrated by improvements in outcome in patients receiving hyperbaric oxygen.222, 223

Figure 39-19. Oxygen diffusion through tissue from a capillary resulting in hypoxic cells.

Figure 39-19

Oxygen diffusion through tissue from a capillary resulting in hypoxic cells. Oxygen diffuses an average of 150 Ci from the capillary. Cells beyond this region are anoxic and nonviable. Cells at the periphery of this radius are hypoxic but viable.

Numerous chemicals have been shown to modify the effects of ionizing radiation. An important class of compounds are the hypoxic cell sensitizers, including metronidazole, misonidazole, and etanidazole.224 These agents mimic oxygen and have been shown in vitro to increase cell kill of hypoxic cells.225 Clinical experience, however, with these agents have been mixed. Only two prospective trials have demonstrated a benefit to their use in conjunction with RT.226, 227 Toxicity is common, particularly peripheral neuropathy. More promising results with less toxicity have been noted with the newest hypoxic sensitizer nimorazole.228

A second class of radiation sensitizers are the thymidine analogues iododeoxyuridine (IUdR) and bromodeoxyuridine (BudR).229 Both are incorporated into DNA in the place of thymidine and render DNA more suspectible to radiation damage. Although several nonrandomized trials have been promising,230, 231 no prospective trial has demonstrated a significant benefit to their use. Of note, both are associated with considerable acute toxicity.

Multiple chemotherapeutic agents sensitize cells to radiation including 5-fluorouracil, actinomycin D, cisplatin, gemcitabine, fludarabine, paclitaxel, doxorubicin, hydroxyurea, mitomycin C, topotecan, and vinorelbine. The mechanism of radiosensitization varies among the different agents. Cisplatin inhibits both SLDR and PLDR.232 Inhibition of repair may also help explain the radiosensitizing properties of topotecan.233 Doxorubicin increases cellular oxygen levels by inhibiting mitochondrial and tumor cell respiration.234 Hydroxyurea is toxic to cells in the S-phase and inhibits entry of cells into the G1- from the S-phase.235 Mitomycin C is preferentially toxic to hypoxic cells.236 Paclitaxel synchronizes cells into the G2- and M-phases.237

Other drugs act as radioprotectors by protecting normal tissues from radiation damage while reportedly not affecting tumor radiosensitivity. The best known radioprotector is amifostine, a derivative of cysteamine which acts as a free radical scavenger. Following administration, amifostine quickly penetrates into normal tissues, but only slowly into tumors, and thereby results in a preferential protection of normal tissues. Promising results have been reported in head and neck cancer patients undergoing RT.238 Amifostine has also been used to reduce chemotherapy-related sequelae.239 Various endogenous biologic response modifiers also act as radio protectors, including interleukin-1 and granulocyte macrophage colony-stimulating factor (GM-CSF);240, 241 neither are classic radioprotectors, since they do not directly scavenge free radicals but rather improve bone marrow tolerance compartment.

Growth Factor and Cytokine Induction Following Radiation Exposure Growth and Regeneration Kinetics

The percentage of cycling tumor cells is called the growth fraction (GF). In human solid tumors, it is usually a small proportion of the total number of cells. If the GF remains constant with time, the growth rate of the tumor is proportional to the GF. If the GF decreases with time, the rate of the tumor growth slows. Solid tumors usually grow at a slower rate as they enlarge and so growth is approximated by the Gompertz formula.81, 242 In circumstances of equilibrium in normal tissues, each mitotic division results in the average of only one new cell. Usually, one daughter cell is lost by desquamation or metastasis. By definition, the cell loss factor (CLF) in a steady state is 1. Maximum growth occurs if the CLF is reduced to 0. The only requirement for growth is a reduction from 1.0 in the CLF, that is, an average of < 1 of two daughter cells of a division is lost. A CLF of < 1 is characteristic of regeneration of both normal tissue and malignant growth. Tumor growth is usually characterized by CLFs that are closer to 1 than 0.

An index for the potential regeneration of tumors and normal tissue populations is the proliferative activity of the cell population.81 One common measurement of tumor growth is the potential doubling time (PDT), which is defined as the time required to double the number of clonogenic cells if the CLF decreases to 0. In this concept, the doubling time is equal to the cell cycle time. Tumors with a high rate of both cell production and loss have the potential for early and rapid regeneration after irradiation or other cytotoxic treatment. Thus, even though a tumor may exhibit slow pretreatment growth, it may regenerate rapidly. Excessive protraction in the time of radiation fractionation or split-course regimens may give inferior local control results if accelerated proliferation occurs during the period when radiation is not given. Clonal proliferation during tumor regression after irradiation was demonstrated by Hermens and Barendsen, who showed an exponential increase in clonogen number in a rat rhabdomyosarcoma during a time of tumor shrinkage (Figure 39-20).243 Accelerated repopulation of irradiated tumors and tissues may be associated with the recruitment of quiescent cells into the cell cycle. This effect is associated with radiation-mediated induction of the immediate early genes c-Jun and EGR1.244

Figure 39-20. Growth curves of rat rhabdosarcoma tumors irradiated in vivo demonstrating accelerated repopulation.

Figure 39-20

Growth curves of rat rhabdosarcoma tumors irradiated in vivo demonstrating accelerated repopulation. A. Volume change in the tumor after a single dose of 2,000 cGy. Curve 1 is the growth of an unirradiated tumor. Curve 2 represents regression and regrowth (more...)

Although the induction of genes following DNA damage was well known in bacteria, it was only relative recently that induction of the transcription of the proinflammatory cytokine TNFα gene provided support for the theory that gene activation mediates the mammalian cell response to DNA damage.245 TNFα is transcriptionally induced following irradiation of human tumor cells in vitr,o246 and has paracrine and autocrine effects on tumor cell killing and might account for some of the systemic effects of localized RT. bFGF is induced in endothelial cells and mediates protection against apoptosis. The concept of the induction and release of growth factors following IR was initially reported in tumor cells and in normal endothelial cells.247 Subsequent studies showed that IR induces expression of genes encoding other growth factors, such as the platelet-derived growth factor, interleukin-1187 and bFGF.183 IR-induction of TGFβ, which is proposed to mediate fibrosis following IR and thereby to mediate some of the late effects of RT on normal tissues. Other cytokines have been also reported to be induced following radiation.248 Potential applications of these observations is the use of TNFα or other cytokines as potential radioenhancing agents in gene therapy combined with ionizing radiation (see below). It has subsequently been discovered that b-FGF protects some of the lung endothelium from radiation-mediated apoptosis and has potential as a radioprotector.249 Also, secretion of growth factors and cytokines may be an important step in radiation carcinogenesis in normal cells. Inhibition of molecular mediators of deleterious late radiation effects on normal tissues may increase the therapeutic ratio and presents the possibility of genetic manipulation in clinical radiotherapy.

The recent demonstration of EGFR activation by ionizing radiation makes this receptor and other members of ErbB receptor tyrosine kinase family important targets for radiosensitizing therapeutic interventions.250 The radiation-induced activation of EGFR results in cytoprotective signaling dominantly involving MAPK and PI3K. The inhibition of EGFR by tyrphostin TK inhibitors or through over-expression of a dominant-negative (DN) EGFR inhibit the radiation-induced cytoprotective response and results in significant radiosensitization of xenograft tumors in mice. The monoclonal antibody, IMC-C225, which blocks the binding of EGF, exhibits antitumor activity in EGFR+ tumors and enhances radiation toxicity in cultured human squamous cell carcinoma; it also increases taxane-, platinum-induced cytotoxicity in non-small cell lung carcinoma xenografts.251 In A431 head and neck squamous cell xenografts, IMC-C225 administered in conjunction with irradiation yielded a radiation enhancement factor of 3.62, attributable to both tumor necrosis and anti-angiogenesis.252 In phase I pharmacokinetic studies, IMC-C225 has a long half-life, lending itself to convenient weekly administration. The reported toxicity profile of IMC-C225 is limited to allergic and dermatologic reactions. An ongoing phase III international multicenter randomized study in locally advanced squamous cell carcinoma of the head and neck is evaluating therapeutic radiation therapy, either alone or in conjunction with IMC-C225. Preclinical studies have shown that ZD1839, an oral anilinoquinazoline, inhibits the epidermal growth factor receptor-associated TK, and has an additive to synergistic effect in combination with radiation or chemotherapy in colon, head and neck, and non-small cell lung cancers. Phase I clinical trials have shown modest dose-related toxicity, and antitumor activity has been reported in a variety of malignancies including lung cancer.253

Induction of Apoptosis

The response of eukaryotic cells to IR and other DNA-damaging agents includes the induction of apoptosis. For more than 40 years, radiobiologists have been aware of cells in irradiated specimens that display the features of apoptosis. Despite this knowledge, the role of apoptosis in terms of end points important in radiation therapy are not clear. The current model suggests that tumor cells with apoptotic propensity are more sensitive to radiation. If this hypothesis is confirmed, strategies can be envisioned that may restore apoptotic propensity to radioresistant tumor cells for therapeutic benefit.

Direct evidence for the activation of caspases in the induction of apoptosis comes from studies with peptide inhibitors,254 the cowpox virus protein CRMA,255 and the baculovirus protein P35.256 Overexpression of CRMA inhibits the induction of apoptosis in diverse settings, including activation of the Fas receptor and treatment with TNFα.257, 258 Similarly, the P35 protein functions as an inhibitor of caspases and blocks apoptosis in insect and mammalian cells.259 The recent finding that IR-induced apoptosis involves activation of a CRMA-insensitive pathway has supported the existence of apoptotic signals that are distinct from those activated by Fas and TNK.260 In this context, caspase-3 is inhibited by P35 but not CRMA in vitro, and IR-induced activation of caspase-3, like the induction of apoptosis, involves a P35-sensitive, CRMA-insensitive pathway.260 Whereas caspase-3 is activated by IR, as well as Fas ligand and TNF, these findings are explained by involvement of a CRMA-sensitive caspase in the Fas- and TNF-induced, but not the IR-induced, cascade. IR-induced activation of caspase-3 is associated with the proteolytic cleavage of poly-(ADP-ribose) polymerase (PARP), 261 DNA-PK,262 protein kinase Cδ,263, 264 and protein kinase C.265 The activation of caspase 3 and the subsequent substrate cleavage in irradiated cells are regulated by members of the Bc1-2/Bc1-x1 family.266 Bc1-2 and Bc1-xL block the release of cytochrome c from the mitochondria of cells exposed to IR and other agents.267-269 Whereas cytochrome c is not released from the mitochondria of cells induced to undergo apoptosis with Fas ligand,82 this event upstream to activation of caspase-3 117 and the insensitivity of IR-induced caspase-3 activity to CRMA260 support distinct apoptotic signals in Fas- and IR-treated cells.

IR-induced apoptosis is mediated at least in part by c-Abl dependent activation of P53 and its homolog, P73.270 IR also induces translocation of SAPK to mitochondria, interaction of SAPK and Bcl-xL, and thereby release of cytochrome c.271, 272 In turn, cytochrome c activates caspase-3 and the cleavage of diverse proteins that confer the apoptotic response.273, 274 Bcl-2 and Bcl-xL block IR-induced cytochrome c release and apoptosis.271, 275, 276 These findings have supported the transduction of DNA damage-induced signals to mitochondria as a determinant of cell fate in the IR response.

Dose Response and the Therapeutic Ratio

Various levels of radiation yield different tumor-control probabilities, depending on the size and anatomic extent of the lesion. The total number of surviving cells is proportional to the initial number and biologic characteristics of clonogenic cells and the total cell kill achieved with a specified dose of radiation. Dose-response relationships for local control of homogeneous tumor groups have been empirically determined. The higher the doses of radiation delivered, the more likely is tumor control (Table 39-1).

Table 39-1. Relationship of Tumor Diameter and Dose to Percent Local Control.

Table 39-1

Relationship of Tumor Diameter and Dose to Percent Local Control.

The dose of radiation that can be delivered to a tumor is limited by the probability of serious normal tissue complications. Therefore, the choice of a tumor dose is based on the relative probability of tumor control and normal tissue complications. The potential therapeutic gain can be estimated for an average group of patients on the basis of tumor size, histologic type, and the normal tissues that will be included in the treatment fields. Figure 39-21 shows a theoretic dose-response relationship for tumor control and normal tissue complications. The therapeutic ratio is defined as the percentage of tumor cures obtained at a given level of toxicity for normal tissues. Hill suggests that the therapeutic ratio is better defined in terms of a ratio of radiation doses required to produce a given percentage of tumor control and complications.242 Figure 39-21A depicts a favorable therapeutic ratio and Figure 39-21B depicts an unfavorable therapeutic ratio. The greater the displacement between the two curves (in the favorable situation), the more radiocurable is the tumor.

Figure 39-21. Dose control and complication curves in curable and noncurable tumors treated with radiotherapy.

Figure 39-21

Dose control and complication curves in curable and noncurable tumors treated with radiotherapy. The percentages of tumor control and normal tissue damage are sigmoidal. In a radiocurable tumor, such as Hodgkin’s disease, the dose required to (more...)

Gene Therapy and Radiation Therapy

Experimental RT has been combined with gene therapy in a variety of different strategies. These include the use of a various different viral and nonviral vectors to transduce tumor cells with enzymes that convert specific prodrugs to radiosensitizers, cytotoxins and/or immunomodulatory cytokines and other molecules that disrupt signaling pathways. Prodrug converting enzymes include herpes simplex virus (HSV), thymidine kinase and bacterial cytosine deaminase. In another approach, the tumor suppressor gene P53 is employed to modulate radiation-mediated apoptosis in cells that lack P53. Limitations of gene therapy include the lack of transduction of the entire tumor cell population as well as the lack of control of gene expression. In an attempt to compensate for lack of uniform tumor transduction and to achieve spatial and temporal control of gene therapy, Weichselbaum and co-workers delineated a strategy whereby a radiation inducible promotor is ligated to a therapeutic gene of interest.277 Radiation-inducible DNA sequences from the EGR-1 promoter were ligated to a DNA encoding TNFα and cloned into a nonreplication competent adenoviral vector. Tumors treated with ionizing radiation and EGR–TNFα regressed more rapidly and to a greater extent than tumors treated with EGR–TNFα or radiation alone. TNFα was induced seven-fold over background levels in the tumor, but not in the blood, thus achieving control over gene expression. This approach to gene therapy overcomes some of the limitations of limited viral transduction by employing a diffusible cytokine. This strategy has been recently evaluated for safety in phase I clinical trials and is currently being tested for efficacy in phase II trials in patients with pancreatic and esophageal cancer.

In another strategy which targets gene therapy by IR, it has been demonstrated that genetically engineered herpes viruses can be induced to proliferate in irradiated tumors. Herpes virus can cause lethal encephalitis and has been modified to be less neurovirulent by one of two strategies. Herpes genes that encode enzymes necessary for viral DNA synthesis or the gamma1 34.5 gene are deleted. The 34.5 gene prevents host protein syntheses shut-off, which has evolved to defeat HSV proliferation. Advani and co-workers employed a herpes virus (with gamma 34.5 deleted) combined with radiation and demonstrated a superior antitumor effect when compared with radiation or herpes virus alone in a flank model of human gliomas.278 These findings were extended and confirmed by Bradley and collegaues who employed IR and genetically engineered (GSE) herpes in an intracranial model of glioma and demonstrated a prolongation of survival in animals treated with the combination of herpes and ionizing radiation.279

Antiangiogenic Therapy Combined with Radiation Therapy

Ionizing radiation has been reported to mediate vascular collapse and thrombosis of very small tumor vessels. However, ionizing radiation has not been considered to target the tumor vessels. Tumor endothelial cells arise from host endothelium and are genetically stable, compared with tumor cells, and therefore are less likely to become resistant to DNA-damaging agents. Also one tumor vessel may supply up to 106 tumor cells with nutrients, thus amplifying the cytotoxic antitumor effects of IR. Teicher and colleagues conducted investigations combining radiation and synthetic antiangiogenesis compounds and demonstrated an increase in tumor cure and tumor growth delay.280 In spite of potential toxicities of the antiangiogenesis drugs, these investigators validated the tumor vasculature as a potential target for RT. In concert with these findings, the synthetic angiogenesis inhibitor, TNP-470, has been found to potentiate the effects of radiotherapy in the treatment of human glioblastoma multiforme xenografts in the nude mouse.281 Other studies have demonstrated that the endogenous angiogenesis inhibitor, angiostatin, enhances radiation-mediated tumor regressions without an increase in toxicity.282 The potentiation of radiotherapy by angiostatin was observed in several human tumor xenografts, including a human glioma. A brief exposure to angiostatin enhances tumor regression when delivered at the same time as radiation.283 These findings also showed that a prolonged course of angiostatin is not necessary for optimal therapeutic effects. The demonstration that endostatin similarly potentiates the anti-tumor effects of radiotherapy indicates that the interaction is broadly applicable to angiogenesis inhibitors.284

Another strategy has been to potentiate the effectiveness of radiotherapy with agents that function by directly blocking positive regulators of angiogenesis. For example, administration of antibodies against the vascular endothelial growth factor (VEGF) potentiates radiotherapy of murine and human tumor models, including that of a glioma xenograft.283 These findings have been corroborated in other studies using anti-VEGF antibodies or inhibitors (SU5416, SU6668, PTK787) of the VEGF signaling pathway in combination with radiotherapy.285-289 In addition, COX-2 inhibitors have been reported to enhance the anti-tumor efficacy of radiotherapy without increasing toxicity.290-292 These findings collectively demonstrate that inhibition of tumor angiogenesis potentiates the efficacy of radiation therapy.

Clinical Radiation Oncology

RT alone is used as curative therapy in a variety of tumor types. Treatment may consist of external beam alone, brachytherapy alone, or a combination of the two. Although combined modality therapy is more common today, definitive RT is still used in early-stage head and neck as well as gynecologic tumors. RT alone is associated with results comparable with that obtained with surgery for tumors of the oral cavity,293 oropharynx,294 supraglottic larynx,295 and glottis.296 Moreover, definitive RT is often associated with better long-term functional outcome than surgery. Definitive RT was recently compared in a large prospective randomized trial with radical surgery in women with early-stage operable cervical cancer. RT was associated with identical tumor control rates with less long-term sequelae.297 Another tumor type commonly treated with RT alone is early stage Hodgkin’s disease.298

RT alone is also the treatment of choice in elderly patients. A common belief is that the elderly are at higher risk for acute and chronic RT sequelae. However, numerous investigators have demonstrated that age per se is not associated with increased toxicity.299, 300 Instead, comorbidities present in the elderly may increase their risk.301

Adjuvant Therapy

A more common use of RT is in combination with surgery and/or chemotherapy. When combined with surgery, RT may be given prior to (preoperative), following (postoperative), or during (intraoperative) surgery. Although common in the past, preoperative RT is less used today except in large borderline resectable tumors, for example, rectal cancer302 and soft tissue sarcomas.303 In contrast, postoperative RT is used in many tumor sites including tumors of the central nervous system,304 head and neck,305 breast,306 lung,307 genitourinary,308 and gastrointestinal tract.309 In patients with resectable disease, postoperative RT is preferred because it allows treatment to be tailored to the pathology findings, and higher doses are possible. Moreover, there is reduced potential for interference in normal wound healing. Indications for postoperative RT include close/positive margins, residual disease, and lymph node involvement. Potential disadvantages of postoperative RT include delaying therapy until wound healing is complete and reduced vascularity of tissues following surgery. Intraoperative RT is the delivery of a single large fraction during surgery with either electrons or low energy photons.310 This is accomplished with either a dedicated treatment machine in the operating room or by transporting the patient to the RT department during surgery. An important benefit is that normal tissues, for example small bowel, can be displaced out of the treatment field. A disadvantage is that the total treatment is delivered in a single fraction which obviates the benefit of fractionation. Promising results have been reported in retroperitoneal soft tissue sarcoma.311 Brachytherapy has also been used at the time of surgery. It is imperative, however, to delay loading for several days to allow for adequate wound healing.312

When combined with chemotherapy, chemotherapy may be administered prior to (neoadjuvant), during (concomitant) or following RT (maintenance). Chemoradiotherapy approaches have been shown to improve local control and eradicate micrometastatic disease. Neoadjuvant chemotherapy has been used in a number of sites including early stage non-Hodgkin’s lymphoma313 and small cell lung cancer.314 A potential advantage is that bulky disease sites can be cytoreduced allowing for smaller treatment volumes. However, increasing evidence suggests that concomitant chemoradiotherapy is preferable in a variety of disease sites. Concomitant chemoradiotherapy is used in locally advanced cancer of the lung,315 head and neck,316 esophagus,317 bladder,318 and cervix.319 Possible interactions between chemotherapeutic drugs and radiation are summarized in Table 39-2.

Table 39-2. Interaction of Radiotherapy and Chemotherapy.

Table 39-2

Interaction of Radiotherapy and Chemotherapy.

In select sites, all three modalities are combined. A variety of schedules have been used. Examples include neoadjuvant chemotherapy, surgery, and postoperative RT (locally advanced breast cancer)320 and surgery followed by concomitant postoperative chemoradiotherapy (cancer of the pancreas and rectum).321, 322

Prophylactic Therapy

The most common example is the prophylactic treatment of regional, clinically uninvolved lymph nodes. Prophylactic cranial irradiation (PCI) is used in patients with limited-stage small cell lung cancer323 and children with high-risk acute leukemia.324 Other examples include breast irradiation in men with prostate cancer who receive diethylstilbesterol (DES), whole lung RT in patients with bulky mediastinal Hodgkin’s disease, and liver RT in advanced Hodgkin’s disease and pancreatic cancer.325-327

Palliative Therapy

RT is an important means of providing rapid and effective palliation due to local and/or metastatic disease. Osseous metastases secondary to breast, prostate, and other cancers are treated with localized fields and short-course regimens, for example, 30 Gy in 10 fractions. Pain relief is achieved in more than 70% of patients.328 The optimal fractionation schedule, however, remains unclear. Rapid large fractions, for example, 20 Gy over 5 days, are equivalent to more protracted regimens using smaller daily doses.329 Such approaches are indicated in patients with symptomatic long-bone sites that are not in close proximity to critical organs. More rapid schedules may be possible. The Radiation Therapy Oncology Group (RTOG) is currently comparing 8 Gy in 1 fraction to 30 Gy in 10 fractions in a randomized trial. Large-field (hemibody) irradiation has been used in patients with widespread bone metastases.330 Promising results have also been reported with intravenous 144Sr331 and samarium.332***

Whole brain RT is indicated in patients with cerebral metastases. Treatment is typically delivered over 10 days to total dose of 30 Gy. As with osseous metastases, controversy exists over the optimal treatment regimen in these patients. Borgelt and colleagues reviewed various regimens ranging from 20 Gy in 5 fractions to 40 Gy in 20 fractions on two randomized trials. No differences were seen in terms of frequency or duration of response. Overall, 50% of patients had significant improvement in neurologic symptoms. However, the less protracted regimens resulted in more rapid overall response rates.333 Protracted regimens are indicated, however, in patients with controlled primaries and solitary metastases.

Other indications for palliative RT include spinal cord compression,334 liver metastases,335 orbital metastases,336 and carcinomatous meningitis.337 Palliative RT is also used in symptomatic locally advanced lung338 and ovarian cancer.339 Brachytherapy can be used in palliative treatment as well, for example, bronchial,340 biliary,341 and esophageal342 obstructions.

Therapy for Benign Disease

RT is used in a wide variety of benign tumors and conditions, such as keloids,343 hemangiomas,344 des-moids,345 and pterygium.346 Other indications include renal347 and cardiac348 transplant rejection, macular degeneration,349 and heterotopic bone prophylaxis following arthroplasty.350 Promising results have recently been reported in the prevention of re-stenosis in patients undergoing coronary angioplasty.351

Radiation Sequelae

Acute Sequelae

Acute radiation sequelae, such as skin desquamation, mucositis, and diarrhea, occur during or immediately following treatment. Such sequelae are believed to be due to the interruption of repopulation of rapidly proliferating tissues.352 The type of reaction is dependent on the site irradiated. The one exception is fatigue, which occurs in almost all patients. Most acute sequelae are self-limited and respond to pharmacologic management, such as diphenoxylate hydrochloride with atropine sulfate (diarrhea) and viscous lidocaine (esophagitis). It is imperative to control symptoms and avoid prolonged treatment breaks, since treatment protraction has been correlated with worse tumor control in several disease sites.353, 354 Prophylactic medication may also be helpful. Promising results have been reported using sulcralfate in patients undergoing thoracic irradiation to decrease the severity of esophagitis.355

The severity of acute sequelae is dependent on a variety of factors. Two major factors are fraction size and treatment volume. Whenever large treatment volumes are used, it is thus imperative to reduce the daily fraction size to minimize acute sequelae. A commonly held belief is that older patients are at higher risk for acute sequelae. However, recent reports have disputed this belief.299, 300

Chronic Sequelae

Chronic reactions, such as fibrosis, fistulae, and necrosis, occur months to years following treatment and are due, in part, to damage to slowly proliferating tissues. Other factors including vascular damage may also play a part in their development. Chronic reactions, like acute reactions, are dependent upon the irradiated site. Chronic reactions, however, are often permanent. Sequelae vary widely in severity ranging from mild fibrosis to small bowel obstruction, fistulae, and second malignancies. Overall, the risk of a second malignancy following RT is low. The notable exception is osteosarcoma arising in irradiated bones in children treated for retinoblastoma, particularly the hereditary type.356

Select chronic radiation sequelae are responsive to medical management, for example, pneumonitis is managed with bronchodilators and, if necessary, a course of corticosteroids. Recent evidence supports the role of angiotensin II receptor antagonist in the treatment and prevention of radiation nephritis.357 Prophylactic medications may also decrease the risk of select late sequelae. Promising results have been reported with pilocarpine in head and neck cancer to decrease the incidence of xerostomia.358 Recently, zinc sulfate has been found to reduce the risk of significant taste alterations in patients with head and neck cancer.359 The most important means of reducing the risk of chronic sequelae, however, is prevention. Strict attention to optimal technique is imperative. Soft tissue sarcoma patients, for example, should never receive treatment to the entire circumference of the extremity, in order to reduce the risk of chronic edema. The risk of late sequelae is also reduced by avoiding the use of large daily fractions, since fraction size is a major determinant of late effects.360 As noted earlier, new approaches including 3D CRT, IMRT, and inverse treatment planning should further aid in reducing the risk of late sequelae.

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2003, BC Decker Inc.
Bookshelf ID: NBK12902