Radiosensitization by Alteration of Energy Absorption

The probability of ionization is essentially proportional to the number of electrons in the target molecule, regardless of chemical composition for the types of ionizing radiation that are normally used in radiation therapy (high energy photons or electrons). However, preferential energy absorption by particular elements can occur with certain energies or types of radiation. Two such cases are boron neutron capture and k-edge absorption of photons by atoms of high atomic weight.

2.2.1 Boron Neutron Capture Therapy (BNCT). Certain isotopes, such as capture low energy (thermal) neutrons very efficiently. This property is expressed as the thermal neutron cross section, in units of barns. The thermal neutron cross section for 10B is 3837 barns. As early as 1936 (354), it was suggested that preferential incorporation of 10B into tumors could be a useful strategy for selectively radiosensitizing tumor cells. 10B itself is not radioactive, but neutron capture by 10B is followed by radioactive decay of the resulting 11B nucleus. 11B splits into 4He nuclei (alpha particles) and 7Li ions, both of which are densely ionizing [i.e., high linear energy transfer (LET)], and consequently very cytotoxic. The ranges of these particles are such that their energy is deposited within one cell diameter of the neutron-capture event. These characteristics make 10B a very desirable isotope for neutron-capture therapy, even though other isotopes have higher thermal neutron capture cross sections.

There are two challenges to implementation of this strategy. One is to deliver thermal neutrons to the tumor, given that low energy neutrons do not travel very far in tissue. The other is to design drugs that will selectively deliver 10B to tumors. The first challenge is being met with advances in instrumentation, such that BNCT is now regarded as a more realistic possibility than in the past (355). Neutrons of a sufficiently high energy to penetrate tissue (epithermal neutrons) are used in such a way that they become thermal neutrons at the depth corresponding to the tumor.

It is estimated that 10 parts per million (ppm) 10B would be enough to increase cytotoxicity twofold over that seen with neutrons alone (355).BSH (Na^H^S!!; 44) was one

of the first compounds synthesized for this purpose (356). BSH and its disulfide, BSSB, are reported to accumulate in animal tumors, but the clinical biodistribution results from patients with brain tumors have been variable and unpredictable (357). In animal studies, very high tumor to normal brain tissue ratios were achieved, but there was no evidence of a therapeutic gain, suggesting that normal brain injury may be more related to the dose than to the vasculature (358).Clinical studies of brain tumor BNCT with sodium tetraborate and BSH did not result in improved survival and there was evidence of increased normal brain injury, consistent with preferential damage to the vasculature (359). Other types of BNCT agents are currently under development. Encouraging results were obtained with BPA (p-boronophenylalanine, 45) in the treat

ment of melanoma (360). 157Gd has been proposed as another isotope that could be useful in neutron-capture therapy (361). Although BNCT is a promising idea, the selective delivery of atoms of high neutron cross section (such as boron) to the target cells is still a limiting factor.

2.2.2 k-Edge Absorption and Photoactiva-tion of Elements of High Atomic Number. The probability of absorption of a photon is highest when the energy of the photon is close to the binding energy of an electron in the target molecule (362). This effect is particularly noticeable for k-shell electrons of elements cf high atomic number. A strategy for radiosen-sitization based on this effect is to incorporate atoms of high atomic number into DNA (363). The k-shell binding energy is characteristic for each element, and irradiation at energies just above this k-edge characteristic energy will result in selective absorption by a particular element. Iodine is particularly attractive for this purpose because: (1)the optimal energy of the activating photon is in a range that is reasonably achievable; (2) iodine is easily incorporated into compounds that can be delivered to tumors; and (3)photoactivation can occur, further enhancing the biological effect. Photoactivation is a process whereby toner shell electrons are ejected from an atom as a consequence of photon absorption, and cascading outer shell electrons fill the successively vacated orbitals, resulting in the emission of multiple low energy X-ray photons and electrons (364).

2.2.3 Photodynamic Therapy. Photodynamic therapy (PDT) consists of administration of a photosensitive compound and illumination of the tumor with visible light. Recent advances in light-delivery technology have provided methods for selective and thorough illumination of the tumor (365), although light delivery continues to be the principal limitation of this therapeutic approach.

PDT has been found to be effective in the treatment of several types of solid tumors in humans. Most of the clinical experience has been with hematoporphyrin derivative (HPD, 46) or porfimer sodium (Photofrin),which is a derivative of HPD (366).A problem with these

Photofrin Absorption

compounds is photosensitization of skin that persists for 6-8 weeks, (367). meta-tetraiMy-droxyphenyl)chlorin (m-THPC) (368-370) has a shorter plasma half-life and a higher tumor/normal tissue ratio, and is currently in Phase 1/11 trials. 8-Aminolaevulinic acid (ALA) has been found to be effective clinically by topical application (371,372) and in animal studies by systemic administration (373). ALA is converted in vivo to protoporphyrin IX, which is the active photosensitizer. Other tet-rapyrroles that are being considered for PDT include purpurins (47) (374) and phthalocya-

nines (375, 376). Cationic dyes have been investigated (377), given that there is evidence that a common feature of many types of tumor cells is the ability to concentrate moderately lipophilic cationic dyes, attributed to differences in mitochondrial membrane potential (378).

The tumoricidal mechanism of PDT has two components: direct tumor cell killing and damage to the vasculature leading to tumor necrosis. The number of tumor cells from excised murine tumors that produce colonies in vitro decreases markedly with time between treatment and excision (379,380), in support of the importance of damage to the vasculature. In contrast, tumor cell killing by ionizing radiation is evident when tumors are excised immediately after irradiation and plated for clonogenic assay (381). Hypoxia induced by PDT can be exploited by concurrent treatment with drugs that are metabolized under hypoxia to toxic species (380, 382).

2.3 Alteration of the Primary Radiolytic Products

Ionization of water is the most common consequence of irradiation of biological systems because they are composed mostly of water. Figure 4.1 shows the relative yields of products of water radiolysis (383). These reactive species differ considerably in their chemical properties. For example, H* is a reducing radical and HO* is highly oxidizing. A potential approach to radiosensitization is to convert the initial radiolytic products into more reactive or more selective species. This can be accomplished by including a substance that can react with the primary radiolytic products before they have a chance to react with other

Figure 4.1. Products of radiolysis of water. Ionizing radiation causes an electron to be ejected from the water molecule (ionization), forming H20 , which dissociates into H+ and HO*. The electron becomes solvated (eaq~). Excitations also occur, and the excited water molecule (H20*) can dissociate into HO' and H'. Some of the reactions of these primary radicals are shown. The numbers in parentheses are the yields of the various species per 100 eV (G-values) (433).

Figure 4.1. Products of radiolysis of water. Ionizing radiation causes an electron to be ejected from the water molecule (ionization), forming H20 , which dissociates into H+ and HO*. The electron becomes solvated (eaq~). Excitations also occur, and the excited water molecule (H20*) can dissociate into HO' and H'. Some of the reactions of these primary radicals are shown. The numbers in parentheses are the yields of the various species per 100 eV (G-values) (433).

molecules. For example, 0, reacts readily with the aqueous electron (eaq~) and h* to convert them to <V~ (Fig. 4.1). n20, on the other hand, can react with eaq~ to convert it to HO' (Fig. 4.1). Hyperbaric H, can be used to convert ho* radicals into h*. Isopropyl alcohol can be used to scavenge ho* and h*, leaving the aqueous electron.

Such methods have been used to demonstrate differences in reactivity of these radio-lytic products toward model biological targets, such as DNA (383). Studies with bacteria and mammalian cells have yielded variable results (384, 385), possibly because the modifiers can act by multiple mechanisms (385). In general, such studies have been consistent with lack of toxicity of the reducing radicals, suggesting that ho* is the only diffusible primary radical that contributes to toxicity. One explanation is that multiple attacks on DNA must occur within a very small volume to produce a cytotoxic lesion (386). Only radicals with the reactivity of ho* can contribute to these clustered lesions because other radicals diffuse away from the critical volume before reacting with DNA (387). For example, the conversion of eH„-

aq to ho* by n20 more than 10 9 s after irradiation should not affect cytotoxicity because the ho* would be formed outside the critical volume (387).

A variation on this theme is to use modifiers that react with products of water radiolysis, to produce a species that is more selective in reacting with cellular targets (388) (Fig. 4.1). For example, Hiller et al. (389)found that the effectiveness of ionizing radiation in inactivating bacteriophage T2 is five times greater if ho* is converted to cc1302* by inclusion of cc14 in the medium, whereas it is decreased by a factor of 20 by conversion of ho* to 02*~ ty inclusion of formate. Similarly, uric acid can enhance radiation damage to alcohol dehydrogenase (in dilute solution), even though it protects lactate dehydrogenase (390). The reaction of HO' scavengers with HO' results in the formation of a more selective radical that may be more effective in damaging a specific cellular target. Dimethylsulfoxide (DMSO) reacts with HO' to form h3c*, or in the presence of oxygen, ch300\ These are relatively non-reactive radicals, and DMSO is therefore considered to be an HO' scavenger. DMSO protects against radiation-induced DNA damage. However, DMSO does not protect against ho*-initiated membrane damage, whereas other ho* scavengers do (391).

0 0

Post a comment