1.1 FOREWORD
An ideal therapy for cancer would be one whereby all tumor cells were selectively destroyed without damaging normal tissues. Most of the cancer cells should be destroyed, either by the treatment itself or with the help from the body's immune system, otherwise the danger exists that the tumor may reestablish itself. Although today's standard treatments - surgery, radiation therapy and chemotherapy - have successfully cured many kinds of cancers, there are still many treatment failures. The promise of a new experimental cancer therapy with some indication of its potential efficacy has led many scientists from around the world to work on an approach called boron neutron capture therapy (BNCT).[1,2]BNCT is a binary radiation therapy modality that brings together two components that when kept separate have only minor effects on cells. The first component is a stable isotope of boron (boron-10) that can be concentrated in tumor cells by attaching it to tumor seeking compounds. The second is a beam of low-energy neutrons. Boron-10 in or adjacent to the tumor cells disintegrates after capturing a neutron and the high energy heavy charged particles produced destroy only the cells in close proximity to it, primarily cancer cells, leaving adjacent normal cells largely unaffected.
1.2 RATIONALE OF NEUTRON CAPTURE THERAPY
Four years after the discovery of neutrons in 1932 by J. Chadwick of Cambridge University, a biophysicist, G.L. Locher of the Franklin Institute at Pennsylvania introduced the concept of Neutron Capture Therapy (NCT). [3] The physical principle of NCT is simple and elegant. It is a two component or binary system, based on the nuclear reaction that occurs when the stable isotope 10B is irradiated with low energy or thermal neutrons to yield highly energetic helium-4 (4He) nuclei (i.e.,alpha particles) and recoiling Lithium-7 (7Li) ions. Figure 1 illustrates the sequence of nuclear events. 
Figure 1: Schematic of Boron-10 Neutron Interaction
BNCT relies on initial targeting of tumor cells by an appropriate chemical compound tagged with 10B which preferentially concentrates in tumor cells. During irradiation of the tumor site by neutrons, the 10B absorbs a low energy neutron and ejects an energetic short-range alpha particle and lithium ion which deposit most of their energy within the cell containing the original 10B atom. Therefore, if a higher concentration of 10B exists in tumor cells relative to other normal tissues, a concomitantly higher dose will be delivered to tumor cells during neutron irradiation.[3]
There are a number of nuclides that have a high propensity for absorbing low energy or thermal neutrons. Of the various nuclides that have high neutron capture cross-sections, 10B is the most attractive for the following reasons: 1) it is non radioactive and readily available, comprising approximately 20% of naturally occurring boron; 2) the particles emitted by the capture reaction 10B(n, [[alpha] ])7 Li are largely high "Linear Energy Transfer", dE/dx, (LET); 3) their combined path lengths are approximately one cell diameter; i.e., about 12 microns, theoretically limiting the radiation effect to those tumor cells that have taken up a sufficient amount of 10B, and simultaneously sparing normal cells; and 4) the well understood chemistry of boron allows it to be readily incorporated into a multitude of different chemical structures. 1,2
Although the neutron capture cross-sections for the elements in normal tissue are several orders of magnitude lower than for 10 B, two of these,hydrogen and nitrogen, are present in such high concentrations that their neutron capture contributes significantly to the total absorbed dose. In order to reduce this "background" dose it is essential that the tumor attain high10B concentrations so that the neutron fluence delivered(neutrons/cm 2) can be held to a minimum, thereby minimizing the (n,p) reaction with nitrogen [14N(n,p)14C] and the neutron-gamma (n, gamma) reaction with hydrogen[1H(n,gamma) 2H] and maximizing the 10B(n,alpha)7Li reaction in the tumor cells.
Alpha particles and lithium ions, from the 10B(n, alpha) 7Li reaction,give rise to closely spaced ionizing events. They have a combined path length of approximately 12 microns., and have high LET. There is, therefore, little if any cellular repair from the induced radiation injury. Since the10B(n, alpha)7Li reaction will produce a significant radiobiological effect only when there is a sufficient fluence of thermal neutrons and a sufficient amount of 10B near to, on, or within the cell. The radiation effect or damage produced can be extremely localized,thereby sparing normal tissue components. Thus, selectivity is simultaneously one of the advantages and disadvantages of NCT, since it requires delivery of 10B to tumor cells in greater amounts than normal cells. In contrast to the ionizing radiation produced by radionuclides, little or no radiation is delivered to bystander cells, which have no 10 B, if the 10B is selectively localized on or within the tumor cells. If the 10B is not localized adverse effects may be produced in normal tissues. [4]
Alpha particles, with high LET, have other biological advantages. Unlike some forms of ionizing radiation, such as X-rays, alpha particles do not require oxygen to enhance their biological effectiveness. A rapidly expanding tumor frequently outgrows its blood supply, so that some regions receive less oxygen than normal tissues do. As a result of this oxygen depletion, the tumor can become more resistant to the effects of conventional photon or electron (i.e.,low LET) radiation therapy. Tumor sensitivity to alpha particles is retained however, even when the tumor has limited oxygen supply.
One more advantage of alpha particles and lithium ions is that they can kill dividing and non dividing tumor cell alike, this is important because tumors are known to have a large number of viable but inactive cells. Other forms of radiation treatment and chemotherapy tend to work best only on the cells that are dividing.[5]
A major advantage of a binary system is that each component can be manipulated independently of the other. With NCT one can adjust the interval between administration of the capture agent and neutron irradiation to an optimum time when there is the highest differential 10B concentrations between normal tissues and the tumor. Furthermore, the neutron beam itself can be collimated so that the field of irradiation is circumscribed and normal tissues with high 10B concentration can be excluded from the treatment volumes. Protection of normal tissues near and within the treatment volume is achieved by selective targeting of 10 B to the tumor.
1.3 CLINICAL CONSIDERATIONS
In 1951, Sweet[25] first suggested that NCT might be useful for the treatment of brain tumors, and in particular, the treatment of the most highly malignant and therapeutically persistent of all brain tumors, glioblastoma multiforme (GBM). Figure 2 shows a schematic for the principle of NCT for the treatment of brain tumors.
Figure 2: Schematic Depicting Concept of BNCT
An epithermal beam of neutrons is directed towards a patients' head, during their passage through tissue these neutrons rapidly lose energy by elastic scattering (a process called thermalization) until they end up as thermal neutrons. The thermal neutrons thus formed, are captured by the 10B atoms which become 11B atoms in the excited state for a very short time (~ 10-12 seconds). The 11 B atoms then fissions producing alpha particles, 7Li recoil nuclei and in 94% of the reactions, gamma rays. Tumor cells are killed selectively by the energetic alpha particles and 7Li fission products .
Sweet and colleagues[26] first demonstrated that certain boron compounds would concentrate in human brain tumor relative to normal brain tissue. Shortly thereafter, a clinical trial of NCT was initiated at Brookhaven National Laboratory [27,28] during1951 and 1952 and at the Massachusetts Institute of T echnology research reactor MITR-I,[29] during 1961 and 1962, using a thermal neutron beam and sodium tetraborate, borax (Na2B4O7*10H2O), as the capture agents.
Unfortunately, these trials failed to show any evidence of therapeutic efficacy. It became clear later, that there were two major reasons for their lack of success. 1) thermal neutrons are attenuated rapidly in tissue due to absorption and scattering, and their useful depth of penetration for NCT therapy is limited to 3-4 cm. This means that only superficial tumors would be destroyed by the 10B capture reaction. [30,31] 2) the boron compounds that were used were freely diffusible, low molecular weight substances that did not achieve selective localization in the tumor. Those which did had high blood values, and this explains why so much radiation was delivered to adjacent normal brain.4 Stimulated by more encouraging clinical studies done in Japan by Hatanaka etal .32 for the treatment of malignant gliomas and those of Mishima et al .13 for melanoma,there has been renewed national and international interest for NCT.
The history of NCT is inextricably linked to GBM. This is a cancer of the glial supportive tissues of the Central Nervous System (CNS). Glial cells provide the environment, in the form of chemical and physical support, which sustains the neurons. Ninety percent of the cells of the CNS are glial cells.They constitute 50% of the volume of the nervous system. Unlike neurons, glial cells are constantly undergoing the cell cycle of birth, differentiation and procreation (mytosis).[33] It is this difference that makes the likelihood of cancerous glial cell far greater than the likelihood of cancerous neurons in adults.
Macroscopically, in GBM, the evidence of anaplasia (i.e., a reversion of the cells or tissue to more primitive, embryonic or undifferentiated form often with increase of capacity for multiplication) is manifest where the smooth,homogeneous texture and color of normal tissue is replaced with a more friable granular gray tumor tissue with areas of necrosis and edema. [34] Figure 3 shows a Computerized Tomography (CT) scan of a patient with glioblastoma, in which the tumor and edema have been marked.
Figure 3: Contrast-enhanced CT scan of a patient with high-grade GBM
Microscopically, as the name "multiforme" suggests the salient feature is the variety of cell forms encountered. The anaplastic areas vary within a wide range, but collectively make up the familiar picture of the glioblastoma multiforme. The important features of the malignant process are increase cellularity, obvious polymorphism of the tumor cells associated with mitosis, alterations in the architectural arrangements of the cells and a variety of secondary changes. These features together with the confirmation of the various macroscopic features, often make the diagnosis clear at first glance even with a low-power optical microscope.
The variability from one field to another and between different tumors makes a single composite picture impossible to describe. Figure 4 shows the characteristic dense cellularity of GBM. Dark circular objects are nuclei of the rapidly growing tumor cells, the cytoplasm is the lighter gray uniform background.
Figure 4: Characteristic dense cellularity of GBM.
Figure 5 shows GBM cells predominantly spindle-shaped. Polygonal cells of different shapes and sizes are also characteristic of GBM. Note that the architecture of the tissue has been altered.
Figure 5: Spindle-shaped GBM cells showing altered architecture of tissue.
Standard radiotherapy for the treatment of high-grade brain tumors (GBM)following a biopsy or subtotal resection is to give external beam radiation with high-energy X-rays (4 - 6 MeV) to a dose of approximately 60 Gy in fractions of 1.8 or 2.0 Gy daily, five days a week. [35] There have been a number of series published [36,37] in which tumor doses up to 160Gy were used (mostly due to interstitial radiation). Unfortunately, despite these high doses, a clear therapeutic benefit has not yet proven.
Average results of conventional treatment show that the median survival for GBM ranges from eight to fourteen months, and untreated GBM results in a median survival of approximately three months. [38] The clinical trials of BNCT being conducted at MIT and elsewhere are still under way, and it is too early to reach a conclusion on whether or not BNCT will be effective for treating GBM's and other types of disease. Early results from these trials, however, indicate that BNCT appears to be as effective as conventional therapy for GBM and it is clearly a therapy which does not require as great an investment in time by the patient as conventional radiotherapy. Results from BNCT trials of melanoma are encouraging and complete or partial tumor control has been observed in several cases. Larger number of patients will have to be treated before a decision can be made about the efficacy of BNCT treatments.
At present, there are several groups in the U.S. and abroad working on this approach. The MIT/Harvard group makes use of a fission converter based epithermal neutron beam at the MITR-II Research Reactor that is filtered by aluminum, Teflon, cadmium, and Lead. This arrangement provides a broad spectrum epithermal beam with low incident gamma and fast neutron contamination while maintaining an incident neutron flux of ~5 x 10 9 neutron/ cm2-sec. This beam permits irradiations for clinical trials to be conducted in 1 - 4 fractions in 10 minutes or less
In describing the performance of NCT beams, three figures of merit are used.These were developed by the MIT/Harvard BNCT group [41,42] and used at the beam design workshop held at MIT. [43] First is the advantage depth (AD) which provides a measure of the maximum useful depth for therapeutic benefit. Advantage depth is defined as the depth in tissue at which the total therapeutic dose is equal to the maximum total background dose. The total therapeutic dose is the sum of the total background dose and the 10B(n,alpha)7Li dose.Figure 6 shows a graphical representation of AD. A maximum advantage depth (ADmax) occurs when the boron dose ratio between the tumor and the healthy tissue/blood is infinite. However, more realistically this ratio is 3:1 to 4:1. Moreover, it has been shown that there exists a geometrical dose absorption factor of about three to one for those boron reactions that are initiated in the microvasculature of the brain. [44,45] By using an effectivetumor-to-blood ratio of 10:1, one can represent a reasonable boron dose partition between tumor and normal tissue, for boron compounds which have a tumor-to-blood ratio of 10:3 and negligible concentration in normal tissue surrounding the capillaries. The currently accepted figure of merit is simply called the Advantage Depth (AD). The Advantage Depth is the depth at which the maximum dose to normal tissue equals the dose to tumor. The dose to normal tissue includes an estimate of dose from B-10 in normal tissue.
The second figure of merit in characterizing a NCT beam, is the concept ofadvantage ratio (AR). The AR gives a measure of a particular treatment beam's ability to minimize integral dose to normal brain when a tumoricidal dose is delivered to brain tumor. The one-dimensional AR is define as the integral dose tha t would be delivered to tumor tissue if it were uniformly distributed within the brain, divided by the integral dose that would be delivered to normal brain, along a particular one-dimensional axis through the brain. Normally, the axis of greatest interest corresponds to the central axis of the treatment beam.
The third figure of merit is the advantage depth dose rate (ADDR),which is the RBE dose rate to tumor defined at the advantage depth. From the previous definition of AD, the ADDR is the maximum RBE dose rate to normal tissue. The ADDR was developed primarily as a clinically meaningful neutron beam intensity criterion for epithermal neutron beam design studies.
These three figures of merit provide a method for comparing and evaluating the neutron beams for BNCT. Figure 6 gives the dose-depth distribution curves in a head phantom for the 12 cm field in the MITR-II's currently available epithermal neutron beam, referred to as the Fission Converter Beam (FCB). The AD for this beam is 8.9 cm, the AR is 5.0, and the ADDR is 126 RBE cGy/min using the capture compound BPA, tumor and normal tissue concentrations of 52.5 and 15 ug/g respectively and RBEs of 1.0, 3.2, 1.3 and 3.8, respectively, for photons, fast and thermal neutrons, boron in normal brain, and boron in tumor.
Figure 6: Typical Depth-Dose Curve for Fission Converter Beam at MITR-II using BPA