The boron-containing melanin precursor analogue p-boronophenylalanine (BPA) has previously been shown to selectively deliver boron to pigmented murine melanomas when administered in a single intragastric dose. If boron neutron capture therapy is to become a clinically useful method of radiation therapy for human malignant melanoma, the boron carrier must be capable of delivering useful amounts of boron to remote tumor sites (metastases) and to poorly pigmented melanomas. We have now determined the ability of BPA to accumulate in several nonpigmented melanoma models including human melanoma xenografts in nude mice. The absolute amount of boron in the nonpigmented melanomas was about 50% of that observed in the pigmented counterparts but was still selectively concentrated in the tumor relative to normal tissues in amounts sufficient for effective neutron capture therapy. Single intragastric doses of BPA resulted in selective localization of boron in the amelanotic Greene melanoma carried in the anterior chamber of the rabbit eye and in a pigmented murine melanoma growing in the lungs. The ratio of the boron concentration in these tumors to the boron concentration in the immediately adjacent normal tissue was in the range of 3:1 to 4:1. These distribution studies support the proposal that boron neutron capture therapy may be useful as a regional therapy for malignant melanoma.

The conditions for the possible initiation of clinical trials with neutron capture therapy at a number of locations in the U.S. is reviewed. There are several new technical developments or plans at the Brookhaven Medical Research Reactor (BMRR), the Power Burst Facility (PBF) at INEL, the Massachusetts Institute of Technology Reactor (MITR) and the Georgia Institute of Technology Research Reactor (GTRR). Emphasis is on the development of epithermal beams for the treatment of deepseated tumors with neutron fluxes in between 10(9) to 10(10) n/cm2s. Therapeutic dose gains, defined as the ratio of tumour dose to maximum normal tissue dose in the treatment volume are expected to be between 2 and 4, depending on the degree of suppression of fast neutron dose. Boron concentrations considered in this case in the tumour are around 35 micrograms 10B/g and tumour/normal tissue concentrations are around 10. The compound development throughout three generations is discussed. The compound proposed nowadays, Na2B12H11SH (or BSH), employed in the treatments in Japan, will likely be replaced in the future by analogous of biomolecules being enriched in the tumour by physiological pathways. Examples are p-boronophenylalanine or boronated porphyrius. The most promising solution envisaged would be the employment of tumour cell specific brononated monoclonal antibodies. Finally the mode of therapy is discussed which will likely be based on a fractioned scheme, to achieve optimized results.

The dose-limiting factor in cancer radiation therapy is the tolerance level of normal tissues within the radiation field. In boron neutron capture therapy (BNCT), thermal neutrons interact with boron via the 10B(n,a)/Li reaction (Taylor, 1935) to produce short-range (~ 5–9 µm), high-linear energy transfer radiations which have a large relative biological effectiveness (RBE) (Gabel, 1984; Fukuda, 1987). In theory, selective localization of 10B within the tumor should allow most of the dose to be restricted to the tumor (Locher, 1936). The failure of the initial clinical trials of BNCT, carried out between 1953 and 1961 at Brookhaven National Laboratory and the Massachusetts General Hospital (Farr, 1954; Goodwin, 1955; Asbury, 1972) was attributed to two major factors: (i) the use of boron-containing compounds which showed no selective accumulation in tumor and (ii) the rapid attenuation in tissue of the incident thermal neutron beam. The high neutron doses employed resulted in excessive surface tissue exposure; viable tumor was found at depth following the neutron irradiations. The substantial levels of boron in blood during irradiation contributed to the damage to normal brain vasculature (tumor/blood ratio <1).

Thiouracil and various derivatives are selectively incorporated into the melanin pigment of melanomas during biosynthesis by serving as false melanin precursors. Using the transplantable Harding-Passey melanoma carried in BALB/c mice, we have extended our previous studies with sulfur-35 (35S) thiouracil. The persistence of high levels of [35S]thiouracil in tumor for periods of up to 2 wk has been demonstrated; during this time the drug content in normal tissues returned to near background levels. The variety of iodine isotopes available makes iodothiouracil a particularly promising melanoma-localizing agent. Tumor uptake and biodistribution of [35S]thiouracil and iodothiouracil (both iodine-127 (127I) and iodine-125 (125I) labeled) have been compared and were found to be essentially the same. The selectivity of [125I]thiouracil for melanoma has been qualitatively demonstrated by autoradiography of whole-body sections and quantitated by analysis of tumor and selected tissues. Iodothiouracil was also shown to localize in remote secondary metastases using a metastatic variant of the Harding-Passey melanoma currently being developed in our laboratory. These studies confirm the melanoma localizing capabilities of an iodinated thiouracil, and therefore the potential of using iodinated thiouracil derivatives for diagnosis and therapy of melanotic melanomas.

We have previously reported that buthionine sulfoximine (BSO) enhances sodium borocaptate (BSH) uptake by down regulating glutathione (GSH) synthesis in cultured cells. This study investigated the influence of BSO on tissue BSH uptake in vivo and the efficacy of BSH–BSO-mediated boron neutron capture therapy (BNCT) on tumor growth using a Fisher-344 rat subcutaneous tumor model. With BSO supplementation, boron uptake in subcutaneous tumor, blood, skin, muscle, liver, and kidney was significantly enhanced and maintained for 12 h. Tumor growth was significantly delayed by using BSO. With further improvement in experimental conditions, radiation exposure time, together with radiation damage to normal tissues, could be reduced.

Human tissues have carbon-isotope ratios (13C/12C) that reflect dietary ratios. This observation has been used to determine the extent of metabolic turnover of DNA in cells of the adult human cerebellum (90 percent of which are neuronal). If adult human neuronal DNA were metabolically stable, its 13C/12C would reflect that in the maternal diet during fetal development as nearly all neurons are formed during maturation of the fetal brain and do not undergo cell division thereafter. The 13C/12C ratios in the food chains and body tissues of Europeans differ from corresponding American ratios by about 50 parts per million on the average. Therefore, turnover was studied by comparing 13C/12C ratios in cerebellar DNA of American-born Americans, European-born Americans, and European-born Europeans. The 13C/12C ratios in cerebellar DNA from European-born Americans were closer to 13C/12C ratios in cerebellar DNA from European-born Europeans than from American-born Americans, indicating that there was little or no turnover of neuronal DNA.