Synthesis and evaluation of a series of 3,5-disubstituted benzisoxazole-4,7-diones. Potent radiosensitizers in vitro.

A series of 3,5-disubstituted-2,1-benzisoxazole-4,7-diones was synthesized and evaluated as radiosensitizers both in vitro and in vivo. These compounds were designed as non-nitro electron-affinic agents in an effort to alleviate some of the toxicities seen with the 2-nitroimidazole radiosensitizers evaluated in the clinic. Several compounds in this series were potent radiosensitizers in vitro, with sensitizer enhancement ratios of 2.0-2.3 at concentrations less than 0.5 mM. Compounds with potent in vitro activity were also evaluated in vivo. However, none of these compounds showed radiosensitizing activity in vivo. The reduction potentials of these compounds were determined by cyclic voltammetry and compared to other electron-affinic radiosensitizers. In general, the reduction potentials of this series of compounds was slightly more positive than the 2-nitroimidazoles, but they fell within the range postulated as acceptable to yield in vivo activity. The results suggest that factors other than reduction potential may be responsible for the lack of in vivo radiosensitizing activity observed for this class of radiosensitizers.

A new class of analogues of the bifunctional radiosensitizer alpha-(1-aziridinylmethyl)-2-nitro-1H-imidazole-1-ethanol (RSU 1069): the cycloalkylaziridines.

A series of compounds related to alpha-(1-aziridinylmethyl)-2-nitro-1H-imidazole-1-ethanol (RSU 1069, 1) were synthesized and evaluated as selective hypoxic cell cytotoxic agents and as radiosensitizers. The aziridine moiety was replaced with a number of other potential alkylating groups including cycloalkylaziridines and azetidines. The data indicated that modification of the aziridine of 1 resulted in a substantial decrease in the ability of the compounds to selectively kill hypoxic cells. However, these modifications did not affect the compounds' in vitro radiosensitizing activity since many of the derivatives were as potent as 1. All of the compounds that were evaluated in vivo were less toxic than 1, and several members of this series had significant activity. The best compound was trans-alpha-[[(4-bromotetrahydro-2H-pyran-3-yl) amino]methyl]-2-nitro-1H-imidazole-1-ethanol (18), which, due to its activity and log P value, is a candidate for additional in vivo studies.