Neurotoxicity of misonidazole in rats following intravenous administration.

Misonidazole is a hypoxic cell radiosensitizer that induces a peripheral neuropathy in humans after exceeding a schedule-dependent cumulative threshold dose. Clinical studies of misonidazole have been conducted using oral administration, whereas most other radiosensitizers have been administered intravenously. Since route of exposure can potentially influence the toxicity of xenobiotics, the objective of this study was to assess the neurotoxicity of misonidazole in rats following intravenous dosing using a battery of routine clinical, neurofunctional, biochemical, and histopathologic screening methods. Male Sprague-Dawley rats were administered intravenous doses of misonidazole at 0 (vehicle control), 100, 200, 300, or 400 mg kg-1 once per day, 5 days per week, for 2 weeks. Animals were evaluated for neurofunctional and pathological changes following termination of treatment (Days 15-17) and at the end of a 4 week observation period (Days 43-45). During the dosing phase, hypoactivity, salivation, rhinorrhea, chromodacryorrhea, rough pelage and ataxia were observed at 400 mg kg-1, and body weight gain of the 300 and 400 mg kg-1 groups was significantly decreased relative to the vehicle controls by 24% and 49%, respectively. Corresponding reductions in food consumption were 8% and 23%, respectively. Although most 400 mg kg-1 animals appeared normal on Day 15 prior to the neurofunctional evaluations, rotorod testing precipitated a number of clinical signs including: ataxia, impaired righting reflex, excessive rearing, tremors, vocalization, circling, head jerking, excessive sniffing and hyperactivity. All of these animals recovered and appeared normal from Day 17 through study termination. There were no treatment-related effects on motor activity, acoustic startle response, rotorod performance, forelimb group strength, toe and tail pinch reflexes, tibial nerve beta-glucuronidase activity or tail nerve conduction velocity. Although hindlimb grip strength of the 400 mg kg-1 group was significantly decreased by 17% relative to the vehicle controls on Day 15, this finding appeared related to the reduced food consumption and body weight gain in these animals. No microscopic changes were detected in peripheral nerves. Necrosis and proliferation of fibrillary astrocytes (gliosis) were seen in the cerebellum and medulla of the 400 mg kg-1 animals on Day 16. Gliosis in these same brain regions was observed in the 300 and 400 mg kg-1 groups on Day 44. The results show that intravenous administration of misonidazole to rats causes dose-limiting central nervous system toxicity without effects on peripheral nervous tissue. The lack of peripheral neurotoxicity was most likely due to a combination of several interrelated factors including route of administration, duration and intensity of the dosing regimen, and total cumulative dose.

A comparison of the effects of methotrexate and misonidazole on the germinal cells of the subependymal plate of the rat.

The cytotoxic effects of the drugs methotrexate (MTX) and misonidazole have been assessed in the rat brain by quantifying changes in the constituent cell populations of a glial cell progenitor layer, the subependymal plate (SEP). Three distinct cell types can be identified in the SEP on the basis of their nuclear morphology: cells with small dark (SD), small light (SL) or large light (LL) nuclei. The cells with SD nuclei may present pluripotential glial cell precursors. A reduction in the total nuclear density of the SEP, after the local ventricular administration of MTX, could be accounted for largely by a loss of cells with SD nuclei; to approximately 45% of control values 2 days after MTX followed by a full recovery in numbers by day 5. A further decline in the number of cells with SD nuclei occurred at 12 weeks after MTX administration. The pattern of changes in the cellularity of the SEP, after misonidazole administration, were similar to those observed after MTX treatment, although the magnitude of the response was reduced. It was concluded that both drugs, but MTX in particular, could have a potential additive effect on glial progenitor cells when used in combination with other forms of cancer therapy including radiation.

Detection of hypoxia by measurement of DNA damage in individual cells from spheroids and murine tumours exposed to bioreductive drugs. II. RSU 1069.

The ability of the dual-function bioreductive drug, RSU 1069, to identify hypoxic cells in multicell spheroids and murine SCCVII squamous cell carcinomas was examined using the alkaline comet method. This method applies fluorescence microscopy and image analysis to measure the amount of migration of DNA from individual cells embedded in agarose and exposed to an electric field. Chinese hamster V79 spheroids, exposed for 1 h to RSU 1069, were disaggregated and individual cells were analysed for DNA damage. Following exposure to RSU 1069, aerobic cells exhibited DNA single-strand breaks while DNA interstrand cross-links were produced in hypoxic cells. Spheroids containing 40-50% radiobiologically hypoxic cells exhibited 20-30% cells with cross-links and the remainder showed only strand breaks. Similar patterns of damage were observed in SCCVII tumours growing in C3H mice exposed to 25-200 mg kg-1. Subsequent irradiation of cells in vitro greatly improved the distinction between aerobic and hypoxic cells from spheroids or SCCVII murine tumours exposed to RSU 1069, especially after treatment with low drug doses. The pattern of damage was relatively stable for at least 4 h after drug injection. Results indicate that detection of hypoxic cells in solid tumours may be practical using this agent or a prodrug, PD 144872, selected for phase I clinical testing as a hypoxic cell radiosensitiser and cytotoxin in human tumours.

Use of the comet assay to detect hypoxic cells in murine tumours and normal tissues exposed to bioreductive drugs.

The alkaline comet assay was applied to individual cells from mice exposed to two bioreductive drugs, tirapazamine and RSU 1069, with the goal of comparing DNA damage to tumours and normal tissues. More DNA single-strand breaks (SSBs) and a greater heterogeneity in DNA damage were observed in tumour cells than in spleen and marrow cells of mice exposed to 10-100 mg/kg tirapazamine, consistent with the presence of hypoxic cells and the greater bioreductive capacity of tumours. In mice injected with 25-200 mg/kg RSU 1069, aerobic cells exhibited large numbers of SSBs while toxic DNA interstrand crosslinks were produced only in hypoxic cells. Cells from bone marrow and spleen showed extensive numbers of SSBs, but minimal crosslinking compared to tumours where 10-20% of cells were heavily crosslinked. DNA damage produced by these two bioreductive drugs may be useful in estimating the range of individual cell oxygen contents within tumours and normal tissues.

Dose-dependent increase in the frequency of micronuclei and chromosomal aberrations by misonidazole in mouse bone marrow.

Cytogenetic effects produced in bone marrow of mice by various doses of misonidazole (MISO, 100 to 1000 mg/kg bodyweight) were studied by assaying the induction of micronuclei (MN) and chromosomal aberrations. Misonidazole increased the frequency of both micronuclei and chromosomal aberrations over normal at 24 h after treatment, however, the values were statistically significant from normal only at drug doses above 250 mg/kg. The increase was proportional to the drug dose with a best fit to linear quadratic model for MN induction. For chromosomal aberrations the data fitted equally well to linear as well as linear quadratic models.

Metabolic studies and neurotoxicity in tumors and brain of mice after hypoxic cell sensitizers.

PURPOSE: The effects of the radiosensitizers RK-28 and RP-170, both 2-nitroimidazole nucleoside analogues, and KU-2285, a fluorinated 2-nitroimidazole, as well as etanidazole (ETA) on glucose metabolism in mouse tumors and brain were studied to assess their degree of neurotoxicity. METHODS AND MATERIALS: Adult male C57Bl mice received differing doses of the above sensitizers IP. Blood, brain, and tumor samples were removed at various times and the levels of glycolytic metabolites determined. Glucose uptake and phosphorylation in brain was measured by the 2-deoxyglucose method of Sokoloff et al. (6). RESULTS: RP-170 showed neither signs of toxicity nor significant alterations in glucose metabolism in brain or tumor at doses up to 4 g/kg b.w. up to 4 h. By contrast, RK-28 was extremely neurotoxic at a dose of 1 g/kg b.w. with a high degree of lethality, resulting in a highly significant increase in the brain glucose level from 0.38 mumol/g to 2.20 mumol/g (p < 0.001) 2 h after administration, whereas that in the tumor was decreased. KU-2285 and ETA were significantly (p < 0.01) less toxic than RK-28 at this dose, as reflected in a lower increase in the brain glucose level (0.60 mumol/g), although KU-2285 approached that of RK-28 (1.43 mumol/g; p < 0.01) after 2 h following a dose of 2 g/kg b.w. However, in contrast to the other sensitizers, KU-2285 concomitantly also resulted in a highly significant continuous increase (p < 0.01) in tumor glucose levels. Labeled 3H-2-deoxyglucose studies showed that RP-170 neither markedly affected the uptake of total radioactivity into the brain nor its degree of phosphorylation whereas, KU-2285 (2 g/kg) and RK-28 (1 g/kg) decreased uptake by approximately 50% and phosphorylation approximately 3 and 4-fold, respectively. At doses of 1 g/kg, ETA and KU-2285 showed no significant changes in these parameters. This indicates a decreased level of neurotoxicity. CONCLUSION: Since the adult brain relies solely on glucose metabolism for its energy supply, interference to this pathway may be instrumental in the development of neurotoxicity, thus, underlining the need for such metabolic studies to assess the level of toxicity by radiosensitizers.

Assessing the bioreductive effectiveness of the nitroimidazole RSU1069 and its prodrug RB6145: with particular reference to in vivo methods of evaluation.

The nitroimidazole, RSU1069, has been shown to have a very high differential toxicity towards hypoxic cells compared to oxic cells both in in vitro and in vivo experimental conditions. However, in the clinic it was found to cause severe emesis and had to be withdrawn. After an extensive drug development programme an analogue of RSU1069, RB6145, which acts as a pro-drug for RSU1069, was found to be the most suitable candidate for further investigation. In in vivo studies with murine tumour models, when RB6145 was used in combination with X-rays it was shown to produce a similar level of toxicity towards hypoxic cells as that observed for RSU1069. Its activity was the same whether it was administered interperitoneally or orally and the same level of anti-tumour effect was observed if the drug was given before or after X-rays. RB6145 is better tolerated systemically in mice than RSU1069 and canine studies have shown that it is less emetic than the parent drug. Bioreductive drugs can also be used in combination with treatments that preferentially increase tumour hypoxia. Photodynamic therapy (PDT) causes extensive vascular damage in tumours. If either RSU1069 or RB6145 are administered during PDT, very large increases in the growth delay induced by PDT alone are seen for the RIF-1 murine tumour. RB6145 has been accepted for clinical toxicity trials with the prospect of using it in combination with X-rays. In the future it may also be of clinical use with treatments such as PDT.

Effects of local hyperthermia on the tissue levels and toxicity of three radiosensitizers in mice.

The toxicity of the radiosensitizers misonidazole (MISO), demethylmisonidazole (DEMISO) and pimonidazole (PIM) in mice can be affected differently when combined with local hyperthermia at 43 degrees C for 30 min. At a dose of 1 mg/g, only MISO plus heat resulted in 50% lethality in animals over a period of 7 days post-treatment, whereas 100% survival was observed in the case of DEMISO and PIM. The enhanced lethality may be associated with the production of toxic intermediates of MISO. Heat did not affect the levels of DEMISO in the tissues studied (plasma, brain and tumour), whereas those of PIM were markedly lowered in tumour but not affected in brain for up to 4 h after combined treatment. MISO was found to be decreased in the tumour at all times but affected differently in brain after 1 and 2 h, initially decreasing and then increasing significantly. In all cases the treatment sequence, i.e. sensitizer plus heat or vice-versa, did not affect the rate of survival. At a dose of 2 mg/g, DEMISO plus heat was found to be more toxic when DEMISO was given first (25% survival) compared to 58% on reversal. However, the levels of DEMISO in the tissues were not affected by heat. Thus, it would appear that there is no correlation between parent drug levels measured in plasma, tumour or brain and hyperthermia-induced drug lethality.

Inactivation and repair of double-stranded DNA damaged by the aziridinyl nitroimidazole RSU 1069.

Incubation of RSU 1069 in the presence of biologically active double-stranded phi X174 DNA resulted in, depending on pH, ionic strength and concentration of drug, inactivation of the DNA. A variety of lesions are induced including a high number of single-strand breaks and alkali-labile lesions, which are at most partly lethal. The main inactivating damage consists probably of base damage, induced by alkylation. A considerable part of the damage induced by RSU 1069 can be repaired by the various repair enzymes of the bacterial host of the phi X174 DNA. Finally the damage (pattern) depends considerably on the ionic composition of the reaction solution, which can be explained by an equilibrium model presented in this paper.

Metabolism and radiosensitization of 4,5-dimethylmisonidazole, a ring-substituted analog of misonidazole.

4,5-Dimethylmisonidazole (DMM) is a ring-substituted derivative of the 2-nitroimidazole, misonidazole. 2-Nitroimidazoles are able to sensitize radioresistant hypoxic cells, and to kill them outright through bioreductive metabolism. The toxic process is believed to reflect the consequences of reductively activated drugs forming adducts with cellular (macro)molecules. Both this process and the radiosensitizing activity are thought to correlate with the electron affinity of radiosensitizing agents. In the present study, methyl groups were added to the imidazole ring of misonidazole in order to hinder adduct formation with cellular molecules after reductive-activation of the compound. It was anticipated that this would substantially decrease the hypoxic-cell toxicity of the parent drug. The presence of the two methyl groups reduced the half-wave reduction potential of DMM by about 70 mV, so we expected that its radiosensitizing ability would also decrease. In direct comparison with misonidazole, DMM, at equimolar concentrations, showed dramatically reduced binding to cellular macromolecules under bioreductive conditions, both in vivo, using a liver perfusion system, and in vitro, using tissue culture cells incubated under extreme hypoxia. However, DMM was only moderately less toxic than the parent compound, and showed greatly diminished radiation sensitization capacity. Since the decrease in toxicity was much less than expected, and the decrease in radiosensitization was much more than expected, this compound may be an important drug for continuing studies on the mechanisms of radiation sensitization, binding and cytotoxicity caused by electron affinic drugs.

The same chemicals induce different neurotoxicity when administered in high doses for short term or low doses for long term to rats and dogs.

Dose- and term-dependent differences in the location and nature of brain lesions induced in rats and dogs by 2,5-hexanedione (2,5-HD), misonidazole, clioquinol, and acrylamide are reported. Subchronic neuropathies ("distal axonopathy") were induced by low-dose administration of these neurotoxicants and at high doses, lesions caused by acute or subacute neurotoxicity were found in the central nervous system (CNS). In rats, 2,5-HD induced extracellular edema, nerve cell degeneration, and axonal degeneration in the cerebellar and vestibular nuclei. Similar lesions were observed in misonidazole-treated dogs and clioquinol induced nerve cell degeneration in the hippocampus and malacia in the piriform lobes of these animals. In rats, acrylamide induced degeneration of Purkinje cells. Although the mechanism(s) underlying the differential neurotoxicity of high and low doses of these neurotoxicants remains unclear, we suggest certain biochemical mechanisms, cytotoxic edema and excitotoxicity, as factors in the production of such lesions after high-dose treatment.

Pharmacologic/pharmacokinetic evaluation of emesis induced by analogs of RSU 1069 and its control by antiemetic agents.

RB 6145, the ring-opened analog of RSU 1069, and PD 130908, the desoxy ring-opened analog of RSU 1069, were compared to RSU 1069 for their emetic potential in dogs. When RB 6145 and PD 130908 were administered intravenously at doses ranging from 20% to 50% of the mouse equivalent maximum tolerated dose (MTD), both analogs were less emetic than RSU 1069 on a molar basis. Furthermore, the 5HT3 antagonist ondansetron prevented emesis at doses as high as 75% of the MTD. The reactivity, and hence the emetic liability of these compounds, is thought to be mediated by formation of the corresponding aziridine intermediate. In mouse plasma, both analogs rapidly converted to two products, the reactive aziridine and a stable oxiazolidinone species formed upon reaction with bicarbonate in the blood. A positive correlation exists between the amounts of aziridine formed by these analogs and their emetic potential.

Mechanism of oncogenicity for bioreductive drugs.

The oncogenic transforming potential of a series of bifunctional bioreductive drugs were examined under either aerated or hypoxic conditions to assess the contribution of side chains or nitroreduced products toward their carcinogenic mechanisms. Both the cytotoxicity and transforming effects of these drugs increased as a function of dose under hypoxia. In air and at doses that resulted in comparable cell killing, RSU-1069 and RB-88716 were substantially more oncogenic than RSU-1164 or SR-4233. In nitrogen, the oncogenicity of SR-4233 as a function of survival increased, whereas the transforming effect for the aziridine-containing drugs, RSU-1969 and RB-88716, decreased. These data suggest that, among the drugs examined, the transforming moiety in air is largely a function of the alkylating aziridine group. In hypoxia, the reduction of the nitro-moiety to the corresponding active metabolites may be responsible for much of the transformation observed.

Dual function nitroimidazoles less toxic than RSU 1069: selection of candidate drugs for clinical trial (RB 6145 and/or PD 130908.

Following the toxicity and synthetic difficulties encountered with the hypoxic cell radiosensitizer RSU 1069, efforts have focused on development of a superior analogue. Two compounds, RB 6145 and PD 130908, have emerged from this program which overcome the instability and synthetic problems associated with RSU 1069 while retaining favorable biological activity. Both agents show comparable radiosensitizing activity to RSU 1069 following oral or i.p. administration to mice bearing the KHT or RIF-1 tumors. Sensitizing efficiency is about 10 X greater than that observed for misonidazole or etanidazole. Toxicity toward hypoxic tumor cells in vivo is demonstrated by clamping tumors (for 60 min) following administration of PD 130908 or RB 6145. Both are effective hypoxic cytotoxins, but less potent than RSU 1069. Systemic toxicity is substantially reduced following oral drug administration. Further, doses achievable following fractionated drug treatments are sufficiently high to produce significant levels of radiosensitization.

Radiosensitizing, toxicological, and pharmacokinetic properties of hydroxamate analogues of nitroimidazoles as bifunctional radiosensitizers/chemical modifiers.

We examined the pharmacokinetics of KIH-802, potassium 2-nitroimidazole-1-acetohydroxamate, using its radioisotope-labelled compound and the acute toxicity in mice. We discovered that the concentration of KIH-802 was very low in the brain and its LD50 was nearly half the value of that of MISO. We also present here new 2-nitroimidazole radiosensitizers/chemical modifiers (KIN-804, 811, 831, 841, 844, 821, 823 and 824) designed to enhance their sensitizing ability intensely by substituting various biologically active groups, such as hydroxamic acids and oximes, with moderate lipophilicity to the aromatic ring, if necessary, through some spacers. The sensitizing effects of all compounds were estimated to be almost equal to or better than that of MISO. The results of their toxicities shows that new hydroxamates KIN-804 and 831 are less toxic than KIH-802 and MISO. Their in vitro enhancement ratios are 2.00 and 1.75, respectively, compared with those of KIH-802, MISO and SR-2508, 1.77, 1.72 and 1.72, respectively, at each dose of 1 mM for EMT6/KU single cell. We concluded that they may be superior radiosensitizers of hydroxamic acid analogues to KIH-802.

Oral (po) dosing with RSU 1069 or RB 6145 maintains their potency as hypoxic cell radiosensitizers and cytotoxins but reduces systemic toxicity compared with parenteral (ip) administration in mice.

RB 6145 is a pro-drug of the hypoxic cell radiosensitizer RSU 1069 with reduced systemic toxicity. The maximum tolerated dose (MTD) of RSU 1069 for C3H/He mice was 80 mg/kg (0.38 mmol/kg) ip but 320 mg/kg (1.5 mmol/kg) following po administration. The MTD values of RB 6145 were 350 mg/kg (0.94 mmol/kg) ip and 1 g/kg (2.67 mmol/kg) po. Toxicity of RSU 1069 toward bone marrow stem cells was also less after po administration than after ip administration; 0.1 mmol/kg ip RSU 1069 and 0.38 mmol/kg po RSU 1069 both reduced the surviving fraction of clonogenic CFU-A cells by 50%. Oral administration of RSU 1069 resulted in lower spermatogenic toxicity. No loss of intestinal crypts was detected after ip or po administration of RSU 1069. Some nephrotoxicity was observed in half of the mice given the highest po dose of 1.5 mmol/kg of RSU 1069; this was not observed following the highest ip dose of drug. For RSU 1069 and RB 6145, administered by either route, the maximum hypoxic cell radiosensitization in murine KHT sarcomas, occurred when the drugs were given 45-60 min before 10 Gy of X rays. The degree of radiosensitization produced by a particular dose of either compound was largely independent of the route of administration. Preliminary pharmacokinetic studies, using 3H-RSU 1069, suggested that anti-tumor efficacy correlated with peak blood level of label and concentration in the tumor at the time of irradiation, which were not reduced by po compared with ip administration. Normal tissue toxicity tended to correlate with total exposure over time, which was reduced approximately two-fold by po administration. Oral administration of RSU 1069 or RB 6145, as well as being convenient, may give therapeutic benefit since dose-limiting toxicity in mice was reduced compared with parenteral administration, whereas radiosensitizing activity was less affected.

The effect of hydralazine on blood flow and misonidazole toxicity in human tumour xenografts.

The effect of post-irradiation hypoxia induced by 5 or 30 mg/kg hydralazine has been studied in three human tumour xenografts (two rectocolic adenocarcinomas and one melanoma) treated with two doses of misonidazole similar to those used in patients (0.1 and 0.2 mg/g). Only a small sensitization was detected using an in vitro colony assay. These results are in marked contrast to the results obtained with rodent tumours. This difference between human tumour xenografts and rodent tumours might be explained by differences in the reduction of tumour blood flow after hydralazine administration (5 and/or 10 mg/kg). Using the laser Doppler technique, the tumour blood flow reduction was 33% and 25% of the control for NA11 and HRT18 tumours, respectively. In contrast, hydralazine induced a 60-70% reduction in blood flow in the murine SCCVII tumour. Using the fluorescent marker Hoechst 33342, the reduction in perfusion was again more pronounced in the murine tumour as compared to the Na11 and HRT18 xenografts. The differences between human tumour xenografts and rodent tumours are not linked to the mouse strain used (nude versus C3H) nor to a tumour bed effect.

A comparison of the relative activities of 8 radiosensitizers in the SOS chromotest.

Misonidazole, and RSU 1069 and 6 of its analogues are all reported to show increased cytotoxicity towards hypoxic cells compared to oxic cells. DNA is considered to be the target through which these drugs exert their cytotoxic activity. Therefore we monitored induction of the SOS response in uvrABC excinuclease proficient and deficient strains of E. coli, under oxic and hypoxic conditions, as an indirect method of assessing the activity of these drugs towards DNA in a biological system. This was done using the SOS chromotest which utilizes E. coli strains which possess a sfiA::lacZ fusion allowing induction of the SOS response to be monitored by assaying beta-galactosidase activity. All of the drugs tested here show some induction of the SOS response in both uvrABC excinuclease proficient and deficient strains. Data shown here suggests that the uvrABC excinuclease is important in the production of a SOS induction signal from RSU 1069-induced DNA lesions and that RSU 1069 may act as a crosslinking agent. The data also shows that SOS induction activity and toxicity do not necessarily correlate and that production of a SOS induction signal may occur via a different pathway for RSU 1069 than for its analogues.

Evaluation of a new 2-nitroimidazole nucleoside analogue, RK-28 as a radiosensitizer for clinical use.

The experimental data previously reported on RK-28, a hypoxic cell sensitizer which is now being tested in a phase I clinical trial, are confusing. Some data indicate superiority of RK-28 over misonidazole (MISO), whereas others do not. This paper presents our experimental data on the efficacy, toxicity, and pharmacokinetics of RK-28, in comparison with those of MISO, and also summarizes the data of other investigators. In our experiments, RK-28 had a 1.5-2.5 times higher sensitizing activity in vitro on EMT6 and SCCVII cells than MISO, and the difference was larger when the pre-irradiation incubation time was longer. The latter was considered to be due to the time-dependent cellular uptake and reactivity of RK-28 with non-protein sulphydryls. In vivo, RK-28 was almost as efficient as or slightly inferior to MISO against SCCVII and EMT6 tumours when assayed with an in vivo/in vitro assay and a growth delay time assay. The LD50/7 by a single injection of RK-28 was half that of MISO, but when 60% of LD50/7 was injected into mice every day, the total dose that could be given was higher for RK-28 than for MISO. Pharmacokinetic studies using mice, rats, rabbits, and a dog showed that RK-28 was rapidly eliminated from the blood and various tissues. From our results it was concluded that the possible success of the clinical trial of RK-28 depends on its low cumulative toxicity.

Studies with bifunctional bioreductive drugs. I. In vitro oncogenic transforming potential.

The oncogenic transforming potential of a series of bioreductive drugs including RSU-1069 and its various alkyl-substituted derivatives, RB-7040, RB-88716, RSU-1164, and RB-88712, has been compared using the C3H 10T1/2 cell system. While the aziridine moiety at the terminal end of the side chain confers greater cytotoxicity to both the 2-nitroimidazole (RSU-1069) and the 5-nitrofuran (RB-88716), it also increases the oncogenic transforming potential of the drugs correspondingly. By substituting the aziridine ring with methyl groups, the cytotoxicity and oncogenicity of these bioreductive drugs decrease in a way that is proportional to the degree of methylation. A clear structure-activity relationship can be demonstrated from these methyl-substituted derivatives such that a tetramethyl-substituent (RB-7040) is much less cytotoxic and oncogenic than a dimethyl-substituent (RSU-1164). RB-7040, which has in vitro and in vivo sensitizing efficiency comparable to the parental compound RSU-1069, is roughly tenfold less cytotoxic and, at concentrations that achieve an in vitro enhancement ratio of 2.9, induces a transforming frequency that is indistinguishable from the spontaneous rate.