Search

Tumour oxygenation status in individual patients may be assessed using the bioreduction and linkage of 2-nitroimidazole markers to viable hypoxic cells in vivo with subsequent detection by conventional nuclear medicine techniques. Iodoazomycin arabinoside (IAZA) was radiolabelled with Iodine-123 and administered i.v. to 51 patients with newly diagnosed malignancies whose tumours were subsequently imaged by planar and single-photon emission computed tomographic (SPECT) procedures. Quantitative analyses of radiotracer avidity were performed at 24 h post-injection and tumour-normal tissue ratios of greater than 1.10 were deemed positive for tumour hypoxia. By this criterion, the frequencies of hypoxia in small-cell lung cancer, squamous cell carcinomas of head and neck and malignant gliomas were 60% (9/15), 40% (6/15) and 0% (0/11) respectively. The correlation of positive IAZA scans with tumour control and survival in patients with lung cancer and head and neck tumours is currently under study. Preliminary observations in neck metastases from squamous cell carcinoma of head and neck tumours indicates decreased local control at 3 months post-treatment in tumours with IAZA avidity. This study concludes that: (1) 123I-IAZA can be administered safely and repeatedly as an outpatient routine imaging procedure in cancer patients during initial work-up and follow-up; (2) that retained drug can be detected by conventional nuclear medicine procedures in inaccessible deep-seated tumours; and (3) that this technique could prove useful for identifying those patients for whom hypoxia-directed therapy is indicated.

Tumor perfusion and oxygenation status have been suggested as factors which may influence treatment outcome in cancer patients. Nuclear medicine assays of tumor perfusion [99mTc-hexamethylpropylenamine oxime (HMPAO)] and tumor hypoxia [123I-iodoazomycin arabinoside (IAZA)] have recently been developed and described. We report on measurements of perfusion and oxygenation status of 27 tumors in 22 patients using these probes. An inverse correlation between tumor uptake of HMPAO and IAZA was measured (p < 0.05), with severe perfusion deficit usually associated with an increased uptake of the hypoxic marker. This trend was observed for limited stage small-cell lung carcinoma, squamous-cell carcinoma of the head and neck, soft-tissue sarcoma, brain metastases from small-cell lung carcinoma and adenocarcinoma of the prostate as a group, but not for glioblastoma multiforme. Whereas each imaging agent can yield information about the physiological status of tumor and normal tissue, the information resulting from their combined use could be important in cancer therapy.

Non-invasive predictive assays which can confirm the presence or absence of hypoxic cells in human tumours show promise for understanding the natural history of tumour oxygenation, and improving the selection of patient subsets for novel radiotherapeutic strategies. Sensitiser adducts have been proposed as markers for hypoxic cells. Misonidazole analogues radiolabelled with iodine-123 have been developed for the detection of tumour hypoxia using conventional nuclear medicine techniques. In this pilot study, we have investigated one such potential marker, 123I-iodoazomycin arabinoside (123I-IAZA). Patients with advanced malignancies have undergone planar and single-photon emission computed tomographic (SPECT) imaging after intravenous administration of 123I-IAZA. We have observed radiotracer avidity in three out of ten tumours studied to date. Normal tissue activity of variable extent was also seen in the thyroid and salivary glands, upper aerodigestive tract, liver, intestine, and urinary bladder. Quantitative analysis of those images showing radiotracer avidity revealed tumour/normal tissue (T/N) ratios of 2.3 (primary small cell lung carcinoma), 1.9 (primary malignant fibrous histiocytoma) and 3.2 (brain metastasis from small cell lung carcinoma) at 18-24 h post injection. These preliminary data suggest that the use of gamma-emitter labelled 2-nitroimidazoles as diagnostic radiopharmaceuticals is feasible and safe, and that metabolic binding of 123I-IAZA is observed in some, but not all tumours. The inference that tumour 123I-IAZA avidity could be a non-invasive measure of tumour hypoxia deserves independent confirmation with needle oximetry.

In a Phase I trial SR 2508 was administered by rapid intravenous infusion to 102 patients receiving radiation therapy. The dose-limiting toxicity was peripheral sensory neuropathy (PN) which was related to the cumulative dose administered. The highest single daily dose, 3.7 g/m2, was tolerated without toxicity. The lowest cumulative toxic dose was 21.6 g/m2, and the highest non-toxic dose was 40.8 g/m2. Grade 1 neuropathies were mild and self-limited; grade 2 neuropathies were long-lasting and debilitating. In a retrospective analysis, the risk of developing neurotoxicity was related to the cumulative drug exposure calculated by the area-under-the-curve (AUC) of plasma concentration versus time. There was an increased incidence of neuropathy in patients with a cumulative AUC of greater than or equal to 36 mM-hr. At a total dose of 34 g/m2 over 6 weeks, the incidence of Grade 1 neuropathy was approximately 30%; no grade 2 neuropathy occurred at this dose and schedule. Additional toxicities observed included nausea and vomiting (6%), skin rash (4%), and transient arthralgias (3%). One patient had transient abnormalities in liver function tests of unknown etiology. (In a more recent Phase II trial neutropenia has been observed which may be related to SR2508). Approximately three times more SR 2508 is tolerable compared to misonidazole, and it appears that severe neuropathy can be avoided by monitoring individual patient pharmacokinetic parameters. Evaluation of the efficacy of this hypoxic cell sensitizer is in progress.

Neurotoxicity induced by misonidazole (MISO) and desmethylmisonidazole (DMM) has become the dose limiting factor in clinical work. In 1981, we reported a preliminary study suggestive that Dexamethasone (DEXA) does have a protective effect against peripheral neuropathies (PN) resulting from toxicity of misonidazole. Furthermore, in that study we have observed that DEXA did not alter the plasma pharmacokinetics of misonidazole. We are presently investigating the use of DEXA (2 mg t.i.d. during treatment), with escalating doses of MISO in an attempt to modify its neurotoxicity. To date, 16 patients have been registered to receive total doses of MISO ranging from 13.5 to 17.5 gm/M2 given in 9 equally divided doses over 3 weeks. DEXA, 2 mg t.i.d. is given 3 days prior to the first dose and continues for the duration of therapy. All patients receive palliative radiation. No toxicity was seen at the total dose of 13.5 gm/M2. One grade I PN occurred in the first four patients receiving 15.5 gm/M2. Six additional patients were entered at this dose level and no further incidence of PN was observed. Patients are being entered at the next step of 17.5 gm/M2.

Although considerable laboratory in vitro and in vivo evidence is now available suggesting that misonidazole (MISO) enhances chemotherapy tumor responses, experience with human tumors is limited. Further, the mechanism of this enhancement is not definitely known. One possible mechanism is that MISO alters the pharmacokinetics of the chemotherapeutic agent, vice versa or both. We studied a group of patients with recurrent malignant gliomas, following radiotherapy. After proven recurrence, they were treated with i.v. BCNU in combination with oral MISO in an 8 week cycle. Our aims were: 1. To obtain a second remission; 2. To assess the toxicity of this combination; 3. To assess the plasma pharmacokinetics of each drug alone and in combination. Six patients entered the protocol. Four of six patients obtained either a partial or subpartial response. Prolonged moderate myelosuppression was observed in 2/6 patients after 3 cycles; 2/6 patients experienced seizures after the first cycle of chemotherapy for the first time in the course of their disease. The plasma pharmacokinetic data indicates no evidence of a MISO-BCNU drug interaction.

With a view to modifying misonidazole (MISO) neurotoxicity, we initiated a randomized clinical study to assess a possible drug interaction and toxicity protection when dexamethasone (DXM) is administered concomittantly with MISO. The ongoing study consists of: 1. Pharmacokinetic evaluation; 2. Assessment of toxicity. Fourteen patients undergoing radiation therapy for different types of malignant neoplasia (excluding brain tumors) have been randomized to receive either MISO alone, or DXM one week prior and during treatment with MISO. Five of seven patients receiving MISO alone developed peripheral neuropathies while only one out of 7 patients that received MISO with DXM coverage developed a transient and mild neuropathy. Pharmacokinetic evaluation of MISO in plasma and urine of those patients receiving DXM has shown no evidence of drug interaction. It is postulated that the mechanism of action of DXM is at the nerve cell membrane level, restoring and stabilizing cell surface properties. In future studies we will investigate the use of DXM with increasing doses of MISO above the recommended maximum dose of 12 gm/m2, hoping to achieve a higher tumor tissue level of MISO while avoiding unacceptable toxicity. The effect of Allopurinol on the plasma kinetics of MISO was studied in four additional patients, observing also no evidence of drug interaction.

From March 15, through August 31, 1983, 37 patients have been entered on the RTOG Phase I trial of SR-2508. The drug was given intravenously three times weekly for three weeks. The starting total dose was 11.7 g/m2 and the highest total dose given was 32 g/m2. The lower lipophilicity of SR-2508 has produced the expected decrease in terminal half-life (5.4 hrs) of drug excretion and increase in total drug excreted unchanged in the urine (71%) compared to misonidazole or desmethylmisonidazole. The maximum single dose (3.7 g/m2) administered was well tolerated. With multiple doses peripheral neuropathy is the dose-limiting toxicity. The lowest cumulative dose producing toxicity was 21.6 g/m2, the highest non-toxic dose was 29.7 g/m2. The use of an individual patient's drug exposure as measured by the area under the curve of drug concentration vs time may be an excellent predictor of toxicity. This may eventually permit individualization of dose and prevention of serious toxicity. A single dose of 2 g/m2 will produce a tumor concentration of drug (approx. 100 micrograms/ml) that will yield a sensitizer enhancement ratio of 1.5 to 1.7. Using a starting dose of 2 g/m2 three times weekly, patients are now being studied on a five week drug schedule to further evaluate predictability of drug toxicity in preparation for clinical trials of drug efficacy.

The human tolerance to multiple dosages of misonidazole (Ro-07-0582) was studied in 28 patients with different types of malignant neoplasias. The mean total dose for this group of patients was 16.2 g. The main toxicity was peripheral neuropathy with an overall incidence of 35%. This neuropathy occurred more frequently and with greater severity when the drug was administered 3 times a week and when patients received total doses of over 18 g. The best tolerated schedule appears to be once or twice a week up to total dosages of 18 g or less (approximately 11 g/m2). Electron microscopy of a sural nerve biopsy from an affected patient revealed residual of previous distal axonal degeneration, with some segmental demyelination and remyelination, which affected both large and small myelinated nerve fibres.

Since January 1981, 52 patients have entered the Radiation Therapy Oncology Group Phase I trial with intravenous (i.v.) desmethylmisonidazole (DMM). DMM is less lipophilic than misonidazole (MISO) and theoretically will be less neurotoxic due to lower penetration into neural tissue and more rapid elimination. The drug is administered intravenously to achieve the maximum drug concentration in tumor for a given dose. The protocol slowly escalates the total dose of drug administered. At this time the planned dose on the three week schedule is 1g/m2 five times per week to a total of 15g/m2, and on the seven week schedule is 1.25g/m2 twice weekly to a total dose of 17.5g/m2. The preliminary plasma pharmacokinetic data demonstrates high peak plasma levels within five minutes of the end of the drug infusion. Compared to MISO the percent of DMM excreted in the urine is increased, 63% vs 10%, and the elimination half-life is decreased: DMM, i.v. 5.3h; MISO, i.v. 9.3h; MISO, oral 10 to 13h. Neurotoxicity has been observed in approximately 30% of patients given a cumulative dose of greater than 11g/m2. This is in comparison to a 50% incidence in the RTOG Phase I study with oral MISO at doses of 12g/m2. There is not sufficient data to evaluate the relationship between neurotoxicity and drug exposure. Further patient accrual on this study is required to better define the properties of DMM.

Various attempts have been made to improve the effectiveness of radiation in the treatment of cerebral malignant astrocytomas. A trend favoring multiple daily fractionated (MDF) radiation therapy over conventional single daily fractionated (CF) radiation therapy was identified in our previous study. In order to assess the effect of MDF with and without misonidazole, a province-wide prospective randomized trial was initiated in January 1981. By March 1984, 124 patients with histologically verified grade III and IV astrocytomas were randomized to CF (5800 cGy/6 weeks/30 fractions), MDF (6141 cGy/4.5 weeks/69 fractions at 89 cGy every 3-4 hours, three times a day) and MDF in combination with misonidazole (1.25 g/m2 three times weekly for the first 3 weeks). Thirty-eight patients were randomized to CF, 43 patients to MDF, and 43 patients to MDF and misonidazole. At the preliminary assessment in July 1984, the median survival time was 27 weeks for the CF group, 39 weeks for the MDF group and 49 weeks for MDF and misonidazole group. The 1-year actuarial survival rate from surgery was 20% for CF group, 41% for MDF group, and 45% for MDF and misonidazole group. There is a statistically significant difference (P less than 0.001) between the CF and MDF group. However, the addition of misonidazole does not significantly alter survival.

Misonidazole is a 2-nitroimidazole compound currently being assessed as a radiosensitizing agent. The effects of misonidazole on human bone marrow hematopoiesis were assayed by culture of committed granulocyte-macrophage progenitor cells (CFU-C). Three groups of patients with nonhematologic malignancies were selected for study. The first group of patients received a single, large dose of misonidazole; the second group received smaller, multiple doses of misonidazole; and the third group who did not receive any misonidazole served as irradiation controls. In 14 of 16 patients who received single or multiple doses of misonidazole, there was a significant decrease in the number of CFU-C present in the bone marrow after misonidazole therapy. In five patients who received irradiation only, there was no difference in the number of pre- and post-treatment bone marrow CFU-C. In misonidazole treated patients, extensive washing of post-treatment bone marrow samples failed to return CFU-C growth to control values. Suppression of CFU-C growth persisted for 3 weeks and returned to control values by 8 weeks. This reduction in the proliferative capacity of human bone marrow progenitor cells suggests that misonidazole may add to the myelotoxicity already associated with radiotherapy, systemic chemotherapy, or as combination of the two.

As part of a phase II study with the hypoxic cell radiosensitizer, misonidazole, we have encountered two cases of dermatitis. When all drugs were withdrawn and the patients were challenged with a small single dose of misonidazole, identical adverse reactions reappeared demonstrating a drug-related skin hypersensitivity.

Complete pharmacological data from 71 patients treated on the phase I trial of SR 2508 were analyzed to see if the dose-limiting toxicity of peripheral neuropathy is related to the individual patient's pharmacokinetic profile. In a retrospective analysis, the risk of toxicity was best predicted by using the bivariate model of total drug exposure and the time over which the treatment course was given. Drug exposure [area under the curve (AUC)] for a single treatment was calculated by the integral over time of the serum concentration of SR 2508. Since the AUC was constant during the course of a patient's treatment, the total drug exposure (total-AUC) was estimated by the product of the AUC times the number of drug administrations. While the clinical efficacy of hypoxic cell sensitizers remains to be proven, SR 2508 is better tolerated than its predecessors, misonidazole and desmethylmisonidazole, as three times the amount of SR 2508 can be given. If this model is confirmed in the current phase II and III trials, the probability of developing neuropathy would be predictable for an individual patient from measurements made at the time of the first drug dose, allowing for the adjustment of drug schedule to avoid all but minor toxicity.

Sixty-five patients were entered on the long schedule of the Phase I trial of SR-2508. The planned total doses ranged from 30 to 40.8 g/m2 using various treatment schema including daily, split course, and every-other-day schedules. The individual dose size was 2 g/m2 for 56 patients and 1.7 g/m2 for nine. In contrast to misonidazole and desmethylmisonidazole, more SR-2508 can be administered as the duration of therapy is lengthened. All six patients on the 30 g/m2 step tolerated the drug without toxicity. This total dose was not achievable in the three week schedule. Additionally, a number of patients did not develop neuropathy at a cumulative dose of 40.8 g/m2. Although the analysis is not yet complete, a given patient's drug exposure as measured by their total AUC (mMxhr), defined as the area-under-the-curve of serum concentration of SR-2508 vs. time for a single dose times the number of doses given, is useful in predicting toxicity for that patient. The recommended starting schedule for the Phase II and III trials is 34 g/m2 over a 6 week period (2 g/m2 every other day). A total AUC of approximately 39 mMxhr should be tolerable. The drug regimen must be altered for patients who have a high AUC. Therefore, it is mandatory to have an accurate and rapid pharmacokinetic analysis for each patient. The clinical efficacy of the hypoxic cell sensitizers remains to be proven. However, using the guidelines derived from the Phase I trial, SR-2508 should be a relatively safe drug, producing minor or no toxicity.

Multiple daily fractionated radiation therapy (MDF) may be more effective than conventionally fractionated radiation therapy (CF) in the treatment of malignant glioma. The hypoxic cell sensitizer misonidazole (MISO) could be more effective when employed with small fractions of radiation every 4 hours to take advantage of the long half-life of the drug. To evaluate MDF and MDF in combination with MISO, a randomized prospective trial was initiated. Between January 1981, and December 1982, patients with histologically verified astrocytoma with anaplastic foci or glioblastoma multiforme were randomized to CF (5800 cGy, 30 fractions, 6 weeks), MDF (6141 cGy, 69 fractions, 4 1/2 weeks, at 89 cGy every 4 hours 3 times daily) and MDF in combination with MISO (1.25 gm/M2 three times weekly for the first 3 weeks). In January 1983, the CF arm was dropped and a high dose MDF arm added (7120 cGy, 80 fractions, 5 1/2 weeks, at 89 cGy per fraction every 4 hours 3 times daily). CCNU chemotherapy was given at the time of tumor progression. One hundred and twenty-eight patients were evaluated (38 CF, 42 MDF, 37 MDF plus MISO, and 11 high dose MDF). Median survival was 29 weeks for CF, 45 weeks for MDF and 50 weeks for MDF plus MISO. Survival was significantly improved for patients treated with MDF compared to patients treated with CF (p less than .002). The addition of MISO to MDF did not result in further improvement in survival. Acute toxicity was acceptable. No clinically apparent delayed toxicity was observed.

Results of world-wide clinical trials with misonidazole are discussed. An attempt is made to assess the reasons for the lack of positive results and the cost-benefit analysis is critically reviewed. The data on the clinical investigations of the second generation misonidazole analogues SR-2508 and RO-03-8799 are presented. Emphasis is placed on future work such as tumor selection for clinical trials, reduction of drug toxicity and methods to increase the drug radiosensitizing properties. Because of the large amount of knowledge, experience, productivity and good scientific clinical data accummulated with nitroimidazoles over the past five years, it is recommended that this is the time to push forward with the work on the newest, more efficient compounds.

RO-07-0582 toxicity studies were performed in 12 patients for a total of 16 assays. Single and multifraction dose schedules were used, and drug concentrations in solid tumor tissue, cerebrospinal fluid, and blood (at 14 to 24 hr.) were established. A severe peripheral neuropathy occurred in 1 patient on the multifraction regimen when the total dosage reached 24 g. Drug absorption and concentration in the blood do not significantly differ from that of metronidazole. Maximum blood levels were reached at from 2 to 4 hr., and several days subsequent to administration had stabilized to levels only slightly above control levels.