Hypoxic cell radiosensitizers: the end of an era? Regarding Lee et al., IJROBP 32:567-576; 1995

Nitroimidazoles as hypoxic cell radiosensitizers and hypoxia probes: misonidazole, myths and mistakes

Nitroimidazoles have been extensively explored as hypoxic cell radiosensitizers but have had limited clinical success, with efficacy restricted by toxicity. However, they have proven clinically useful as probes for tumour hypoxia. Both applications, and probably much of the dose-limiting toxicities, reflect the dominant chemical property of electron affinity or ease of reduction, associated with the nitro substituent in an aromatic structure. This single dominant property affords unusual, indeed extraordinary flexibility in drug or probe design, suggesting further development is possible in spite of earlier limitations, in particular building on the benefit of hindsight and an appreciation of errors made in earlier studies. The most notable errors were: the delay in viewing cellular thiol depletion as a likely common artefact in testing in vitro; slow recognition of pH-driven concentration gradients when compounds were weak acids and bases; and a failure to explore the possible involvement of pH and ascorbate in influencing hypoxia probe binding. The experience points to the need to involve a wider range of expertise than that historically involved in many laboratories when studying the effects of chemicals on radiation response or using diagnostic probes.

Reevaluation of the radiosensitizing effects of sanazole and nimorazole in vitro and in vivo

Sanazole (AK-2123, 3-nitrotriazole derivative, N1-(3-methoxypropyl)-2-(3-nitro-1 H-1,2,4-triazol-1-yl)acetamide) and nimorazole (5-nitroimidazole derivative, 4-(2-(5-nitro-1H-1-imidazolyl)ethyl)morpholine) have been tested clinically as hypoxic cell radiosensitizers, mainly outside Japan. To determine if these sensitizers deserve clinical investigation in Japan, we reevaluated the radiosensitizing effects of these compounds in vitro and in vivo, in comparison with a fluorinated 2-nitroimidazole derivative KU-2285 (N1-(2-hydroxyethyl)-1,2-difluoro-3-(2-nitro-1 H-1-midazolyl)propanamide). KU-2285 is a known and established radiosensitizer, but is not suitable for clinical studies because of the high cost of synthesis. In vitro, the radiosensitizing effects of the three compounds on SCCVII (squamous cell carcinoma line in C3H mice) tumor cells were examined at 0.5 and 1 mM under aerobic or hypoxic conditions, using a colony assay. In vivo, SCCVII tumors grown subcutaneously in the hind legs of C3H/HeN mice were irradiated with or without prior intraperitoneal administration of 100, 200 or 400 mg/kg of the drugs. Thereafter, tumor growth delay was measured. In vitro, no sensitizing effect was observed under aerobic conditions at 1 mM. Under hypoxic conditions, the sensitizer enhancement ratio (SER) determined at 1% cell survival level for sanazole, nimorazole and KU-2285 was 1.55, 1.45 and 1.95, respectively, at 1 mM, and 1.40, 1.40 and 1.75, respectively, at 0.5 mM. In vivo, all three compounds had significant radiosensitizing effects; their effects appeared to decrease in the order of KU-2285, sanazole, and nimorazole. It was suggested that sanazole may be more suitable for clinical trials than nimorazole.

Effects of hypoxic cell radiosensitizer doranidazole (PR-350) on the radioresponse of murine and human tumor cells in vitro and in vivo

We have investigated the radiosensitizing effect of doranidazole, a hypoxic cells radiosensitizer, using SCCVII tumor cells of C3H mice and CFPAC-1 and MIA PaCa-2 human pancreatic tumor cells. The radiosensitivity of hypoxic SCCVII cells in vitro increased with 1 mM doranidazole by a factor of 1.34 and 1.68, when determined by clonogenic survival and micronucleus (MN) formation, respectively. The radiation-induced growth delay of SCCVII tumors was significantly enhanced and the TCD(50/120) was reduced by a factor of 1.33 when 200 mg/kg doranidazole was injected, i.v., 20 min prior to tumor irradiation. The in vivo-in vitro excision assay showed that radiosensitivity of SCCVII cells in vivo increased by a factor of 1.47 with 200 mg/kg doranidazole. The radiation-induced growth delay of CFPAC-1 xenografts in nude mice was significantly enhanced and the TCD(50/90) was reduced by a factor of 1.30 by 200 mg/kg doranidazole. On the other hand, 200 mg/kg of doranidazole exerted no influence on the radiation-induced growth delay in MIA PaCa-2 xenografts. The tumor oxygenation status, as determined with an oxygen sensitive needle probe and the immunohistological study using pimonidazole, indicated that MIA PaCa-2 tumors are better oxygenated than CFPAC-1 tumors. The relatively well-oxygenated status in MIA PaCa-2 tumor may account for the lack of radiosensitization by doranidazole. It is concluded that the magnitude of radiosensitization of tumors by doranidazole is dependent on the oxygenation status of the tumors and that doranidazole may be useful in increasing the response of hypoxic human pancreatic tumor to IORT.

Effect of nitroimidazole sensitizers on in vitro glycolytic metabolism of hypoxic squamous cell carcinoma

Two nitroimidazole compounds, misonidazole (MISO) and nimorazole (NIMO), were evaluated for their potential to modify uptake of [5,6-3H] 2-fluoro-2-deoxy-D-glucose (3H-FDG) in the human squamous carcinoma cell line UT-SCC-5 exposed to increasing levels of hypoxia. UT-SCC-5 cells were incubated with 0-10 mM of MISO or NIMO under normal or reduced oxygen concentrations of 20%, 1.5%, or 0% with 5% CO2 for 6 h, after which 74 KBq of 3H-FDG was added in media for 1 h. In the presence of normal concentrations of O2, both sensitizers increased 3H-FDG uptake by up to 178% (MISO) or 84% (NIMO) when compared with untreated cells. In anoxia, MISO decreased 3H-FDG uptake to 35% of that of control whereas NIMO-treated cells showed a respective decrease in tracer uptake to 62%. Clonogenic assays clearly indicated that MISO was toxic and NIMO moderately toxic for hypoxic cells, whereas both sensitizers exerted only a very modest effect on survival of fully oxygenated cells. Our findings indicate that nitroimidazole treatment consistently increases 3H-FDG uptake into UT-SCC-5 cells under normal oxygen concentrations. In hypoxia, the observed decrease in tracer uptake is dependent on both the level of ambient oxygen and drug concentration and may reflect both direct toxicity and inhibition of glycolysis. The observations may be useful for further applications of 18F-FDG positron emission tomography (PET) to monitor effects of hypoxic cell radiosensitizers on tumor metabolism in vivo.

Comparison of the effectiveness of tirapazamine and carbogen with nicotinamide in enhancing the response of a human tumor xenograft to fractionated irradiation

The goal of this study was to compare, with a human tumor xenograft, two different strategies for increasing tumor response to fractionated irradiation, namely, oxygenating the hypoxic tumor cells with carbogen and nicotinamide, or killing these cells with the hypoxic cytotoxin, tirapazamine (TPZ). We used the human hypopharyngeal squamous cell carcinoma cell line FaDu implanted in immune-deficient SCID mice and assessed its response to radiation by cell survival and by growth delay. The tumors were irradiated either once or twice daily with 2 or 2.5 Gy/fraction with either TPZ (0.08 mmol/kg) or nicotinamide (1,000 mg/kg) with carbogen breathing. We also tested the effect of giving TPZ on alternate days, or daily during the first half of the course, the second half, or for the whole course of radiation. We found that adding TPZ or nicotinamide with carbogen to the fractionated radiation regimen enhanced the response of the human xenograft. The enhancement was somewhat greater (though not significantly so) for TPZ, especially when given with each radiation dose. In conclusion, adding TPZ, or nicotinamide plus carbogen, to fractionated irradiation enhanced the response of this human tumor xenograft to fractionated irradiation. Consistent with theoretical modeling, there was a greater enhancement of the radiation response of the tumor when TPZ was given with each radiation dose than when given with only half of the radiation doses.

Angiogenesis vs. response after combined chemoradiotherapy of squamous cell head and neck cancer

Oxygen/drug supply to cancer cells is an important factor defining response to radiotherapy and chemotherapy. Although tumor angiogenesis is considered an important prognostic marker, its role in the outcome of chemotherapy and radiotherapy is unknown. In the present study we examined the possible correlation of the degree of angiogenesis with response to cytotoxic therapy in locally advanced squamous cell head and neck cancer (HNC). Vascular grade (VG) was assessed immunohistochemically using the JC70 monoclonal antibody (MAb) in tumor specimens from 76 patients treated with platinum/5-fluorouracil (with or without methotrexate) induction chemotherapy (ICT) (n = 37) or concurrent chemoradiotherapy (CCRT) with cisplatin or carboplatin (n = 39). Seventeen of 76 analyzed patients had an overall microvessel score of < 11 (VGI), 25/76 of 11-30 (VG2), 16/76 of 31-50 (VG3) and 18/76 of > 50 (VG4). Complete response rate after ICT or after CCRT was higher in cases with an intermediate vascularization (VG2,3). Both local relapse-free and overall survival were significantly better in the VG2 group. Patients treated with CCRT had a better survival compared to those treated with ICT. This was mainly observed in VG1 tumors. Multivariate analysis showed that VG and treatment modality were independent prognostic factors for local relapse and survival. Intratumoral angiogenesis correlation with the cytotoxic therapy outcome is likely to follow a bell-shaped relation, the response being better in cases with an intermediate VG. This may be the consequence of 2 vasculature-dependent factors, i.e., the drug/oxygen availability and the ability of cancer cells to undergo rapid repopulation in optimally oxygenated conditions. Our pilot study stresses the importance of individualization of therapy according to VG.

Concurrent tirapazamine and radiotherapy for advanced head and neck carcinomas: a Phase II study

PURPOSE

To evaluate the efficacy and toxicity of tirapazamine, a hypoxic cytotoxin, combined with conventional radiotherapy (RT) for advanced head and neck carcinomas.

MATERIALS AND METHODS

From Oct. 1994 to Nov. 1996, 40 patients with stage III or IV carcinomas of the head and neck were enrolled in a Phase II trial to receive conventional RT (70 Gy in 7 weeks) with concurrent tirapazamine (159 mg/m2 intravenously, 3 times per week for 12 doses). One patient subsequently withdrew from the protocol treatment, and was excluded from analyses. Among the 39 cases, the primary sites were located in the oropharynx (n = 28), supraglottic larynx (n = 6), or hypopharynx (n = 5). Twenty-seven patients had T3 or T4, and 27 had N2 or N3 disease.

RESULTS

Thirty-two (82%) patients received full 12 drug doses. Thirty-two patients (82%) received full 70 Gy of RT. The most frequent drug toxicities were muscle cramps (77%) and nausea/vomiting (62%), usually grade 1 or 2. Overall, 13 patients (33%) experienced grade 3 or 4 drug-related toxicities. No excessive RT-associated acute normal tissue reactions were observed. With a median follow-up of 13 months, the 1-year and 2-year local control rate was 64% and 59% respectively.

CONCLUSION

The tirapazamine regimen was well tolerated with a compliance rate of 82%. The toxicity of RT with concurrent tirapazamine was acceptable in treating advanced head and neck carcinomas. The disease control trend was encouraging. Further clinical studies are warranted.

Evaluation of cervical nodal necrosis in nasopharyngeal carcinoma by computed tomography: incidence and prognostic significance

PURPOSE

The purpose was to study the prognostic value of contrast-enhanced computed tomography (CT) nodal necrosis in nasopharyngeal carcinoma.

PATIENTS AND METHODS

One hundred sixty-one patients with newly diagnosed nasopharyngeal carcinoma and nodal metastases were reviewed. Forty patients also received cisplatin-based neoadjuvant chemotherapy in addition to radiotherapy. Nodal necrosis was defined as presence of hypodense areas in more than 33% of the node. Nodal response rate to chemotherapy, overall nodal control rate, local control rate, distant failure rate, overall relapse-free survival rate, and overall and cause-specific survival rates were compared between patients with and without nodal necrosis. Multivariate analysis was also performed.

RESULTS

The incidence of nodal necrosis was 22.9%. Overall nodal response rates to chemotherapy were 88.9% (8/9) in patients with nodal necrosis and 74.2% (23/31) in those without. No significant differences in nodal control rate, local control rate, distant failure rate, and overall and cause-specific survival rates were found. Five-year overall relapse-free survival rate was lower in patients with cervical nodal necrosis (36%) as compared with those without (53%, p = .04). Multivariate analysis, however, did not confirm cervical nodal necrosis to be an independent prognostic factor.

CONCLUSIONS

Presence of nodal necrosis in nasopharyngeal carcinoma does not affect nodal response to chemotherapy and nodal control by radiotherapy with or without chemotherapy. Cervical nodal necrosis does not appear to be an independent factor in predicting treatment outcome. Further studies to correlate nodal density with oxygenation status as well as tumor cell kinetics are warranted.

A phase I/II study of a hypoxic cell radiosensitizer KU-2285 in combination with intraoperative radiotherapy

A fluorinated 2-nitroimidazole radiosensitizer KU-2285 was given before intraoperative radiotherapy (IORT) to 30 patients with unresectable, unresected or macroscopic residual tumours. Twenty-three patients had pancreatic cancer and five had osteosarcoma. The IORT dose was 30 Gy for unresectable pancreatic cancer and 60 Gy for osteosarcoma. The dose of KU-2285 administered ranged from 1 to 9 g m-2. Four patients received a dose of 9 g m-2, and ten received 6.8-7 g m-2. All patients tolerated KU-2285 well, and no drug-related toxicity was observed. The average tumour concentration of KU-2285 immediately after IORT was 166 microg g-1 at dose of 6.8-7 g m-2 and 333 microg g-1 at 9 g m-2. The average tumour-plasma ratio was > or = 0.82. Eleven patients with unresectable but localized pancreatic cancer treated with KU-2285 plus IORT and external beam radiotherapy had a median survival time of 11 months and 1-year local control rate of 50%, which compares favourably with those of 8 months (P = 0.26) and 28% (P = 0.10) for 22 matched historical control patients. The five patients with osteosarcoma attained local control. The results of this first study on KU-2285 and IORT appear encouraging, and further studies of this compound seem to be warranted.

Evaluation of the radiosensitizing effects of RK28 intravenous, intraarterial, and intratumoral injections on the rabbit VX2 tumor system

PURPOSE

To evaluate the differences in the radiosensitizing effects of intravenous (i.v.) injection, intraarterial (i.a.) injection, and intratumoral (i.t.) injections of the hypoxic cell radiosensitizer RK28 ([1-(4'-hydroxy-2'-butenoxy)methyl-2-nitroimidazole], a 2-nitroimidazole with an acyclic sugar analogue substituted at the N-1 position of the imidazole ring) using an animal experimental system.

METHODS AND MATERIALS

Rabbit VX2 tumors, which were implanted in the muscle of left hind legs and grown to 3 cm in diameter, were treated with RK28 (80 mg/kg x b.wt.) before 15 Gy of local x-ray irradiation. The auricular vein and the left saphenous artery were used for systemic injection and regional injection, respectively. For i.t. injection, a 21-gauge needle with three lateral holes was positioned in the central area of the tumor. Tumor regression was precisely evaluated by computed tomograpy (CT), and survival time was also studied. Using high-performance liquid chromatography (HPLC), pharmacokinetic studies for RK28 and its seven major metabolites were performed in tumor and serum at 0, 10, 20, 30, and 60 min after drug injection was completed.

RESULTS

Radiosensitizing effects of RK28 were considered present after i.a. injection (p < 0.05) and i.t. injection (p < 0.05) after analyzing tumor volumes on day 21 after treatment. Increased survival was not observed in any group with RK28 injection compared with survival in the group treated by x-ray irradiation alone. Pharmacokinetic studies showed the average concentration of RK28 in the tumor during x-ray irradiation was 1.3 times higher after i.a. injection and 3.5 times higher after i.t. injection than that after i.v. injection. The time modifying factor50 (TMF50: ratio of time for tumor to decrease by 50%, radiation alone vs. radiation plus drug) was calculated to be 1.5 after i.v. injection, 1.7 after i.a. injection, and 2.3 after i.t. injection. The values of TMF50 correlated to the average concentrations of RK28 in the tumor. As to metabolites of RK28, beta-glucuronated compound and cysteine conjugate were highly detected. The concentrations of cysteine conjugate were higher in the tumor than in serum via i.v. injection.

CONCLUSIONS

Radiosensitizing effects of RK28 were observed on the rabbit VX-2 tumor system after i.a. or i.t. injection. Pharmacokinetic studies proved that radiosensitizing effects depended on the concentration in the tumor, though the administration routes were different. Combined forms with nonprotein thiols were detected. However, survival benefits were not obtained by RK28. For clinical applications of RK28, i.a. or i.t. injection could facilitate better local control of cancer.

Modulating the radiation response

This review describes some current and future approaches designed to modulate the response of tumors and normal tissues to cell killing by ionizing radiation. The emerging knowledge of tumor, cellular and molecular biology is providing a better understanding of the clinical results with the hypoxic cell sensitizers and novel approaches to radiation sensitization and protection.