Iodine-124-labeled iodo-azomycin-galactoside imaging of tumor hypoxia in mice with serial microPET scanning.

Tumor hypoxia, present in many human cancers, can lead to resistance to radiation and chemotherapy, is associated with a more aggressive tumor phenotype and is an independent prognostic factor of clinical outcome. It is therefore important to identify and localize tumor hypoxia in cancer patients. In the current study, serial microPET imaging was used to evaluate iodine-124-labeled iodo-azomycin-galactoside ((124)I-IAZG) (4.2-day physical half-life) as a hypoxia imaging agent in 17 MCa breast tumors and six FSaII fibrosarcomas implanted in mice. For comparison, another promising hypoxic-cell PET radiotracer, fluorine-18-labeled fluoro-misonidazole ((18)F-FMISO), was also imaged in the same tumor-bearing animals. Twelve animals were also imaged with (18)F-labeled fluoro-deoxyglucose ((18)F-FDG). In addition, histological examination was performed, and direct measurement of tumor oxygenation status carried out with the Oxylite probe system. Two size groups were used, relatively well-oxygenated tumors in the range of 80-180 mg were designated as small, and those >300 mg and highly hypoxic, as large. Based on the data from 11 MCa and six FSaII tumors, both (124)I-IAZG and (18)F-FMISO images showed high tracer uptake in the large tumors. In (18)F-FMISO images at 1, 3-4, and 6-8 h post-injection (p.i.), there was considerable whole-body background activity. In contrast, (124)I-IAZG imaging was optimal when performed at 24-48 h p.i., when the whole-body background had dissipated considerably. As a result, the (124)I-IAZG images at 24-48 h p.i. had higher tumor to whole-body activity contrast than the (18)F-FMISO images at 3-6 h p.i. Region-of-interest analysis was performed as a function of time p.i. and indicated a tumor uptake of 5-10% (of total-body activity) for FMISO at 3-6 h p.i., and of ~17% for IAZG at 48 h p.i. This was corroborated by biodistribution data in that the tumor-to-normal tissue (T/N, normal tissues of blood, heart, lung, liver, spleen, kidney, intestine, and muscle) activity ratios of IAZG at 24 h p.i. was 1.5-2 times higher than those of FMISO at 3 h p.i., with the exception of stomach. Statistical analysis indicated that these differences in T/N ratios were significant. The small tumors were visualized in the (18)F-FDG images, but not in the (124)I-IAZG or (18)F-FMISO images. This was perhaps due to the combined effect of a smaller tumor volume and a lower hypoxic fraction. Oxylite probe measurement indicated a lesser proportion of regions with pO(2)<2.5 mmHg in the small tumors (e.g., pO(2) was <2.5 mmHg in 28% and 67% of the data in small and large FSaII tumors, respectively), and the biodistribution data showed lower uptake of the tracers in the small tumors than in the large tumors. In the first study of its kind, using serial microPET imaging in conjunction with biodistribution analysis and direct probe measurements of local pO(2) to evaluate tumor hypoxia markers, we have provided data showing the potential of (124)I-IAZG for hypoxia imaging.