HIF-1A, pimonidazole, and iododeoxyuridine to estimate hypoxia and perfusion in human head-and-neck tumors.

PURPOSE: Tumor hypoxia measured by microelectrodes has been shown to indicate poor patient outcome. Here we investigated four potentially more widely applicable immunohistochemical parameters of tumor oxygenation and perfusion in human head-and-neck tumors. METHODS: Twenty patients with squamous cell carcinomas of the head and neck treated with primary surgery were injected with pimonidazole and IdUrd the evening before operation. Consecutive paraffin-embedded sections were stained for blood vessels, pimonidazole, IdUrd, and HIF-1alpha. IdUrd labeling and Ki-67 labeling around individual blood vessels were scored. The spatial relationship between HIF-1alpha and pimonidazole was studied, as well as the distribution of both markers as a function of distance from the nearest blood vessel. RESULTS: Measurement of all four parameters (diffusion-limited fraction, pimonidazole fraction, HIF-1alpha fraction, IdUrd-negative vessels) was feasible, and a significant difference between tumors was found for all parameters. IdUrd-labeled cells were absent around some vessels, indicating lack of perfusion, because these regions were positive for Ki-67. There was a positive correlation between diffusion-limited fraction and pimonidazole area for all images from all tumors, although no correlation for mean values per tumor. Colocalization of pimonidazole and HIF-1alpha was low (0.02%-25%). Most expression profiles showed a more homogenous distribution for HIF-1alpha than pimonidazole. There was no significant correlation between the pimonidazole and HIF-1alpha fractions in the 10 tumors studied. CONCLUSIONS: Simultaneous immunohistochemical measurements related to hypoxia and perfusion are feasible (and easily applicable) in resected human tumors. The different geographic distributions of HIF-1alpha and pimonidazole indicate that HIF-1alpha might not be suitable as a marker for chronic hypoxia. Each parameter will be correlated with outcome in a larger ongoing study on head-and-neck tumors treated with surgery with or without postoperative radiotherapy.

Differentiation-associated staining with anti-pimonidazole antibodies in head and neck tumors.

BACKGROUND AND PURPOSE: Hypoxia is a strong negative prognostic factor for all three major treatment modalities for cancer. The bioreductive drug pimonidazole is currently under clinical investigation as a hypoxia marker. In human head and neck tumors, in addition to staining patterns typical of chronic hypoxia, staining was seen specifically around areas of keratinization, raising the question of whether this is hypoxia-related. This could influence quantitative hypoxia estimates using this marker. We investigated here whether the differentiation-related staining was caused by locally high reductive enzyme levels. PATIENTS AND METHODS: The nitrotetrazolium compound NBT was used, which is reduced by nitroreductases to yield a blue color. The assay was validated on three genetically related MDA231 human mammary carcinoma cell lines: wildtype, overexpressing DT-diaphorase (DT1), and overexpressing cytochrome p450 reductase (R4). Increased NBT staining under normoxia was indeed seen for both R4 and DT1 lines. Pimonidazole staining under normoxia was only seen in the R4 line. RESULTS: Frozen tumor sections from 20 patients with head and neck cancer injected with pimonidazole were incubated with NBT. Parallel sections were stained for pimonidazole. Staining patterns were then compared on matched images, and areas of keratinization scored for the presence or absence of pimonidazole and NBT. Pimonidazole staining was seen in 56% of keratinized areas, and of these, 78% showed increased NBT staining, indicating that high reductase levels are not a necessary requirement for differentiation-associated pimonidazole staining. In a second series, frozen sections of tumors from 15 patients not receiving pimonidazole were incubated with NBT and compared with staining after incubation with pimonidazole under both oxic and hypoxic conditions. Pimonidazole staining of some keratinizing areas under oxic conditions was seen. Of these areas, only a proportion (70%) showed increased NBT staining, confirming the lack of correspondence between keratin-associated pimonidazole staining and reductase levels. CONCLUSION: Hypoxia-independent pimonidazole staining can occur in more differentiated head and neck tumors, necessitating caution in hypoxia quantification. These data argue against a causative role for locally high reductase levels in differentiation-associated staining. DT-diaphorase appears to play no role in pimonidazole reduction.

Phase 1 study to identify tumour hypoxia in patients with head and neck cancer using technetium-99m BRU 59-21.

The aim of this study was to assess the safety and biodistribution of technetium-99m BRU 59-21, a novel radioactively labelled 2-nitro-imidazole hypoxic marker, in head and neck cancer patients and to correlate uptake with pimonidazole staining. (99m)Tc-BRU 59-21 was administered intravenously (mean dose 824 MBq, range 780-857 MBq) to ten head and neck cancer patients scheduled for primary surgery, and whole-body images and SPET scans were then obtained. Uptake of radioactivity in the regions of interest was determined and tumour to normal tissue ratios were calculated after correlative evaluation with MRI/CT. Twelve to 16 h before surgery (up to 2 weeks after the scan), patients received pimonidazole intravenously. Tumour sections were stained immunohistochemically for pimonidazole binding. No serious adverse events were reported. In five patients there were ten adverse events, which were mild in intensity and resolved completely without intervention. Uptake of (99m)Tc-BRU 59-21 was observed in eight of the ten primary tumours. Tumour to normal tissue ratios on the SPET scans for primary tumour and lymph nodes increased from 1.8 (range 0.9-2.7) to 2.1 (range 0.8-3.7) between 30 min and 3 h post injection. Tumour to normal tissue ratios in the primary tumour were significantly correlated with pimonidazole staining for SPET scans performed 30 min and 3 h post injection ( P=0.016 and P=0.037, respectively). When primary tumour and involved lymph nodes were considered in conjunction, correlation between the tumour to normal tissue ratio and pimonidazole staining was observed for early ( P<0.001) but not for late SPET scans ( P=0.076). However, late scans showed better tumour delineation than early scans. Administration of (99m)Tc-BRU 59-21 in head and neck cancer patients appears to be safe and feasible. Uptake and retention in tumour tissue was observed, suggestive of tumour hypoxia, and this was supported by correlations with staining for the hypoxic marker pimonidazole.