Dynamics of tumor hypoxia measured with bioreductive hypoxic cell markers.

Hypoxic cells are common in tumors and contribute to malignant progression, distant metastasis and resistance to radiotherapy. It is well known that tumors are heterogeneous with respect to the levels and duration of hypoxia. Several strategies, including high-oxygen-content gas breathing, radiosensitizers and hypoxic cytotoxins, have been developed to overcome hypoxia-mediated radioresistance. However, with these strategies, an increased tumor control rate is often accompanied by more severe side effects. Consequently, development of assays for prediction of tumor response and early monitoring of treatment responses could reduce both over- and undertreatment, thereby avoiding unnecessary side effects. The purpose of this review is to discuss different assays for measurement of hypoxia that can be used to detect changes in oxygen tension. The main focus is on exogenous bioreductive hypoxia markers (2-nitroimidazoles) such as pimonidazole, CCI-103F, EF5 and F-misonidazole. These are specifically reduced and bind to macromolecules in viable hypoxic cells. A number of these bioreductive drugs are approved for clinical use and can be detected with methods ranging from noninvasive PET imaging (low resolution) to microscopic imaging of tumor sections (high resolution). If the latter are stained for multiple markers, hypoxia can be analyzed in relation to different microenvironmental parameters such as vasculature, proliferation and endogenous hypoxia-related markers, for instance HIF1alpha and CA-IX. In addition, temporal and spatial changes in hypoxia can be analyzed by consecutive injection of two different hypoxia markers. Therefore, bioreductive exogenous hypoxia markers are promising as tools for development of predictive assays or as tools for early treatment monitoring and validation of potential endogenous hypoxia markers.

Dynamics of hypoxia, proliferation and apoptosis after irradiation in a murine tumor model.

Proliferation and hypoxia affect the efficacy of radiotherapy, but radiation by itself also affects the tumor microenvironment. The purpose of this study was to analyze temporal and spatial changes in hypoxia, proliferation and apoptosis after irradiation (20 Gy) in cells of a murine adenocarcinoma tumor line (C38). The hypoxia marker pimonidazole was injected 1 h before irradiation to label cells that were hypoxic at the time of irradiation. The second hypoxia marker, CCI-103F, and the proliferation marker BrdUrd were given at 4, 8 and 28 h after irradiation. Apoptosis was detected by means of activated caspase 3 staining. After immunohistochemical staining, the tumor sections were scanned and analyzed with a semiautomatic image analysis system. The hypoxic fraction decreased from 22% in unirradiated tumors to 8% at both 8 h and 28 h after treatment (P < 0.01). Radiation did not significantly affect the fraction of perfused vessels, which was 95% in unirradiated tumors and 90% after treatment. At 8 h after irradiation, minimum values for the BrdUrd labeling index (LI) and maximum levels of apoptosis were detected. At 28 h after treatment, the BrdUrd labeling and density of apoptotic cells had returned to pretreatment levels. At this time, the cell density had decreased to 55% of the initial value and a proportion of the cells that were hypoxic at the time of irradiation (pimonidazole-stained) were proliferating (BrdUrd-labeled). These data indicate an increase in tumor oxygenation after irradiation. In addition, a decreased tumor cell density without a significant change in tumor blood perfusion (Hoechst labeling) was observed. Therefore, it is likely that in this tumor model the decrease in tumor cell hypoxia was caused by reduced oxygen consumption.

Pimonidazole binding and tumor vascularity predict for treatment outcome in head and neck cancer.

Hypoxia is associated with tumor aggressiveness and is an important cause of resistance to radiation treatment. Assays of tumor hypoxia could provide selection tools for hypoxia-modifying treatments. This study correlated the exogenous 2-nitroimidazole hypoxia marker 1-[(2-hydroxy-3-piperidinyl)propyl]-2-nitroimidazole hydrochloride (pimonidazole) with the endogenous hypoxia-related marker carbonic anhydrase 9 (CA9) and with vascular parameters using immunohistochemical techniques and a computerized image analysis system. Tumor biopsies were obtained from patients with head and neck carcinomas that were potential candidates for a Phase II trial with accelerated radiotherapy combined with carbogen and nicotinamide (ARCON). If, after completion of the diagnostic workup, the eligibility criteria were met and informed consent was obtained, patients were treated with ARCON. Those patients that were not eligible or refused ARCON were treated with radiotherapy, surgery, or a combined modality. Forty-three biopsies were analyzed, and the results were related with treatment outcome. The distribution patterns of pimonidazole and CA9 were similar, although the CA9 signal was generally observed already at shorter distances from blood vessels. There was a weak but significant correlation between the relative tumor areas positive for pimonidazole binding and areas with CA9 expression. Locoregional tumor control was significantly lower for patients who had hypoxic tumors or tumors with low vascular density. The 2-year control rates were 48 versus 87% for tumors with high and low pimonidazole binding levels (stratified by median, P = 0.01) and 48 and 88% for tumors with low and high vascular density (stratified by median, P = 0.01). These associations disappeared in the subgroup of patients treated with ARCON. There was no relationship between the level of CA9 expression and treatment outcome. It is concluded that pimonidazole binding and vascular density can predict treatment outcome in head and neck cancer and may be useful as selection tools for hypoxia-modifying treatments. Pimonidazole and CA9 demonstrate concordant staining patterns, but the latter is a less specific marker for hypoxia.

Hypoxic cell turnover in different solid tumor lines.

PURPOSE: Most solid tumors contain hypoxic cells, and the amount of tumor hypoxia has been shown to have a negative impact on the outcome of radiotherapy. The efficacy of combined modality treatments depends both on the sequence and timing of the treatments. Hypoxic cell turnover in tumors may be important for optimal scheduling of combined modality treatments, especially when hypoxic cell targeting is involved. METHODS AND MATERIALS: Previously we have shown that a double bioreductive hypoxic marker assay could be used to detect changes of tumor hypoxia in relation to the tumor vasculature after carbogen and hydralazine treatments. This assay was used in the current study to establish the turnover rate of hypoxic cells in three different tumor models. The first hypoxic marker, pimonidazole, was administered at variable times before tumor harvest, and the second hypoxic marker, CCI-103F, was injected at a fixed time before harvest. Hypoxic cell turnover was defined as loss of pimonidazole (first marker) relative to CCI-103F (second marker). RESULTS: The half-life of hypoxic cell turnover was 17 h in the murine C38 colon carcinoma line, 23 h and 49 h in the human xenograft lines MEC82 and SCCNij3, respectively. Within 24 h, loss of pimonidazole-stained areas in C38 and MEC82 occurred concurrent with the appearance of pimonidazole positive cell debris in necrotic regions. In C38 and MEC82, most of the hypoxic cells had disappeared after 48 h, whereas in SCCNij3, viable cells that had been labeled with pimonidazole were still observed after 5 days. CONCLUSIONS: The present study demonstrates that the double hypoxia marker assay can be used to study changes in both the proportion of hypoxic tumor cells and their lifespan at the same time. The present study shows that large differences in hypoxic cell turnover rates may exist among tumor lines, with half-lives ranging from 17-49 h.

Vascular architecture, hypoxia, and proliferation in first-generation xenografts of human head-and-neck squamous cell carcinomas.

PURPOSE: To quantify the physiologic status of human tumor cells in relation to the tumor vasculature. METHODS AND MATERIALS: Fourteen tumors of 11 first-generation xenograft lines of human head-and-neck squamous cell carcinoma were injected with the hypoxic cell marker pimonidazole, the proliferation marker BrdUrd, and the perfusion marker Hoechst 33342. Consecutive tissue sections were processed with immunohistochemical methods and analyzed with image-analysis techniques. RESULTS: Three different hypoxic patterns were found: patchy, ribbon-like, and mixed. An image-analysis method was developed to quantify these, and an elongation index (length/width) was calculated for hypoxia. The mean elongation indices ranged from 2.0 to 28.3 and showed a good correlation with the visual scoring of hypoxic patterns. Comparative analysis of hypoxic and proliferating cells in zones around the tumor vasculature showed the presence of both hypoxic and proliferating cells in all zones up to 250 microm from the vessels. The largest coexistence of hypoxic and proliferating cells seemed to occur at 50-100 microm from the vessels. CONCLUSIONS: The three hypoxic patterns could be quantified by an elongation index, which is an additional parameter that allows distinction of tumors with similar fractions of hypoxic cells. The analysis of hypoxic and proliferating cells as a function of distance from the tumor vasculature indicates that proliferation does occur also at low oxygen tensions.