A noninvasive method for assessment of the extent of hypoxia in experimental and human tumors is highly needed. In this study, the potential usefulness of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) was investigated, using gadopentetate dimeglumine (Gd-DTPA) as contrast agent and A-07 human melanoma xenografts as tumor model. DCE-MRI was performed at a voxel size of 0.3 ؋ 0.6 ؋ 2.0 mm 3 with spoiled gradient-recalled sequences. Images of E ⅐ F (E is the initial extraction fraction of Gd-DTPA and F is perfusion) and (the partition coefficient of Gd-DTPA, which is proportional to extracellular volume fraction) were obtained by Kety analysis of DCE-MRI data. The study was based on the hypothesis that hypoxic tissue would have low E ⅐ F (i.e., poor oxygen supply) and/or low (i.e., high cell density and, hence, high oxygen consumption rate). Twenty-two tumors were first subjected to DCE-MRI and then to measurement of fraction of hypoxic cells, using a radiobiological assay. E ⅐ F was found to be strongly correlated to fraction of hypoxic cells (P < 0.000001), whereas significant correlation between and fraction of hypoxic cells could not be detected. It is thus possible that E ⅐

Purpose:To investigate whether gadopentetate dimeglumine (Gd-DTPA)-based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) may be a useful method for assessing fraction of radiobiologically hypoxic cells in tumors. Materials and Methods:A-07 and R-18 human melanoma xenografts were used as preclinical tumor models. DCE-MRI was performed at a voxel size of 0.23 ϫ 0.47 ϫ 2.0 mm 3 . Tumor images of E ⅐ F (E is the initial extraction fraction of Gd-DTPA and F is blood perfusion) and (the partition coefficient of Gd-DTPA) were produced by subjecting DCE-MRI series to Kety analysis. Microvascular density and extracellular volume fraction (ECVF) were determined by analysis of histological preparations. The fraction of radiobiologically hypoxic cells was measured by the paired survival curve method.Results: E ⅐ F correlated with microvascular density, and correlated with ECVF. The fraction of hypoxic cells was ϳ6.5-fold higher in R-18 tumors than in A-07 tumors, consistent with the observation that A-07 tumors showed higher values for E ⅐ F and microvascular density and lower cell density (i.e., higher values for and ECVF) than R-18 tumors.Conclusion: E ⅐ F and images obtained by Kety analysis of DCE-MRI series contain information that may be utilized to estimate the extent of radiobiological hypoxia in tumors.

This study aimed at developing immunohistochemical assays for different subpopulations of hypoxic cells in tumors. BALB/c-nu/nu mice bearing A-07 or R-18 tumors were given a single dose of 90 mg/kg body weight or three doses (3 h apart) of 30 mg/kg body weight of pimonidazole hydrochloride intravenously. The fraction of pimonidazole-labeled cells was assessed in paraffin-embedded and frozen tumor sections and compared with the fraction of radiobiologically hypoxic cells. The staining pattern in paraffin-embedded sections indicated selective staining of chronically hypoxic cells. Frozen sections showed a staining pattern consistent with staining of both chronically and acutely/repetitively hypoxic cells. Fraction of pimonidazole-labeled cells in paraffin-embedded sections was lower than the fraction of radiobiologically hypoxic cells (single-dose and triple-dose experiment). In frozen sections, fraction of pimonidazole-labeled cells was similar to (single-dose experiment) or higher than (triple-dose experiment) fraction of radiobiologically hypoxic cells. Three different subpopulations of hypoxic cells could be quantified by pimonidazole immunohistochemistry: the fraction of cells that are hypoxic because of limitations in oxygen diffusion, the fraction of cells that are hypoxic simultaneously because of fluctuations in blood perfusion, and the fraction of cells that are exposed to one or more periods of hypoxia during their lifetime because of fluctuations in blood perfusion.