Mechanisms of blood flow and hypoxia production in rat 9L-epigastric tumors.

Classical descriptions of tumor physiology suggest two origins for tumor hypoxia; steady-state (diffusion-limited) hypoxia and cycling (perfusion-modulated) hypoxia. Both origins, primarily studied and characterized in murine models, predict relatively small, isolated foci or thin shells of hypoxic tissue interspersed with contrasting oxic tissue. These foci or shells would not be expected to scale with overall tumor size since the oxygen diffusion distance (determined by oxygen permeability and tissue oxygen consumption rate) is not known to vary dramatically from tumor to tumor. We have identified much larger (macroscopic) regions of hypoxia in rat gliosarcoma tumors and in larger human tumors (notably sarcomas and high-grade glial tumors), as indicated by biochemical binding of the hypoxia marker, EF5. Thus, we considered an alternative cause of tumor hypoxia related to a phenomenon first observed in window-chamber tumor models: namely longitudinal arteriole gradients. Although longitudinal arteriole gradients, as originally described, are also microscopic in nature, it is possible for them to scale with tumor size if tumor blood flow is organized in an appropriate manner. In this organization, inflowing blood would arise from relatively well-oxygenated sources and would branch and then coalesce to poorly-oxygenated outflowing blood over distances much larger than the length of conventional arterioles (multi-millimeter scale). This novel concept differs from the common characterization of tumor blood flow as disorganized and/or chaotic. The organization of blood flow to produce extended longitudinal gradients and macroscopic regional hypoxia has many important implications for the imaging, therapy and biological properties of tumors. Herein, we report the first experimental evidence for such blood flow, using rat 9L gliosarcoma tumors grown on the epigastric artery/vein pair.

In vivo profiling of hypoxic gene expression in gliomas using the hypoxia marker EF5 and laser-capture microdissection.

Hypoxia is a key determinant of tumor aggressiveness, yet little is known regarding hypoxic global gene regulation in vivo. We used the hypoxia marker EF5 coupled with laser-capture microdissection to isolate RNA from viable hypoxic and normoxic regions of 9L experimental gliomas. Through microarray analysis, we identified several mRNAs (including the HIF targets Vegf, Glut-1, and Hsp27) with increased levels under hypoxia compared with normoxia both in vitro and in vivo. However, we also found striking differences between the global in vitro and in vivo hypoxic mRNA profiles. Intriguingly, the mRNA levels of a substantial number of immunomodulatory and DNA repair proteins including CXCL9, CD3D, and RAD51 were found to be downregulated in hypoxic areas in vivo, consistent with a protumorigenic role of hypoxia in solid tumors. Immunohistochemical staining verified increased HSP27 and decreased RAD51 protein levels in hypoxic versus normoxic tumor regions. Moreover, CD8(+) T cells, which are recruited to tumors upon stimulation by CXCL9 and CXCL10, were largely excluded from viable hypoxic areas in vivo. This is the first study to analyze the influence of hypoxia on mRNA levels in vivo and can be readily adapted to obtain a comprehensive picture of hypoxic regulation of gene expression and its influence on biological functions in solid tumors.

The Relationship among Hypoxia, Proliferation, and Outcome in Patients with De Novo Glioblastoma: A Pilot Study.

The hypoxia and proliferation index increase with grade in human glial tumors, but there is no agreement whether either has prognostic importance in glioblastomas. We evaluated these end points individually and together in 16 de novo human glioblastomas using antibodies against the 2-nitroimidazole hypoxia detection agent EF5 and the proliferation detection agent Ki-67. Frozen tumor tissue sections were fluorescence-stained for nuclei (Hoechst 33342), hypoxia (anti-EF5 antibodies), and proliferation (anti-Ki-67 antibodies). EF5 binding adjacent to Ki-67+ cells, overall EF5 binding, the ratio of these values, and the proliferation index were evaluated. Patients were classified using recursive partitioning analysis and followed up until recurrence and/or death. Recursive partitioning analysis was statistically significant for survival (P = .0026). Overall EF5 binding, EF5 binding near Ki-67+ cells, and proliferation index did not predict recurrence. Two additional survival analyses based on ratios of the overall EF5 binding to EF5 binding near Ki-67+ cells were performed. High and low ratio values were determined by two cutoff points: (a) the 50% value for the ratio [EF5/Ki-67(Binding)]/[Tumor(binding)] = Ratio(EF5 50%) and (b) the median EF5 value (75.6%) of the ratio [EF5/Ki-67(Binding)]/[Tumor(binding)] = Ratio(patients median). On the basis of the Ratio(EF5 50%), recurrence (P = .0074) and survival (P = .0196) could be predicted. Using the Ratio(patients median), only survival could be predicted (P = .0291). In summary, patients had a worse prognosis if the [EF5/Ki-67(Binding)]/[Tumor(binding)] ratio was high. A hypothesis for the mechanisms and translational significance of these findings is discussed.

Imaging and analytical methods as applied to the evaluation of vasculature and hypoxia in human brain tumors.

Tissue hypoxia results from the interaction of cellular respiration, vascular oxygen carrying capacity, and vessel distribution. We studied the relationship between tumor vasculature and regions of low pO(2) using quantitative analysis of binding of the 2-nitroimidazole EF5 given to patients intravenously (21 mg/kg) approximately 24 h preceding surgery. We describe new computer algorithms for determining EF5 binding as a function of radial distance from individual blood vessels and converting this value to tissue pO(2). Tissues from six human brain tumors were assessed. In a hemangiopericytoma, a WHO Grade 2 and WHO Grade 3 glial brain tumor, all tissue pO(2) values calculated by EF5 binding were >20 mmHg (described as "physiologically oxygenated"). In these three tumors, EF5 binding gradients (measured as a function of distance from each observed vessel) were low, with small positive and negative values averaging close to zero. Much lower tissue oxygen levels were found, including near some vessels, in glioblastomas. Gradients of EF5 binding away from vessels were larger in glioblastomas than in the low-grade tumors, but positive and negative values again averaged to near zero. Based on these preliminary data, we hypothesize a new paradigm for tumor blood flow in human brain tumors whereby in-flowing and out-flowing blood patterns may have contrasting effects on average tissue EF5 (and by inference, oxygen) gradients. Our studies also imply that neither distance to the nearest blood vessel nor distance from each observed blood vessel provide reliable estimates of tissue pO(2).