Evaluation of the Relationship between MR Estimates of Blood Oxygen Saturation and Hypoxia: Effect of an Antiangiogenic Treatment on a Gliosarcoma Model

Abstract

To assess the reproducibility of the magnetic resonance (MR) estimate of blood oxygen saturation (sO(2)) in the rat brain, to evaluate the relationship between low MR estimate of sO(2) values and tissue hypoxia in a hypoxic and necrotic glioscarcoma model (9L gliosarcoma cells), and to evaluate the capability of the MR estimate of sO(2) parameter to help identify modifications induced by an antiangiogenic treatment (sorafenib) in 9L gliosarcoma tumors. Experiments were performed with permits from the French Ministry of Agriculture. Forty-eight male rats bearing a 9L gliosarcoma were randomized in untreated and treated (sorafenib) groups. MR blood volume fraction and MR estimate of sO(2) parameters were estimated 1 day before and 1, 3, 5, and 8 days after the start of the treatment. The in vivo MR estimate of sO(2) measurement was correlated with the ex vivo hypoxia assessment by using pimonidazole staining. Paired and unpaired t tests, as well as parametric Pearson tests, were used for the statistical analyses. In healthy tissues, MR estimate of sO(2) measurements were comparable to literature values and were reproducible (mean across all animals, 68.0% ± 6.5 [standard deviation]). In untreated tumors, MR estimate of sO(2) and immunohistochemical analysis yielded correlated fractional hypoxic-necrotic areas (R(2) = 0.81). In tumors treated with antiangiogenic therapy, tumor MR estimate of sO(2) was decreased with respect to the healthy tissue (P< .001). Results of this study suggest that the MR estimate of sO(2) is a reproducible estimate that could be used as an in vivo probe of hypoxia in brain tumors and as a sensitive reporter of the hypoxic effects of antiangiogenic therapies.