Regional cervical tumor microcirculatory response: Are there systematic differences in contrast uptake patterns with respect to the uterus?

Purpose/Objective: Dynamic Contrast Enhanced (DCE) Magnetic Resonance Imaging (MRI) has been reported to be valuable in the evaluation of perfusion patterns throughout the entire tumor bed. Tumor cell heterogeneity over the entire structure has been well described in the literature. However, regional heterogeneity of DCE-MRI within the tumor has not been previously investigated until this point. The purpose of this study was to assess the topographical pattern of DCE-MRI among different subdivisions within tumors evaluated by DCE-MRI in cervical cancer patients. Materials/Methods: The DCE-MRI data from subdivisions of cervical cancer patients were evaluated prior to radiation therapy (n = 18). Each tumor was sectioned into eight three-dimensional subdivision wedge "regions" arrayed around the tumor center-of-volume. The orientation of the wedges was based upon uterine direction as determined by the projection of the endometrial canal and spatial orientation of the uterus. Tumor perfusion was estimated by the Relative Signal Intensity (RSI), which was defined by the ratio of the plateau post-injection signal intensity (SI) (at 75 seconds post-injection) divided by the pre-contrast SI on a pixel-by-pixel basis. Perfusion in each of eight regions of the cervical cancers was estimated by the mean RSI within each region. Statistical differences among the various regions were computed from RSI distributions within each three-dimensional wedge evaluated in cervical cancer patients pre-radiotherapy. Results: Significant differences over the eight regional subdivisions were found via a repeated-measures analysis of variance (ANOVA) design that produced an F-value = 18.37 with Degrees of Freedom - DOF(7,119) such that p< 0.0001. In addition, a planned comparison between tumor segments closest to the uterine corpus as compared to segments closest to the cervix demonstrated significant differences t-value = 3.19, p< 0.002.