The outcome of radiation therapy of patients with locally advanced squamous cell carcinoma of the uterine cervix is determined primarily by characteristic features of the tumor microenvironment. For these patients, there is a dual problem of overtreatment and high mortality rates of 30-50%. On the one hand, the addition of chemotherapy to radiation therapy increases the survival rates with only 10% while inflicting all patients with added severe side effects. On the other hand, a large group of patients would probably benefit from additional or alternative treatments. Therefore, novel biomarkers predicting the outcome of radiation therapy would be highly beneficial for personalized treatment of locally advanced cervical cancer. Magnetic Resonance Imaging (MRI) is an established method for anatomical characterization of cervical carcinomas and may be developed to be an attractive method for characterization of the tumor microenvironment and, hence, for prognostic evaluation.
In this study, several cervical carcinoma xenograft lines were investigated by dynamic contrast enhanced MRI and diffusion weighted MRI. The radioresponsiveness of the tumors and their propensity to metastasize to lymph nodes were also determined. Moreover, characteristics of the microenvironment of the tumors were assessed by probe measurements and immunohistochemistry. Statistically significant correlations were found between MRI-derived parametric images (Ktrans, the volume transfer constant of the MR contrast agent and ADC, the apparent diffusion coefficient), response to radiation treatment, metastatic propensity, and characteristic features of the tumor microenvironment, including tumor oxygenation, fraction of hypoxic tissue, lactate concentration, glucose concentration, interstitial fluid pressure, tumor cell density, and the density of the tumor stroma.