Key words: cervical cancer, MRI, parametrial invasion, treatment response

"Magnetic resonance imaging of cervical cancer: prediction of parametrial invasion and treatment response"

The choice of therapy (surgery or radiation) in early-stage cervical cancer is based on the stage of the disease. In this thesis we show that a new MRI technique, diffusion-weighted MRI, is able to determine more precisely whether there is or isn’t invasion to surrounding tissue. This results in more patients receiving the correct treatment.

Late-stage cervical cancer is usually treated with radiation and chemotherapy. In a minority (6-7%) of patients, there is an incomplete reaction to the radiation. In selected cases, this group can still be cured through surgery. In this thesis, imaging techniques were used to demonstrate local persistence of cancer. The decrease in volume of the cancer during radiation, as shown on MRIs, was able to predict an incomplete reaction to the radiation. After the radiation, it appeared that predetermined MRI criteria and PET data increased the precision of detecting local persistence of disease. Hopefully these options can be used in the future to select patients who are eligible for additional treatment.