CARCINOMA OF THE CERVIX - PREDICTIVE VALUE OF CLINICAL AND MAGNETIC-RESONANCE (MR) IMAGING ASSESSMENT OF PROGNOSTIC FACTORS

Purpose: This retrospective study assesses the predictive value of mag netic resonance imaging (MRI) to identify high risk cervical cancer pa tients. Methods and Materials: The MRI evaluation of morphologic risk factors in patients with invasive cervical carcinoma treated with defi nitive radiation therapy were correlated with clinical factors and wit h complete tumor regression (CTR) at 6 months, tumor local control (TL C), and patient outcome at 12 months after irradiation. Sixty-six pati ents, median age 44.5 years, with bulky Stage I or greater disease wer e included in the study. Results: In univariate analysis, clinical Int ernational Federation of Gynecology and Obstetrics (FIGO) stage had si gnificant correlation with patient outcome, but it correlated poorly w ith complete tumor regression and tumor local control. In contrast, MR I stage showed significant correlation with complete tumor regression, tumor local control, and disease-free survival at 12 months. When eac h stage was analyzed separately, the greatest difference was demonstra ted between clinical and MRI assignment of stage Ib disease. MRI Stage Ib disease significantly correlated with all three categories analyze d, while clinical Stage Ib did not. Superiority of MRI assessment of l ow stage disease was also evident in the detection of lymph node metas tasis. Significant risk for nodal metastasis was related to tumor size greater than 4 cm, invasion of the parametria and urinary bladder, an d stage of the disease. Conclusion: The multivariate analysis demonstr ated that the most related variables in order of significance were the presence of juxtaregional and paraaortic lymph nodes, patient age, tu mor size, and MRI tumor stage. This study demonstrates the value of MR imaging as an adjunct to clinical assessment of bulky invasive cervic al cancer, rendering more complete assessment of morphologic risk fact ors important in patient prognosis.