Consecutive magnetic resonance imaging during brachytherapy for cervical carcinoma: predictive value of volume measurements with respect to persistent disease and prognosis

BACKGROUND Cervical cancer is associated with a high yearly mortality. The presence of persistent disease after radiotherapy is a significant predictor of patient survival. The aim of our study was to assess if tumor volume regression measured with MR imaging at the time of brachytherapy can discriminate between patients who eventually will achieve a complete response to radiotherapy from those who will not. The second objective was to evaluate whether tumor volume regression predicts overall treatment failure. METHODS MRI was evaluated quantitatively in 35 patients; by means of tumor volumetry on T2-weighted MR images before treatment, at the first BCT application, and at the final BCT. The MR images were independently analyzed by two investigators. As a reference standard histopathologic confirmation of residual tumor and/or clinical exam during follow-up > 1 year were used. Area under the curve were compared, P-values <0.05 were considered significant. RESULTS There was a good correlation between volume measurements made by the two observers. A residual tumor volume >9.4 cm(3) at final BCT and tumor volume regression < 77 % of the pre-treatment volume were significantly associated with local residual tumor after completion of therapy (p < 0.02) (AUC, 0.98-1.00). A volume >2.8 cm(3) at final BCT was associated with overall treatment failure (p < 0.03). CONCLUSION Our study shows that volume analysis during BCT is a predictive tool for local tumor response and overall treatment outcome. The potential of local response assessment to identify patients at high risk of overall treatment failure is promising.