BACKGROUND: Imaging findings of residual cervical tumor after chemoradiotherapy can closely resemble those of post-irradiation inflammation. PURPOSE: To determine the diagnostic performance of magnetic resonance imaging (MRI) in evaluating residual disease after chemoradiotherapy in patients with locally advanced cervical carcinoma (LACC). MATERIAL AND METHODS: Retrospective analysis of prospectively collected data from 41 patients with histopathologically proven LACC (International Federation of Gynecology and Obstetrics stage ≥IB2) who underwent MRI before and after chemoradiotherapy. At each examination, a qualitative and semi-quantitative analysis of primary tumor, including tumor volume and signal intensity were assessed on T2-weighted (T2W) images. All patients had surgery after post-chemoradiotherapy MRI. MRI and histopathologic results were compared. RESULTS: All patients showed significant difference in tumor volume and signal intensity between pre- and post-chemoradiotherapy MRI (P < 0.0001). According to pathology, 27/41 (66%) patients had true negative and 2/41 (5%) had true positive post-chemoradiotherapy MRI. Eleven out of 41 (27%) patients showed inflammation with false positive post-chemoradiotherapy MRI and 1/41 (2%) had a false negative post-chemoradiotherapy MRI. Sensitivity, specificity, accuracy, positive predictive values, and negative predictive values of post-chemoradiotherapy MRI in predicting residual disease were 69%, 71%, 71%, 15%, and 96%, respectively. CONCLUSION: The differentiation of residual tumor from post-irradiation inflammation with early post- chemoradiotherapy MRI (within 28-60 days) is difficult with a high risk of false positive results. Combination of qualitative and semi-quantitative analysis does not improve the accuracy. Conversely, post-chemoradiotherapy MRI has a high negative predictive value with a low risk of false negative results. The role of conventional MRI combined with functional techniques should be evaluated.
Cervical cancer response to neoadjuvant chemoradiotherapy: MRI assessment compared with surgery.
FANFANI, Francesco;
2016-01-01
Abstract
BACKGROUND: Imaging findings of residual cervical tumor after chemoradiotherapy can closely resemble those of post-irradiation inflammation. PURPOSE: To determine the diagnostic performance of magnetic resonance imaging (MRI) in evaluating residual disease after chemoradiotherapy in patients with locally advanced cervical carcinoma (LACC). MATERIAL AND METHODS: Retrospective analysis of prospectively collected data from 41 patients with histopathologically proven LACC (International Federation of Gynecology and Obstetrics stage ≥IB2) who underwent MRI before and after chemoradiotherapy. At each examination, a qualitative and semi-quantitative analysis of primary tumor, including tumor volume and signal intensity were assessed on T2-weighted (T2W) images. All patients had surgery after post-chemoradiotherapy MRI. MRI and histopathologic results were compared. RESULTS: All patients showed significant difference in tumor volume and signal intensity between pre- and post-chemoradiotherapy MRI (P < 0.0001). According to pathology, 27/41 (66%) patients had true negative and 2/41 (5%) had true positive post-chemoradiotherapy MRI. Eleven out of 41 (27%) patients showed inflammation with false positive post-chemoradiotherapy MRI and 1/41 (2%) had a false negative post-chemoradiotherapy MRI. Sensitivity, specificity, accuracy, positive predictive values, and negative predictive values of post-chemoradiotherapy MRI in predicting residual disease were 69%, 71%, 71%, 15%, and 96%, respectively. CONCLUSION: The differentiation of residual tumor from post-irradiation inflammation with early post- chemoradiotherapy MRI (within 28-60 days) is difficult with a high risk of false positive results. Combination of qualitative and semi-quantitative analysis does not improve the accuracy. Conversely, post-chemoradiotherapy MRI has a high negative predictive value with a low risk of false negative results. The role of conventional MRI combined with functional techniques should be evaluated.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
