Agreement Between Magnetic Resonance Imaging and Pathologic Findings in the Tumor Size Evaluation Before and After Neoadjuvant Chemotherapy Treatment: A Prospective Study.

OBJECTIVES

To compare the agreement between magnetic resonance imaging (MRI) results and postsurgical pathologic findings for tumor size evaluation in cervical cancer patients before and after neoadjuvant chemotherapy (NACT) treatment.

METHODS

The study analyzed the agreement between pretreatment MRI results and postsurgical pathologic findings about the tumor size in 100 cervical cancer patients without NACT and 397 cervical cancer patients with NACT, respectively.

RESULTS

In general, the agreement between pretreatment MRI results and postsurgical pathologic findings of tumor size was 0.855 (95% confidence interval [CI], 0.763-0.909) in cervical cancer patients without NACT, whereas the agreement between posttreatment MRI results and postsurgical pathologic findings was 0.503 (95% CI, 0.421-0.576). Only 62.72% (249/397) of patients who underwent NACT treatment have the same chemotherapy response evaluation results; the κ coefficient was 0.384(95% CI, 0.310-0.457) between posttreatment MRI and postsurgical pathologic findings. We still found International Federation of Gynecology and Obstetrics stage is associated with the chemotherapy response evaluation.

CONCLUSIONS

Our data suggest that pretreatment MRI can be a surrogate indicator for postsurgical pathologic findings. However, posttreatment MRI could not be a surrogate indicator for postsurgical pathologic findings. The chemotherapy response evaluation based on only MRI is not so reliable. More indicators should be developed for chemotherapy response evaluation.