ABSTRACT
Objective To evaluate the predictive value of diffusion weighted imaging ( DWI) for progres-sion free survival(PFS)in patients with cervix cancer after neoadjuvant chemotherapy (NACT).Methods Thirty two consecutive patients with pathologically confirmed cervical cancer underwent MRI including DWI before and after NACT.Pathologic results and MR images were reviewed .Univariate analysis was applied by Kaplan -Meier method.The Cox proportional hazard model was used to evaluate prognostic variables for multivariate analysis .Re-ceiver operating characteristic ( ROC) curves were used to find optimal cutoff values of independent prognostic fac -tors for disease progression.Results Of 32 patients,10 had disease progression during follow -up period.In uni-variate analysis,FIGO stage,tumor size,the depth of tumor invasion,lymph node metastasis and ADC changes be-tween before and after neoadjuvant chemotherapy (ΔADC) were significantly associated with the disease progres-sion.In multivariate analysis , FIGO stage and ΔADC were the independent prognostic factors for PFS .Optimal cutoff values that determined by ROC curves analyses were FIGO stage ⅡandΔADC 0.31.Area under the ROC curve(AUC)of FIGO stage and ΔADC were 0.841(sensitivity 90.0%,specificity 68.2%)and 0.864(sensitivity 80.0%,specificity 81.8%),respectively.Conclusion In patients with cervix cancer after NACT ,FIGO stage andΔADC are significant predictive values for PFS .