| Objective:To assess the ability of 18F-FDG PET/CT to evaluate tumor response after 125I brachytherapy,investigate the proper follow-up time interval,and explore the correlation between metabolic heterogeneity along with the parameters distracted from 18F-FDG PET/CT image and the progression of the lesion,then propose the corresponding suggestions.Methods:A total of 34 patients(73 lesions)were included in this study.The basic information of patients,the first postoperative 18F-FDG PET/CT image and postoperative multimodal image follow-up image were collected.The prognosis of every lesion was obtained from the multimodal follow-up image after the first postoperative 18F-FDG PET/CT scan.Chi square test and Mann-Whitney U test were used to compare the differences of clinical characteristics between progressive group and non-progressive group.Authenticity,predictability and synthetic index are calculated to analyze the diagnostic ability of tumor response assessment after 125I brachytherapy by 18F-FDG PET/CT.Detected a proper follow-up time interval based on the change-point theory.The distribution pattern of FDG metabolic concentration in lesions and semi-quantitative parameters such as tumor metabolic volume,total lesion glycolysis,standardized uptake value and standardized uptake value normalized by lean body mass were extracted from the image and analyzed by univariate and multivariate binary logistic regression,based on which one qualitative model and one quantitative model were constructed respectively.A 18F-FDG PET/CT diagnostic model containing two models above was constructed by binary logistic regression.Finally,a combine models were constructed of variables related to three models above and clinical information of the patients.Analysis and comparation of four models was made and the corresponding suggestions were proposed.Results:1.The sensitivity,specificity,positive predictive value,negative predictive value and accuracy of 18F-FDG PET/CT in evaluating the therapeutic effect of 125I brachytherapy were89.5%,72.0%,54.8%,94.7%and 76.8%respectively;the diagnostic accuracy of lesions distributed in muscle,solid organs and bones were 100%,86.7%and 81.8%respectively;for the lesions distributed in the lung,the diagnostic sensitivity of 18F-FDG PET/CT was 100%,and the positive predictive value and accuracy were 75%;the accuracy and specificity of 18F-FDG PET/CT for residual tumor activity in target lesions reached the peak at 200 days;2.Mild increased FDG uptake inside the focus(OR,6.000;95%CI:1.935,18.603;P=0.002<0.05)and FDG uptake at or below the peritumoral background(OR,0.179;95%CI: 0.056,0.574;P=0.004<0.05)were significantly correlated with the residual activity of the lesions in univariate analysis,and the former(OR,5.340;95%CI:1.210,23.577;P=0.027<0.05)was significantly correlated with the state of tumor progression in multivariate analysis with a cut-off of 0.357;the ratio of SUVmin from the lesion to the SUVavg from the surrounding background(OR,4.166;95%CI:1.155,15.032;P=0.029<0.05)and the ratio of SULmin from the lesion to the SUVavg from the surrounding background(OR,4.158;95%CI:1.154,14.978;P=0.029<0.05)were significantly correlated with tumor activity in univariate analysis;the former(OR,4.166;95%CI:1.155,15.032;P=0.029<0.05)was significantly correlated with postoperative lesion efficacy in multivariate analysis;3.The AUCs of FDG metabolic concentration distribution pattern model,multi 18F-FDG PET/CT semi-quantitative parameter model,18F-FDG PET/CT diagnostic model and clinical 18F-FDG PET/CT diagnostic model were 0.790,0.666,0.813 and 0.926 respectively,which were significantly correlated with the progression of postoperative lesions.Conclusion:the qualitative evaluation of FDG metabolic concentration heterogeneity distribution plays a relatively more significant role than the quantitative evaluation of metabolic parameters in the evaluation of the tumor response by 18F-FDG PET/CT after 125I brachytherapy;18F-FDG PET/CT scan are capable of monitoring the response status of lesions after 125I brachytherapy with a combination of qualitative evaluation,quantitative evaluation and clinical information such as the location of the lesions and the interval between the procedure. |