| Background:Rectal mucinous adenocarcinoma(RMAC)is a histological subtype of rectal cancer.It is characterized by the secretion of extracellular mucus and mucus lake components account for more than 50%of the tumor volume.RMAC recommends neoadjuvant chemoradiotherapy,and then surgical resection.However,compared with conventional rectal adenocarcinoma,mucinous adenocarcinoma indicates that neoadjuvant therapy is less sensitive,pathological remission,and tumor downgrading rates are relatively low,so patients with mucinous adenocarcinoma who are not sensitive to neoadjuvant are screened out very early Group selection of individualized treatment is particularly important.Magnetic resonance imaging(MRI)is one of the most commonly used tools for the evaluation of rectal cancer.The curative effect can be evaluated by referring to the tumor volume shrinkage and signal changes after neoadjuvant treatment.However,the volume changes and signal changes of the mucus lake after neoadjuvant treatment are not obvious,and the research on evaluating the efficacy of chemoradiotherapy for mucinous adenocarcinoma is obviously insufficient.Therefore,there is an urgent need to propose a new prediction method for the neoadjuvant efficacy evaluation of RMAC patients.Objective:1.To compare the MRI differences between rectal mucinous adenocarcinoma and non-mucinous adenocarcinoma after neoadjuvant therapy.2.To explore the correlation between the characteristics of mucus distribution based on T2weighted imaging(T2WI)and the efficacy of neoadjuvant therapy.3.To explore the application of apparent diffusion coefficient(ADC)value and early changes in the evaluation of the efficacy of neoadjuvant therapy for RMAC.4.Based on T2WI,T1 weighted imaging(T1WI)enhancement,and diffusion weighted imaging(DWI)radiomics,to construct a prediction model for the efficacy of neoadjuvant chemoradiotherapy for RMAC.Methods:1.36 cases of rectal mucinous adenocarcinoma and 129 non-mucinous adenocarcinoma patients were included in the study.These patients were all based on MRI to obtain the characteristics of tumor changes before and after neoadjuvant radiotherapy and chemotherapy.The postoperative pathological response was used as the standard to evaluate the difference in tumor changes between the two groups.2.118 RMAC patients who underwent surgery directly(without any neoadjuvant treatment)were included.Based on the comparison between T2WI and pathology,an image distinguishing method based on the distribution characteristics of mucus lakes was trained.Then,this method was used in 160 RMAC patients undergoing adjuvant therapy,and the mucus distribution was divided into:mixed type of mucus pool(MTMP)and separate type of mucus pool(STMP).Finally,analyze the correlation between the mucus classification and the efficacy of neoadjuvant therapy.3.59 RMAC patients were enrolled,all of whom underwent MRI scans during radiotherapy and chemotherapy before surgery.Based on the post-processing workstation,the ADC values of the mucus area and the non-mucus area are obtained respectively.The receiver operating characteristic(ROC)and the area under curve(AUC)were used to evaluate the effectiveness of ADC values in different regions in predicting the efficacy of chemoradiotherapy.4.79 patients of RMAC were included in the study.Based on the baseline T2WI,DWI and enhanced T1WI,the tumor region of interest was acquired based on the ITK-SNAP software platform,and the radiomics feature extraction was performed based on the Matlab2016b platform self-edited software.The extracted features include first-order statistics features,morphological features,texture features,and features based on wavelet transform.After repeatability test and feature dimensionality reduction,multivariate logistic regression was finally used to construct the radiomics prediction model.At the same time,the routine tumor characteristics from baseline MRI were obtained.ROC and AUC were used to distinguish predictive performance of these models.Results:1.There was no significant difference between the two groups of mucinous adenocarcinoma and non-mucinous adenocarcinoma in terms of baseline tumor staging,mesorectal fascia(MRF)involvement and lymph node staging(P=0.654,0.298,0.735).After neoadjuvant chemoradiotherapy,the reduction in the distance between the lowermost edge of the mucinous adenocarcinoma tumor and the anal margin,the reduction in longitudinal diameter of the tumor,the reduction in the maximum lateral infiltration thickness,and the tumor downstaging rate were significantly lower than those in the non-mucinous adenocarcinoma group(P=0.005,<0.001,=0.001,=0.012).2.72 patients received neoadjuvant chemoradiotherapy and 88 patients received chemotherapy alone.The tumor response rate of the chemoradiotherapy group was higher than that of the chemotherapy group(58.3%vs.21.6%,P<0.001).In the chemotherapy group,the tumor response rate of MTMP patients was lower than that of STMP patients(4.9%vs.25.5%,P=0.002),with an odds ratio of 11.050 and a 95%CI of 2.368-51.571.3.The optimal threshold for early ADC changes in the mucus area during neoadjuvant therapy is 0.325×10-3 mm2/s,and the AUC used to predict pathological remission was 0.881(95%CI,0.770-0.951).4.The AUC of the three models of T2WI,DWI,and T1WI enhancementwere 0.816(95%CI:0.714-0.917),0.821(95%CI:0.725-0.918)and 0.819(95%CI:0.723-0.916),which were higher than the conventional features respectively(0.607[95%CI:0.483-0.732])(P<0.05).Conclusions:1.The sensitivity of rectal mucinous adenocarcinoma to neoadjuvant therapy is lower than that of non-mucinous adenocarcinoma.2.T2WI can be used to distinguish the classification of mucus,which is a predictor of the response to neoadjuvant therapy;in the neoadjuvant chemotherapy group,the tumor response of MTMP patients is low.3.The early changes of ADC in mucus area can be used to reflect the final efficacy of chemoradiotherapy.4.Baseline T2WI,DWI,and enhanced T1WI imaging radiomics model can predict the efficacy of RMAC. |
PDF Full Text Request