Study On Relationship Of Image To Pathology Based Target In Cervical Cancer Radiotherapy And Clinical Results

objective:Radiotherapy is the main method for treating cervical cancer. It is very important to identify and determine the tumor volume, in the processing of precise radiotherapy, especially in the intracavitary brachytherapy. The only way to improve local control rate and reduce the incidence of proctitis and cystitis is to identify the tumor and the subclinical invasion of tumor accurately.and to give high dose to this areas. CT, MRI and other imaging techniques are commonly used in clinical diagnose, which can provide the anatomical information of tumor and its surrounding tissues. More information on molecular biology of tumor can be provided through metabolic imaging of molecular, with the rapid development of PET / CT (Positron emission tomography / Computer tomography)technology. But which method would be relativelyaccurate in determining the scope and border of tumor? Although there are a lot of research on comparing biological with pathological target in the lung cancer and esophageal cancer , there are no similar report on cervical cancer. In this study, we compared the images of CT, MRI and PET/CT functional imaging information of cervical cancer before operation with its pathologic volume being reconstructed after operation by the technology of whole-mount serial sections, to analyze the correlation of the anatomical image in target volume delineation of cervical cancer radiotherapy and explore the feasibility of using the reconstructed pathologic volume as the gold standard to determine the optimal SUV cutoff-value of PET imaging and establish a scientifical basis for delineating biological target volume of cervical cancer. At the same time, the range of microscopic extension in the different directions of cervical cancer was observed under microscope to study the boundary of clinical target volume. To our knowledge, there are not any relative reports on the correlation of radiotherapy target volume imageology delineating volume of cervical cancer and the reconstructed pathologic volume both at home and abroad. In addition, we made a retrospective analysis of the 10 patients withⅢB stage cervical cancer who were treated in our department from march to may 2008 ,To evaluate the relationship of the therapeutic efficacy and target volume delineation principle which in GTV based extension 2cm bouncary into subclinical irradiated volume of CT-based images of three-dimensional intracavitary brachytherapy .Materials and Methods 9 patients with stageⅠB-ⅡB cervical cancer who are sensitive to radiation and pathological tumor type of which were cauliflower and squamous cell carcinoma were chosen to be the objects. All of them have not been treated by any cancer-related therapy ,and they do not have tuberculosis, inflammation, sarcoidosis, paragonimiasis and any other diseases which may cause PET false-positive results.Before the study, they all completed informed consents.1. Done MRI and ultrasonic examination before surgery and delineated each layer image into GTV_MRI in MRI workstation.2. Done 18F-FDG PET/CT scan one week before surgery and measured the distribution, SUV cutoff-value and volume parameters of positive imaging region through PET image. Measured the tumor diameter and calculated tumor volume through CT, GTV_PET was delineated automatically by use of various percentages relative to the maximal SUV and various absolute SUV.3. Extensive resection of hysterectomy double attachment (or keep one side of attachment) + pelvic lymph node dissection.4. The whole uterine and cervical tumors were fixed by formalin,Then the specimen was cut consecutively to make whole-mount serial sections and make HE staining, experienced pathologist outlined pathologic gross target volume under the microscope and got the image and then calculated the volume of the tumor, took this as the―golden standard‖and compared the GTV_CT,GTV_MRI, determined the optimal SUV cutoff-value.Retrospective analysis of the10 patients withⅢB stage cervical cancer who were treated in our department from march to may 2008,aged from 41 to 56 years old (with a median age of 47 years old) were selected. Pathological types of squamous cell carcinoma, Ten patients with FIGO stageⅢB cervical cancer.the tumor largest diameter was 6.22±0.92cm, Cauliflower type 6 cases, tubercle type 2 cases and usur 2 cases. The four patients had extensionin vaginal.They were treated with CT-based three-dimensional 192Ir intracavity and external beam radiotherapy.The target volume delineated principle of GTV include Cervical residual lesions,whole cervix and invasion Uterine,the CTV include GTV base on 2cm margin of Uterine ang vagin . Regular follow-up after treatment . The overall survival rate and the local control rate by Kaplan-Meier analysis . Results:1. The averages of GTV_CT,GTV_MRI and Vpath were 18.42±11.10cm3, 13.56±10.95 cm3 and 9.57±8.48cm3 respectively, Wilcoxon test results showed that the differences between GTV_CT,GTV_MRI and GTV_path are significant ( p=0.021, p=0.008) , the averages of GTV_MRI and GTV_CT were both significantiy more than GTV_path, but GTV_MRI was closer to the actual pathologic volume,the average of GTV_MRI was larger than GTV_path by approximately 41.6 % ,and the average of GTV_CT was larger than GTV_path by closely two times. Spearman correlation analysis showed that GTV_CT,GTV_MRI and GTV_path were correlated with each other (r=0.933, p=0.000;r=0.833, p=0.005), the DCT which was the maximum diameter of the tumor based on CT imaging, was not correlated with GTV_path (r=4.19, p=0.192), but was significant correlated with DMRI and GTV_path (r=0.80, p=0.010).2. PET/CT showed that primary tumor lesions have unusual FDG radioactivity uptake, the imaging is very clearly. The average SUVmax,SUVavg and SUVmin were 18.71±9.33,7.64±2.91 and 2.29±0.68 respectively, The optimal threshold SUV and absolute SUV were 40.32%±7.28% and 6.97±3.68 respectively. Correlation analysis showed The optimal threshold SUV and GTV_path were negative correlation (r=-0.650,p=0.045). The absolute SUV and GTV_path were not correlated with each other (r= -0.678, p=0.058).The optimal threshold SUV and the largest diameter of the tumor was not correlated (r=-0.444,p=0.235). The absolute SUV and the largest diameter of the tumor was not correlated ( r=0.633,p=0.067). The optimal threshold SUV and absolute SUV were not correlated with SUVmax (r=0.540,p=0.515.r=-0.370,0.327).3. Pathological Whole-mount serial sections and HE staining after microscopically observation, 9 cases specimens in 3 cases visible Microscopic extension tumor lesion, the range of Microscopic extension was 0¹9mm, average 6.9±5.08 mm.4. Clinical research results showed that 10 patients of locally adanced cervical cancer were followed up for 2 rears and following rate of 100%. The2-years total survival rate, pelvic control rates were 90% and 90% respectively. and grade 1-2 late complication of sigmoid colon and rectum was 50%. No grade 3 or more complications developed.Conclusion:1.Taken the tumor's GTV_path which was getting from the technique of WMSS as a―golden standard‖to analysis the correlation of CT, MRI and 18F-FDG PET/CT imaging in determining the cervical cancer target volume. The volume calculated by CT and MRI imaging was significant different from GTV_path (p=0.021,p=0.008), but the GTV_MRI was more close to GTV_path. 18F-FDG PET-CT determined that the optimal threshold and absolute SUV of cervical cancer were 40.32%±7.28% and 6.97±3.68 respectively.2 . CT-based three-dimensional 192Ir intracavity and external beam radiotherapy for patient with locally advanced cervical cancer.The target volume delineated principle which were the delineated residual tumor and the cervix based on the CT as GTV, CTV was the uterus and vagina border increased 2cm.give higher radioation dose in CTV can gain better local control rate, but the long-term efficacy needs further observation.