.3 T High Field Strength Functional Magnetic Resonance Imaging Of Advanced Cervical Cancer Clinical Applications Research

Part I Application of Diffusion Weighted Imaging (DWI) in locally advanced uterine cervical cancer and ADC values in predicting treatment responseObjective To investigate the application of Diffusion Weighted Imaging (DWI) in locally advanced uterine cervical cancer and the value of ADC in predicting treatment response. Methods 28 patients confirmed by pathology and 10 cases of normal uterine cervix for contrast were recruited in this prospective clinical trial.DWI was perfomed in 3T magnetic resonance scanner with the b value of 800 and 2000 s/mm2,17 patients were done with the same protocol before and after the treatment. Results The mean age of patient cohort is 52 years. Pathological types including:squmaous cell carcinoma in 23 cases, adenocarcinaom in 4 cases, and one case of neuroendocrine carcinoma (small cell carcinoma). Clinical stage (FIGO):Ⅱa stage in 6 cases;Ⅱb stage in 8 cases;Ⅲa stage in 3 cases;Ⅲb stage in 9 cases;Ⅳin stage 2 cases.The dispiction of tumor lesion and metastasis pelvic lymph node in high-resolution T2WI and DWI were superior to conventional T2WI and T1 WI. The DWI of higher b value (2000 s/mm2) had the better contrast with slight compromised SNR. DWI had advantage in showing the exact borders of lesions, and the DWI based tumor volume measurement method was more convenient and accurate in monitoring tumor shrinkage, compared to the current response evaluation criteria in solid tumors (RECIST).There were significant difference between the cervical cancer and normal cervix tissue or myometrium(P<0.001).Different pathological types of cervical cancer showed no significant difference in ADC values.The ADC value increased in tumor regions after treatement in both b values (800 s/mm2:9.85±1.55Vs13.408±2.9, P=0.000;2000 s/mm2:7.38±0.98Vs 8.933±1.92,P=0.008).The ADC histogram analysis was better than single whole ROI measurement with less bias.The ADC25%,ADC50%,ADC75% of tumor before treatment showed statistical significance in univariate logistical regression for predicting tumor response(P=0.023,0.034,0.041). ROC curves showed no significance among the area of under curve (AZADC25%=0.818,AZADC50%=0.775,AZadc75%=0.716).Conclusions DWI helps to confirme the biological target zone of tumor for radiotherapy, ADC value analysised with histogram has the potential in predicting the tumor response to treatment. 英文摘要PartⅡMonitoring of tumor oxygenation changes in uterine cervical carcinoma paitents breathing Carbogen gas mixture in 3T MR scannerObjective To investigate the application of BOLD imging in advanced uterine cervical cancer and the hypoxia related MR parameters R2* andΔR2* value in predicting treatment response.Methods 28 patients with locally advanced uterine cervical cancer were studied prospectively. Using multi-echo gradient sequences, blood oxygen level dependent imaging was perfomed in 3T magnetic resonance scanner with the Carbogen mixture gas (95%02+5%C02) inhalation,14 cases were done with the same protocol before and after the treatment. Results Cervical cancer lesions presented as low signal intensity in R2* map while high signal intensity in T2* map. There were sharp contrast between the tumor and myometrium on both functional maps. No significant difference was found in baseline tumor R2* values of mean, center and peripheral region of tumor. Tumor R2* values decreased after Carbogen mixture was administrated. After treatment, tumor R2* values increased from 16.110±2.88 to 20.844±3.49 (P=0.002),while theΔR2*(R2* change before and after Carbogen inhalation) dropped down (2.472±2.07 Vs 0.516±1.87 P=0.045).The hypoxia related parameter of baseline tumor R2* values had correlation with tumor volumes(r=-0.474,P=0.013) as well as maximum diameters (r=-0.410, P=0.034). There was no statistical significant difference between the good and poor response subgroup after therapy in both the baseline R2* (P=0.216) andΔR2*(P=0.868).Conclusions The evaluation of tumor hypoxia status is essential for treatment. The R2* andΔR2* derived of MR BOLD imaging can reflect the tumor oxygenation indirectly. The fluctuation of R2* andΔR2* values before and after therapy reflects the oxygenation changes intra the tumor region. Part III A pharmacokinetic assessment of dynamic contrast-magnetic resonance imaging predicts response to chemoradiation in locally advanced cervical cancerObjective. To investigate the pharmacokinetic assessment of dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) in tumor angiogenesis of uterine cervical carcinoma and as a response indicator in whom receving concurrent chemoradiation. Methods twenty eight women with locally advanced cervical cancer were included in a prospective cohort study. Dynamic contrast enhanced MRI was carried out before and at the conclusion of therapy for a subgroup of 15 cases using a 3T MRI scanner. Pharmacokinetic analysis using a two-compartment model was performed with designed software.Results Three patients could not complete the PWI for technique reasons. The the quality of perfusion imaging was fine with good time resolution. Cervical cancer presented dramatically enhanced in early phase and typical "wash out" enhancement modality in radiological appearance. There was no significant difference among pathlogical type of cervical cancer in physiopathological parameter. The perfusion parameters Ktrans and kep dropped down while contrast concentration enhancement ratio (CER) increased after therapy, and the latter also had weak correlation with the maximum diameter of tumor. Both Ktrans(P=0.039)and CER(P=0.042)showed statistical significance in univariate logistical regression for predicting tumor response. ROC curves showed no significance between the area of under curves(AZKtrans=0.859,AZCER=0.844).There was no significant correlation between perfusion and hypoxia parameters. Reproducibility of pharmacokinetic modeling was confirmed. Conclusions Quantitive measurement of the perfusion imaging might reflect the tumor angiogenesis and be useful to monitor the blood supply status after therapy, as well as to predict treatment response.