Analysis Of Prognostic Factors Of Stage ⅡB-ⅢB Cervical Cancer Based On DCE-MR And Clinical Features

Objective:To investigate the predictive value of imaging parameters and clinical parameters of DCE-MR before treatment on the efficacy of CCRT in stageⅡB-ⅢB cervical cancer,and to obtain independent influencing factors for predicting disease-free survival time.Materials and Methods:Patients with stageⅡB-ⅢB cervical cancer who received standard CCRT treatment at Liaoning Cancer Hospital（Dalian Medical University Clinical Oncology College）from January 2015 to December 2017 were retrospectively analyzed.Follow-up was conducted once every 3 to 6 months in 2 years,once every 6 to12 months in 3 to 5 years,and once every year after 5 years.The primary endpoint of follow-up was disease-free Survival（DFS）,defined as the time between the date of first confirmation of CR by radiographic examination and Disease progression（local recurrence or distant metastasis）or the occurrence of death.Follow-up was performed by outpatient reexamination and telephone follow-up.Patients were divided into progression-free group and progression-free group according to the follow-up.Clinical factors,including age,birth,abortion,menopausal age,maximum palpation diameter,clinical stage,tumor markers,and pathological types,were obtained from clinical records of patients.Siemens Magnetom Trio 3.0T MRI scanner was used.A diagnostic radiologist with 3 years of working experience used PACS system to comprehensively evaluate and record routine MRI parameters such as the size of the primary lesion and the presence or absence of positive lymph nodes before treatment.The volume transfer constant K^trans(min^-1),the extracellular space and the plasma rate constant K_ep(min^-1),and the extracellular and extracellular space were calculated using Omnikinetics（GE Healthcare,China）Extended Tofts Liner two-compartment model of hemodynamics.The quantitative and semi-quantitative parameters of DCE-MRI,such as volume ratio Ve,maximum rise slope Maxslope,maximum contrast agent concentration value Max Conc,etc.Kolmogorov-Smirnov test was used to determine whether the measurement data fit the normal distribution.The data with normal distribution were compared by independent sample t test.Data that did not conform to normal distribution were compared by Mann-Whitney U test.Enumeration data（including menopause,enlarged lymph nodes,clinical FIGO-stage,etc.）were compared by chi-square test.X-TILE3.6.1 software was used to calculate the optimal cut-off value of the measurement data,and patients were divided into the high value group and the low value group according to the optimal cut-off value.Univariate survival analysis used univariate Cox regression analysis to include the clinical features and imaging features of P<0.15 in univariate analysis into the multivariate Cox regression analysis,and Kaplan-Meier method was used to draw the DFS survival curve.Results:1.The follow-up time ranged from 3.6 to 71.6 months,with a median follow-up time of44.8 months.The one-year,two-year and three-year disease-free survival rates were91%,84%and 81%,respectively.Of the 144 patients,29 patients（20%）has died,recurrence,or metastasis:10 patients（34%）has died,3 patients（10%）had in-situ recurrence,10 patients（34%）had pelvic recurrence,and 6 patients（21%）had distant metastasis.2.There were statistically significant differences in the number of births,pathological types and clinical stages between the non-progressive group and the progressive group（P value was 0.003,0.006 and 0.047,respectively）.There were no significant differences in age,menopause or not,number of abortions,maximum diameter of palpation tumor and SCC-Ag level between the two groups（P>0.05）.The largest tumor diameter before treatment was significantly larger in the progression group than in the progression-free group（P<0.001）.Before treatment,the values of K^trans,K_ep,Max Conc and Max Slope in the progression-free group were significantly higher than those in the progression group（P values were 0.001,0.003,0.015 and 0.002,respectively）.The difference in lymph node positivity between the two groups was statistically significant（P<0.001）.Before treatment,there was no significant difference in Ve and Vp values between the two groups（P>0.05）.3.In univariate Cox regression analysis,the number of births,the maximum diameter of imaging tumor,lymph node positivity,K^transvalue,K_epvalue,Max Conc value and Max Slope value were significantly correlated with DFS（P<0.05）.4.In multivariate Cox regression analysis,lymph node positivity（P<0.001,HR4.195[1.904-9.240]）,K^transvalue<0.54（P=0.016,HR 0.349[0.148-0.825]）,and K_epvalue<1.12（P=0.014,HR 0.390[0.166-0.917]）were independent influencing factors for 3-year disease-free survival in this study.Conclusion:1.DCE-MR imaging characteristics and clinical factors before treatment were correlated with CCRT efficacy of stage IIB-IIIB cervical cancer,including number of births,2009FIGO stage,maximum size of imaging tumor,lymph node positivity,K^transvalue,K_epvalue,Max Conc value and Max Slope value.2.Lymph node positivity,K^transvalue<0.54,and K_epvalue<1.12 were independent risk factors affecting 3-year disease-free survival after CCRT treatment for cervical cancer of IIB-IIIB.