Predicting Response And Survival In Patients With Infiltrative HCC After Initial TACE-Effect Of MR Bias Field Correction

Background and Aims:To explore the application value of intensive 3D measurement of infiltrative hepatocellular carcinoma in patients with invasive hepatocellular carcinoma after correction of magnetic resonance deflection field in the evaluation and survival rate prediction after catheter arterial chemoembolization.Methods:A retrospective analysis of the MRI images of 101 infiltrative hepatocellular carcinoma patients who received conventional or drug-eluting bead chemotherapy from 2001 to 2013 at the Johns Hopkins University School of Medicine hospital before the initial hepatic artery chemoembolization and one month after the embolization was used.Semi-automatic three-dimensional quantitative software performs liver segmentation on multi-phase enhanced magnetic resonance imaging,and based on the segmentation,uses the deflection correction software to perform deflection correction of the liver part in the contour,and performs the whole liver tumor enhancement signal on the corrected image.The three-dimensional measurement of the liver and the calculation of liver enhanced tumor volume(ETV),including ETV(in cm3)during baseline imaging and the relative change in ETV before and after TACE(in%change,ETV%)were used to predict response and survival rates,respectively.Statistical analysis includes Q statistical method to find the cutoff value of variables,Kaplan-Meier method to analyze survival time and Cox proportional hazard model for single-factor and multi-factor risk analysis.Results:The Q statistical method can be used to find the cut-off point for the relative change of baseline ETV and ETV.For the baseline ETV,the cut-off point is 415 cm3(P=0.003),and the ETV%cut-off point is 41%(p=0.01).When using baseline ETV the mortality rate of patients in the "baseline ETV?415 cm3" group was twice as high as that of the patients in the"baseline ETV?415 cm3" group:(hazard ratio:2.00,95%confidence interval:1.23-3.26,p=0.01),Can significantly improve survival prediction(baseline ETV<415cm3 patients'median survival time:19.66 months:baseline ETV?415cm3 patients' median survival time 9.21 months,p<0.001 log-rank test).Using 41%as the ETV%cut-off point,the mortality rate of patients with a relative tumor enhancement reduction of more than 41%is 58%lower than that of patients with a reduction of less than 41%(HR:0.58,95%CI:0.37-0.93,p=0.02),from baseline to follow-up MR imaging,a relative reduction of at least 41%in ETV is significant in predicting survival:(relative reduction in ETV ?41%:median survival rate of patients 19.20 months:relative reduction in ETV<41%:The median survival rate of patients was 8.71 months,p=0.008).Univariate and multivariate survival analysis of baseline ETV showed that:Child-Pugh and BCLC staging are related to survival,P<0.05:Child-Pugh grade"A"patients have a higher mortality rate than Pugh grade "B" or "C",58%lower in patients.The mortality rate of BCLC "B" stage patients is 47%lower than that of BCLC "C" or "D"stage patients.Univariate and multivariate analysis of ETV%showed that the mortality rate of Child-Pugh "A" patients is 70%lower than that of Child-Pugh "B" or "C" patients.The mortality rate of BCLC "B" stage patients is 42%lower than that of BCLC "C" or "D" stage patients.Of note,univariate analysis did not demonstrate the type of TACE to be a statistically significant prognostic factor.Conclusion:After the first hepatic arterial chemoembolization in patients with infiltrative hepatocellular carcinoma,3D measurement of whole liver tumor enhancement volume in MRI images corrected by magnetic resonance bias field can be used to predict the survival rate and accurately evaluate the effectiveness of TACE treatment,which is beneficial to the formulation and adjustment of treatment plan in clinic as early as possible.Background and Aims:The change of T1 relaxation time can be used to evaluate liver diseases such as liver fibrosis and inflammation.In the commonly used T1 mapping scanning technology,the look-locker sequence is not sensitive to the unevenness of the B1 field,but only limited images can be obtained;the variable flip angle T1 mapping can collect whole liver information,but during the image formation process,due to the pulse of the magnetic field causes an eddy current field to be induced in the conductor structure around the coil,which causes the total magnetic field to change,deforms or artifacts the image,and limits the accuracy of image analysis,segmentation,and quantification.This study aims to measure T1 realaxtion time by B1 field-corrected variable flip angle(B1 inhomogeneity-corrected VFA)sequence scan T1 mapping" comparing with modified look-locker sequence(modified look-locker inversion recovery,MOLLI)T1 mapping to discusses the effect and clinical application of B1 field correction.Methods:This study was approved by the the Shandong Provincial Hospital Review Committee.From October 2019 to December 2019,52 healthy peaple without focal or diffuse liver disease and 20 patients with liver cirrhosis underwent 3T MRI liver scans.All healthy peaple had an average age of 24.35±2.19 years old,without history of liver disease or alcohol abuse,and had normal clinical laboratory examination indexes.All patients received a comprehensive clinical laboratory examination.All patients used a 3.0T Siemens(MAGNETOM PRISMA,Siemens,Germany)whole-body magnetic resonance scanner,and the scanning range was from the top of the diaphragm to the lower edge of the liver.The 18-channel phased array body coil is used to obtain T1 mapping,and the MOLLI sequence and B1 inhomogeneity-corrected VFA sequence scan are used.The imaging parameters were kept consistent,and the T1 mapping was obtained by scanning with MOLLI and B1 inhomogeneity-corrected VFA sequence scan.The scanning scheme includes:for MOLLI sequence T1 mapping,a two-dimensional TRUFI(True Fast Imaging with Steady-State Free Precession)sequence is obtained in one breath-hold.Three axial slices were obtained at the level of the confluence of the inferior vena cava,the portal vein and the gallbladder fossa.For the B1 inhomogeneity-corrected VFA sequence,a three-dimensional(3D)T1 weighted VIBE(volume interpolation using double flip angle breath-hold inspection)sequence.Two radiologists placed the region of interest on the corresponding T1 mapping,and measured the T1 relaxation time on the right and left lobes of the liver.Bland-Altman,linear regression,Student t test and one-way analysis of variance were used for statistical analysis.Summarize the continuous data of the normal distribution into the mean and standard deviation(SDs),and calculate the correlation coefficient(ICC)of the quantitative data of the two methods.A Bland-Altman diagram was constructed to evaluate the consistency of MOLLI and B1 inhomogeneity-corrected VFA sequences,.In order to evaluate the relationship between T1 relaxation time and them,regression analysis was performed.Paired t-test was used to compare the average T1 relaxation time of the left and right lobe.The independent t test was used to compare the average T1 relaxation time between men and women.We also used one-way analysis of variance(ANOVA)to make multiple comparisons among Asian ethnic groups.All inference tests are statistically significant with P<0.05.Results:The mean T1 relaxation times(ms)in the whole liver were 900.0±36.39 by MOLLI and 947.72±31.31 by B1 inhomogeneity-corrected VFA.The correlation coefficient was 0.9(P<0.0001).Regression analysis displayed a strong correlation between the two methods in the whole liver(R2=0.80,P<0.0001).Paired t-test showed a significant difference between the two methods in the whole liver(P<0.0001).The Bland-Altman plot between the MOLLI and B1 inhomogeneity-corrected VFA revealed the mean difference of-47.63(95%confidence interval(CI)=-52.11 to-43.15).The mean native T1 relaxation times(ms)in the right lobe were 906.91± 37.79 by MOLLI and 952.13 ± 33.58 by B1 inhomogeneity-corrected VFA.The mean native T1 relaxation times(ms)in the left lobe were 893.28 ± 42.94 by MOLLI and 943.31±36.78 by B1 inhomogeneity-corrected VFA.There were significant differences in T1 relaxation times between right and left lobes,using both MOLLI and B1 inhomogeneity-corrected VFA methods(P<0.05).There was no significant difference of native T1 relaxation times between male and female,using MOLLI(894 ± 36.75 vs.902.7± 34.60 ms,P=0.29)or B1 inhomogeneity-corrected VFA(941.4 ±31.68 vs.950.54 ± 28.44 ms,P=0.20).One-way Analysis of Variance(ANOVA)indicated no association of native T1 relaxation times among Asian ethnic groups,using MOLLI(P=0.18)or B1 inhomogeneity-corrected VFA(P=0.19).Native T1 values in cirrhotic patients were significantly higher than healthy volunteers in both methods(P<0.05).Conclusion:Native T1 relaxation times of the normal liver were established by MOLLI and B1 inhomogeneity-corrected VFA T1 mapping methods.It may provide a reference value for determining normal versus abnormal liver.The results obtained by the two scanning methods are relatively correlated.B1 inhomogeneity-corrected VFA scanning method can correct the B1 inhomogeneity well.