| BACKGROUND AND PURPOSE: renal carcinoma is a particularly aggressive and chemoresistant malignancy that accounts for approximately 3% of all cancers in adult.The higher percentage of all renal cancers is renal cell carcinomas(RCCs)originating from the renal epithelium.And approximately around 25% of all renal cell carcinoma patients have advanced disease at the time of the diagnosis and patients with metastatic renal cell carcinoma face especially poor prognoses including a median survival of just 13 months.The clinical behavior of cc RCCs is highly variable,ranging from slow-growing localized tumors to aggressive metastatic disease.cc RCCs are resistant to both chemotherapy and radiotherapy,with surgery(nephrectomy)for localised disease,being the only suitable treatment.The increase in our understanding of the genetic factors underlying cc RCC,has translated to improvements in target identification and therapies.The understanding of VEGF and m TOR being central to cc RCC biology,led to the implementation of multiple antiangiogenic drugs(sunitinib,sorafenib,pazopanib,everolimus,and bevacizumab plus interferon-α)for cc RCC treatment [35].Furthermore,as for the prognosis,Stage I patients have the best prognosis with 5-year survival rates of ~80-95%.The survival rates progressively worsen with stage,with Stage II patients having survival rates of ~80% and Stage III ~60%.Even with advances in targeted therapies,Stage IV patients have survival rates of just over two years [36].The majority of renal tumors are the outcome of incidental findings,it often occurs when imaging is done for other medical reasons [37].The most commonly used imaging techniques to then assess these incidental renal masses are ultrasound,computed tomography,and magnetic resource imaging [38].In most cases an accurate characterization of renal tumors is possible just by imaging [38].Nevertheless,there are a number of factors that can hamper a reliable diagnosis.With the growing number of incidentally detected renal cell carcinomas,also the amount of unexpected benign renal masses at resection has increased [39].Not only technical factors,but also errors in image interpretation as well as certain pathologic features can lead to misdiagnosis and in some cases even unnecessary surgery [40].GE’s Discovery CT750HD(GE Healthcare,Milwaukke,IL,USA)has let one x-ray tube shift its energy from one to another within 1 ms;this approach has been called "fast k V-switching".The fundamental base for a fast k V-switching CT is that the scintillator in the detector must use a very short time to acquire the x-ray photons.The acquisition time consists of three components: the rise time of the output signal of the detector,which is the time between when the x-ray source is turned on and when the detector reaches the constant value,the time when the signal is constant and the fall time of the output signal of the detector,which is the decay of the signal after the x-ray source is turned off.A short fall time is needed to ensure that there is very little information left in the detector which can be transferred to the next view.This remaining information will otherwise blur the image and reduce the resolution.This can be corrected with a software,but the image will then have a reduced signal to noise ratio [41].Spectral computed tomography has developed much in the recent year and beside not having some of the traditional computed tomography issues such as tissue characterization and beam hardening artifact,it has also improved the detection rate of diseases [60].And perhaps the most essential characteristic of it is the fact that it has two layers of detection;Whileas the bottom layer absorbs high energy photons,the top layer absorbs the high energy photons,all through out single x-ray source.One of the other major advantages of the spectral CT is its ability to generate virtual nonenhanced images that not only decrease the cost and the radiation but also the effective atomic number Z,the iodine map and the virtual monoenergetic images at various levels of ke V throughout the C-ray spectrum.70 ke V choice was based upon previous reports that demonstrated and showed that it has a high contrast to noise ratio and has the lowest image noise among various other ke V monoenergetic images ranged between 40-140 ke V [60].All images were generated from spectral CT.The objective of this study is to evaluate the role of histogram analysis assessed by SingleSource Dual-Energy CT on tumor grades of Clear Cell Renal Cell Carcinoma(cc RCC).MATERIALS AND METHOD: The target population for this study was patients who were reported to the Radiology department from 2011 to 2015 and were diagnosed with Renal Cell Carcinoma(RCC).Histogram parameters were evaluated to check its ability to differentiate grades of clear cell renal cell carcinomas(cc RCCs)on non-enhanced Single-Source Dual-Energy CT.Approximately 122 patients were diagnosed with Clear Cell Renal Cell Carcinoma(cc RCC)and 72(I=5,II=24,III=8,I-II=17,II-III=16,III-IV=2)included in our study and merged into three groups;group one consists of grade I and grade I-II(22 patient;14 males,and 8 females,59.32±10.531);group two consists of grade II only(24 patients;12 males,and 12 females,61.4±8.256);Whiles,group three consists of grade II-III,III and III-IV combined(26 patients;22 males,and 4 females,58.65±11.296).Comparisons were made within the three groups;and between two groups(group 1 and 2,group 1 and 3,and group 1 and 3)Single-source dual energy CT machine(GE Discovery HD 750,USA)was used to preform upper abdominal scan.All of the Single-Source Dual-Energy CT images were subtracted from Advanced Workstation 4.6 software and were pre-evaluated.Single-energy images with Digital Imaging and Communication in Medicine(DICOM)formatted were then transferred from PACS(picture archiving and communication system)workstation into Omni-Kintetics workstation for further analyses were all images were then processed.All ROIs(regions on interest)were then drawn on the largest slices tumor area(cystic areas were excluded).Histogram analysis was preformed once the ROIs delaminated areas were determined.Histogram parameters include: min intensity,max intensity,mean value,median value,standard deviation(SD),skewness,kurtosis,uniformity,energy,entropy,and the cumulative frequency distributions(5th,25 th,75th,and 95 th percentiles).The statistical analysis was done on SPSS24 software;One-way ANOVA and Independent t–test were used for the normally distributed variables.Whereas,Kruskal-Wallis and Mann-Whitney U test were used on the not normally distributed variables in order to determine the significant texture parameters to differentiate cc RCC according to three groups.Receiver Operating Characteristics(ROC)was used to assess sensitivity and specificity of significant parameters to differentiate our study tumors that were compared according to their grades on non-enhanced phase of SingleSource Dual-Energy CT.P < 0.05 symbolized statistical significance.RESULTS: The comparisons between the three groups resulted in the highest value of AUC=0.741 was achieved by skewness(73% sensitivity);Whileas,the highest sensitivity was at 85% in both median Intensity(AUC=0.691,p=0.023)and quantile 95(AUC=0.677,p=0.045);and the highest specificity of 66% was achieved by skewness.Moreover,comparisons between group 1 and group 2 of CCRCC shows that the highest AUC=0.615 was achieved by Variance(p=0.024);Whileas,the highest value of sensitivity of 59% was achieved by Energy(p=0.049).And the comparisons between groups 1 and 3 shows the highest AUC=0.741 and highest specificity of 66% was achieved by Skewness(p=0.011),However,highest sensitivity of 85% was achieved by Quantile 75(AUC=0.691,P=0.018),Quantile 90(AUC=0.678,p=0.022),and Quantile 95(AUC=0.677,p=0.020).Finally,comparisons made between group 2 and 3 reveals that the highest AUC=0.691 was achieved by both Median Intensity(p=0.015),and Quantile 75(p=0.027).Highest specificity of 63% was achieved by Quantile 50(p=0.019).Whileas,Quantile 75 got the highest sensitivity of 88%.Monoenergetic images were used.The size of the tumors is positively associated to higher tumor grade.Mean value,mean intensity,and energy show to be the best parameters that have differential diagnosis values.CONCLUSION: Histogram analysis on non-enhanced phase on Spectral CT(Single-Source DualEnergy Computed Tomography)images showed great significance in differentiating tumors grade of clear cell renal cell carcinoma,and can be used in grading ccRCC. |
PDF Full Text Request