A Transformer‐Based microvascular invasion classifier enhances prognostic stratification in HCC following radiofrequency ablation
Wentao WangYueyue WangDanjun SongYingting ZhouRongkui LuoSi-Qi YingLi YangWei SunJia‐Bin CaiXi WangZhen BaoJiaping ZhengMengsu ZengQiang GaoXiaoying WangJian ZhouManning WangGuoliang ShaoShengxiang RaoKai Zhu
4
Citation
33
Reference
10
Related Paper
Citation Trend
Abstract:
Abstract Background & Aims We aimed to develop a Transformer‐based deep learning (DL) network for prognostic stratification in hepatocellular carcinoma (HCC) patients undergoing RFA. Methods A Swin Transformer DL network was trained to establish associations between magnetic resonance imaging (MRI) datasets and the ground truth of microvascular invasion (MVI) based on 696 surgical resection (SR) patients with solitary HCC ≤3 cm, and was validated in an external cohort ( n = 180). The multiphase MRI‐based DL risk outputs using an optimal threshold of .5 was employed as a MVI classifier for prognosis stratification in the RFA cohort ( n = 180). Results Over 90% of all enrolled patients exhibited hepatitis B virus infection. Liver cirrhosis was significantly more prevalent in the RFA cohort compared to the SR cohort (72.2% vs. 44.1%, p < .001). The MVI risk outputs exhibited good performance (area under the curve values = .938 and .883) for predicting MVI in the training and validation cohort, respectively. The RFA patients at high risk of MVI classified by the MVI classifier demonstrated significantly lower recurrence‐free survival (RFS) and overall survival rates at 1, 3 and 5 years compared to those classified as low risk ( p < .001). Multivariate cox regression modelling of a‐fetoprotein > 20 ng/mL [hazard ratio (HR) = 1.53; 95% confidence interval (95% CI): 1.02–2.33, p = .047], high risk of MVI (HR = 3.76; 95% CI: 2.40–5.88, p < .001) and unfavourable tumour location (HR = 2.15; 95% CI: 1.40–3.29, p = .001) yielded a c‐index of .731 (bootstrapped 95% CI: .667–.778) for evaluating RFS after RFA. Among the three risk factors, MVI was the most powerful predictor for intrahepatic distance recurrence. Conclusions The proposed MVI classifier can serve as a valuable imaging biomarker for prognostic stratification in early‐stage HCC patients undergoing RFA.The present studt was completed to assess the clinical utility of B protein, as a tumor marker of hepatocellular carcinoma. The association of B protein and liver cirrhosis was also evaluated because hepatocellular carcinoma is usually combined with cirrhosis. The serum levels of B protein were studied by a Latex-agglutination test. One hundred and twenty-nine patients including 23 hepatocellular carcinoma, 50 hepatocellular carcinoma combined with liver cirrhosis, 40 liver cirrhosis, and 16 chronic hepatitis were tested. The positive rates of B protein in various diseases were as follows: 30.4% (7/23) in patients with hepatocellular carcinoma; 68.6% (35/50) in patients with hepatocellular carcinoma combined with cirrhosis; 82.5% (33/40) in patients with cirrhosis; and 62.5% (10/16) in patients with chronic hepatitis. When B protein was used as a tumor marker of hepatocellular carcinoma, the sensitivity (57.5%) and specificity (25%) were very low. Furthermore, patients with hepatocellular carcinoma, usually combined with cirrhosis; which carried the highest positive rate on B protein determination. This also limited clinical utilization of B protein as a tumor marker of hepatocellular carcinoma. Moreover, there was no correlation of B protein with the serum alpha-fetoprotein level or tumor size. On the contrary, positive correlation of the B protein level with Child's staging (Tau-c value = 0.392, p = 0.008), and death during follow-up (Tau-c value = 0.456, p = 0.021), were discovered in patients with cirrhosis.(ABSTRACT TRUNCATED AT 250 WORDS)
Hepatitis B
Cite
Citations (0)
We have measured the serum erythropoietin concentrations in 14 patients with liver cirrhosis and in 14 patients with a hepatocellular carcinoma. Among these patients, 2 with liver cirrhosis (14.3%) and 7 with a hepatocellular carcinoma (50.0%) were found to have raised serum erythropoietin concentrations, ranging up to 40 mU/ml. Negative correlations were found between erythropoietin and the RBC, and the Hb and Ht in the cases with liver cirrhosis. In contrast, a positive correlation which was not significant was found only between the erythropoietin and the RBC in cases involving a hepatocellular carcinoma. This has suggested that the relationship between the erythropoietin and the RBC in cases of a hepatocellular carcinoma differs from the relationship seen under the usual physiological circumstances of those with liver cirrhosis.
Cite
Citations (0)
In a retrospective study, 42 out of 44 cases of hepatocellular carcinoma were investigated immunohistochemically for chronic hepatitis B infection. Surface antigen was found in the liver tissue of only 4 of these cases. In 41 of the patients, mildly to moderately active cirrhosis of the liver was found to be underlying the carcinoma. The age distribution and case histories showed that hepatocellular carcinoma often developed from low-complication cirrhosis of long standing and of various etiologies, and must thus be considered a late complication of cirrhosis.
Hepatitis B
Etiology
Cite
Citations (9)
Objective To investigate the expression and relationship of TRA1 mRNA among hepatocellular carcinoma,liver cirrhosis and normal liver tissues. Methods RT-PCR was used to detect the expression of TRA1 mRNA in 34specimens of hepatocellular carcinoma,liver cirrhosis and normal liver tissues. Results The expression rates of TRA1in hepatocellular carcinoma,liver cirrhosis and normal liver tissues were 95. 00%,84. 21% and 50. 00%,respectively,there were significant differences between hepatocellular carcinoma and normal liver tissues(P 0. 05),but no significant differences between liver cirrhosis and normal liver tissues(P 0. 05). The expression levels of TRA1 mRNA were 1. 67 ± 0. 96,1. 49 ± 0. 53,0. 57 ± 0. 27,respectively,which was significantly higher in hepatocellular carcinoma and liver cirrhosis tissues than that in normal liver tissues(P 0. 05). Conclusion TRA1 mRNA is involved in the progression of hepatocellular carcinoma and liver cirrhosis,and it may be a potential biomarker for diagnosis and target for therapy.
Cite
Citations (0)
Hepatitis B
Cite
Citations (1)
Background Although most people with relapsing onset multiple sclerosis (R-MS) eventually transition to secondary progressive multiple sclerosis (SPMS), little is known about disability progression in SPMS. Methods All R-MS patients in the Cardiff MS registry were included. Cox proportional hazards regression was used to examine a) hazard of converting to SPMS and b) hazard of attaining EDSS 6.0 and 8.0 in SPMS. Results 1611 R-MS patients were included. Older age at MS onset (hazard ratio [HR] 1.02, 95%CI 1.01–1.03), male sex (HR 1.71, 95%CI 1.41–2.08), and residual disability after onset (HR 1.38, 95%CI 1.11–1.71) were asso- ciated with increased hazard of SPMS. Male sex (EDSS 6.0 HR 1.41 [1.04–1.90], EDSS 8.0 HR 1.75 [1.14–2.69]) and higher EDSS at SPMS onset (EDSS 6.0 HR 1.31 [1.17–1.46]; EDSS 8.0 HR 1.38 [1.19–1.61]) were associated with increased hazard of reaching disability milestones, while older age at SPMS was associated with a lower hazard of progression (EDSS 6.0 HR 0.94 [0.92–0.96]; EDSS 8.0: HR 0.92 [0.90–0.95]). Conclusions Different factors are associated with hazard of SPMS compared to hazard of disability progres- sion after SPMS onset. These data may be used to plan services, and provide a baseline for comparison for future interventional studies and has relevance for new treatments for SPMS RobertsonNP@cardiff.ac.uk
Cite
Citations (0)
The hazard ratio and median survival time are the routine indicators in survival analysis. We briefly introduced the relationship between hazard ratio and median survival time and the role of proportional hazard assumption. We compared 110 pairs of hazard ratio and median survival time ratio in 58 articles and demonstrated the reasons for the difference by examples. The results showed that the hazard ratio estimated by the Cox regression model is unreasonable and not equivalent to median survival time ratio when the proportional hazard assumption is not met. Therefore, before performing the Cox regression model, the proportional hazard assumption should be tested first. If proportional hazard assumption is met, Cox regression model can be used; if proportional hazard assumption is not met, restricted mean survival times is suggested.风险比（hazard ratio，HR）和中位生存时间是生存分析时的常规分析和报告指标。本文简要介绍了HR和中位生存时间的关系以及比例风险假定在这两者之间的作用，分析了检索出的58篇文献中的110对风险比和中位生存时间比的差异，并通过实例阐明了产生这种差异的原因。结果表明，在不满足比例风险假定时，Cox回归模型计算得到的风险比是不合理的，且与中位生存时间之比不等价。因此，在使用Cox回归模型前，应先进行比例风险假定的检验，只有符合比例风险假定时才能使用该模型；当不符合比例风险假定时，建议使用限制性平均生存时间。.
Cite
Citations (0)
Objective To evaluate the diagnostic value of diffusion weighted imaging(DWI) at 3.0 T magnetic resonance in liver cirrhosis and hepatocellular carcinoma.Methods DWI was performed in 10 healthy volunteers,21 patients with liver cirrhosis and 30 patients with hepatocellular carcinoma at 3.0T magnetic resonance. DWI and ADC values in normal liver, liver cirrhosis and hepatocellular carcinoma were analysed.The diagnostic sensitivity and specificity of ADC values in normal liver, liver cirrhosis and hepatocellular carcinoma were also analysed.Results When b value was 100 s/mm2,the ADC values of normal liver were higher than liver cirrhosis, hepatocellular carcinoma(P0.05),but there was not significance between liver cirrhosis and hepatocellular carcinoma(P0.05).When b value was 300 s/mm2,1000 s/mm2,the ADC values of three groups had significant differences(P0.01). When b value was 300 s/mm2, the diagnostic sensitivity and specificity were 82.6% and 100% in liver cirrhosis ,95.2% and 90.4% in hepatocellular carcinoma ,respectively. When b value was 1000 s/mm2, the diagnostic sensitivity and specificity of ADC values were 86.9% and 100% in liver cirrhosis,95.2% and 66.7% in hepatocellular carcinoma,respectively. Conclusion DWI at 3.0 T magnetic resonance in examination of liver could synthetically and quantitatively analyze the rule of ADC values in liver cirrhosis and hepatocellular carcinoma.
Cite
Citations (0)
The aim of this study was to determine the characteristics of hepatocellular carcinoma at a major health center in southern Turkey. Computed tomography was compared to the combination of ultrasonography and serum alpha-fetoprotein determination in the diagnosis of hepatocellular carcinoma.Of 226 patients with liver cirrhosis, 35 were diagnosed with hepatocellular carcinoma on first admission or during follow-up in the period between 1999 and 2002. The features investigated were, age at time of hepatocellular carcinoma diagnosis, etiology of cirrhosis, severity of cirrhosis at presentation, tumor pattern, stage of hepatocellular carcinoma, serum alpha-fetoprotein level, and dynamic computed tomography findings. Results were compared to previous findings in Turkey and elsewhere.In the hepatocellular carcinoma patients, the male:female ratio was 4:1 and the mean age at presentation was 61 years. Chronic hepatitis B virus infection (65.7%) and chronic hepatitis C virus infection (28.6%) were the most frequently identified risk factors for hepatocellular carcinoma. Forty percent of the patients had Child-Pugh A cirrhosis when they were diagnosed with hepatocellular carcinoma. Sixty-seven percent of patients had fewer than three hepatocellular carcinoma nodules in the liver at the time of diagnosis. Only three of the hepatocellular carcinoma cases were Okuda stage I. The combination of ultrasonography and serum alpha-fetoprotein >20 ng/ml identified hepatocellular carcinoma in 32 of the 35 total cases.The results indicate that hepatitis B virus infection in patients with cirrhosis is still the leading risk factor for the development of hepatocellular carcinoma. Also, early-stage hepatocellular carcinoma is rarely diagnosed in cirrhosis patients from this region of Turkey. Surveillance with computed tomography for early diagnosis of hepatocellular carcinoma seems not to be mandatory.
Hepatitis B
Hepatitis C
Cite
Citations (25)