An interpretable machine learning model based on contrast-enhanced CT parameters for predicting treatment response to conventional transarterial chemoembolization in patients with hepatocellular carcinoma.

OBJECTIVE: To explore the potential of pre-therapy computed tomography (CT) parameters in predicting the treatment response to initial conventional TACE (cTACE) in intermediate-stage hepatocellular carcinoma (HCC) and develop an interpretable machine learning model. METHODS: This retrospective study included 367 patients with intermediate-stage HCC who received cTACE as first-line therapy from three centers. We measured the mean attenuation values of target lesions on multi-phase contrast-enhanced CT and further calculated three CT parameters, including arterial (AER), portal venous (PER), and arterial portal venous (APR) enhancement ratios. We used logistic regression analysis to select discriminative features and trained three machine learning models via 5-fold cross-validation. The performance in predicting treatment response was evaluated in terms of discrimination, calibration, and clinical utility. Afterward, a Shapley additive explanation (SHAP) algorithm was leveraged to interpret the outputs of the best-performing model. RESULTS: The mean diameter, ECOG performance status, and cirrhosis were the important clinical predictors of cTACE treatment response, by multiple logistic regression. Adding the CT parameters to clinical variables showed significant improvement in performance (net reclassification index, 0.318, P < 0.001). The Random Forest model (hereafter, RF-combined model) integrating CT parameters and clinical variables demonstrated the highest performance on external validation dataset (AUC of 0.800). The decision curve analysis illustrated the optimal clinical benefits of RF-combined model. This model could successfully stratify patients into responders and non-responders with distinct survival (P = 0.001). CONCLUSION: The RF-combined model can serve as a robust and interpretable tool to identify the appropriate crowd for cTACE sessions, sparing patients from receiving ineffective and unnecessary treatments.

Real-time automatic prediction of treatment response to transcatheter arterial chemoembolization in patients with hepatocellular carcinoma using deep learning based on digital subtraction angiography videos.

BACKGROUND: Transcatheter arterial chemoembolization (TACE) is the mainstay of therapy for intermediate-stage hepatocellular carcinoma (HCC); yet its efficacy varies between patients with the same tumor stage. Accurate prediction of TACE response remains a major concern to avoid overtreatment. Thus, we aimed to develop and validate an artificial intelligence system for real-time automatic prediction of TACE response in HCC patients based on digital subtraction angiography (DSA) videos via a deep learning approach. METHODS: This retrospective cohort study included a total of 605 patients with intermediate-stage HCC who received TACE as their initial therapy. A fully automated framework (i.e., DSA-Net) contained a U-net model for automatic tumor segmentation (Model 1) and a ResNet model for the prediction of treatment response to the first TACE (Model 2). The two models were trained in 360 patients, internally validated in 124 patients, and externally validated in 121 patients. Dice coefficient and receiver operating characteristic curves were used to evaluate the performance of Models 1 and 2, respectively. RESULTS: Model 1 yielded a Dice coefficient of 0.75 (95% confidence interval [CI]: 0.73-0.78) and 0.73 (95% CI: 0.71-0.75) for the internal validation and external validation cohorts, respectively. Integrating the DSA videos, segmentation results, and clinical variables (mainly demographics and liver function parameters), Model 2 predicted treatment response to first TACE with an accuracy of 78.2% (95%CI: 74.2-82.3), sensitivity of 77.6% (95%CI: 70.7-84.0), and specificity of 78.7% (95%CI: 72.9-84.1) for the internal validation cohort, and accuracy of 75.1% (95% CI: 73.1-81.7), sensitivity of 50.5% (95%CI: 40.0-61.5), and specificity of 83.5% (95%CI: 79.2-87.7) for the external validation cohort. Kaplan-Meier curves showed a significant difference in progression-free survival between the responders and non-responders divided by Model 2 (p = 0.002). CONCLUSIONS: Our multi-task deep learning framework provided a real-time effective approach for decoding DSA videos and can offer clinical-decision support for TACE treatment in intermediate-stage HCC patients in real-world settings.

Predicting hyperprogressive disease in patients with advanced hepatocellular carcinoma treated with anti-programmed cell death 1 therapy.

BACKGROUND: Hyperprogressive disease (HPD) is a new progressive pattern in patients with advanced hepatocellular carcinoma (HCC) treated with programmed cell death 1 (PD-1) inhibitors. We aimed to investigate risk factors associated with HPD in advanced HCC patients undergoing anti-PD-1 therapy. METHODS: A total of 69 patients treated with anti-PD-1 therapy between March 2017 and January 2020 were included. HPD was determined according to the time to treatment failure, tumour growth rate, and tumour growth rate ratio. Univariate and multivariate analyses were performed to identify clinical variables significantly associated with HPD. A risk model was constructed based on clinical variables with prognostic significance for HPD. FINDINGS: Overall, 10 (14·49%) had HPD. Haemoglobin level, portal vein tumour thrombus, and Child-Pugh score were significantly associated with HPD. The risk model had an area under the curve of 0·931 (95% confidence interval, 0·844-1·000). Patients with HPD had a significantly shorter overall survival (OS) than that of the patients with non-HPD (p < 0·001). However, there was no significant difference in OS between PD (progressive disease) patients with and without HPD (p = 0·05). INTERPRETATION: We identified three clinical variables as risk factors for HPD, providing an opportunity to aid the pre-treatment evaluation of the risk of HPD in patients treated with immunotherapy. FUNDING: This study was funded by the National Natural Science Foundation of China (81571664, 81871323, and 81801665); National Natural Science Foundation of Guangdong Province (2018B030311024); Scientific Research General Project of Guangzhou Science Technology and Innovation Commission (201707010,328); and China Postdoctoral Science Foundation (2016M600145).

Stent placement combined with intraluminal radiofrequency ablation and hepatic arterial infusion chemotherapy for advanced biliary tract cancers with biliary obstruction: a multicentre, retrospective, controlled study.

OBJECTIVE: To evaluate the efficacy and safety of stent placement combined with intraluminal radiofrequency ablation (intra-RFA) and hepatic arterial infusion chemotherapy (HAIC) for patients with advanced biliary tract cancers (Ad-BTCs) and biliary obstruction (BO). METHODS: We retrospectively reviewed data for patients with Ad-BTCs and BO who underwent stent placement with or without intra-RFA and HAIC in three centres between November 2013 and November 2018. The stent patency time (SPT), overall survival (OS), and adverse events (AEs) were analysed. RESULTS: Of the 135 enrolled patients, 64 underwent stent placement combined with intra-RFA and HAIC, while 71 underwent only stent placement. The median SPT was significantly longer in the combination group (8.2 months, 95% confidence interval [CI]: 7.1-9.3) than in the control group (4.3 months, 95% CI: 3.6-5.0; p < 0.001). A similar result was observed for OS (combination: 13.2 months, 95% CI: 11.1-16.5; control: 8.5 months, 95% CI: 7.6-9.6; p < 0.001). The incidence of AEs related to biliary tract operation was not significantly different between the two groups (p > 0.05). The most common AE and serious AE related to HAIC were alanine aminotransferase elevation (24/64; 37.5%) and thrombocytopenia (8/64; 12.5%), respectively. All AEs were tolerable, and there was no death from AEs. CONCLUSIONS: Stent placement combined with intra-RFA and HAIC may be a safe, potential treatment strategy for patients with Ad-BTCs and BO. KEY POINTS: • Advanced biliary cancers (Ad-BTCs) with biliary obstruction (BO) can rapidly result in liver failure and cachexia with an extremely poor prognosis. • Stent placement combined with intraluminal radiofrequency ablation and hepatic arterial infusion chemotherapy may be safe and effective for patients with Ad-BTCs and BO. • The long-term efficacy and safety of the combined treatment is promising.

Relationship between circulating tumor cells undergoing EMT and short-term efficacy following interventional treatment in patients with hepatocellular carcinoma.

OBJECTIVE: A growing number of studies have indicated that epithelial-mesenchymal transition (EMT) phenotypes and the number of circulating tumor cells (CTCs) are significant indicators of tumor characteristics and treatment efficacy, and thus have a broad range of potential applications in the diagnosis and treatment of malignant tumors. The value of data on CTC phenotypes and CTC counts in the diagnosis of hepatocellular carcinoma (HCC) and assessment of efficacy after comprehensive interventional therapy remains unclear. METHODS: Data of 107 patients who exhibited space-occupying lesions in the liver on enhanced CT/MRI scans at the Guangdong Provincial People's Hospital (a tertiary medical center) between August 2017 and October 2018, were retrospectively analyzed. All enrolled patients were treated with transcatheter arterial chemoembolization (TACE) combined with microwave ablation (MWA). An imFISH CTC assay was used to isolate and count CTCs with different EMT phenotypes in the patients' peripheral blood, which facilitated an analysis of the value of CTC phenotype and CTC count data in the diagnosis or treatment of HCC. RESULTS: The CTC count and EMT phenotypes in HCC patients were not associated with patient characteristics such as age, sex, Hepatitis B Virus (HBV)-DNA status, alcohol consumption history, Aspartate Transaminase (AST) to Platelet Ratio Index (APRI) score, Eastern Cooperative Oncology Group (ECOG) score, Child-Pugh score, alpha-fetoprotein (AFP), number and size of tumors, vascular invasion, or metastasis (P â€‹> â€‹0.05). The CTC count and EMT phenotypes in HCC patients before treatment were not predictive of short-term efficacy (P â€‹> â€‹0.05). Comprehensive interventional therapy reduced the total CTC count and mesenchymal CTC count (P â€‹= â€‹0.034 and 0.022, respectively). CONCLUSION: TACE in combination with ablation reduced the total CTC count and mesenchymal CTC count. The CTC count and EMT phenotypes may be associated with long-term efficacy.

[Effects of Ringer's solution with different concentrations of alcohol on biphasic compound action potentials of frog sciatic nerve trunk].

OBJECTIVE: To quantitatively investigate the effects of Ringer's solution with different concentrations of alcohol (1%～80%) on biphasic compound action potentials (AP) from frog sciatic nerve trunk, and their recoveries from alcohol effects. METHODS: Individual segments of frog sciatic nerve trunk with a length of 6 to 8 cm were prepared. Ringer's solution with different concentrations of alcohol (0%, 1%, 2%, 4%, 8%, 16%, 32%, 48%, 64% and 80%) was applied onto the segment of the trunk between the stimulus and ground electrodes via an agent reservoir which was newly armed in a nerve trunk shielded chamber for 5 minutes. The nerve trunk was respectively electro-stimulated to generate the biphasic compound AP which was recorded using the experimental system of BL-420F. This was followed by 5 times washout plus 5 min administration with Ringer's solution before recovery recording of AP. RESULTS: Compared to normal Ringer's solution, Ringer's solution with alcohol at ≤4% did not have dramatic impacts on the AP amplitude and conduction velocity, while Ringer's solution with alcohol at ≥8% there was significant decrease in these two parameters. Ringer's solution with alcohol at the conentrations of 16%, 32% and ≥48% could prevent a small proportion (30%), a large proportion (90%) and all (100%) of sciatic nerve trunks, respectively, from generating AP. Washout with normal Ringer's solution after alcohol application at the concentration of ≤32%, AP could totally recover to normal status. While alcohol at the concentration of 48%, 64% and 80%, the probabilities to regenerate APs were 90%, 40% and 0%, and the AP amplitudes were decreased to 60%, 36% and 0%, respectively. After washout, AP conduction velocity showed no difference with alcohol at the concentration of ≤8% when compared with that before washout, while it could not be recovered to normal under alcohol at ≥16%. CONCLUSION: Ringer's solution with different concentrations of alcohol exerts different effects on biphasic compound AP amplitude and conduction velocity. Hopefully, our findings could be helpful for the alcoholic usage and its recovery from alcoholic damage.