Prediction of early recurrence of hepatocellular carcinoma after resection using digital pathology images assessed by machine learning

Combining a deep learning model with clinical data better predicts hepatocellular carcinoma behavior following surgery

Hepatocellular carcinoma (HCC) is among the most common cancers worldwide, and tumor recurrence following liver resection or transplantation is one of the highest contributors to mortality in HCC patients after surgery. Using artificial intelligence (AI), we developed an interdisciplinary model to predict HCC recurrence and patient survival following surgery. We collected whole-slide H&E images, clinical variables, and follow-up data from 300 patients with HCC who underwent transplant and 169 patients who underwent resection at the Cleveland Clinic. A deep learning model was trained to predict recurrence-free survival (RFS) and disease-specific survival (DSS) from the H&E-stained slides. Repeated cross-validation splits were used to compute robust C-index estimates, and the results were compared to those obtained by fitting a Cox proportional hazard model using only clinical variables. While the deep learning model alone was predictive of recurrence and survival among patients in both cohorts, integrating the clinical and histologic models significantly increased the C-index in each cohort. In every subgroup analyzed, we found that a combined clinical and deep learning model better predicted post-surgical outcome in HCC patients compared to either approach independently.

Exploring the Role of GITR/GITRL Signaling: From Liver Disease to Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is the most common primary liver cancer and presents a continuously growing incidence and high mortality rates worldwide. Besides advances in diagnosis and promising results of pre-clinical studies, established curative therapeutic options for HCC are not currently available. Recent progress in understanding the tumor microenvironment (TME) interactions has turned the scientific interest to immunotherapy, revolutionizing the treatment of patients with advanced HCC. However, the limited number of HCC patients who benefit from current immunotherapeutic options creates the need to explore novel targets associated with improved patient response rates and potentially establish them as a part of novel combinatorial treatment options. Glucocorticoid-induced TNFR-related protein (GITR) belongs to the TNFR superfamily (TNFRSF) and promotes CD8+ and CD4+ effector T-cell function with simultaneous inhibition of Tregs function, when activated by its ligand, GITRL. GITR is currently considered a potential immunotherapy target in various kinds of neoplasms, especially with the concomitant use of programmed cell-death protein-1 (PD-1) blockade. Regarding liver disease, a high GITR expression in liver progenitor cells has been observed, associated with impaired hepatocyte differentiation, and decreased progenitor cell-mediated liver regeneration. Considering real-world data proving its anti-tumor effect and recently published evidence in pre-clinical models proving its involvement in pre-cancerous liver disease, the idea of its inclusion in HCC therapeutic options theoretically arises. In this review, we aim to summarize the current evidence supporting targeting GITR/GITRL signaling as a potential treatment strategy for advanced HCC.

Nomogram for preoperative estimation of microvascular invasion risk in hepatocellular carcinoma

Microvascular invasion (MVI) is an adverse prognostic indicator of tumor recurrence after surgery for hepatocellular carcinoma (HCC). Therefore, developing a nomogram for estimating the presence of MVI before liver resection is necessary. We retrospectively included 260 patients with pathologically confirmed HCC at the Fifth Medical Center of Chinese PLA General Hospital between January 2021 and April 2024. The patients were randomly divided into a training cohort (n = 182) for nomogram development, and a validation cohort (n = 78) to confirm the performance of the model (7:3 ratio). Significant clinical variables associated with MVI were then incorporated into the predictive nomogram using both univariate and multivariate logistic analyses. The predictive performance of the nomogram was assessed based on its discrimination, calibration, and clinical utility. Serum carnosine dipeptidase 1 ([CNDP1] OR 2.973; 95 % CI 1.167-7.575; p = 0.022), cirrhosis (OR 8.911; 95 % CI 1.922-41.318; p = 0.005), multiple tumors (OR 4.095; 95 % CI 1.374-12.205; p = 0.011), and tumor diameter ≥3 cm (OR 4.408; 95 % CI 1.780-10.919; p = 0.001) were independent predictors of MVI. Performance of the nomogram based on serum CNDP1, cirrhosis, number of tumors and tumor diameter was achieved with a concordance index of 0.833 (95 % CI 0.771-0.894) and 0.821 (95 % CI 0.720-0.922) in the training and validation cohorts, respectively. It fitted well in the calibration curves, and the decision curve analysis further confirmed its clinical usefulness. The nomogram, incorporating significant clinical variables and imaging features, successfully predicted the personalized risk of MVI in HCC preoperatively.

A clinical-radiomic-pathomic model for prognosis prediction in patients with hepatocellular carcinoma after radical resection

PURPOSE

Radical surgery, the first-line treatment for patients with hepatocellular cancer (HCC), faces the dilemma of high early recurrence rates and the inability to predict effectively. We aim to develop and validate a multimodal model combining clinical, radiomics, and pathomics features to predict the risk of early recurrence.

MATERIALS AND METHODS

We recruited HCC patients who underwent radical surgery and collected their preoperative clinical information, enhanced computed tomography (CT) images, and whole slide images (WSI) of hematoxylin and eosin (H & E) stained biopsy sections. After feature screening analysis, independent clinical, radiomics, and pathomics features closely associated with early recurrence were identified. Next, we built 16 models using four combination data composed of three type features, four machine learning algorithms, and 5-fold cross-validation to assess the performance and predictive power of the comparative models.

RESULTS

Between January 2016 and December 2020, we recruited 107 HCC patients, of whom 45.8% (49/107) experienced early recurrence. After analysis, we identified two clinical features, two radiomics features, and three pathomics features associated with early recurrence. Multimodal machine learning models showed better predictive performance than bimodal models. Moreover, the SVM algorithm showed the best prediction results among the multimodal models. The average area under the curve (AUC), accuracy (ACC), sensitivity, and specificity were 0.863, 0.784, 0.731, and 0.826, respectively. Finally, we constructed a comprehensive nomogram using clinical features, a radiomics score and a pathomics score to provide a reference for predicting the risk of early recurrence.

CONCLUSIONS

The multimodal models can be used as a primary tool for oncologists to predict the risk of early recurrence after radical HCC surgery, which will help optimize and personalize treatment strategies.

Proposal of an integrated staging system using albumin-bilirubin grade and serum alpha-fetoprotein values for predicting postoperative prognosis of recurrent hepatocellular carcinoma

BACKGROUND

Because repeat hepatectomy for recurrent hepatocellular carcinoma is a potentially invasive procedure, it is necessary to identify patients who truly benefit from repeat hepatectomy. Albumin-bilirubin grading has been reported to predict survival in patients with hepatocellular carcinoma. However, as prognosis also depends on tumor factors, a staging system that adds tumor factors to albumin-bilirubin grading may lead to a more accurate prognostication in patients with recurrent hepatocellular carcinoma.

METHODS

Albumin-bilirubin grading and serum alpha-fetoprotein levels were combined and the albumin-bilirubin-alpha-fetoprotein score was created ([albumin-bilirubin grading = 1; 1 point, 2 or 3; 2 points] + [alpha-fetoprotein<75 ng/mL, 0 points; ≥5, 1 point]). Patients were classified into three groups, and their characteristics and survival were evaluated. The predictive ability of the albumin-bilirubin-alpha-fetoprotein score was compared with that of the Cancer of the Liver Italian Program and the Japan Integrated Stage scores.

RESULTS

Albumin-bilirubin-alpha-fetoprotein score significantly stratified postoperative survival (albumin-bilirubin-alpha-fetoprotein score = 1/2/3: 5-year recurrence-free survival [%]: 22.4/20.7/0.0, p < 0.001) and showed the highest predictive value for survival among the integrated systems (albumin-bilirubin-alpha-fetoprotein score/Japan Integrated Stage/Cancer of the Liver Italian Program: 0.785/0.708/0.750).

CONCLUSIONS

Albumin-bilirubin-alpha-fetoprotein score is useful for predicting the survival of patients with recurrent hepatocellular carcinoma undergoing repeat hepatectomy.

Preoperative Bone Loss Predicts Decreased Survival Associated with Microvascular Invasion after Resection of Hepatocellular Carcinoma

BACKGROUND

Osteopenia is a well-known risk factor for survival in patients with hepatocellular carcinoma; however, it is unclear whether osteopenia can apply to both genders and how osteopenia is associated with cancer progression. The aim of this study was to elucidate whether osteopenia predicts reduced survival in regression models in both genders and whether osteopenia is associated with the pathological factors associated with reduced survival.

METHODS

This study included 188 consecutive patients who underwent hepatectomy. Bone mineral density was assessed using computed tomography (CT) scan images taken within 3 months before surgery. Non-contrast CT scan images at the level of the 11th thoracic vertebra were used. The cutoff value of osteopenia was calculated using a threshold value of 160 Hounsfield units. Overall survival (OS) curves and recurrence-free survival (RFS) were constructed using the Kaplan-Meier method, as was a log-rank test for survival. The hazard ratio and 95% confidence interval for overall survival were calculated using Cox's proportional hazard model.

RESULTS

In the regression analysis, age predicted bone mineral density. The association in females was greater than that in males. The OS and RFS of osteopenia patients were shorter than those for non-osteopenia patients. According to univariate and multivariate analyses, osteopenia was an independent risk factor for OS and RFS. The sole pathological factor associated with osteopenia was microvascular portal vein invasion.

CONCLUSION

Models suggest that osteopenia may predict decreased OS and RFS in patients undergoing resection of hepatocellular carcinoma due to the mechanisms mediated via microvascular portal vein invasion.

Deep learning and digital pathology powers prediction of HCC development in steatotic liver disease

BACKGROUND AND AIMS

Identifying patients with steatotic liver disease who are at a high risk of developing HCC remains challenging. We present a deep learning (DL) model to predict HCC development using hematoxylin and eosin-stained whole-slide images of biopsy-proven steatotic liver disease.

APPROACH AND RESULTS

We included 639 patients who did not develop HCC for ≥7 years after biopsy (non-HCC class) and 46 patients who developed HCC <7 years after biopsy (HCC class). Paired cases of the HCC and non-HCC classes matched by biopsy date and institution were used for training, and the remaining nonpaired cases were used for validation. The DL model was trained using deep convolutional neural networks with 28,000 image tiles cropped from whole-slide images of the paired cases, with an accuracy of 81.0% and an AUC of 0.80 for predicting HCC development. Validation using the nonpaired cases also demonstrated a good accuracy of 82.3% and an AUC of 0.84. These results were comparable to the predictive ability of logistic regression model using fibrosis stage. Notably, the DL model also detected the cases of HCC development in patients with mild fibrosis. The saliency maps generated by the DL model highlighted various pathological features associated with HCC development, including nuclear atypia, hepatocytes with a high nuclear-cytoplasmic ratio, immune cell infiltration, fibrosis, and a lack of large fat droplets.

CONCLUSIONS

The ability of the DL model to capture subtle pathological features beyond fibrosis suggests its potential for identifying early signs of hepatocarcinogenesis in patients with steatotic liver disease.

Role of Fibrinogen-like Protein 1 in Tumor Recurrence Following Hepatectomy

Partial hepatectomy is a first-line treatment for hepatocellular carcinoma. Within 2 weeks following partial hepatectomy, specific molecular pathways are activated to promote liver regeneration. Nevertheless, residual microtumors may also exploit these pathways to reappear and metastasize. Therapeutically targeting molecules that are differentially regulated between normal cells and malignancies, such as fibrinogen-like protein 1 (FGL1), appears to be an effective approach. The potential functions of FGL1 in both regenerative and malignant cells are discussed within the ambit of this review. While FGL1 is normally elevated in regenerative hepatocytes, it is normally downregulated in malignant cells. Hepatectomy does indeed upregulate FGL1 by increasing the release of transcription factors that promote FGL1, including HNF-1α and STAT3, and inflammatory effectors, such as TGF-β and IL6. This, in turn, stimulates certain proliferative pathways, including EGFR/Src/ERK. Hepatectomy alters the phase transition of highly differentiated hepatocytes from G0 to G1, thereby transforming susceptible cells into cancerous ones. Activation of the PI3K/Akt/mTOR pathway by FGL1 allele loss on chromosome 8, a tumor suppressor area, may also cause hepatocellular carcinoma. Interestingly, FGL1 is specifically expressed in the liver via HNF-1α histone acetylase activity, which triggers lipid metabolic reprogramming in malignancies. FGL1 might also be involved in other carcinogenesis processes such as hypoxia, epithelial-mesenchymal transition, immunosuppression, and sorafenib-mediated drug resistance. This study highlights a research gap in these disciplines and the necessity for additional research on FGL1 function in the described processes.

RNF125‑mediated ubiquitination of MCM6 regulates the proliferation of human liver hepatocellular carcinoma cells

Hepatocellular carcinoma (HCC) is the third leading cause of cancer-associated mortality worldwide. Minichromosome maintenance proteins (MCMs), particularly MCM2-7, are upregulated in various cancers, including HCC. The aim of the present study was to investigate the role of MCM2-7 in human liver HCC (LIHC) and the regulation of the protein homeostasis of MCM6 by a specific E3 ligase. Bioinformatics analyses demonstrated that MCM2-7 were highly expressed in LIHC compared with corresponding normal tissues at the mRNA and protein levels, and patients with LIHC and high mRNA expression levels of MCM2, MCM3, MCM6 and MCM7 had poor overall survival rates. Cell Counting Kit-8 and colony formation assays revealed that the knockdown of MCM2, MCM3, MCM6 or MCM7 in Huh7 and Hep3B HCC cells inhibited cell proliferation and colony formation. In addition, pull-down, co-immunoprecipitation and ubiquitination assays demonstrated that RNF125 interacts with MCM6 and mediates its ubiquitination. Furthermore, co-transfection experiments indicated that RNF125 promoted the proliferation of HCC cells mainly through MCM6. In summary, the present study suggests that the RNF125-MCM6 axis plays an important role in the regulation of HCC cell proliferation and is a promising therapeutic target for the treatment of LIHC.

Radiomics analysis of R2* maps to predict early recurrence of single hepatocellular carcinoma after hepatectomy

Objectives

This study aimed to evaluate the effectiveness of radiomics analysis with R2* maps in predicting early recurrence (ER) in single hepatocellular carcinoma (HCC) following partial hepatectomy.

Methods

We conducted a retrospective analysis involving 202 patients with surgically confirmed single HCC having undergone preoperative magnetic resonance imaging between 2018 and 2021 at two different institutions. 126 patients from Institution 1 were assigned to the training set, and 76 patients from Institution 2 were assigned to the validation set. A least absolute shrinkage and selection operator (LASSO) regularization was conducted to operate a logistic regression, then features were identified to construct a radiomic score (Rad-score). Uni- and multi-variable tests were used to assess the correlations of clinicopathological features and Rad-score with ER. We then established a combined model encompassing the optimal Rad-score and clinical-pathological risk factors. Additionally, we formulated and validated a predictive nomogram for predicting ER in HCC. The nomogram's discrimination, calibration, and clinical utility were thoroughly evaluated.

Results

Multivariable logistic regression revealed the Rad-score, microvascular invasion (MVI), and α fetoprotein (AFP) level > 400 ng/mL as significant independent predictors of ER in HCC. We constructed a nomogram based on these significant factors. The areas under the receiver operator characteristic curve of the nomogram and precision-recall curve were 0.901 and 0.753, respectively, with an F1 score of 0.831 in the training set. These values in the validation set were 0.827, 0.659, and 0.808.

Conclusion

The nomogram that integrates the radiomic score, MVI, and AFP demonstrates high predictive efficacy for estimating the risk of ER in HCC. It facilitates personalized risk classification and therapeutic decision-making for HCC patients.

From prediction to prevention: Machine learning revolutionizes hepatocellular carcinoma recurrence monitoring

In this editorial, we comment on the article by Zhang et al entitled Development of a machine learning-based model for predicting the risk of early postoperative recurrence of hepatocellular carcinoma. Hepatocellular carcinoma (HCC), which is characterized by high incidence and mortality rates, remains a major global health challenge primarily due to the critical issue of postoperative recurrence. Early recurrence, defined as recurrence that occurs within 2 years posttreatment, is linked to the hidden spread of the primary tumor and significantly impacts patient survival. Traditional predictive factors, including both patient- and treatment-related factors, have limited predictive ability with respect to HCC recurrence. The integration of machine learning algorithms is fueled by the exponential growth of computational power and has revolutionized HCC research. The study by Zhang et al demonstrated the use of a groundbreaking preoperative prediction model for early postoperative HCC recurrence. Chall-enges persist, including sample size constraints, issues with handling data, and the need for further validation and interpretability. This study emphasizes the need for collaborative efforts, multicenter studies and comparative analyses to validate and refine the model. Overcoming these challenges and exploring innovative approaches, such as multi-omics integration, will enhance personalized oncology care. This study marks a significant stride toward precise, effi-cient, and personalized oncology practices, thus offering hope for improved patient outcomes in the field of HCC treatment.

Systemic Drugs for Hepatocellular Carcinoma: What Do Recent Clinical Trials Reveal About Sequencing and the Emerging Complexities of Clinical Decisions?

Liver cancer was the fourth leading cause of cancer death in 2015 with increasing incidence between 1990 and 2015. Orthotopic liver transplantation, surgical resection and ablation comprise the only curative therapy options. However, due to the late manifestation of clinical symptoms, many patients present with intermediate or advanced disease, resulting in no curative treatment option being available. Whereas intermediate-stage hepatocellular carcinoma (HCC) is usually still addressable by transarterial chemoembolization (TACE), advanced-stage HCC is amenable only to pharmacological treatments. Conventional cytotoxic agents failed demonstrating relevant effect on survival also because their use was severely limited by the mostly underlying insufficient liver function. For a decade, tyrosine kinase inhibitor (TKI) sorafenib was the only systemic therapy that proved to have a clinically relevant effect in the treatment of advanced HCC. In recent years, the number of substances for systemic treatment of advanced HCC has increased enormously. In addition to tyrosine kinase inhibitors, immune checkpoint inhibitors (ICI) and antiangiogenic drugs are increasingly being applied. The combination of anti-programmed death ligand 1 (PD-L1) antibody atezolizumab and anti-vascular endothelial growth factor (VEGF) antibody bevacizumab has become the new standard of care for advanced HCC due to its remarkable response rates. This requires more and more complex clinical decisions regarding tumor therapy. This review aims at summarizing recent developments in systemic therapy, considering data on first- and second-line treatment, use in the neoadjuvant and adjuvant setting and combination with locoregional procedures.

The Molecular Twin artificial-intelligence platform integrates multi-omic data to predict outcomes for pancreatic adenocarcinoma patients

Contemporary analyses focused on a limited number of clinical and molecular biomarkers have been unable to accurately predict clinical outcomes in pancreatic ductal adenocarcinoma. Here we describe a precision medicine platform known as the Molecular Twin consisting of advanced machine-learning models and use it to analyze a dataset of 6,363 clinical and multi-omic molecular features from patients with resected pancreatic ductal adenocarcinoma to accurately predict disease survival (DS). We show that a full multi-omic model predicts DS with the highest accuracy and that plasma protein is the top single-omic predictor of DS. A parsimonious model learning only 589 multi-omic features demonstrated similar predictive performance as the full multi-omic model. Our platform enables discovery of parsimonious biomarker panels and performance assessment of outcome prediction models learning from resource-intensive panels. This approach has considerable potential to impact clinical care and democratize precision cancer medicine worldwide.

Factors affecting overall survival and disease-free survival after surgery for hepatocellular carcinoma: a nomogram-based prognostic model-a Western European multicenter study

Few studies have assessed the clinical implications of the combination of different prognostic indicators for overall survival (OS) and disease-free survival (DFS) of resected hepatocellular carcinoma (HCC). This study aimed to evaluate the prognostic factors in HCC patients for OS and DFS outcomes and establish a nomogram-based prognostic model to predict the DFS of HCC. A multicenter, retrospective European study was conducted through the collection of data on 413 consecutive treated patients with a first diagnosis of HCC between January 2010 and December 2020. Univariate and multivariate Cox regression analyses were performed to identify all independent risk factors for OS and DFS outcomes. A nomogram prognostic staging model was subsequently established for DFS and its precision was verified internally by the concordance index (C-Index) and externally by calibration curves. For OS, multivariate Cox regression analysis indicated Child-Pugh B7 score (HR 4.29; 95% CI 1.74-10.55; p = 0.002) as an independent prognostic factor, along with Barcelona Clinic Liver Cancer (BCLC) stage ≥ B (HR 1.95; 95% CI 1.07-3.54; p = 0.029), microvascular invasion (MVI) (HR 2.54; 95% CI 1.38-4.67; p = 0.003), R1/R2 resection margin (HR 1.57; 95% CI 0.85-2.90; p = 0.015), and Clavien-Dindo Grade 3 or more (HR 2.73; 95% CI 1.44-5.18; p = 0.002). For DFS, multivariate Cox regression analysis indicated BCLC stage ≥ B (HR 2.15; 95% CI 1.34-3.44; p = 0.002) as an independent prognostic factor, along with multiple nodules (HR 2.04; 95% CI 1.25-3.32; p = 0.004), MVI (HR 1.81; 95% CI 1.19-2.75; p = 0.005), satellite nodules (HR 1.63; 95% CI 1.09-2.45; p = 0.018), and R1/R2 resection margin (HR 3.39; 95% CI 2.19-5.25; < 0.001). The C-Index of the nomogram, tailored based on the previous significant factors, showed good accuracy (0.70). Internal and external calibration curves for the probability of DFS rate showed optimal consistency and fit well between the nomogram-based prediction and actual observations. MVI and R1/R2 resection margins should be considered as significant OS and DFS predictors, while satellite nodules should be included as a significant DFS predictor. The nomogram-based prognostic model for DFS provides a more effective prognosis assessment for resected HCC patients, allowing for individualized treatment plans.

Current Status and Analysis of Machine Learning in Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is a common tumor. Although the diagnosis and treatment of HCC have made great progress, the overall prognosis remains poor. As the core component of artificial intelligence, machine learning (ML) has developed rapidly in the past decade. In particular, ML has become widely used in the medical field, and it has helped in the diagnosis and treatment of cancer. Different algorithms of ML have different roles in diagnosis, treatment, and prognosis. This article reviews recent research, explains the application of different ML models in HCC, and provides suggestions for follow-up research.

Cathelicidin Antimicrobial Peptide Acts as a Tumor Suppressor in Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is associated with high rates of metastasis and recurrence, and is one of the most common causes of cancer-associated death worldwide. This study examined the protein changes within circulating exosomes in patients with HCC against those in healthy people using isobaric tags for a relative or absolute quantitation (iTRAQ)-based quantitative proteomics analysis. The protein levels of von Willebrand factor (VWF), cathelicidin antimicrobial peptide (CAMP), and proteasome subunit beta type-2 (PSMB2) were altered in HCC. The increased levels of VWF and PSMB2 but decreased CAMP levels in the serum of patients with HCC were validated by enzyme-linked immunosorbent assays. The level of CAMP (the only cathelicidin found in humans) also decreased in the circulating exosomes and buffy coat of the HCC patients. The serum with reduced levels of CAMP protein in the HCC patients increased the cell proliferation of Huh-7 cells; this effect was reduced following the addition of CAMP protein. The depletion of CAMP proteins in the serum of healthy people enhances the cell proliferation of Huh-7 cells. In addition, supplementation with synthetic CAMP reduces cell proliferation in a dose-dependent manner and significantly delays G1-S transition in Huh-7 cells. This implies that CAMP may act as a tumor suppressor in HCC.

The mechanism of ferroptosis and its related diseases

Ferroptosis, a regulated form of cellular death characterized by the iron-mediated accumulation of lipid peroxides, provides a novel avenue for delving into the intersection of cellular metabolism, oxidative stress, and disease pathology. We have witnessed a mounting fascination with ferroptosis, attributed to its pivotal roles across diverse physiological and pathological conditions including developmental processes, metabolic dynamics, oncogenic pathways, neurodegenerative cascades, and traumatic tissue injuries. By unraveling the intricate underpinnings of the molecular machinery, pivotal contributors, intricate signaling conduits, and regulatory networks governing ferroptosis, researchers aim to bridge the gap between the intricacies of this unique mode of cellular death and its multifaceted implications for health and disease. In light of the rapidly advancing landscape of ferroptosis research, we present a comprehensive review aiming at the extensive implications of ferroptosis in the origins and progress of human diseases. This review concludes with a careful analysis of potential treatment approaches carefully designed to either inhibit or promote ferroptosis. Additionally, we have succinctly summarized the potential therapeutic targets and compounds that hold promise in targeting ferroptosis within various diseases. This pivotal facet underscores the burgeoning possibilities for manipulating ferroptosis as a therapeutic strategy. In summary, this review enriched the insights of both investigators and practitioners, while fostering an elevated comprehension of ferroptosis and its latent translational utilities. By revealing the basic processes and investigating treatment possibilities, this review provides a crucial resource for scientists and medical practitioners, aiding in a deep understanding of ferroptosis and its effects in various disease situations.

Pre-operative gamma-glutamyl transferase levels predict outcomes in hepatitis B-related hepatocellular carcinoma after curative resection

BACKGROUND

Surgical resection is a curative therapy for early-stage hepatocellular carcinoma (HCC); however, HCC recurrence is not uncommon. Identifying outcome predictors helps to manage the disease. Gamma-glutamyl transferase (GGT) may predict the development of HCC, but its role to predict the outcomes after surgical resection of HCC was unclear. This study aimed to investigate pre-operative GGT levels for outcome prediction in patients with hepatitis B virus (HBV)-related HCC.

METHODS

We conducted a retrospective cohort study to include patients with HBV-related HCC receiving surgical resection. Clinical information, HCC characteristics and usage of antiviral therapy were collected. A time-dependent Cox proportional hazard regression analysis were used to predict HCC recurrence and survival.

RESULTS

A total of 699 consecutive patients with HBV-related HCC who received surgical resection with curative intent between 2004 and 2013 were included. After a median of 4.4 years, 266 (38%) patients had HCC recurrence. Pre-operative GGT positively correlated with cirrhosis, tumor burden and significantly increased in patients to develop HCC recurrence. Multivariable analysis demonstrated that pre-operative GGT ≥38 U/L increased 57% risk (hazard ratio [HR]: 1.57, 95% confidence interval [CI]: 1.20-2.06) of recurrent HCC after adjustment for confounding factors. Specifically, pre-operative GGT ≥38 U/L predicted early (<2 years) HCC recurrence (HR: 1.94, 95% CI: 1.30-2.89). Moreover, pre-operative GGT ≥38 U/L predicted all-cause mortality (HR: 1.73, 95% CI: 1.06-2.84) after surgery.

CONCLUSION

Pre-operative GGT levels ≥38 U/L independently predict high risks of HCC recurrence and all-cause mortality in HBV-related HCC patients receiving surgical resection.

Artificial Intelligence-Driven Morphology-Based Enrichment of Malignant Cells from Body Fluid

Cell morphology is a fundamental feature used to evaluate patient specimens in pathologic analysis. However, traditional cytopathology analysis of patient effusion samples is limited by low tumor cell abundance coupled with the high background of nonmalignant cells, restricting the ability of downstream molecular and functional analyses to identify actionable therapeutic targets. We applied the Deepcell platform that combines microfluidic sorting, brightfield imaging, and real-time deep learning interpretations based on multidimensional morphology to enrich carcinoma cells from malignant effusions without cell staining or labels. Carcinoma cell enrichment was validated with whole genome sequencing and targeted mutation analysis, which showed a higher sensitivity for detection of tumor fractions and critical somatic variant mutations that were initially at low levels or undetectable in presort patient samples. Our study demonstrates the feasibility and added value of supplementing traditional morphology-based cytology with deep learning, multidimensional morphology analysis, and microfluidic sorting.

FOS-Like Antigen 1 Expression Was Associated With Survival of Hepatocellular Carcinoma Patients

BACKGROUND

Early diagnosis and proper management of hepatocellular carcinoma (HCC) improve patient prognosis. Several studies attempted to discover new genes to understand the pathogenesis and identify the prognostic and predictive factors in HCC patients, to improve patient's overall survival (OS) and maintain their physical and social activity. The transcription factor FOS-like antigen 1 (FOSL1) acts as one of the important prognostic factors in different tumors, and its overexpression correlates with tumors' progression and worse patient survival. However, its expression and molecular mechanisms underlying its dysregulation in human HCC remain poorly understood. Our study was conducted to evaluate the expression of FOSL1 in HCC tissues and its relationship with various clinicopathological parameters besides OS.

METHODS

This study is a retrospective cohort study conducted among 113 patients with a proven diagnosis of HCC, who underwent tumor resection and received treatment at South Egypt Cancer Institute. Immunohistochemistry for FOSL1 expression and survival curves were conducted followed by statistical analysis.

RESULTS

HCC occurred at older age group and affected males more than females. There was a statistically significant correlation between combined cytoplasmic and nuclear expression of FOSL1 and worse prognosis in HCC patients. There was a statistically significant correlation of FOSL1 expression with histological grade, lymphovascular embolization, and tumor budding where high expression indicated potential deterioration of HCC patients. There was statistically significant correlation between tumor size, tumor grade and FOSL1 expression with the cumulative OS.

CONCLUSIONS

Combined cytoplasmic and nuclear FOSL1 expression has significant prognostic association with HCC and diagnostic importance, as it can identify cirrhosis and premalignant lesions that can progress to HCC. Furthermore, Kaplan-Meier survival analysis found that overexpressed FOSL1 was correlated with poor OS.

Recurrence prediction in clear cell renal cell carcinoma using machine learning of quantitative nuclear features

The recurrence of non-metastatic renal cell carcinoma (RCC) may occur early or late after surgery. This study aimed to develop a recurrence prediction machine learning model based on quantitative nuclear morphologic features of clear cell RCC (ccRCC). We investigated 131 ccRCC patients who underwent nephrectomy (T1-3N0M0). Forty had recurrence within 5 years and 22 between 5 and 10 years; thirty-seven were recurrence-free during 5-10 years and 32 were for more than 10 years. We extracted nuclear features from regions of interest (ROIs) using a digital pathology technique and used them to train 5- and 10-year Support Vector Machine models for recurrence prediction. The models predicted recurrence at 5/10 years after surgery with accuracies of 86.4%/74.1% for each ROI and 100%/100% for each case, respectively. By combining the two models, the accuracy of the recurrence prediction within 5 years was 100%. However, recurrence between 5 and 10 years was correctly predicted for only 5 of the 12 test cases. The machine learning models showed good accuracy for recurrence prediction within 5 years after surgery and may be useful for the design of follow-up protocols and patient selection for adjuvant therapy.

Early warning and diagnosis of liver cancer based on dynamic network biomarker and deep learning

Background

Early detection of complex diseases like hepatocellular carcinoma remains challenging due to their network-driven pathology. Dynamic network biomarkers (DNB) based on monitoring changes in molecular correlations may enable earlier predictions. However, DNB analysis often overlooks disease heterogeneity.

Methods

We integrated DNB analysis with graph convolutional neural networks (GCN) to identify critical transitions during hepatocellular carcinoma development in a mouse model. A DNB-GCN model was constructed using transcriptomic data and gene expression levels as node features.

Results

DNB analysis identified a critical transition point at 7 weeks of age despite histological examinations being unable to detect cancerous changes at that time point. The DNB-GCN model achieved 100% accuracy in classifying healthy and cancerous mice, and was able to accurately predict the health status of newly introduced mice.

Conclusion

The integration of DNB analysis and GCN demonstrates potential for the early detection of complex diseases by capturing network structures and molecular features that conventional biomarker discovery methods overlook. The approach warrants further development and validation.

A convolutional neural network STIFMap reveals associations between stromal stiffness and EMT in breast cancer

Intratumor heterogeneity associates with poor patient outcome. Stromal stiffening also accompanies cancer. Whether cancers demonstrate stiffness heterogeneity, and if this is linked to tumor cell heterogeneity remains unclear. We developed a method to measure the stiffness heterogeneity in human breast tumors that quantifies the stromal stiffness each cell experiences and permits visual registration with biomarkers of tumor progression. We present Spatially Transformed Inferential Force Map (STIFMap) which exploits computer vision to precisely automate atomic force microscopy (AFM) indentation combined with a trained convolutional neural network to predict stromal elasticity with micron-resolution using collagen morphological features and ground truth AFM data. We registered high-elasticity regions within human breast tumors colocalizing with markers of mechanical activation and an epithelial-to-mesenchymal transition (EMT). The findings highlight the utility of STIFMap to assess mechanical heterogeneity of human tumors across length scales from single cells to whole tissues and implicates stromal stiffness in tumor cell heterogeneity.

Artificial Intelligence-Based Opportunities in Liver Pathology-A Systematic Review

BACKGROUND

Artificial Intelligence (AI)-based Deep Neural Networks (DNNs) can handle a wide range of applications in image analysis, ranging from automated segmentation to diagnostic and prediction. As such, they have revolutionized healthcare, including in the liver pathology field.

OBJECTIVE

The present study aims to provide a systematic review of applications and performances provided by DNN algorithms in liver pathology throughout the Pubmed and Embase databases up to December 2022, for tumoral, metabolic and inflammatory fields.

RESULTS

42 articles were selected and fully reviewed. Each article was evaluated through the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool, highlighting their risks of bias.

CONCLUSIONS

DNN-based models are well represented in the field of liver pathology, and their applications are diverse. Most studies, however, presented at least one domain with a high risk of bias according to the QUADAS-2 tool. Hence, DNN models in liver pathology present future opportunities and persistent limitations. To our knowledge, this review is the first one solely focused on DNN-based applications in liver pathology, and to evaluate their bias through the lens of the QUADAS2 tool.

Combining a protein-targeting small molecule and a thiol-targeting small molecule for detecting a serum risk marker of liver tumor recurrence

This work proposes a novel bioassay designed to detect the 2B receptor of serotonin in serum samples, which can serve as a risk marker for cancer recurrence after surgical resection. Traditional methods for detecting this marker are often costly and time-consuming, requiring specialized reagents and equipment. The new bioassay is designed to enable direct and reagent-less detection of the 2B receptor in serum samples, without the need of antibodies or enzymes. The assay uses a small molecule ligand for the 2B receptor combined with a thiol-targeting fluorescent dye on a compact peptide-based molecular frame. This design allows for a rapid and specific readout of the fluorescent signal upon probe-protein interaction. In addition, the covalent biosensing process used in the assay allows for signal enhancement by electrochemical cross-linking of serum proteins. The bioassay was successfully used to detect the 2B receptor in serum samples from hepatocarcinoma patients, indicating its potential as a powerful tool for early cancer detection and monitoring.

A Paradigm Shift in Primary Liver Cancer Therapy Utilizing Genomics, Molecular Biomarkers, and Artificial Intelligence

Primary liver cancer is the sixth most common cancer worldwide and the third leading cause of cancer-related death. Conventional therapies offer limited survival benefit despite improvements in locoregional liver-directed therapies, which highlights the underlying complexity of liver cancers. This review explores the latest research in primary liver cancer therapies, focusing on developments in genomics, molecular biomarkers, and artificial intelligence. Attention is also given to ongoing research and future directions of immunotherapy and locoregional therapies of primary liver cancers.

Significance of nucleologenesis, ribogenesis, and nucleolar proteome in the pathogenesis and recurrence of hepatocellular carcinoma

INTRODUCTION

Emerging evidence suggests that enhanced ribosome biogenesis, increased size, and quantitative distribution of nucleoli are associated with dysregulated transcription, which in turn drives a cell into aberrant cellular proliferation and malignancy. Nucleolar alterations have been considered a prognostic histological marker for aggressive tumors. More recently, advancements in the understanding of chromatin network (nucleoplasm viscosity) regulated liquid-liquid phase separation mechanism of nucleolus formation and their multifunctional role shed light on other regulatory processes, apart from ribosomal biogenesis of the nucleolus.

AREAS COVERED

Using hepatocellular carcinoma as a model to study the role of nucleoli in tumor progression, we review the potential of nucleolus coalescence in the onset and development of tumors through non-ribosomal biogenesis pathways, thereby providing new avenues for early diagnosis and cancer therapy.

EXPERT OPINION

Molecular-based classifications have failed to identify the nucleolar-based molecular targets that facilitate cell-cycle progression. However, the algorithm-based tumor risk identification with high-resolution medical images suggests prominent nucleoli, karyotheca, and increased nucleus/cytoplasm ratio as largely associated with tumor recurrence. Nonetheless, the role of the non-ribosomal functions of nucleoli in tumorigenesis remains elusive. This clearly indicates the lacunae in the study of the nucleolar proteins pertaining to cancer. [Figure: see text].

ScRNA-seq revealed targeting regulator of G protein signaling 1 to mediate regulatory T cells in Hepatocellular carcinoma

BACKGROUND

Regulatory T cells (Tregs) are central to determine immune response outcomes, thus targeting Tregs for immunotherapy is a promising strategy against tumor development and metastasis.

OBJECTIVES

The objective of this study was to identify genes for targeting Tregs to improve the outcome of HCC.

METHODS

We integrated expression data from different samples to remove batch effects and further applied embedding function in Scanpy to conduct sub-clustering of CD4+ T cells in HCC for each of two independent scRNA-seq data. The activity of transcription factors (TFs) was inferred by DoRothEA. Gene expression network analysis was performed in WGCNA R package. We finally used R packages (survminer and survival) to conduct survival analysis. Multiplex immunofluorescence analysis was performed to validate the result from bioinformatic analyses.

RESULTS

We found that regulator of G protein signaling 1 (RGS1) expression was significantly elevated in Tregs compared to other CD4+ T cells in two independent public scRNA-seq datasets, and increased RGS1 predicted inferior clinical outcome of HCC patients. Multiplex immunofluorescence analysis supported that the higher expression of RGS1 in HCC Tregs in tumor tissue compared to it in adjacent tissue. Moreover, RGS1 expression in Tregs was positively correlated with the expression of marker genes of Tregs, C-X-C chemokine receptor 4 (CXCR4), and three CXCR4-dependent genes in both scRNA-seq and bulk RNA-seq data. We further identified that these three genes were selectively expressed in Tregs as compared to other CD4+ T cells. The activities of two transcription factors, recombination signal binding protein for immunoglobulin kappa J region (RBPJ) and yin yang 1 (YY1), were significantly different in HCC Tregs with RGS1 high and RGS1 low.

CONCLUSIONS

Our findings suggested that RGS1 may regulate Treg function possibly through CXCR4 signaling and RGS1 could be a potential target to improve responses for immunotherapy in HCC.

"SMART" cytology: The next generation cytology for precision diagnosis

Cytology plays an important role in diagnosing and managing human diseases, especially cancer, as it is often a simple, low cost yet effective, and non-invasive or minimally invasive diagnostic tool. However, traditional morphology-based cytology practice has limitations, especially in the era of precision diagnosis. Recently there have been tremendous efforts devoted to apply computational tools and to perform molecular analysis on cytological samples for a variety of clinical purposes. Now is probably the appropriate juncture to integrate morphology, machine learning, and molecular analysis together and transform cytology from a morphology-driven practice to the next level - "SMART" Cytology. In this article we will provide a rather brief review of the relevant works for computational analysis on cytology samples, focusing on single-cell-based multiplex quantitative analysis of biomarkers, and introduce the conceptual framework of "SMART (Single cell, Multiplex, AI-driven, and Real Time)" Cytology.

Potential Molecular Targeted Therapy for Unresectable Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is one of the most prevalent and lethal cancers, representing a serious worldwide health concern. The recurrence incidence of hepatocellular carcinoma (HCC) following surgery or ablation is as high as 70%. Thus, the clinical applicability of standard surgery and other locoregional therapy to improve the outcomes of advanced HCC is restricted and far from ideal. The registered trials did not identify a treatment that prolonged recurrence-free survival, the primary outcome of the majority of research. Several investigator-initiated trials have demonstrated that various treatments extend patients' recurrence-free or overall survival after curative therapies. In the past decade, targeted therapy has made significant strides in the treatment of advanced HCC. These targeted medicines produce antitumour effects via specific signals, such as anti-angiogenesis or advancement of the cell cycle. As a typical systemic treatment option, it significantly improves the prognosis of this fatal disease. In addition, the combination of targeted therapy with an immune checkpoint inhibitor is redefining the paradigm of advanced HCC treatment. In this review, we focused on the role of approved targeted medicines and potential therapeutic targets in unresectable HCC.

Prediction of chemotherapy-related complications in pediatric oncology patients: artificial intelligence and machine learning implementations

Although the overall incidence of pediatric oncological diseases tends to increase over the years, it is among the rare diseases of the pediatric population. The diagnosis, treatment, and healthcare management of this group of diseases are important. Prevention of treatment-related complications is vital for patients, particularly in the pediatric population. Nowadays, the use of artificial intelligence and machine learning technologies in the management of oncological diseases is becoming increasingly important. With the advancement of software technologies, improvements have been made in the early diagnosis of risk groups in oncological diseases, in radiology, pathology, and imaging technologies, in cancer staging and management. In addition, these technologies can be used to predict the outcome in chemotherapy treatment of oncological diseases. In this context, this study identifies artificial intelligence and machine learning methods used in the prediction of complications due to chemotherapeutic agents used in childhood cancer treatment. For this purpose, the concepts of artificial intelligence and machine learning are explained in this review. A general framework for the use of machine learning in healthcare and pediatric oncology has been drawn and examples of studies conducted on this topic in pediatric oncology have been given. IMPACT: Artificial intelligence and machine learning are advanced tools that can be used to predict chemotherapy-related complications. Algorithms can assist clinicians' decision-making processes in the management of complications. Although studies are using these methods, there is a need to increase the number of studies on artificial intelligence applications in pediatric clinics.

Role of Machine Learning in Precision Oncology: Applications in Gastrointestinal Cancers

Gastrointestinal (GI) cancers, consisting of a wide spectrum of pathologies, have become a prominent health issue globally. Despite medical imaging playing a crucial role in the clinical workflow of cancers, standard evaluation of different imaging modalities may provide limited information. Accurate tumor detection, characterization, and monitoring remain a challenge. Progress in quantitative imaging analysis techniques resulted in "radiomics", a promising methodical tool that helps to personalize diagnosis and treatment optimization. Radiomics, a sub-field of computer vision analysis, is a bourgeoning area of interest, especially in this era of precision medicine. In the field of oncology, radiomics has been described as a tool to aid in the diagnosis, classification, and categorization of malignancies and to predict outcomes using various endpoints. In addition, machine learning is a technique for analyzing and predicting by learning from sample data, finding patterns in it, and applying it to new data. Machine learning has been increasingly applied in this field, where it is being studied in image diagnosis. This review assesses the current landscape of radiomics and methodological processes in GI cancers (including gastric, colorectal, liver, pancreatic, neuroendocrine, GI stromal, and rectal cancers). We explain in a stepwise fashion the process from data acquisition and curation to segmentation and feature extraction. Furthermore, the applications of radiomics for diagnosis, staging, assessment of tumor prognosis and treatment response according to different GI cancer types are explored. Finally, we discussed the existing challenges and limitations of radiomics in abdominal cancers and investigate future opportunities.

Postoperative Adjuvant Hepatic Arterial Infusion Chemotherapy With FOLFOX in Hepatocellular Carcinoma With Microvascular Invasion: A Multicenter, Phase III, Randomized Study

PURPOSE

To report the efficacy and safety of postoperative adjuvant hepatic arterial infusion chemotherapy (HAIC) with 5-fluorouracil and oxaliplatin (FOLFOX) in hepatocellular carcinoma (HCC) patients with microvascular invasion (MVI).

PATIENTS AND METHODS

In this randomized, open-label, multicenter trial, histologically confirmed HCC patients with MVI were randomly assigned (1:1) to receive adjuvant FOLFOX-HAIC (treatment group) or routine follow-up (control group). The primary end point was disease-free survival (DFS) by intention-to-treat (ITT) analysis while secondary end points were overall survival, recurrence rate, and safety.

RESULTS

Between June 2016 and August 2021, a total of 315 patients (ITT population) at five centers were randomly assigned to the treatment group (n = 157) or the control group (n = 158). In the ITT population, the median DFS was 20.3 months (95% CI, 10.4 to 30.3) in the treatment group versus 10.0 months (95% CI, 6.8 to 13.2) in the control group (hazard ratio, 0.59; 95% CI, 0.43 to 0.81; P = .001). The overall survival rates at 1 year, 2 years, and 3 years were 93.8% (95% CI, 89.8 to 98.1), 86.4% (95% CI, 80.0 to 93.2), and 80.4% (95% CI, 71.9 to 89.9) for the treatment group and 92.0% (95% CI, 87.6 to 96.7), 86.0% (95% CI, 79.9 to 92.6), and 74.9% (95% CI, 65.5 to 85.7) for the control group (hazard ratio, 0.64; 95% CI, 0.36 to 1.14; P = .130), respectively. The recurrence rates were 40.1% (63/157) in the treatment group and 55.7% (88/158) in the control group. Majority of the adverse events were grade 0-1 (83.8%), with no treatment-related death in both groups.

CONCLUSION

Postoperative adjuvant HAIC with FOLFOX significantly improved the DFS benefits with acceptable toxicities in HCC patients with MVI.

Ensemble learning based on efficient features combination can predict the outcome of recurrence-free survival in patients with hepatocellular carcinoma within three years after surgery

Preoperative prediction of recurrence outcome in hepatocellular carcinoma (HCC) facilitates physicians' clinical decision-making. Preoperative imaging and related clinical baseline data of patients are valuable for evaluating prognosis. With the widespread application of machine learning techniques, the present study proposed the ensemble learning method based on efficient feature representations to predict recurrence outcomes within three years after surgery. Radiomics features during arterial phase (AP) and clinical data were selected for training the ensemble models. In order to improve the efficiency of the process, the lesion area was automatically segmented by 3D U-Net. It was found that the mIoU of the segmentation model was 0.8874, and the Light Gradient Boosting Machine (LightGBM) was the most superior, with an average accuracy of 0.7600, a recall of 0.7673, a F1 score of 0.7553, and an AUC of 0.8338 when inputting radiomics features during AP and clinical baseline indicators. Studies have shown that the proposed strategy can relatively accurately predict the recurrence outcome within three years, which is helpful for physicians to evaluate individual patients before surgery.

Utility of Sonazoid-Enhanced Ultrasound for the Macroscopic Classification of Hepatocellular Carcinoma: A Meta-analysis

We assessed the diagnostic value of Sonazoid-enhanced ultrasound (SEUS) in determining the macroscopic classification of hepatocellular carcinoma (HCC) because of its strong relevance to the poor prognosis of the non-simple nodular (non-SN) type. The PubMed, EMBASE, Web of Science and Cochrane Library databases were searched for studies investigating patients who underwent surgery for HCC after undergoing SEUS pre-operatively. Five studies involving a total of 334 patients met the inclusion criteria. The summary sensitivity and specificity were 0.74 (95% confidence interval [CI]: 0.63-0.83) and 0.92 (95% CI: 0.82-0.97), respectively. The positive and negative likelihood ratios of SEUS for determining the macroscopic classification of HCC in Kupffer phase were 9.21 (95% CI: 4.02-21.13) and 0.28 (95% CI: 0.19-0.41), respectively. The diagnostic odds ratio of SEUS for determining the macroscopic classification of HCC was 34.2 (95% CI: 11.64-100.51), and the area under the summary receiver operating characteristic curve was 0.87 (95% CI: 0.84-0.90). Subgroup analysis suggested that small HCCs (≤30 mm) and studies including fewer than 70 patients may be associated with a higher diagnostic odds ratio than the corresponding subsets. SEUS had moderate diagnostic value for determining the macroscopic classification of HCC in the Kupffer phase.

Early extrahepatic recurrence as a pivotal factor for survival after hepatocellular carcinoma resection: A 15-year observational study

BACKGROUND

Surgical resection is one of the most widely used modalities for the treatment of hepatocellular carcinoma (HCC). Early extrahepatic recurrence (EHR) of HCC after surgical resection is considered to be closely associated with poor prognosis. However, data regarding risk factors and survival outcomes of early EHR after surgical resection remain scarce.

AIM

To investigate the clinical features and risk factors of early EHR and elucidate its association with survival outcomes.

METHODS

From January 2004 to December 2019, we enrolled treatment-naïve patients who were ≥ 18 years and underwent surgical resection for HCC in two tertiary academic centers. After excluding patients with tumor types other than HCC and/or ineligible data, this retrospective study finally included 779 patients. Surgical resection of HCC was performed according to the physicians' decisions and the EHR was diagnosed based on contrast-enhanced computed tomography or magnetic resonance imaging, and pathologic confirmation was performed in selected patients. Multivariate Cox regression analysis was performed to identify the variables associated with EHR.

RESULTS

Early EHR within 2 years after surgery was diagnosed in 9.5% of patients during a median follow-up period of 4.4 years. The recurrence-free survival period was 5.2 mo, and the median time to EHR was 8.8 mo in patients with early EHR. In 52.7% of patients with early EHR, EHR occurred as the first recurrence of HCC after surgical resection. On multivariate analysis, serum albumin < 4.0 g/dL, serum alkaline phosphatase > 100 U/L, surgical margin involvement, venous and/or lymphatic involvement, satellite nodules, tumor necrosis detected by pathology, tumor size ≥ 7 cm, and macrovascular invasion were determined as risk factors associated with early EHR. After sub-categorizing the patients according to the number of risk factors, the rates of both EHR and survival showed a significant correlation with the risk of early EHR. Furthermore, multivariate analysis revealed that early EHR was associated with substantially worse survival outcomes (Hazard ratio, 6.77; 95% confidence interval, 4.81-9.52; P < 0.001).

CONCLUSION

Early EHR significantly deteriorates the survival of patients with HCC, and our identified risk factors may predict the clinical outcomes and aid in postoperative strategies for improving survival.

Role of interventional oncology in hepatocellular carcinoma: Future best practice beyond current guidelines

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related deaths globally. Liver transplant remains the goal of curative treatment, but limited supply of organs decreases accessibility and prolongs waiting time to transplantation. Therefore, interventional oncology therapies have been used to treat the majority of HCC patients, including those awaiting transplant. The Barcelona Clinic Liver Cancer (BCLC) classification is the most widely used staging system in management of HCC that helps allocate treatments. Since its inception in 1999, it was updated for the fifth time in November 2021 and for the first time shaped by expert opinions outside the core BCLC group. The most recent version includes additional options for early-stage disease, substratifies intermediate disease into three groups, and lists alternates to Sorafenib that can double the expected survival of advanced-stage disease. The group also proposed a new BCLC staging schema for disease progression, and endorsed treatment stage migration (TSM) directly into the main staging and treatment algorithm. This article reviews the recent developments underlying the current BCLC guidelines and highlights ongoing research, particularly involving radioembolization, that will shape future best practice.

Multi-modality artificial intelligence in digital pathology

In common medical procedures, the time-consuming and expensive nature of obtaining test results plagues doctors and patients. Digital pathology research allows using computational technologies to manage data, presenting an opportunity to improve the efficiency of diagnosis and treatment. Artificial intelligence (AI) has a great advantage in the data analytics phase. Extensive research has shown that AI algorithms can produce more up-to-date and standardized conclusions for whole slide images. In conjunction with the development of high-throughput sequencing technologies, algorithms can integrate and analyze data from multiple modalities to explore the correspondence between morphological features and gene expression. This review investigates using the most popular image data, hematoxylin-eosin stained tissue slide images, to find a strategic solution for the imbalance of healthcare resources. The article focuses on the role that the development of deep learning technology has in assisting doctors' work and discusses the opportunities and challenges of AI.

Clinical relevance of biomarkers in cholangiocarcinoma: critical revision and future directions

Cholangiocarcinoma (CCA) is a malignant tumour arising from the biliary system. In Europe, this tumour frequently presents as a sporadic cancer in patients without defined risk factors and is usually diagnosed at advanced stages with a consequent poor prognosis. Therefore, the identification of biomarkers represents an utmost need for patients with CCA. Numerous studies proposed a wide spectrum of biomarkers at tissue and molecular levels. With the present paper, a multidisciplinary group of experts within the European Network for the Study of Cholangiocarcinoma discusses the clinical role of tissue biomarkers and provides a selection based on their current relevance and potential applications in the framework of CCA. Recent advances are proposed by dividing biomarkers based on their potential role in diagnosis, prognosis and therapy response. Limitations of current biomarkers are also identified, together with specific promising areas (ie, artificial intelligence, patient-derived organoids, targeted therapy) where research should be focused to develop future biomarkers.

Artificial intelligence for the prevention and clinical management of hepatocellular carcinoma

Hepatocellular carcinoma (HCC) currently represents the fifth most common malignancy and the third-leading cause of cancer-related death worldwide, with incidence and mortality rates that are increasing. Recently, artificial intelligence (AI) has emerged as a unique opportunity to improve the full spectrum of HCC clinical care, by improving HCC risk prediction, diagnosis, and prognostication. AI approaches include computational search algorithms, machine learning (ML) and deep learning (DL) models. ML consists of a computer running repeated iterations of models, in order to progressively improve performance of a specific task, such as classifying an outcome. DL models are a subtype of ML, based on neural network structures that are inspired by the neuroanatomy of the human brain. A growing body of recent data now apply DL models to diverse data sources - including electronic health record data, imaging modalities, histopathology and molecular biomarkers - to improve the accuracy of HCC risk prediction, detection and prediction of treatment response. Despite the promise of these early results, future research is still needed to standardise AI data, and to improve both the generalisability and interpretability of results. If such challenges can be overcome, AI has the potential to profoundly change the way in which care is provided to patients with or at risk of HCC.

Fast Classification of Thyroid Nodules with Ultrasound Guided-Fine Needle Biopsy Samples and Machine Learning

A rapid classification method was developed for the malignant and benign thyroid nodules with ultrasound guided-fine needle aspiration biopsy (FNAB) samples. With probe electrospray ionization mass spectrometry, the mass-scan data of FNAB samples were used as datasets for machine learning. The patients were marked as malignant (98 patients), benign (110 patients) or undetermined (42 patients) by experienced doctors in terms of ultrasound, the B-Raf (BRAF) gene, and cytopathology inspections. Pairwise coupling was performed on 163 ions to generate 3630 ion ratios as new features for classifier training. With the new features, the performance of the multilayer perception (MLP) classifier is much better than that with the 163 ions as features directly. After training, the accuracy of the MLP classifier is as high as 92.0%. The accuracy of the single-blind test is 82.4%, which proved the good generalization ability of the MLP classifier. The overall concordance is 73.0% between prediction and six-month follow-up for patients in the undetermined group. Especially, the classifier showed high accuracy for the undetermined patients with suspicious for papillary carcinoma diagnosis (90.9%). In summary, the machine learning method based on FNAB samples has potential for real clinical applications.

Deep Learning-Based Universal Expert-Level Recognizing Pathological Images of Hepatocellular Carcinoma and Beyond

Background and Aims

We aim to develop a diagnostic tool for pathological-image classification using transfer learning that can be applied to diverse tumor types.

Methods

Microscopic images of liver tissue with and without hepatocellular carcinoma (HCC) were used to train and validate the classification framework based on a convolutional neural network. To evaluate the universal classification performance of the artificial intelligence (AI) framework, histological images from colorectal tissue and the breast were collected. Images for the training and validation sets were obtained from the Xiamen Hospital of Traditional Chinese Medicine, and those for the test set were collected from Zhongshan Hospital Xiamen University. The accuracy, sensitivity, and specificity values for the proposed framework were reported and compared with those of human image interpretation.

Results

In the human–machine comparisons, the sensitivity, and specificity for the AI algorithm were 98.0, and 99.0%, whereas for the human experts, the sensitivity ranged between 86.0 and 97.0%, while the specificity ranged between 91.0 and 100%. Based on transfer learning, the accuracies of the AI framework in classifying colorectal carcinoma and breast invasive ductal carcinoma were 96.8 and 96.0%, respectively.

Conclusion

The performance of the proposed AI framework in classifying histological images with HCC was comparable to the classification performance achieved by human experts, indicating that extending the proposed AI’s application to diagnoses and treatment recommendations is a promising area for future investigation.

The Significance of Systemic Inflammation Markers in Intrahepatic Recurrence of Early-Stage Hepatocellular Carcinoma after Curative Treatment

Simple Summary

This retrospective study using the prospectively collected registry data of newly diagnosed, previously untreated hepatocellular carcinoma (HCC) evaluates the significance of systemic inflammatory markers (SIMs) to intrahepatic recurrence (IHR) after curative treatment. Out of 4076 patients who met the inclusion criteria, 52.6% experienced IHR. SIMs, including pre-treatment platelet-to-lymphocyte ratio (PLR), post-treatment changes of neutrophil-to-lymphocyte ratio PLR, and lymphocyte-to-monocyte ratio were significantly associated with the prognosis of early-stage HCC patients who received initial curative treatment. The prognostic significances of SIMs were consistent for IHR-free survival, early and late IHR, and overall survival.

Abstract

Systemic inflammatory markers (SIMs) are known to be associated with carcinogenesis and prognosis of hepatocellular carcinoma (HCC). We evaluated the significance of SIMs in intrahepatic recurrence (IHR) of early-stage HCC after curative treatment. This study was performed using prospectively collected registry data of newly diagnosed, previously untreated HCC between 2005 and 2017 at a single institution. Inclusion criteria were patients with Barcelona Clinic Liver Cancer stage 0 or A, who underwent curative treatment. Pre-treatment and post-treatment values of platelet, neutrophil, lymphocyte, monocyte, neutrophil/lymphocyte ratio (NLR), platelet/lymphocyte ratio (PLR), and lymphocyte/monocyte ratio (LMR) were analyzed with previously well-known risk factors of HCC to identify factors associated with IHR-free survival (IHRFS), early IHR, and late IHR. Of 4076 patients, 2142 patients (52.6%) experienced IHR, with early IHR in 1018 patients (25.0%) and late IHR in 1124 patients (27.6%). Pre-treatment platelet count and PLR and post-treatment worsening of NLR, PLR, and LMR were independently associated with IHRFS. Pre-treatment platelet count and post-treatment worsening of NLR, PLR, and LMR were significantly related to both early and late IHR. Pre-treatment values and post-treatment changes in SIMs were significant factors of IHR in early-stage HCC, independent of previously well-known risk factors of HCC.

Role of three-dimensional printing and artificial intelligence in the management of hepatocellular carcinoma: Challenges and opportunities

Hepatocellular carcinoma (HCC) constitutes the fifth most frequent malignancy worldwide and the third most frequent cause of cancer-related deaths. Currently, treatment selection is based on the stage of the disease. Emerging fields such as three-dimensional (3D) printing, 3D bioprinting, artificial intelligence (AI), and machine learning (ML) could lead to evidence-based, individualized management of HCC. In this review, we comprehensively report the current applications of 3D printing, 3D bioprinting, and AI/ML-based models in HCC management; we outline the significant challenges to the broad use of these novel technologies in the clinical setting with the goal of identifying means to overcome them, and finally, we discuss the opportunities that arise from these applications. Notably, regarding 3D printing and bioprinting-related challenges, we elaborate on cost and cost-effectiveness, cell sourcing, cell viability, safety, accessibility, regulation, and legal and ethical concerns. Similarly, regarding AI/ML-related challenges, we elaborate on intellectual property, liability, intrinsic biases, data protection, cybersecurity, ethical challenges, and transparency. Our findings show that AI and 3D printing applications in HCC management and healthcare, in general, are steadily expanding; thus, these technologies will be integrated into the clinical setting sooner or later. Therefore, we believe that physicians need to become familiar with these technologies and prepare to engage with them constructively.

Artificial intelligence in liver diseases: improving diagnostics, prognostics and response prediction

Clinical routine in hepatology involves the diagnosis and treatment of a wide spectrum of metabolic, infectious, autoimmune and neoplastic diseases. Clinicians integrate qualitative and quantitative information from multiple data sources to make a diagnosis, prognosticate the disease course, and recommend a treatment. In the last 5 years, advances in artificial intelligence (AI), particularly in deep learning, have made it possible to extract clinically relevant information from complex and diverse clinical datasets. In particular, histopathology and radiology image data contain diagnostic, prognostic and predictive information which AI can extract. Ultimately, such AI systems could be implemented in clinical routine as decision support tools. However, in the context of hepatology, this requires further large-scale clinical validation and regulatory approval. Herein, we summarise the state of the art in AI in hepatology with a particular focus on histopathology and radiology data. We present a roadmap for the further development of novel biomarkers in hepatology and outline critical obstacles which need to be overcome.

A Hybrid Machine Learning Model Based on Semantic Information Can Optimize Treatment Decision for Naïve Single 3-5-cm HCC Patients

BACKGROUND

Tumor recurrence is an abomination for hepatocellular carcinoma (HCC) patients receiving local treatment.

PURPOSE

The aim of the study was to build a hybrid machine learning model to recommend optimized first treatment (laparoscopic hepatectomy [LH] or microwave ablation [MWA]) for naïve single 3-5-cm HCC patients based on early recurrence (ER, ≤2 years) probability.

METHODS

This retrospective study collected 20 semantic variables of 582 patients (LH: 300, MWA: 282) from 13 hospitals with at least 24 months follow-up. Both groups were divided into training, validation, and test set, respectively. Five algorithms (logistics regression, random forest, neural network, stochastic gradient boosting, and eXtreme Gradient Boosting [XGB]) were used for model building. A model with highest area under the receiver operating characteristic curve (AUC) in a validation set of LH and MWA was selected to connect as a hybrid model which made decision based on ER probability. Model testing was performed in a comprehensive set comprising LH and MWA test sets.

RESULTS

Four variables in each group were selected to build LH and MWA models, respectively. LH-XGB model (AUC = 0.744) and MWA-stochastic gradient method (AUC = 0.750) model were selected for model building. In the comprehensive set, a treatment confusion matrix was established based on recommended and actual treatment. The predicted ER probabilities were comparable with the actual ER rates for various types of patients in matrix (p > 0.05). ER rate of patients whose actual treatment consistent with recommendation was lower than that of inconsistent patients (LH: 21.2% vs. 46.2%, p = 0.042; MWA: 26.3% vs. 54.1%, p = 0.048). By recommending optimal treatment, the hybrid model can significantly reduce ER probability from 38.2% to 25.6% for overall patients (p < 0.001).

CONCLUSIONS

The hybrid model can accurately predict ER probability of different treatments and thereby provide reliable evidence to make optimal treatment decision for patients with single 3-5-cm HCC.

Prediction of non-muscle invasive bladder cancer recurrence using machine learning of quantitative nuclear features

Non-muscle invasive bladder cancer (NMIBC) generally has a good prognosis; however, recurrence after transurethral resection (TUR), the standard primary treatment, is a major problem. Clinical management after TUR has been based on risk classification using clinicopathological factors, but these classifications are not complete. In this study, we attempted to predict early recurrence of NMIBC based on machine learning of quantitative morphological features. In general, structural, cellular, and nuclear atypia are evaluated to determine cancer atypia. However, since it is difficult to accurately quantify structural atypia from TUR specimens, in this study, we used only nuclear atypia and analyzed it using feature extraction followed by classification using Support Vector Machine and Random Forest machine learning algorithms. For the analysis, 125 patients diagnosed with NMIBC were used; data from 95 patients were randomly selected for the training set, and data from 30 patients were randomly selected for the test set. The results showed that the support vector machine-based model predicted recurrence within 2 years after TUR with a probability of 90% and the random forest-based model with probability of 86.7%. In the future, the system can be used to objectively predict NMIBC recurrence after TUR.

Development of Models to Predict Postoperative Complications for Hepatitis B Virus-Related Hepatocellular Carcinoma

Background

Surgical treatment remains the best option for patients with hepatocellular carcinoma (HCC) caused by chronic hepatitis B virus (HBV) infection. However, there is no optimal tool based on readily accessible clinical parameters to predict postoperative complications. Herein, our study aimed to develop models that permitted risk of severe complications to be assessed before and after liver resection based on conventional variables.

Methods

A total of 1,047 patients treated by hepatectomy for HCC with HBV infection at three different centers were recruited retrospectively between July 1, 2014, and July 1, 2018. All surgical complications were recorded and scored by the Comprehensive Complication Index (CCI). A CCI ≥26.2 was used as a threshold to define patients with severe complications. We built two models for the CCI, one using preoperative and one using preoperative and postoperative data. Besides, CCI and other potentially relevant factors were evaluated for their ability to predict early recurrence and metastasis. All the findings were internally validated in the Hangzhou cohort and then externally validated in the Lanzhou and Urumqi cohorts.

Results

Multivariable analysis identified National Nosocomial Infections Surveillance (NNIS) index, tumor number, gamma-glutamyltransferase (GGT), total cholesterol (TC), potassium, and thrombin time as the key preoperative parameters related to perioperative complications. The nomogram based on the preoperative model [preoperative CCI After Surgery for Liver tumor (CCIASL-pre)] showed good discriminatory performance internally and externally. A more accurate model [postoperative CCI After Surgery for Liver tumor (CCIASL-post)] was established, combined with the other four postoperative predictors including leukocyte count, basophil count, erythrocyte count, and total bilirubin level. No significant association was observed between CCI and long-term complications.

Conclusion

Based on the widely available clinical data, statistical models were established to predict the complications after hepatectomy in patients with HBV infection. All the findings were extensively validated and shown to be applicable nationwide. Such models could be used as guidelines for surveillance follow-up and the design of post-resection adjuvant therapy.