Liver Resection Criteria for Patients with Hepatocellular Carcinoma and Multiple Tumors Based on Total Tumor Volume

BACKGROUND

In many Asian hepatocellular carcinoma (HCC) guidelines, resection is an option for multiple HCCs. It is difficult to compare small but multiple tumors vs. fewer large tumors in terms of the traditional tumor burden definition. We aimed to evaluate the role of liver resection for multiple HCCs and determine factors associated with survival benefits.

METHODS

We reviewed 160 patients with multiple HCCs who underwent liver resection between July 2003 and December 2018. The risk factors for tumor recurrence were assessed using Cox proportional hazards modeling, and survival was analyzed using the Kaplan-Meier method.

RESULTS

In all 160 patients, 133 (83.1%) exceeded the Milan criteria. Total tumor volume (TTV) > 275 cm3 and serum alpha-fetoprotein (AFP) level > 20 ng/mL were associated with disease-free survival. Patients beyond the Milan criteria were grouped into three risk categories: no risk (TTV ≤ 275 cm3 and AFP ≤ 20 ng/mL, n = 39), one risk (either TTV > 275 cm3 or AFP > 20 ng/mL, n = 76), and two risks (TTV > 275 cm3 and AFP > 20 ng/mL, n = 18). No-risk group had comparable disease-free survival (p = 0.269) and overall survival (p = 0.215) to patients who met the Milan criteria.

CONCLUSION

Patients with TTV ≤ 275 cm3 and AFP ≤ 20 ng/mL can have good outcomes even exceed the Milan criteria.

Development of a predictive nomogram for in-hospital death risk in multimorbid patients with hepatocellular carcinoma undergoing Palliative Locoregional Therapy

Patients diagnosed with hepatocellular carcinoma (HCC) often present with multimorbidity, significantly contributing to adverse outcomes, particularly in-hospital mortality. This study aimed to develop a predictive nomogram to assess the impact of comorbidities on in-hospital mortality risk in HCC patients undergoing palliative locoregional therapy. We retrospectively analyzed data from 345 hospitalized HCC patients who underwent palliative locoregional therapy between January 2015 and December 2022. The nomogram was constructed using independent risk factors such as length of stay (LOS), hepatitis B virus (HBV) infection, hypertension, chronic obstructive pulmonary disease (COPD), anemia, thrombocytopenia, liver cirrhosis, hepatic encephalopathy (HE), N stage, and microvascular invasion. The model demonstrated high predictive accuracy with an AUC of 0.908 (95% CI: 0.859-0.956) for the overall dataset, 0.926 (95% CI: 0.883-0.968) for the training set, and 0.862 (95% CI: 0.728-0.994) for the validation set. Calibration curves indicated a strong correlation between predicted and observed outcomes, validated by statistical tests. Decision curve analysis (DCA) and clinical impact curves (CIC) confirmed the model's clinical utility in predicting in-hospital mortality. This nomogram offers a practical tool for personalized risk assessment in HCC patients undergoing palliative locoregional therapy, facilitating informed clinical decision-making and improving patient management.

Comprehensive review of hepatocellular carcinoma with portal vein tumor thrombus: State of art and future perspectives

BACKGROUND

Despite advances in the diagnosis of patients with hepatocellular carcinoma (HCC), 70%-80% of patients are diagnosed with advanced stage disease. Portal vein tumor thrombus (PVTT) is among the most ominous signs of advanced stage disease and has been associated with poor survival if untreated.

DATA SOURCES

A systematic search of MEDLINE (PubMed), Embase, Cochrane Library and Database for Systematic Reviews (CDSR), Google Scholar, and National Institute for Health and Clinical Excellence (NICE) databases until December 2022 was conducted using free text and MeSH terms: hepatocellular carcinoma, portal vein tumor thrombus, portal vein thrombosis, vascular invasion, liver and/or hepatic resection, liver transplantation, and systematic review.

RESULTS

Centers of surgical excellence have reported promising results related to the individualized surgical management of portal thrombus versus arterial chemoembolization or systemic chemotherapy. Critical elements to the individualized surgical management of HCC and portal thrombus include precise classification of the portal vein tumor thrombus, accurate identification of the subgroups of patients who may benefit from resection, as well as meticulous surgical technique. This review addressed five specific areas: (a) formation of PVTT; (b) classifications of PVTT; (c) controversies related to clinical guidelines; (d) surgical treatments versus non-surgical approaches; and (e) characterization of surgical techniques correlated with classifications of PVTT.

CONCLUSIONS

Current evidence from Chinese and Japanese high-volume centers demonstrated that patients with HCC and associated PVTT can be managed with surgical resection with acceptable results.

Novel nomograms based on microvascular invasion grade for early-stage hepatocellular carcinoma after curative hepatectomy

Microvascular invasion (MVI) is a critical risk factor for postoperative recurrence of hepatocellular carcinoma (HCC). This study aimed to firstly develop and validate nomograms based on MVI grade for predicting recurrence, especially early recurrence, and overall survival in patients with early-stage HCC after curative resection. We retrospectively reviewed the data of patients with early-stage HCC who underwent curative hepatectomy in the First Affiliated Hospital of Fujian Medical University (FHFU) and Mengchao Hepatobiliary Hospital of Fujian Medical University (MHH). Kaplan-Meier curves and Cox proportional hazards regression models were used to analyse disease-free survival (DFS) and overall survival (OS). Nomogram models were constructed on the datasets from the 70% samples of and FHFU, which were validated using bootstrap resampling with 30% samples as internal validation and data of patients from MHH as external validation. A total of 703 patients with early-stage HCC were included to create a nomogram for predicting recurrence or metastasis (DFS nomogram) and a nomogram for predicting survival (OS nomogram). The concordance indexes and calibration curves in the training and validation cohorts showed optimal agreement between the predicted and observed DFS and OS rates. The predictive accuracy was significantly better than that of the classic HCC staging systems.

The Use of Artificial Intelligence in the Liver Histopathology Field: A Systematic Review

Digital pathology (DP) has begun to play a key role in the evaluation of liver specimens. Recent studies have shown that a workflow that combines DP and artificial intelligence (AI) applied to histopathology has potential value in supporting the diagnosis, treatment evaluation, and prognosis prediction of liver diseases. Here, we provide a systematic review of the use of this workflow in the field of hepatology. Based on the PRISMA 2020 criteria, a search of the PubMed, SCOPUS, and Embase electronic databases was conducted, applying inclusion/exclusion filters. The articles were evaluated by two independent reviewers, who extracted the specifications and objectives of each study, the AI tools used, and the results obtained. From the 266 initial records identified, 25 eligible studies were selected, mainly conducted on human liver tissues. Most of the studies were performed using whole-slide imaging systems for imaging acquisition and applying different machine learning and deep learning methods for image pre-processing, segmentation, feature extractions, and classification. Of note, most of the studies selected demonstrated good performance as classifiers of liver histological images compared to pathologist annotations. Promising results to date bode well for the not-too-distant inclusion of these techniques in clinical practice.

Surgical resection versus transarterial chemoembolization for patients with hepatocellular carcinoma beyond Milan criteria: prognostic role of tumor burden score

Tumor burden score (TBS) has been recently introduced to indicate the extent of tumor burden in different cancers, but its role in advanced hepatocellular carcinoma (HCC) is unclear. We aimed to determine the prognostic role of TBS in patients with HCC beyond the Milan criteria receiving surgical resection (SR) or transarterial chemoembolization (TACE). A total of 1303 newly diagnosed HCC patients beyond Milan criteria receiving SR or TACE as the primary therapy were retrospectively analyzed. Independent prognostic predictors were examined by the multivariate Cox proportional hazards model. SR was associated with better overall survival compared with TACE in these patients. Multivariate Cox analysis of the entire cohort revealed that age > 66 years (hazard ratio [HR]: 1.145, 95% confidence interval [CI]: 1.004-1.305, p = 0.043), serum α-fetoprotein > 200 ng/mL (HR: 1.602, 95% CI: 1.402-1.831, p < 0.001), performance status 2-4 (HR: 1.316, 95% CI: 1.115-1.553, p < 0.001), medium TBS (HR: 1.225, 95% CI:1.045-1.436, p = 0.012), high TBS (HR: 1.976, 95% CI: 1.637-2.384, p < 0.001), albumin-bilirubin (ALBI) grade 2-3 (HR: 1.529, 95% CI: 1.342-1.743, p < 0.001), presence of vascular invasion (HR: 1.568, 95% CI: 1.354-1.816, p < 0.001), and TACE (HR: 2.396, 95% CI: 2.082-2.759, p < 0.001) were linked with decreased survival. SR consistently predicted a significantly better survival in different TBS subgroups. TBS is a feasible and independent prognostic predictor in HCC beyond the Milan criteria. SR provides better long-term outcome compared with TACE in these patients independent of TBS grade, and should be considered as the primary treatment modality in this special patient group.

Development and validation of a REcurrent Liver cAncer Prediction ScorE (RELAPSE) following liver transplantation in patients with hepatocellular carcinoma: Analysis of the US Multicenter HCC Transplant Consortium

HCC recurrence following liver transplantation (LT) is highly morbid and occurs despite strict patient selection criteria. Individualized prediction of post-LT HCC recurrence risk remains an important need. Clinico-radiologic and pathologic data of 4981 patients with HCC undergoing LT from the US Multicenter HCC Transplant Consortium (UMHTC) were analyzed to develop a REcurrent Liver cAncer Prediction ScorE (RELAPSE). Multivariable Fine and Gray competing risk analysis and machine learning algorithms (Random Survival Forest and Classification and Regression Tree models) identified variables to model HCC recurrence. RELAPSE was externally validated in 1160 HCC LT recipients from the European Hepatocellular Cancer Liver Transplant study group. Of 4981 UMHTC patients with HCC undergoing LT, 71.9% were within Milan criteria, 16.1% were initially beyond Milan criteria with 9.4% downstaged before LT, and 12.0% had incidental HCC on explant pathology. Overall and recurrence-free survival at 1, 3, and 5 years was 89.7%, 78.6%, and 69.8% and 86.8%, 74.9%, and 66.7%, respectively, with a 5-year incidence of HCC recurrence of 12.5% (median 16 months) and non-HCC mortality of 20.8%. A multivariable model identified maximum alpha-fetoprotein (HR = 1.35 per-log SD, 95% CI,1.22-1.50, p < 0.001), neutrophil-lymphocyte ratio (HR = 1.16 per-log SD, 95% CI,1.04-1.28, p < 0.006), pathologic maximum tumor diameter (HR = 1.53 per-log SD, 95% CI, 1.35-1.73, p < 0.001), microvascular (HR = 2.37, 95%-CI, 1.87-2.99, p < 0.001) and macrovascular (HR = 3.38, 95% CI, 2.41-4.75, p < 0.001) invasion, and tumor differentiation (moderate HR = 1.75, 95% CI, 1.29-2.37, p < 0.001; poor HR = 2.62, 95% CI, 1.54-3.32, p < 0.001) as independent variables predicting post-LT HCC recurrence (C-statistic = 0.78). Machine learning algorithms incorporating additional covariates improved prediction of recurrence (Random Survival Forest C-statistic = 0.81). Despite significant differences in European Hepatocellular Cancer Liver Transplant recipient radiologic, treatment, and pathologic characteristics, external validation of RELAPSE demonstrated consistent 2- and 5-year recurrence risk discrimination (AUCs 0.77 and 0.75, respectively). We developed and externally validated a RELAPSE score that accurately discriminates post-LT HCC recurrence risk and may allow for individualized post-LT surveillance, immunosuppression modification, and selection of high-risk patients for adjuvant therapies.

The Role of Artificial Intelligence in the Detection and Implementation of Biomarkers for Hepatocellular Carcinoma: Outlook and Opportunities

Liver cancer is a leading cause of cancer-related death worldwide, and its early detection and treatment are crucial for improving morbidity and mortality. Biomarkers have the potential to facilitate the early diagnosis and management of liver cancer, but identifying and implementing effective biomarkers remains a major challenge. In recent years, artificial intelligence has emerged as a promising tool in the cancer sphere, and recent literature suggests that it is very promising in facilitating biomarker use in liver cancer. This review provides an overview of the status of AI-based biomarker research in liver cancer, with a focus on the detection and implementation of biomarkers for risk prediction, diagnosis, staging, prognostication, prediction of treatment response, and recurrence of liver cancers.

Evaluating the benefit of surgical resection for hepatocellular carcinoma with multifocality or intrahepatic vascular invasion

BACKGROUND

The role of hepatectomy for hepatocellular carcinoma (HCC) with multifocality or intrahepatic vascular involvement remains ill-defined. Our objective was to evaluate benefits of surgical resection for patients with these high-risk features.

METHODS

The National Cancer Database was used to identify HCC patients with vascular involvement and/or multifocality (T2/T3, N-/M-) from 2011 to 2015. Propensity score matching (k-nearest neighbors, no replacement, 1:1) grouped patients by treatment: surgical resection versus non-surgical modalities. Groups were matched using patient, clinical, and liver-specific characteristics. Median overall survival (OS) was calculated using Kaplan-Meier, and adjusted analyses were performed using shared frailty models.

RESULTS

14,557 patients met inclusion criteria, including 1892 (9.4%) treated with surgical resection. Median cohort OS was 20.5 months. After adjustment, surgical resection was associated with survival advantage compared to non-surgical treatment (37.8 versus 15.7 months, log-rank P < .001; adjusted hazard ratio 0.49, 95% confidence interval, 0.45-0.54). Patients with minimal comorbidity, unifocal disease, and age <54 had highest probability of survival one year post-surgery.

CONCLUSIONS

Surgical resection is associated with a survival advantage in HCC with multifocality and/or intrahepatic vascular involvement. The presence of these features should not contraindicate consideration of hepatectomy in suitable surgical candidates.

Using machine learning to preoperatively stratify prognosis among patients with gallbladder cancer: a multi-institutional analysis

BACKGROUND

Gallbladder cancer (GBC) is an aggressive malignancy associated with a high risk of recurrence and mortality. We used a machine-based learning approach to stratify patients into distinct prognostic groups using preperative variables.

METHODS

Patients undergoing curative-intent resection of GBC were identified using a multi-institutional database. A classification and regression tree (CART) was used to stratify patients relative to overall survival (OS) based on preoperative clinical factors.

RESULTS

CART analysis identified tumor size, biliary drainage, carbohydrate antigen 19-9 (CA19-9) levels, and neutrophil-lymphocyte ratio (NLR) as the factors most strongly associated with OS. Machine learning cohorted patients into four prognostic groups: Group 1 (n = 109): NLR ≤1.5, CA19-9 ≤20, no drainage, tumor size <5.0 cm; Group 2 (n = 88): NLR >1.5, CA19-9 ≤20, no drainage, tumor size <5.0 cm; Group 3 (n = 46): CA19-9 >20, no drainage, tumor size <5.0 cm; Group 4 (n = 77): tumor size <5.0 cm with drainage OR tumor size ≥5.0 cm. Median OS decreased incrementally with CART group designation (59.5, 27.6, 20.6, and 12.1 months; p < 0.0001).

CONCLUSIONS

A machine-based model was able to stratify GBC patients into four distinct prognostic groups based only on preoperative characteristics. Characterizing patient prognosis with machine learning tools may help physicians provide more patient-centered care.

Individualized Approach in the Surgical Management of Hepatocellular Carcinoma: Results from a Greek Multicentre Study

Simple Summary

Hepatocellular carcinoma (HCC) is the most common primary liver cancer with expected increasing frequency in the next few decades. The Barcelona Clinic Liver Cancer (BCLC) Staging System is a widely adopted tool for guiding the therapeutic algorithms of patients with HCC. This classification has been guiding clinical practice for the last two decades. However, emerging data demonstrate that patients beyond the traditional criteria of operability or resectability can benefit from surgical treatment. We present the Greek multicentre experience of treating HCC within and beyond BCLC guidelines.

Abstract

Background: Hepatocellular carcinoma (HCC) is the most common primary liver cancer and the third leading cause of death worldwide. The management of HCC is complex, with surgical treatment providing long-term survival in eligible patients. This study aims to present the experience of aggressive surgical management of HCC in Greece. Methods: This is a retrospective multicentre clinical study with 242 patients. Results: Most patients were male (79%) and had a median age of 71 yrs. According to the most recent BCLC criteria, 172 patients (71.1%) were classified as BCLC 0-A stage, 33 patients (13.6%) were classified as BCLC B, and 37 (15.3%) were classified as BCLC C. A total of 54% of the patients underwent major hepatectomy. Major postoperative morbidity was 15.6%, and the 90-day postoperative mortality rate was 4.5%. The median follow-up was 33.5 months. Three- and five-year overall survival was 65% and 48%, respectively. The median overall survival was 55 months. Significantly, five-year survival was 55% for BCLC A, and 34% and 21% for BCLC B and C, respectively. In univariate analysis, cirrhosis, type of resection (R status), and BCLC stage were associated with overall survival. Multivariate analysis indicated that R1 and R2 resections compared to R0, and BCLC C compared to BCLC 0-A, were independently associated with increased mortality. Conclusions: Aggressive surgical treatment of HCC offers satisfactory long-term survival prospects. A significant percentage (29%) of HCCs that underwent liver resection were of the intermediate and advanced BCLC stage. The management of patients with HCC should be discussed in multidisciplinary tumour board meetings on a case-by-case basis to be more effective.

A Novel Nomogram Model to Predict the Recurrence-Free Survival and Overall Survival of Hepatocellular Carcinoma

Background

Treatments for patients with early‐stage hepatocellular carcinoma (HCC) include liver transplantation (LT), liver resection (LR), radiofrequency ablation (RFA), and microwave ablation (MWA), are critical for their long-term survival. However, a computational model predicting treatment-independent prognosis of patients with HCC, such as overall survival (OS) and recurrence-free survival (RFS), is yet to be developed, to our best knowledge. The goal of this study is to identify prognostic factors associated with OS and RFS in patients with HCC and develop nomograms to predict them, respectively.

Methods

We retrospectively retrieved 730 patients with HCC from three hospitals in China and followed them up for 3 and 5 years after invasive treatment. All enrolled patients were randomly divided into the training cohort and the validation cohort with a 7:3 ratio, respectively. Independent prognostic factors associated with OS and RFS were determined by the multivariate Cox regression analysis. Two nomogram prognostic models were built and evaluated by concordance index (C-index), calibration curves, area under the receiver operating characteristics (ROC) curve, time-dependent area under the ROC curve (AUC), the Kaplan–Meier survival curve, and decision curve analyses (DCAs), respectively.

Results

Prognostic factors for OS and RFS were identified, and nomograms were successfully built. Calibration discrimination was good for both the OS and RFS nomogram prediction models (C-index: 0.750 and 0.746, respectively). For both nomograms, the AUC demonstrated outstanding predictive performance; the DCA shows that the model has good decision ability; and the calibration curve demonstrated strong predictive power. The nomograms successfully discriminated high-risk and low-risk patients with HCC associated with OS and RFS.

Conclusions

We developed nomogram survival prediction models to predict the prognosis of HCC after invasive treatment with acceptable accuracies in both training and independent testing cohorts. The models may have clinical values in guiding the selection of clinical treatment strategies.

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.

Advancements of Artificial Intelligence in Liver-Associated Diseases and Surgery

Background and Objectives: The advancement of artificial intelligence (AI) based technologies in medicine is progressing rapidly, but the majority of its real-world applications has not been implemented. The establishment of an accurate diagnosis with treatment has now transitioned into an artificial intelligence era, which has continued to provide an amplified understanding of liver cancer as a disease and helped to proceed better with the method of procurement. This article focuses on reviewing the AI in liver-associated diseases and surgical procedures, highlighting its development, use, and related counterparts. Materials and Methods: We searched for articles regarding AI in liver-related ailments and surgery, using the keywords (mentioned below) on PubMed, Google Scholar, Scopus, MEDLINE, and Cochrane Library. Choosing only the common studies suggested by these libraries, we segregated the matter based on disease. Finally, we compiled the essence of these articles under the various sub-headings. Results: After thorough review of articles, it was observed that there was a surge in the occurrence of liver-related surgeries, diagnoses, and treatments. Parallelly, advanced computer technologies governed by AI continue to prove their efficacy in the accurate screening, analysis, prediction, treatment, and recuperation of liver-related cases. Conclusions: The continual developments and high-order precision of AI is expanding its roots in all directions of applications. Despite being novel and lacking research, AI has shown its intrinsic worth for procedures in liver surgery while providing enhanced healing opportunities and personalized treatment for liver surgery patients.

The application of artificial intelligence in hepatology: A systematic review

The integration of human and artificial intelligence (AI) in medicine has only recently begun but it has already become obvious that intelligent systems can dramatically improve the management of liver diseases. Big data made it possible to envisage transformative developments of the use of AI for diagnosing, predicting prognosis and treating liver diseases, but there is still a lot of work to do. If we want to achieve the 21^st century digital revolution, there is an urgent need for specific national and international rules, and to adhere to bioethical parameters when collecting data. Avoiding misleading results is essential for the effective use of AI. A crucial question is whether it is possible to sustain, technically and morally, the process of integration between man and machine. We present a systematic review on the applications of AI to hepatology, highlighting the current challenges and crucial issues related to the use of such technologies.

Patient stratification in hepatocellular carcinoma: impact on choice of therapy

INTRODUCTION

HCC comprises around 60 to 80% of all primary liver cancers and exhibits wide geographical variability. Appropriate treatment allocation needs to include both patient and tumor characteristics.

AREAS COVERED

Current HCC classification systems to guide therapy are either liver function-centric and evaluate physiologic liver function to guide therapy or prognostic stratification classification systems broadly based on tumor morphologic parameters, patient performance status, and liver reserve assessment. This review focuses on different classification systems for HCC, their strengths, and weaknesses as well as the use of artificial intelligence in improving prognostication in HCC.

EXPERT OPINION

Future HCC classification systems will need to incorporate clinic-pathologic data from a multitude of sources and emerging therapies to develop patient-specific treatment plans targeting a patient's unique tumor profile.

Surgical data science - from concepts toward clinical translation

Recent developments in data science in general and machine learning in particular have transformed the way experts envision the future of surgery. Surgical Data Science (SDS) is a new research field that aims to improve the quality of interventional healthcare through the capture, organization, analysis and modeling of data. While an increasing number of data-driven approaches and clinical applications have been studied in the fields of radiological and clinical data science, translational success stories are still lacking in surgery. In this publication, we shed light on the underlying reasons and provide a roadmap for future advances in the field. Based on an international workshop involving leading researchers in the field of SDS, we review current practice, key achievements and initiatives as well as available standards and tools for a number of topics relevant to the field, namely (1) infrastructure for data acquisition, storage and access in the presence of regulatory constraints, (2) data annotation and sharing and (3) data analytics. We further complement this technical perspective with (4) a review of currently available SDS products and the translational progress from academia and (5) a roadmap for faster clinical translation and exploitation of the full potential of SDS, based on an international multi-round Delphi process.

Artificial Intelligence in hepatology, liver surgery and transplantation: Emerging applications and frontiers of research

The integration of artificial intelligence (AI) and augmented realities into the medical field is being attempted by various researchers across the globe. As a matter of fact, most of the advanced technologies utilized by medical providers today have been borrowed and extrapolated from other industries. The introduction of AI into the field of hepatology and liver surgery is relatively a recent phenomenon. The purpose of this narrative review is to highlight the different AI concepts which are currently being tried to improve the care of patients with liver diseases. We end with summarizing emerging trends and major challenges in the future development of AI in hepatology and liver surgery.

Challenges and opportunities in the application of artificial intelligence in gastroenterology and hepatology

Artificial intelligence (AI) is an umbrella term used to describe a cluster of interrelated fields. Machine learning (ML) refers to a model that learns from past data to predict future data. Medicine and particularly gastroenterology and hepatology, are data-rich fields with extensive data repositories, and therefore fruitful ground for AI/ML-based software applications. In this study, we comprehensively review the current applications of AI/ML-based models in these fields and the opportunities that arise from their application. Specifically, we refer to the applications of AI/ML-based models in prevention, diagnosis, management, and prognosis of gastrointestinal bleeding, inflammatory bowel diseases, gastrointestinal premalignant and malignant lesions, other nonmalignant gastrointestinal lesions and diseases, hepatitis B and C infection, chronic liver diseases, hepatocellular carcinoma, cholangiocarcinoma, and primary sclerosing cholangitis. At the same time, we identify the major challenges that restrain the widespread use of these models in healthcare in an effort to explore ways to overcome them. Notably, we elaborate on the concerns regarding intrinsic biases, data protection, cybersecurity, intellectual property, liability, ethical challenges, and transparency. Even at a slower pace than anticipated, AI is infiltrating the healthcare industry. AI in healthcare will become a reality, and every physician will have to engage with it by necessity.

Significance of liver resection for intermediate stage hepatocellular carcinoma according to subclassification

Background

Patients diagnosed with Barcelona Clinic Liver Cancer (BCLC) intermediate stage hepatocellular carcinoma (HCC) encompass a broad clinical population. Kinki criteria subclassifications have been proposed to better predict prognoses and determine appropriate treatment strategies for these patients. This study validated the prognostic significance within the Kinki criteria substages and analyzed the role of liver resection in patients with intermediate stage HCC.

Methods

Patients with intermediate stage HCC (n = 378) were retrospectively subclassified according to the Kinki criteria (B1, n = 123; B2, n = 225; and B3, n = 30). We analyzed the overall survival (OS) and treatment methods.

Results

The OS was significantly different between adjacent substages. Patients in substage B1 who underwent liver resection had a significantly better prognosis than those who did not, even after propensity score matching (PSM). Patients in substage B2 who underwent liver resection had a significantly better prognosis than those who did not; however, there was no difference after PSM. There was no difference in prognosis based on treatments among patients in substage B3.

Conclusions

The Kinki criteria clearly stratify patients with intermediate stage HCC by prognosis. For substage B1 HCC patients, liver resection provides a better prognosis than other treatment modalities. In patients with substage B2 and B3, an alternative approach is required.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-08421-3.

A simple scoring system to estimate perioperative mortality following liver resection for primary liver malignancy-the Hepatectomy Risk Score (HeRS)

Background

Selection of the optimal treatment modality for primary liver cancers remains complex, balancing patient condition, liver function, and extent of disease. In individuals with preserved liver function, liver resection remains the primary approach for treatment with curative intent but may be associated with significant mortality. The purpose of this study was to establish a simple scoring system based on Model for End-stage Liver Disease (MELD) and extent of resection to guide risk assessment for liver resections.

Methods

The 2005-2015 NSQIP database was queried for patients undergoing liver resection for primary liver malignancy. We first developed a model that incorporated the extent of resection (1 point for major hepatectomy) and a MELD-Na score category of low (MELD-Na =6, 1 point), medium (MELD-Na =7-10, 2 points) or high (MELD-Na >10, 3 points) with a score range of 1-4, called the Hepatic Resection Risk Score (HeRS). We tested the predictive value of this model on the dataset using logistic regression. We next developed an optimal multivariable model using backwards sequential selection of variables under logistic regression. We performed K-fold cross validation on both models. Receiver operating characteristics were plotted and the optimal sensitivity and specificity for each model were calculated to obtain positive and negative predictive values.

Results

A total of 4,510 patients were included. HeRS was associated with increased odds of 30-day mortality [HeRS =2: OR =3.23 (1.16-8.99), P=0.025; HeRS =3: OR =6.54 (2.39-17.90), P<0.001; HeRS =4: OR =13.69 (4.90-38.22), P<0.001]. The AUC for this model was 0.66. The AUC for the optimal multivariable model was higher at 0.76. Under K-fold cross validation, the positive predictive value (PPV) and negative predictive value (NPV) of these two models were similar at PPV =6.4% and NPV =97.7% for the HeRS only model and PPV =8.4% and NPV =98.1% for the optimal multivariable model.

Conclusions

The HeRS offers a simple heuristic for estimating 30-day mortality after resection of primary liver malignancy. More complicated models offer better performance but at the expense of being more difficult to integrate into clinical practice.

The Clinical Value of Postoperative Transarterial Chemoembolization for Resectable Patients with Intermediate Hepatocellular Carcinoma After Radical Hepatectomy: a Propensity Score-Matching Study

BACKGROUND AND AIMS

Surgical resection for patients with intermediate hepatocellular carcinoma (HCC) is preferred in China, but the prognosis remains far from satisfactory. Postoperative transarterial chemoembolization (p-TACE) has been conducted prevalently to prevent recurrence, but its efficacy remains controversial. Hence, we collected the data from primary liver cancer big data (PLCBD) to investigate the clinical value of p-TACE for patients with intermediate HCC and identify the potential beneficiaries.

METHODS

Patients who were diagnosed with intermediate HCC between December 2012 and December 2015 were identified through the PLCBD. Disease-free survival (DFS) of patients who received p-TACE or not following radical resection was evaluated using Kaplan-Meier survival curves before and after 1:1 propensity scoring match (PSM). Subgroup analysis was conducted stratified by risk factors associated with recurrence.

RESULTS

A total of 325 intermediate HCC patients receiving radical resection were eligible in this study, including 123 patients in the p-TACE group and 202 in the non-TACE group. Median DFS in the p-TACE group was significantly longer than in the non-TACE group (23.3 months vs. 18.0 months, P = 0.016) in the whole cohort with no severe complicates, which was confirmed in a well-matched cohort (17.4 months vs. 23.3 months, P = 0.012). In addition, p-TACE was identified as an independent risk factors of DFS by multivariate Cox regression analysis before and after PSM (both P < 0.05). After adjusting for other prognostic variables, patients were found to significantly benefit from p-TACE in DFS if they were male, or had hepatitis, diabetes, cirrhosis, AFP ≤ 400 ng/ml, anatomic hepatectomy, no severe surgical complication, no intraoperative transfusion, tumor number = 2, differentiation grading III, capsule, or had no transfusion (all P < 0.05).

CONCLUSION

With the current data, we concluded that p-TACE was safe and efficient for the patients with intermediate HCC following radical resection, and male patients with hepatitis, diabetes, cirrhosis, AFP ≤ 400 ng/ml, anatomic hepatectomy, no severe surgical complication, no intraoperative transfusion, tumor number = 2, differentiation grading III, and capsule would benefit more from p-TACE.

Clinical value of artificial intelligence in hepatocellular carcinoma: Current status and prospect

Hepatocellular carcinoma (HCC) is the most commonly diagnosed type of liver cancer and the fourth leading cause of cancer-related mortality worldwide. The early identification of HCC and effective treatments for it have been challenging. Due to the sufficient compensatory ability of early patients and its nonspecific symptoms, HCC is more likely to escape diagnosis in the incipient stage, during which patients can achieve a more satisfying overall survival if they undergo resection or liver transplantation. Patients at advanced stages can profit from radical therapies in a limited way. In order to improve the unfavorable prognosis of HCC, diagnostic ability and treatment efficiency must be improved. The past decade has seen rapid advancements in artificial intelligence, underlying its unique usefulness in almost every field, including that of medicine. Herein, we sought and reviewed studies that put emphasis on artificial intelligence and HCC.

Artificial intelligence in transplantation (machine-learning classifiers and transplant oncology)

PURPOSE OF REVIEW

To highlight recent efforts in the development and implementation of machine learning in transplant oncology - a field that uses liver transplantation for the treatment of hepatobiliary malignancies - and particularly in hepatocellular carcinoma, the most commonly treated diagnosis in transplant oncology.

RECENT FINDINGS

The development of machine learning has occurred within three domains related to hepatocellular carcinoma: identification of key clinicopathological variables, genomics, and image processing.

SUMMARY

Machine-learning classifiers can be effectively applied for more accurate clinical prediction and handling of data, such as genetics and imaging in transplant oncology. This has allowed for the identification of factors that most significantly influence recurrence and survival in disease, such as hepatocellular carcinoma, and thus help in prognosticating patients who may benefit from a liver transplant. Although progress has been made in using these methods to analyse clinicopathological information, genomic profiles, and image processed data (both histopathological and radiomic), future progress relies on integrating data across these domains.

Overall Tumor Burden Dictates Outcomes for Patients Undergoing Resection of Multinodular Hepatocellular Carcinoma Beyond the Milan Criteria

OBJECTIVE

The objective of the current study was to define surgical outcomes after resection of multinodular hepatocellular carcinoma (HCC) beyond the Milan criteria, and develop a prediction tool to identify which patients likely benefit the most from resection.

BACKGROUND

Liver resection for multinodular HCC, especially beyond the Milan criteria, remains controversial. Rigorous selection of the best candidates for resection is essential to achieve optimal outcomes after liver resection of advanced tumors.

METHODS

Patients who underwent resection for HCC between 2000 and 2017 were identified from an international multi-institutional database. Patients were categorized according to Milan criteria status. Pre- and postoperative overall survival (OS) prediction models that included HCC tumor burden score (TBS) among patients with multinodular HCC beyond Milan criteria were developed and validated.

RESULTS

Among 1037 patients who underwent resection for HCC, 164 (15.8%) had multinodular HCC beyond the Milan criteria. Among patients with multinodular HCC, 25 (15.2%) patients experienced a serious complication and 90-day mortality was 3.7% (n = 6). Five-year OS after resection of multinodular HCC beyond Milan criteria was 52.8%. A preoperative TBS-based model (5-year OS: low-risk, 73.7% vs intermediate-risk, 45.1% vs high-risk, 13.1%), and postoperative TBS-based model (5-year OS: low-risk, 80.1% vs intermediate-risk, 37.2% vs high-risk, not reached) categorized patients into distinct prognostic groups relative to long-term prognosis (both P < 0.001). Pre- and postoperative models could accurately stratify OS in an external validation cohort (5-year OS; low vs medium vs high risk; pre: 66.3% vs 25.2% vs not reached, P = 0.012; post: 61.4% vs 42.5% vs not reached, P = 0.045) Predictive accuracy of the pre- and postoperative models was good in the training (c-index; pre: 0.68; post: 0.71), internal validation (n = 2000 resamples) (c-index, pre: 0.70; post: 0.72) and external validation (c-index, pre: 0.67; post 0.68) datasets. TBS alone could stratify patients relative to 5-year OS after resection of multinodular HCC beyond Milan criteria (c-index: 0.65; 5-year OS; low TBS: 70.2% vs medium TBS: 54.7% vs high TBS: 16.7%; P < 0.001). The vast majority of patients with low and intermediate TBS were deemed low or medium risk based on both the preoperative (98.4%) and postoperative risk scores (95.3%).

CONCLUSION

Prognosis of patients with multinodular HCC was largely dependent on overall tumor burden. Liver resection should be considered among patients with multinodular HCC beyond the Milan criteria who have a low- or intermediate-TBS.

Using Hepatocellular Carcinoma Tumor Burden Score to Stratify Prognosis after Liver Transplantation

Liver transplantation (LT) remains a mainstay of treatment for hepatocellular carcinoma (HCC). Tumor factors such as size and number of tumors define eligibility for LT using the Milan criteria. The tumor burden score (TBS) incorporates both tumor number and size into a single continuous variable and has been used to differentiate prognosis among patients undergoing resection for HCC. The objective of the present study was to evaluate the ability of the TBS to predict overall and recurrence-free survival in patients undergoing LT for HCC. The Scientific Registry of Transplant Recipients (SRTR) was used to analyze all liver transplants for HCC, with initial tumor size data from 2004 to 2018. There were 12,486 patients in the study period. In the unadjusted analyses, patients with a high TBS had worse overall (p < 0.0001) and recurrence-free (p < 0.0001) survival. In the adjusted analyses, a high TBS was associated with a greater hazard ratio (HR) of death (HR = 1.21; 95%CI, [1.13-1.30]; p < 0.001) and recurrence (HR = 1.49; 95%CI [1.3-1.7]; p < 0.001). When we superimposed the TBS on the Milan criteria, we saw that a higher TBS was associated with a higher hazard of recurrence at values that were either all within (HR = 1.20; 95%CI, [1.04-1.37]; p = 0.011) or variably within (HR = 1.53; 95%CI, [1.16-2.01]; p = 0.002) the Milan criteria. In conclusion, the TBS is a promising tool in predicting outcomes in patients with HCC after LT.

Prognostic role of artificial intelligence among patients with hepatocellular cancer: A systematic review

BACKGROUND

Prediction of survival after the treatment of hepatocellular carcinoma (HCC) has been widely investigated, yet remains inadequate. The application of artificial intelligence (AI) is emerging as a valid adjunct to traditional statistics due to the ability to process vast amounts of data and find hidden interconnections between variables. AI and deep learning are increasingly employed in several topics of liver cancer research, including diagnosis, pathology, and prognosis.

AIM

To assess the role of AI in the prediction of survival following HCC treatment.

METHODS

A web-based literature search was performed according to the Preferred Reporting Items for Systemic Reviews and Meta-Analysis guidelines using the keywords “artificial intelligence”, “deep learning” and “hepatocellular carcinoma” (and synonyms). The specific research question was formulated following the patient (patients with HCC), intervention (evaluation of HCC treatment using AI), comparison (evaluation without using AI), and outcome (patient death and/or tumor recurrence) structure. English language articles were retrieved, screened, and reviewed by the authors. The quality of the papers was assessed using the Risk of Bias In Non-randomized Studies of Interventions tool. Data were extracted and collected in a database.

RESULTS

Among the 598 articles screened, nine papers met the inclusion criteria, six of which had low-risk rates of bias. Eight articles were published in the last decade; all came from eastern countries. Patient sample size was extremely heterogenous (n = 11-22926). AI methodologies employed included artificial neural networks (ANN) in six studies, as well as support vector machine, artificial plant optimization, and peritumoral radiomics in the remaining three studies. All the studies testing the role of ANN compared the performance of ANN with traditional statistics. Training cohorts were used to train the neural networks that were then applied to validation cohorts. In all cases, the AI models demonstrated superior predictive performance compared with traditional statistics with significantly improved areas under the curve.

CONCLUSION

AI applied to survival prediction after HCC treatment provided enhanced accuracy compared with conventional linear systems of analysis. Improved transferability and reproducibility will facilitate the widespread use of AI methodologies.

Prognostic nomograms and risk classifications of outcomes in very early-stage hepatocellular carcinoma patients after hepatectomy

BACKGROUND

Numerous clinical models have been proposed to evaluate and predict recurrence and survival of hepatocellular carcinoma (HCC) patients in different stages after resection, but no model for very early-stage HCC.

METHODS

The data of 661 very early-stage HCC patients after curative resection in our hospital were retrospectively reviewed. Kaplan-Meier curves and Cox proportional hazards regression models were used to analyze recurrence and survival. The risk classifications for recurrence and survival were established by using classification and regression tree analysis. The nomograms were constructed and validated using bootstrap resampling and an independent 186-patient validation cohort from the same institution.

RESULTS

According to the results of multivariate analysis for prognosis after resection, decision trees and 3-stratification classifications that satisfactorily determined the risk of recurrence and survival were established. Based on these two risk classifications, a six-factor nomogram for predicting recurrence and a six-factor nomogram for predicting survival were created. The concordance indexes were 0.64 for recurrence nomogram, with a 95% confidence interval of 0.60-0.67, and 0.76 for survival nomogram, with a 95% confidence interval of 0.70-0.82. The calibration curves showed good agreement between the predictions made by the nomograms and the actual survival outcomes. These predicting results for recurrence and survival were better than three common classical HCC stages and were confirmed in the independent validation cohort.

CONCLUSIONS

The 3-stratification classifications enabled satisfactory risk evaluations of recurrence and survival, and the nomograms showed considerably accurate predictions of the risk of recurrence and survival in very early-stage HCC patients after curative resection.

Prediction of tumor recurrence by α-fetoprotein model after curative resection for hepatocellular carcinoma

BACKGROUND

Preoperative α-fetoprotein (AFP) level levels may help select patients with hepatocellular carcinoma (HCC) for surgery. The objective of the current study was to assess an AFP model to predict tumor recurrence and patient survival after curative resection for HCC.

METHODS

Patients undergoing curative-intent resection for HCC between 2000 and 2017 were identified from a multi-institutional database. AFP score was calculated based on the last evaluation before surgery. Probabilities of tumor recurrence and overall survival (OS) were compared according to an AFP model.

RESULTS

A total of 825 patients were included. An optimal cut-off AFP score of 2 was identified with an AFP score ≥3 versus ≤2 independently predicting tumor recurrence and OS. Net reclassification improvements indicated the AFP model was superior to the Barcelona Clinic Liver Cancer (BCLC) system to predict recurrence (p < 0.001). Among patients with BCLC B-C, AFP score ≤2 identified a subgroup of patients with AFP levels of ≤100 ng/mL with a low 5-year recurrence risk (≤2 45.2% vs. ≥3 61.8%, p = 0.046) and favorable 5-year OS (≤2 54.5% vs. ≥3 39.4%, p = 0.035). In contrast, among patients within BCLC 0-A, AFP score ≥3 identified a subgroup of patients with AFP values > 1000 ng/mL with a high 5-year recurrence (≥3 47.9% vs. ≤2% 38.4%, p = 0.046) and worse 5-year OS (≥3 47.8% vs. ≤2 65.9%, p < 0.001). In addition, the AFP score independently correlated with vascular invasion, tumor differentiation and capsule invasion.

CONCLUSIONS

The AFP model was more accurate than the BCLC system to identify which HCC patients may benefit the most from surgical resection.

Part 1: Artificial intelligence technology in surgery

Artificial intelligence (AI) is one of the disruptive technologies of the fourth Industrial Revolution that is changing our work practices. This technology is in use in highly diverse industries including health care, defence, insurance and e-commerce. This review focuses on the relevance of AI to surgery. AI will aid surgeons with diagnostic decision-making, patient selection for surgery as well as improve patient pre- and post-operative care and management. Ethical considerations of AI with respect to patient rights and data privacy are highlighted. A further challenge is how best to present to national regulators a pragmatic way to assess AI as 'software as a medical device'. This relates to the ramifications for the adoption of AI technology in clinical practice, and its subsequent public funding support and reimbursement. It is evident that AI technology has important applications in surgery in the 21st century. The establishment of a key work programme in this area will be important if surgeons are to fully utilize AI in surgery.

Recurrence beyond the Milan criteria after curative-intent resection of hepatocellular carcinoma: A novel tumor-burden based prediction model

BACKGROUND

Accurate prediction of recurrence patterns of hepatocellular carcinoma (HCC) may allow for prioritization of patients for resection or transplantation as well as guide post-resection surveillance strategies.

METHODS

Patients who underwent curative-intent R0 resection for HCC between 2000 and 2017 were identified using a multi-institutional database. A prognostic model that incorporated HCC tumor burden score (TBS) to predict recurrence beyond the Milan criteria (MC) was developed and validated.

RESULTS

Among 718 patients who underwent R0 resection for HCC, 185 (25.8%) recurred within and 110 (15.3%) beyond the MC. On multivariable analysis, AFP more than 400 ng/mL (hazard ratio [HR] = 2.26; 95% confidence interval [CI]: 1.27-4.02), lymphovascular invasion (HR = 2.00; 95% CI: 1.14-3.50), and TBS (HR = 1.08; 95% CI: 1.03-1.12) were associated with recurrence beyond the MC. A weighted TBS-based score was constructed: [0.074*TBS + 0.692*lymphovascular invasion (yes: 1, no: 0) + 0.816*AFP > 400 (yes:1, no:0)]. Patients with a low, medium, and high TBS-based risk score had a 5-year incidence of recurring beyond the MC of 16.2%, 28.6%, and 47.2%, respectively (P < .001). The predictive accuracy of the model was very good in the training (C-index: 0.761) and validation (C-index: 0.706) datasets and outperformed the previously reported clinical risk score (CRS; C-index: 0.680).

CONCLUSION

A TBS-based model accurately predicted recurrence beyond MC after curative-intent resection of HCC and outperformed the CRS. Incorporating TBS allows for better risk stratification and identifies patients in need of closer surveillance.