Machine learning-based development and validation of a scoring system for progression-free survival in liver cancer

Artificial intelligence techniques in liver cancer

Hepatocellular Carcinoma (HCC), the most common primary liver cancer, is a significant contributor to worldwide cancer-related deaths. Various medical imaging techniques, including computed tomography, magnetic resonance imaging, and ultrasound, play a crucial role in accurately evaluating HCC and formulating effective treatment plans. Artificial Intelligence (AI) technologies have demonstrated potential in supporting physicians by providing more accurate and consistent medical diagnoses. Recent advancements have led to the development of AI-based multi-modal prediction systems. These systems integrate medical imaging with other modalities, such as electronic health record reports and clinical parameters, to enhance the accuracy of predicting biological characteristics and prognosis, including those associated with HCC. These multi-modal prediction systems pave the way for predicting the response to transarterial chemoembolization and microvascular invasion treatments and can assist clinicians in identifying the optimal patients with HCC who could benefit from interventional therapy. This paper provides an overview of the latest AI-based medical imaging models developed for diagnosing and predicting HCC. It also explores the challenges and potential future directions related to the clinical application of AI techniques.

Development of machine learning-based personalized predictive models for risk evaluation of hepatocellular carcinoma in hepatitis B virus-related cirrhosis patients with low levels of serum alpha-fetoprotein

INTRODUCTION AND OBJECTIVES

The increasing incidence of hepatocellular carcinoma (HCC) in China is an urgent issue, necessitating early diagnosis and treatment. This study aimed to develop personalized predictive models by combining machine learning (ML) technology with a demographic, medical history, and noninvasive biomarker data. These models can enhance the decision-making capabilities of physicians for HCC in hepatitis B virus (HBV)-related cirrhosis patients with low serum alpha-fetoprotein (AFP) levels.

PATIENTS AND METHODS

A total of 6,980 patients treated between January 2012 and December 2018 were included. Pre-treatment laboratory tests and clinical data were obtained. The significant risk factors for HCC were identified, and the relative risk of each variable affecting its diagnosis was calculated using ML and univariate regression analysis. The data set was then randomly partitioned into validation (20 %) and training sets (80 %) to develop the ML models.

RESULTS

Twelve independent risk factors for HCC were identified using Gaussian naïve Bayes, extreme gradient boosting (XGBoost), random forest, and least absolute shrinkage and selection operation regression models. Multivariate analysis revealed that male sex, age >60 years, alkaline phosphate >150 U/L, AFP >25 ng/mL, carcinoembryonic antigen >5 ng/mL, and fibrinogen >4 g/L were the risk factors, whereas hypertension, calcium <2.25 mmol/L, potassium ≤3.5 mmol/L, direct bilirubin >6.8 μmol/L, hemoglobin <110 g/L, and glutamic-pyruvic transaminase >40 U/L were the protective factors in HCC patients. Based on these factors, a nomogram was constructed, showing an area under the curve (AUC) of 0.746 (sensitivity = 0.710, specificity=0.646), which was significantly higher than AFP AUC of 0.658 (sensitivity = 0.462, specificity=0.766). Compared with several ML algorithms, the XGBoost model had an AUC of 0.832 (sensitivity = 0.745, specificity=0.766) and an independent validation AUC of 0.829 (sensitivity = 0.766, specificity = 0.737), making it the top-performing model in both sets. The external validation results have proven the accuracy of the XGBoost model.

CONCLUSIONS

The proposed XGBoost demonstrated a promising ability for individualized prediction of HCC in HBV-related cirrhosis patients with low-level AFP.

Development and Validation of a Novel Prognostic Nomogram Based on Platelet and CD8+T Cell Counts in Hepatocellular Carcinoma Patients with Portal Vein Tumor Thrombosis

Purpose

Portal vein tumor thrombosis (PVTT) is one of the hallmarks of advanced Hepatocellular carcinoma (HCC). Platelet (PLT) function parameters and CD8+T cells (CD8+Ts) play an important role in HCC progression and metastasis. This study is committed to establishing an efficient prognosis prediction model and exploring the combined effect of PLT and CD8+Ts on PVTT prognosis.

Patients and Methods

This retrospective study collected 932 HCC patients with PVTT from 2007 to 2017 and randomly divided them into a training cohort (n = 656) and a validation cohort (n = 276). We performed multivariable Cox and Elastic-net regression analysis, constructed a nomogram and used Kaplan-Meier survival curves to compare overall survival and progression-free survival rates in different substrata. Relationships between indicators involved were also analyzed.

Results

We found tumor number, size, treatment, PLT, γ-glutamyl transferase, alpha-fetoprotein, mean platelet volume, and CD8+Ts were related to the 5-year OS of patients with PVTT, and established a nomogram. The area under the receiver operating characteristic curve (AUCs) for predicting the 1-year OS rates were 0.767 and 0.794 in training and validation cohorts. The calibration curve and decision curve indicated its predictive consistency and strong clinical utility. We also found those with low PLT (<100*10^9/L) and high CD8+Ts (>320 cells/μL) had a better prognosis.

Conclusion

We established a well-performing prognostic model for PVTT based on platelet functional parameters and CD8+Ts, and found that PT-8 formed by PLT and CD8+Ts was an excellent predictor of the prognosis of PVTT.

Computed tomography-based radiomics to predict early recurrence of hepatocellular carcinoma post-hepatectomy in patients background on cirrhosis

BACKGROUND

The prognosis for hepatocellular carcinoma (HCC) in the presence of cirrhosis is unfavourable, primarily attributable to the high incidence of recurrence.

AIM

To develop a machine learning model for predicting early recurrence (ER) of post-hepatectomy HCC in patients with cirrhosis and to stratify patients' overall survival (OS) based on the predicted risk of recurrence.

METHODS

In this retrospective study, 214 HCC patients with cirrhosis who underwent curative hepatectomy were examined. Radiomics feature selection was conducted using the least absolute shrinkage and selection operator and recursive feature elimination methods. Clinical-radiologic features were selected through univariate and multivariate logistic regression analyses. Five machine learning methods were used for model comparison, aiming to identify the optimal model. The model's performance was evaluated using the receiver operating characteristic curve [area under the curve (AUC)], calibration, and decision curve analysis. Additionally, the Kaplan-Meier (K-M) curve was used to evaluate the stratification effect of the model on patient OS.

RESULTS

Within this study, the most effective predictive performance for ER of post-hepatectomy HCC in the background of cirrhosis was demonstrated by a model that integrated radiomics features and clinical-radiologic features. In the training cohort, this model attained an AUC of 0.844, while in the validation cohort, it achieved a value of 0.790. The K-M curves illustrated that the combined model not only facilitated risk stratification but also exhibited significant discriminatory ability concerning patients' OS.

CONCLUSION

The combined model, integrating both radiomics and clinical-radiologic characteristics, exhibited excellent performance in HCC with cirrhosis. The K-M curves assessing OS revealed statistically significant differences.

Identification of associated risk factors for serological distribution of hepatitis B virus via machine learning models

BACKGROUND

The provincial-level sero-survey was launched to learn the updated seroprevalence of hepatitis B virus (HBV) infection in the general population aged 1-69 years in Chongqing and to assess the risk factors for HBV infection to effectively screen persons with chronic hepatitis B (CHB).

METHODS

A total of 1828 individuals aged 1-69 years were investigated, and hepatitis B surface antigen (HBsAg), antibody to HBsAg (HBsAb), and antibody to B core antigen (HBcAb) were detected. Logistic regression and three machine learning (ML) algorithms, including random forest (RF), support vector machine (SVM), and stochastic gradient boosting (SGB), were developed for analysis.

RESULTS

The HBsAg prevalence of the total population was 3.83%, and among persons aged 1-14 years and 15-69 years, it was 0.24% and 4.89%, respectively. A large figure of 95.18% (770/809) of adults was unaware of their occult HBV infection. Age, region, and immunization history were found to be statistically associated with HBcAb prevalence with a logistic regression model. The prediction accuracies were 0.717, 0.727, and 0.725 for the proposed RF, SVM, and SGB models, respectively.

CONCLUSIONS

The logistic regression integrated with ML models could helpfully screen the risk factors for HBV infection and identify high-risk populations with CHB.

Evaluation of machine learning algorithms for the prognosis of breast cancer from the Surveillance, Epidemiology, and End Results database

INTRODUCTION

Many researchers used machine learning (ML) to predict the prognosis of breast cancer (BC) patients and noticed that the ML model had good individualized prediction performance.

OBJECTIVE

The cohort study was intended to establish a reliable data analysis model by comparing the performance of 10 common ML algorithms and the the traditional American Joint Committee on Cancer (AJCC) stage, and used this model in Web application development to provide a good individualized prediction for others.

METHODS

This study included 63145 BC patients from the Surveillance, Epidemiology, and End Results database.

RESULTS

Through the performance of the 10 ML algorithms and 7th AJCC stage in the optimal test set, we found that in terms of 5-year overall survival, multivariate adaptive regression splines (MARS) had the highest area under the curve (AUC) value (0.831) and F1-score (0.608), and both sensitivity (0.737) and specificity (0.772) were relatively high. Besides, MARS showed a highest AUC value (0.831, 95%confidence interval: 0.820-0.842) in comparison to the other ML algorithms and 7th AJCC stage (all P < 0.05). MARS, the best performing model, was selected for web application development (https://w12251393.shinyapps.io/app2/).

CONCLUSIONS

The comparative study of multiple forecasting models utilizing a large data noted that MARS based model achieved a much better performance compared to other ML algorithms and 7th AJCC stage in individualized estimation of survival of BC patients, which was very likely to be the next step towards precision medicine.

Development and validation of a nomogram for predicting the cancer-specific survival of fibrolamellar hepatocellular carcinoma patients

Fibrolamellar hepatocellular carcinoma (FLC) is a rare subtype of hepatocellular carcinoma. Our study aimed to construct a nomogram to predict the cancer-specific survival (CSS) of FLC. Data of 200 FLC patients enrolled in the Surveillance, Epidemiology, and End Results (SEER) database were divided into the training group and the validation group. Prognostic factors identified in the univariate and multivariate Cox regression analyses were used to construct the nomogram. The concordance index (C-index), calibration curves, time-dependent receiver operating characteristic curve (ROC), and decision curve analysis (DCA) were used to evaluate the performance of the nomogram. As a result, age ≥ 59, N1 stage, M1 stage, tumor size ≤ 2.0 cm, and no surgery were significantly associated with lower CSS in multivariate Cox regression analysis. The calibration plot showed good consistency of the nomogram between predicted and observed outcomes in the training and validation groups. Compared with the TNM staging system, the prognostic evaluation model (PEM) showed a higher C-index (0.823 vs 0.656). The PEM also showed better predictive performance, with areas under the curve of 0.909 and 0.890 for predicting the 1- and 5-year survival. The AUCs of the TNM stage model for predicting 1- and 5-year survival were 0.629 and 0.787, respectively. In addition, the DCA curve showed that the nomogram had better clinical utility. Finally, we concluded that Age, N stage, M stage, tumor size, and surgery are independent prognostic factors for FLC. PEM established based on these five prognostic indicators can help predict the CSS of patients with FLC.

A Novel Prognostic Score Based on Artificial Intelligence in Hepatocellular Carcinoma: A Long-Term Follow-Up Analysis

Objective

T cell immunity plays an important role in anti-tumor effects and immunosuppression often leads to the development and relapse of cancer. This study aimed to investigate the effect of T cell numbers on the long-term prognosis of patients with hepatocellular carcinoma (HCC) and construct an artificial neural network (ANN) model to evaluate its prognostic value.

Methods

We enrolled 3,427 patients with HCC at Beijing Ditan Hospital, Capital Medical University, and randomly divided them into two groups of 1,861 and 809 patients as the training and validation sets, respectively. Cox regression analysis was used to screen for independent risk factors of survival in patients with HCC. These factors were used to build an ANN model using Python. Concordance index, calibration curve, and decision curve analysis were used to evaluate the model performance.

Results

The 1-year, 3-year, 5-year, and 10-year cumulative overall survival (OS) rates were 66.9%, 45.7%, 34.9%, and 22.6%, respectively. Cox multivariate regression analysis showed that age, white blood cell count, creatinine, total bilirubin, γ-GGT, LDH, tumor size ≥ 5 cm, tumor number ≥ 2, portal vein tumor thrombus, and AFP ≥ 400 ng/ml were independent risk factors for long-term survival in HCC. Antiviral therapy, albumin, T cell, and CD8 T cell counts were independent protective factors. An ANN model was developed for long-term survival. The areas under the receiver operating characteristic (ROC) curve of 1-year, 3-year, and 5-year OS rates by ANNs were 0.838, 0.833, and 0.843, respectively, which were higher than those of the Barcelona Clinic Liver Cancer (BCLC), tumor node metastasis (TNM), Okuda, Chinese University Prognostic Index (CUPI), Cancer of the Liver Italian Program (CLIP), Japan Integrated Staging (JIS), and albumin–bilirubin (ALBI) models (P < 0.0001). According to the ANN model scores, all patients were divided into high-, middle-, and low-risk groups. Compared with low-risk patients, the hazard ratios of 5-year OS of the high-risk group were 8.11 (95% CI: 7.0-9.4) and 6.13 (95% CI: 4.28-8.79) (P<0.0001) in the training and validation sets, respectively.

Conclusion

High levels of circulating T cells and CD8 + T cells in peripheral blood may benefit the long-term survival of patients with HCC. The ANN model has a good individual prediction performance, which can be used to assess the prognosis of HCC and lay the foundation for the implementation of precision treatment in the future.

Alpha-Fetoprotein+Alkaline Phosphatase (A-A) Score Can Predict the Prognosis of Patients with Ruptured Hepatocellular Carcinoma Underwent Hepatectomy

Background

This research is aimed at establishing a scoring system alpha-fetoprotein+alkaline phosphatase (A-A score) based on preoperative serum alpha-fetoprotein (AFP) and alkaline phosphatase (ALP) levels and to investigate its clinical significance in patients with ruptured hepatocellular carcinoma (rHCC) after hepatectomy.

Methods

175 ruptured hepatocellular carcinoma (HCC) patients treated with hepatectomy were included. Survival analysis was assessed by the Kaplan-Meier method. Prognostic factors were analyzed in a multivariate model. Preoperative serum AFP and ALP values are assigned a score of 1 if they exceed the threshold value and 0 if they are below the threshold value, A-A score is obtained by summing the scores of two variables (AFP, ALP), and the predictive values of AFP, ALP, and A-A score were compared by receiver operating characteristic curve (ROC) analysis, and subgroup analyses were performed to further evaluate the power of A-A scores.

Results

Of the 175 patients, 67 (38.3%) had an A-A score of 0, 72 (41.1%) had an A-A score of 1, and 36 (20.6%) had an A-A score of 2. In multivariate analysis, the A-A score, the BCLC stage, and the extent of resection were independent predictors of OS in patients with rHCC. The 1-, 3-, and 5-year OS and RFS in patients with an A-A score of 1 were better than those with an A-A score of 0 and worse than those with an A-A score of 1 (all p < 0.05). Based on the results of ROC analysis, the A-A score is superior to AFP or ALP alone in predicting the prognosis of patients with ruptured HCC. In subgroup analysis, A-A score could accurately predict the prognosis of patients with or without microvascular invasion (MVI) and with different Child-Pugh grades or gender.

Conclusions

The A-A score can effectively predict the prognosis of patients after hepatectomy of ruptured hepatocellular carcinoma. At the same time, it also has good evaluation ability in different subgroups.

Nomogram for prediction of long-term survival with hepatocellular carcinoma based on NK cell counts

INTRODUCTION

Among all immune cells, natural killer (NK) cells play an important role as the first line of defense against tumor. The purpose of our study is to observe whether the NK cell counts can predict the overall survival of patients with hepatocellular carcinoma (HCC).

METHODS

To develop a novel model, from January 2010 to June 2015, HCC patients enrolled in Beijing Ditan hospital were divided into training and validation cohort. Cox multiple regression analysis was used to analyze the independent risk factors for 1-year, 3-year and 5-year overall survival (OS) of patients with HCC, and the nomogram was used to establish the prediction model. In addition, the decision tree was established to verify the contribution of NK cell counts to the survival of patients with HCC.

RESULTS

The model used in predicting overall survival of HCC included six variables (namely, NK cell counts, albumin (ALB) level, alpha-fetoprotein (AFP) level, portal vein tumor thrombus (PVTT), tumor number and treatment). The C-index of nomogram model in HCC patients predicting 1-year, 3-year and 5-year overall survival was 0.858, 0.788 and 0.782 respectively, which was higher than tumor-lymph node-metastasis (TNM) staging system, Okuda, model for end-stage liver disease (MELD), MELD-Na, the Chinese University Prognostic Index (CUPI) and Japan Integrated Staging (JIS) scores (p < 0.001). The decision tree showed the specific 5-year OS probability of HCC patients under different risk factors, and found that NK cell counts were the third in the column contribution.

CONCLUSIONS

Our study emphasizes the utility of NK cell counts for exploring interactions between long-term survival of HCC patients and predictor variables.

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.

Reappraisal of the Role of Alkaline Phosphatase in Hepatocellular Carcinoma

Background: Alkaline phosphatase (ALP) is a marker of liver function and is associated with biliary tract disease. It was reported as a prognostic factor for hepatocellular carcinoma (HCC). The genetic expression in tumor-tissue microarrays and the perioperative serologic changes in ALP have never been studied for their correlation with HCC prognosis. Methods: The genetic expression of ALP isoforms (placental (ALPP), intestinal (ALPI) and bone/kidney/liver (ALPL)) was analyzed in tumor and non-cancerous areas in 38 patients with HCC after partial hepatectomy. The perioperative change in ALP was further analyzed in a cohort containing 525 patients with HCC to correlate it with oncologic outcomes. A total of 43 HCC patients were enrolled for a volumetry study after major and minor hepatectomy. Results: The genetic expression of the bone/kidney/liver isoform was specifically and significantly higher in non-cancerous areas than in tumors. Patients with HCC with a higher ALP (>81 U/dL) had significantly more major hepatectomies, vascular invasion, and recurrence. Cox regression analysis showed that gender, major hepatectomies, the presence of satellite lesions, higher grades (III or IV) and perioperative changes in liver function tests were independent prognostic factors for recurrence-free survival, and a postoperative increase in the ALP ratio at postoperative day (POD) 7 vs. POD 0 > 1.46 should be emphasized. A liver regeneration rate more than 1.8 and correlation analysis revealed that the ALP level at POD 7 and 30 was significantly higher and correlated with remnant liver growth. Conclusions: This study demonstrated that the perioperative ALP change was an independent prognostic factor for HCC after partial hepatectomies, and the elevation of ALP represented a functional biomarker for the liver but not an HCC biomarker. The higher regeneration capacity was possibly associated with the elevation of ALP after operation.

An Inflammation-Index Signature Predicts Prognosis of Patients with Intrahepatic Cholangiocarcinoma After Curative Resection

Background

The prognosis of patients with intrahepatic cholangiocarcinoma (ICC) after resection is at great variance. We aimed to establish a novel prognostic nomogram in facilitating the risk stratification for these patients.

Methods

A total of 82 high-dimensional radiological and pathological data were analyzed by LASSO-penalized Cox regression analyses and the panels with the best predictive performance were selected. Specific nomograms were established based on the selected panels and were validated in both primary (n=292) and validation cohorts (n=107). The area under the receiver operating characteristic curve (AUC) and the concordance index (C-index) were used to compare the predictive ability of nomograms and other staging systems.

Results

The modified Glasgow Prognostic Score (mGPS) was identified as the prognostic factor for both overall survival (OS) and progression-free survival (PFS). The nomograms built on the prognostic factors showed powerful efficacy in survival prediction, with C-indexes of 0.800 (95% CI 0.767-0.833) and 0.752 (95% CI 0.718-0.786) for OS and PFS in the primary cohort, 0.659 (95% CI 0.586-0.732) and 0.638 (95% CI 0.571-0.705) for OS and PFS in the validation cohort, respectively. Compared with tumor-node-metastasis stage, Barcelona Clinic Liver Cancer staging score, Cancer of the Liver Italian Program score, and Okuda staging system, the nomograms had significantly higher values of AUC and C-indexes in survival prediction in the primary and validation cohorts.

Conclusion

Compared with currently used staging systems, the nomograms showed significantly higher efficacy in predicting survival of ICC patients after resection. The nomograms provide new versions of personalized management for these patients.

Current updates in machine learning in the prediction of therapeutic outcome of hepatocellular carcinoma: what should we know?

With the development of machine learning (ML) algorithms, a growing number of predictive models have been established for predicting the therapeutic outcome of patients with hepatocellular carcinoma (HCC) after various treatment modalities. By using the different combinations of clinical and radiological variables, ML algorithms can simulate human learning to detect hidden patterns within the data and play a critical role in artificial intelligence techniques. Compared to traditional statistical methods, ML methods have greater predictive effects. ML algorithms are widely applied in nearly all steps of model establishment, such as imaging feature extraction, predictive factor classification, and model development. Therefore, this review presents the literature pertaining to ML algorithms and aims to summarize the strengths and limitations of ML, as well as its potential value in prognostic prediction, after various treatment modalities for HCC.

Artificial intelligence in the diagnosis and management of hepatocellular carcinoma

Despite recent improvements in therapeutic interventions, hepatocellular carcinoma is still associated with a poor prognosis in patients with an advanced disease at diagnosis. Recently, significant progress has been made in image recognition through advances in the field of artificial intelligence (AI) (or machine learning), especially deep learning. AI is a multidisciplinary field that draws on the fields of computer science and mathematics for developing and implementing computer algorithms capable of maximizing the predictive accuracy from static or dynamic data sources using analytic or probabilistic models. Because of the multifactorial and complex nature of liver diseases, the machine learning approach to integrate multiple factors would appear to be an advantageous approach to improve the likelihood of making a precise diagnosis and predicting the response of treatment and prognosis of liver diseases. In this review, we attempted to summarize the potential use of AI in the diagnosis and management of liver diseases, especially hepatocellular carcinoma.

A Machine Learning Approach Yields a Multiparameter Prognostic Marker in Liver Cancer

A number of staging systems have been developed to predict clinical outcomes in hepatocellular carcinoma (HCC). However, no general consensus has been reached regarding the optimal model. New approaches such as machine learning (ML) strategies are powerful tools for incorporating risk factors from multiple platforms. We retrospectively reviewed the baseline information, including clinicopathologic characteristics, laboratory parameters, and peripheral immune features reflecting T-cell function, from three HCC cohorts. A gradient-boosting survival (GBS) classifier was trained with prognosis-related variables in the training dataset and validated in two independent cohorts. We constructed a 20-feature GBS model classifier incorporating one clinical feature, 14 laboratory parameters, and five T-cell function parameters obtained from peripheral blood mononuclear cells. The GBS model-derived risk scores demonstrated high concordance indexes (C-indexes): 0.844, 0.827, and 0.806 in the training set and validation sets 1 and 2, respectively. The GBS classifier could separate patients into high-, medium- and low-risk subgroups with respect to death in all datasets (P < 0.05 for all comparisons). A higher risk score was positively correlated with a higher clinical stage and the presence of portal vein tumor thrombus (PVTT). Subgroup analyses with respect to Child-Pugh class, Barcelona Clinic Liver Cancer stage, and PVTT status supported the prognostic relevance of the GBS-derived risk algorithm independent of the conventional tumor staging system. In summary, a multiparameter ML algorithm incorporating clinical characteristics, laboratory parameters, and peripheral immune signatures offers a different approach to identify patients with the greatest risk of HCC-related death.