Prediction Of Survival In Hepatocellular Carcinoma Research Articles

Comparative analysis of radiomics and deep-learning algorithms for survival prediction in hepatocellular carcinoma

To examine the comparative robustness of computed tomography (CT)-based conventional radiomics and deep-learning convolutional neural networks (CNN) to predict overall survival (OS) in HCC patients. Retrospectively, 114 HCC patients with pretherapeutic CT of the liver were randomized into a development (n = 85) and a validation (n = 29) cohort, including patients of all tumor stages and several applied therapies. In addition to clinical parameters, image annotations of the liver parenchyma and of tumor findings on CT were available. Cox-regression based on radiomics features and CNN models were established and combined with clinical parameters to predict OS. Model performance was assessed using the concordance index (C-index). Log-rank tests were used to test model-based patient stratification into high/low-risk groups. The clinical Cox-regression model achieved the best validation performance for OS (C-index [95% confidence interval (CI)] 0.74 [0.57–0.86]) with a significant difference between the risk groups (p = 0.03). In image analysis, the CNN models (lowest C-index [CI] 0.63 [0.39–0.83]; highest C-index [CI] 0.71 [0.49–0.88]) were superior to the corresponding radiomics models (lowest C-index [CI] 0.51 [0.30–0.73]; highest C-index [CI] 0.66 [0.48–0.79]). A significant risk stratification was not possible (p > 0.05). Under clinical conditions, CNN-algorithms demonstrate superior prognostic potential to predict OS in HCC patients compared to conventional radiomics approaches and could therefore provide important information in the clinical setting, especially when clinical data is limited.

Prognostic factors and survival prediction in hepatocellular carcinoma: development and validation of a novel nomogram based on the SEER database.

Current staging systems for hepatocellular carcinoma (HCC) still have limitations in clinical practice. Our study aimed to explore the prognostic factors and develop a new nomogram to predict the cancer-specific survival (CSS) for patients with HCC. A total of 6,166 HCC patients were identified from the Surveillance, Epidemiology, and End Results (SEER) database. Patients were randomly grouped into the training cohort (70%) and validation cohort (30%). Multivariate Cox analysis was used to identify prognostics factors for CSS of patients, then we incorporated these variables and presented a new nomogram to predict 2- and 5-year CSS. The performance of the nomogram was assessed with respect to its calibration, concordance index (C-index), area under the receiver operating characteristic (ROC) curve (AUC), and decision curve analysis (DCA). Multivariate Cox analysis revealed that American Joint Committee on Cancer (AJCC) stage, race, grade, surgery, chemotherapy, radiation, tumor size, bone metastasis (BM), and alpha-fetoprotein (AFP) were independently associated with CSS. The prediction nomogram which contained these predictors showed good performance, with a C-index of 0.802 [95% confidence interval (CI), 0.792-0.812] in the training cohort and 0.801 (95% CI, 0.787-0.815) in the validation cohort. The calibration curves demonstrated good agreement between the actual observation and the nomogram prediction. Furthermore, the nomogram showed improved discriminative capacity (AUC, 0.873 and 0.875 for 2- and 5-year CSS in validation set) compared to the 7th tumor-node-metastasis (TNM) staging system (AUC, 0.735 and 0.717). The DCA also indicated good application of the nomogram. This study presents a novel nomogram that incorporates the important prognostic factors of HCC, which can be conveniently used to accurately predict the 2- and 5-year CSS of patients with HCC, thus assisting individualized clinical decision making.

Abstract 6557: A 6-microbial signature risk score for overall survival prediction in hepatocellular carcinoma: a machine learning approach

Abstract Introduction: Hepatocellular carcinoma (HCC) is a severe type of cancer associated with high malignancy and poor prognosis. To reach a better prognosis, several factors affecting HCC patients must be considered. In this study, we aim to investigate the relationship between gut microbiota with the survival of HCC patients, and to build a machine learning (ML) prognostic model based on microbiome signatures. Methods: Microbial and survival data for HCC patients from The Cancer Genome Atlas Program (TCGA) were obtained using the cBioportal database. The microbiome signatures were screened for prognostic effect using the univariate cox regression. The significant microbial signatures were fitted into the Least Absolute Selection and Shrinkage Operator - Cox (LASSO-Cox) model to reduce multicollinearity. Signatures with non-zero LASSO coefficients were screened for independent effect in the multivariate cox regression analysis. Results: A total of 1407 microbial signatures were downloaded from TCGA dataset. 63 prognostic microbial signatures were identified using the univariate cox. A total of 30 prognostic microbial signatures were screened out using LASSO and further evaluated using independent effect to obtain microbiome risk score. Six microbial signatures were identified as independent prognostic markers including: Marichromatium (HR: 1.26; 95%CI: [0.01-1.56]), p=0.038), Paraclostridium (HR: 1.45; 95%CI:[1.15-1.82], p=0.002), Pseudorhodoferax (HR: 1.20; 95%CI:[1.01-1.43], p=0.037), Chitinimonas (HR: 1.47 95%CI: [1.10-1.97], p=0.009), Marinobacter (HR: 0.79; 95%CI: [0.65-0.97], p=0.022), Desulfovermiculus (HR: 1.44; 95%CI: [1.16-1.78, p=0.001]). Kaplan-Meier survival plot showed better overall survival (OS) in the low-risk group compared to the high-risk group (median OS: 83.6 vs. 41.8; log-rank p-value <0.0001). Conclusion: Our results show that the gut microbiota is significantly associated with HCC prognosis, in which 5 of them were associated with poorer prognosis and one was associated with better survival. These microbial signatures provide a potential prognostic marker for HCC survival. Citation Format: Nour N. Al-Bzour, Ayah N. Al-Bzour. A 6-microbial signature risk score for overall survival prediction in hepatocellular carcinoma: a machine learning approach [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6557.

Development and validation of a nomogram for survival prediction in hepatocellular carcinoma after partial hepatectomy

BackgroundThe prognosis for hepatocellular carcinoma (HCC) is complex due to its high level of heterogeneity, even after radical resection. This study was designed to develop and validate a prognostic nomogram for predicting the postoperative prognosis for HCC patients following partial hepatectomy.Patients and methodsWe extracted data on HCC patients and randomly divided them into two groups (primary and validation cohorts), using the Surveillance, Epidemiology and End Results (SEER) database. We developed the prediction model based on the data of the primary cohort and prognostic factors were evaluated using univariate and multivariate Cox regression analysis. A nomogram was constructed for predicting the 1-, 3-, and 5-year survival probability of HCC patients after surgery based on the results of the multivariate Cox regression analysis. The performance of the nomogram was evaluated in terms of its discrimination and calibration. To validated the model, discrimination and calibration were also evaluated in the validation cohort. Decision curve analysis (DCA) was performed to assess the clinical utility of the nomogram.ResultsA total of 890 patients who underwent partial hepatectomy for HCC were included in the study. The primary cohort enrolled 628 patients with a median follow-up time of 39 months, the 1-, 3-, and 5-year survival rate were 95.4%, 52.7% and 25.8% during follow-up. Multivariate Cox regression analysis showed that differentiation, tumor size, AFP and fibrosis were independently association with the prognosis of HCC patients after partial hepatectomy. The nomogram showed a moderate discrimination ith a C-index of 0.705 (95% CI 0.669 to 0.742), and good calibration. Similar discrimination with a C-index of 0.681 (95% CI 0.625 to 0.737), and calibration were also observed in the validation cohort. Decision curve analysis showed that the nomogram could be useful to predicting the prognosis in HCC patients following partial hepatectomy.ConclusionsThe proposed nomogram is highly predictive and has moderate calibration and discrimination, potentially contributing to the process of managing HCC patients after partial hepatectomy in an individualized way.

Machine learning-based transcriptome analysis of lipid metabolism biomarkers for the survival prediction in hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is the most common primary malignancy of the liver with a very high fatality rate. Our goal in this study is to find a reliable lipid metabolism-related signature associated with prognostic significance for HCC. In this study, HCC lipid metabolism-related molecular subtype analysis was conducted based on the 243 lipid metabolism genes collected from the Molecular Signatures Database. Several significant disparities in prognosis, clinicopathological characteristics, and immune and ferroptosis-related status were found across the three subtypes, especially between C1 and C3 subgroups. Differential expression analysis yielded 57 differentially expressed genes (DEGs) between C1 and C3 subtypes. GO and KEGG analysis was employed for functional annotation. Three of 21 prognostic DEGs (CXCL8, SLC10A1, and ADH4) were finally selected through machine-learning-based discovery and validation strategy. The risk score = (0.103) × expression value of CXCL8 + (−0.0333) × expression value of SLC10A1 + (−0.0812) × expression value of ADH4. We used these three to construct a HCC prognostic risk model, which stratified the patients of the validation cohort into two risk subtypes with significantly different overall survival. Our work provides possible significance of the lipid metabolism-associated model in stratifying patient prognosis and its feasibility to guide therapeutic selection.

Preoperative contrast-enhanced computed tomography-based radiomics model for overall survival prediction in hepatocellular carcinoma.

BACKGROUNDHepatocellular carcinoma (HCC) is the most common primary liver malignancy with a rising incidence worldwide. The prognosis of HCC patients after radical resection remains poor. Radiomics is a novel machine learning method that extracts quantitative features from medical images and provides predictive information of cancer, which can assist with cancer diagnosis, therapeutic decision-making and prognosis improvement.AIMTo develop and validate a contrast-enhanced computed tomography-based radiomics model for predicting the overall survival (OS) of HCC patients after radical hepatectomy.METHODSA total of 150 HCC patients were randomly divided into a training cohort (n = 107) and a validation cohort (n = 43). Radiomics features were extracted from the entire tumour lesion. The least absolute shrinkage and selection operator algorithm was applied for the selection of radiomics features and the construction of the radiomics signature. Univariate and multivariate Cox regression analyses were used to identify the independent prognostic factors and develop the predictive nomogram, incorporating clinicopathological characteristics and the radiomics signature. The accuracy of the nomogram was assessed with the concordance index, receiver operating characteristic (ROC) curve and calibration curve. The clinical utility was evaluated by decision curve analysis (DCA). Kaplan–Meier methodology was used to compare the survival between the low- and high-risk subgroups.RESULTSIn total, seven radiomics features were selected to construct the radiomics signature. According to the results of univariate and multivariate Cox regression analyses, alpha-fetoprotein (AFP), neutrophil-to-lymphocyte ratio (NLR) and radiomics signature were included to build the nomogram. The C-indices of the nomogram in the training and validation cohorts were 0.736 and 0.774, respectively. ROC curve analysis for predicting 1-, 3-, and 5-year OS confirmed satisfactory accuracy [training cohort, area under the curve (AUC) = 0.850, 0.791 and 0.823, respectively; validation cohort, AUC = 0.905, 0.884 and 0.911, respectively]. The calibration curve analysis indicated a good agreement between the nomogram-prediction and actual survival. DCA curves suggested that the nomogram had more benefit than traditional staging system models. Kaplan–Meier survival analysis indicated that patients in the low-risk group had longer OS and disease-free survival (all P < 0.0001).CONCLUSIONThe nomogram containing the radiomics signature, NLR and AFP is a reliable tool for predicting the OS of HCC patients.

The Landscape of Using Glycosyltransferase Gene Signatures for Overall Survival Prediction in Hepatocellular Carcinoma.

Hepatocellular carcinoma (HCC) is a heterogeneous disease that occurs in the setting of chronic liver diseases. The role of glycosyltransferase (GT) genes has recently been the focus of research associated with tumor development. However, the prognostic value of GT genes in HCC remains unclear. Therefore, this study aimed to identify GT genes related to HCC prognosis through bioinformatics analysis. We firstly constructed a prognostic signature based on four GT genes using univariate and least absolute shrinkage and selection operator (LASSO) Cox regression analyses in The Cancer Genome Atlas (TCGA) dataset. Next, the risk score of each patient was calculated, and HCC patients were divided into high- and low-risk groups. Kaplan–Meier analysis showed that the survival rate of high-risk patients was significantly lower than that of low-risk patients. Receiver operating characteristic (ROC) curves assessed that risk scores calculated with a four-gene signature could predict 3- and 5-year overall survival (OS) of HCC patients, revealing the prognostic ability of this gene signature. Moreover, univariate and multivariate Cox regression analyses demonstrated that the risk score was an independent prognostic factor of HCC. Finally, functional analysis revealed that immune-related pathways were enriched and the immune status was different between the two risk groups in HCC. In summary, the novel GT gene signature could be used for prognostic prediction of HCC. Thus, targeting the GT genes may serve as an alternative treatment strategy for HCC.

Identification of ferroptosis-related genes for overall survival prediction in hepatocellular carcinoma

Ferroptosis is a novel type of cell death depending on iron and is strongly related to the development of tumors. Hepatocellular carcinoma (HCC) is a malignancy with high incidence. Despite some reports demonstrating the relation between ferroptosis-related genes and HCC, more details have not been excavated. In the present study, we collected and analyzed HCC patients' datasets from the TCGA-LIHC project and ICGC portal, respectively. Through the bioinformatic methods, we screened 126 differentially expressed genes. Then a prognostic model was established with four genes (GPX2, MT3, PRDX1, and SRXN1). PRDX1 is the hub gene of the prognosis model and has a high expression in hepatocellular carcinoma tumor tissue and cell lines. We further found that silencing PRDX1 increased the accumulation of ferrous ions and lipid peroxidation accumulation in HEPG2 cells and promoted ferroptosis in hepatocellular carcinoma. In conclusion, the study demonstrated the four-gene signature can be used to predict HCC prognosis. It also revealed the potential function of the ferroptosis-related gene PRDX1 in HCC, which can be a biomarker of the prediction for HCC outcome.

Drug-Eluting Bead Transarterial Chemoembolization versus Conventional Transarterial Chemoembolization Both Combined Apatinib for Hepatocellular Carcinoma: A Retrospective, Propensity-Score Matched Study

PurposeThis study aims to compare the efficacy of drug-eluting bead transarterial chemoembolization (DEB-TACE) versus conventional TACE (cTACE), both combined with apatinib, and to establish predictive nomograms to support individualized survival prediction in hepatocellular carcinoma (HCC) patients.Patients and MethodsThis retrospective study assessed HCC patients from June 2015 to December 2019. Patients were classified as DEB-TACE plus apatinib (D-apatinib) and cTACE plus apatinib (c-apatinib). The endpoints were overall survival (OS) and progression-free survival (PFS). The nomograms were constructed, and the C-index, receiver operating characteristic (ROC) curve, and calibration curves were used to validate the nomograms. Propensity score matching (PSM) analysis was applied to reduce patient selection bias.ResultsA total of 174 patients were included. After PSM analysis, 58 pairs of patients were selected. Before PSM analysis, the median OS and PFS were 21.0 and 8.0 months in the D-apatinib group, respectively, which were better than the 18.0 and 5.0 months observed in the c-apatinib group (P < 0.05). The complete response (CR) rate and objective response rate (ORR) of the D-apatinib group were higher than those of the c-apatinib group. The C-index values of the nomograms in the D-apatinib group and the c-apatinib group were 0.826 and 0.802, and the area under the curve (AUC) values in the ROC curve were 0.934 and 0.892. After PSM analysis, the survival of patients treated with D-apatinib was better than that of patients treated with c-apatinib (P < 0.05). The C-index values were 0.854 and 0.794 in the D-apatinib group and the c-apatinib group, respectively, and the AUC values were 0.960 and 0.890. The incidence of adverse events was higher in the c-apatinib group.ConclusionDEB-TACE in combination with apatinib showed better treatment effectiveness for unresectable HCC. The nomograms can identify HCC patients who may benefit most from the treatment.

Construction and Validation of a Prognostic Gene-Based Model for Overall Survival Prediction in Hepatocellular Carcinoma Using an Integrated Statistical and Bioinformatic Approach.

Hepatocellular carcinoma (HCC) is one of the most common lethal cancers worldwide and is often related to late diagnosis and poor survival outcome. More evidence is demonstrating that gene-based prognostic models can be used to predict high-risk HCC patients. Therefore, our study aimed to construct a novel prognostic model for predicting the prognosis of HCC patients. We used multivariate Cox regression model with three hybrid penalties approach including least absolute shrinkage and selection operator (Lasso), adaptive lasso and elastic net algorithms for informative prognostic-related genes selection. Then, the best subset regression was used to identify the best prognostic gene signature. The prognostic gene-based risk score was constructed using the Cox coefficient of the prognostic gene signature. The model was evaluated by Kaplan–Meier (KM) and receiver operating characteristic curve (ROC) analyses. A novel four-gene signature associated with prognosis was identified and the risk score was constructed based on the four-gene signature. The risk score efficiently distinguished the patients into a high-risk group with poor prognosis. The time-dependent ROC analysis revealed that the risk model had a good performance with an area under the curve (AUC) of 0.780, 0.732, 0.733 in 1-, 2- and 3-year prognosis prediction in The Cancer Genome Atlas (TCGA) dataset. Moreover, the risk score revealed a high diagnostic performance to classify HCC from normal samples. The prognosis and diagnosis prediction performances of risk scores were verified in external validation datasets. Functional enrichment analysis of the four-gene signature and its co-expressed genes involved in the metabolic and cell cycle pathways was constructed. Overall, we developed a novel-gene-based prognostic model to predict high-risk HCC patients and we hope that our findings can provide promising insight to explore the role of the four-gene signature in HCC patients and aid risk classification.

Four Immune-Related Long Non-coding RNAs for Prognosis Prediction in Patients With Hepatocellular Carcinoma.

BackgroundLong non-coding RNA (LncRNA) plays an important role in the occurrence and development of hepatocellular carcinoma (HCC). This study aims to establish an immune-related LncRNA model for risk assessment and prognosis prediction in HCC patients.MethodsHepatocellular carcinoma patient samples with complete clinical data and corresponding whole transcriptome expression were obtained from the Cancer Genome Atlas (TCGA). Immune-related genes were acquired from the Gene Set Enrichment Analysis (GSEA) website and matched with LncRNA in the TCGA to get immune-related LncRNA. Least Absolute Shrinkage and Selection Operator (LASSO) regression was used for screening the candidate LncRNAs and calculating the risk coefficient to establish the prognosis model. Patients were divided into a high-risk group and a low-risk group depending on the median risk score. The reliability of the prediction was evaluated in the validation cohort and the whole cohort. GSEA and principal component analysis were used for function evaluation.ResultsA total of 319 samples met the screening criteria and were randomly distributed across the training cohort and the validation cohort. After comparison with the IMMUNE_RESPONSE gene set and the IMMUNE_SYSTEM_PROCESS gene set, a total of 3094 immune-related LncRNAs were screened. Ultimately, four immune-related LncRNAs were used to construct a formula using LASSO regression. According to the formula, the low-risk group showed a higher survival rate than the high-risk group in the validation cohort and the whole cohort. The receiver operating characteristic curves data demonstrated that the risk score was more specific than other traditional clinical characteristics in predicting the 5-year survival rate for HCC.ConclusionThe four-immune-related-LncRNA model can be used for survival prediction in HCC and guide clinical therapy.

Development and validation of a RNA binding protein gene pair-associated prognostic signature for prediction of overall survival in hepatocellular carcinoma

BackgroundIncreasing evidence has demonstrated the correlation between hepatocellular carcinoma (HCC) prognosis and RNA binding proteins (RBPs) dysregulation. Thus, we aimed to develop and validate a reliable prognostic signature that can estimate the prognosis for HCC.MethodsGene expression profiling and clinical information of 374 HCC patients were derived from the TCGA data portal. The survival-related RBP pairs were determined using univariate cox-regression analysis and the signature was built based on LASSO analysis. All patients were divided patients into high-and low-risk groups according to the optimal cut off of the signature score determined by time-dependent receiver operating characteristic (ROC) curve analysis. The predictive value of the signature was further validated in an independent cohort.ResultsA 37-RBP pairs signature consisting of 61 unique genes was constructed which was significantly associated with the survival. The RBP-related signature accurately predicted the prognosis of HCC patients, and patients in high-risk groups showed poor survival in two cohorts. The novel signature was an independent prognostic factor of HCC in two cohorts (all P < 0.001). Furthermore, the C-index of the prognostic model was 0.799, which was higher than that of many established risk models. Pathway and process enrichment analysis showed that the 61 unique genes were mainly enriched in translation, ncRNA metabolic process, RNA splicing, RNA modification, and translational termination.ConclusionThe novel proposed RBP-related signature based on relative expression orderings could serve as a promising independent prognostic biomarker for patients with HCC, and could improve the individualized survival prediction in HCC.

THU-103 - Hand-foot-skin reaction of grade two or higher within sixty days as the best response criterion for survival prediction in hepatocellular carcinoma treated by sorafenib

THU-103 - Hand-foot-skin reaction of grade two or higher within sixty days as the best response criterion for survival prediction in hepatocellular carcinoma treated by sorafenib