Dataset Information

Classification of early and late stage liver hepatocellular carcinoma patients from their genomics and epigenomics profiles.

ABSTRACT: BACKGROUND:Liver Hepatocellular Carcinoma (LIHC) is one of the major cancers worldwide, responsible for millions of premature deaths every year. Prediction of clinical staging is vital to implement optimal therapeutic strategy and prognostic prediction in cancer patients. However, to date, no method has been developed for predicting the stage of LIHC from the genomic profile of samples. METHODS:The Cancer Genome Atlas (TCGA) dataset of 173 early stage (stage-I), 177 late stage (stage-II, Stage-III and stage-IV) and 50 adjacent normal tissue samples for 60,483 RNA transcripts and 485,577 methylation CpG sites, was extensively analyzed to identify the key transcriptomic expression and methylation-based features using different feature selection techniques. Further, different classification models were developed based on selected key features to categorize different classes of samples implementing different machine learning algorithms. RESULTS:In the current study, in silico models have been developed for classifying LIHC patients in the early vs. late stage and cancerous vs. normal samples using RNA expression and DNA methylation data. TCGA datasets were extensively analyzed to identify differentially expressed RNA transcripts and methylated CpG sites that can discriminate early vs. late stages and cancer vs. normal samples of LIHC with high precision. Naive Bayes model developed using 51 features that combine 21 CpG methylation sites and 30 RNA transcripts achieved maximum MCC (Matthew's correlation coefficient) 0.58 with an accuracy of 78.87% on the validation dataset in discrimination of early and late stage. Additionally, the prediction models developed based on 5 RNA transcripts and 5 CpG sites classify LIHC and normal samples with an accuracy of 96-98% and AUC (Area Under the Receiver Operating Characteristic curve) 0.99. Besides, multiclass models also developed for classifying samples in the normal, early and late stage of cancer and achieved an accuracy of 76.54% and AUC of 0.86. CONCLUSION:Our study reveals stage prediction of LIHC samples with high accuracy based on the genomics and epigenomics profiling is a challenging task in comparison to the classification of cancerous and normal samples. Comprehensive analysis, differentially expressed RNA transcripts, methylated CpG sites in LIHC samples and prediction models are available from CancerLSP (http://webs.iiitd.edu.in/raghava/cancerlsp/).

SUBMITTER: Kaur H

PROVIDER: S-EPMC6730898 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Classification of early and late stage liver hepatocellular carcinoma patients from their genomics and epigenomics profiles.

Kaur Harpreet H Bhalla Sherry S Raghava Gajendra P S GPS

PloS one 20190906 9

<h4>Background</h4>Liver Hepatocellular Carcinoma (LIHC) is one of the major cancers worldwide, responsible for millions of premature deaths every year. Prediction of clinical staging is vital to implement optimal therapeutic strategy and prognostic prediction in cancer patients. However, to date, no method has been developed for predicting the stage of LIHC from the genomic profile of samples.<h4>Methods</h4>The Cancer Genome Atlas (TCGA) dataset of 173 early stage (stage-I), 177 late stage (st ...[more]

PMID: 31490960

Similar Datasets

Project description:BackgroundLiver zonation is a unique phenomenon in which the liver exhibits distinct functions among hepatocytes along the radial axis of the lobule. This phenomenon can cause the sectionalized initiation of several liver diseases, including hepatocellular carcinoma (HCC). However, few studies have explored the zonation features of HCC.MethodsFour single-cell RNA sequencing datasets were used to identify hepatocyte-specific zonation markers. Integrative analysis was then performed with a training RNA-seq cohort (616 HCC samples) and an external validating microarray cohort (285 HCC samples) from the International Cancer Genome Consortium, The Cancer Genome Atlas, Gene Expression Omnibus, and EMBL's European Bioinformatics Institute for clustering using non-negative matrix factorization consensus clustering based on zonation genes. Afterward, we evaluated the prognostic value, clinical characteristics, transcriptome and mutation features, immune infiltration, and immunotherapy response of the HCC subclasses.ResultsA total of 94 human hepatocyte-specific zonation markers (39 central markers and 55 portal markers) were identified for the first time. Subsequently, three subgroups of HCC, namely Cluster1, Cluster2, and Cluster3 were identified. Cluster1 exhibited a non-zonational-like signature with the worst prognosis. Cluster2 was intensively associated with a central-like signature and exhibited low immune infiltration and sensitivity toward immune blockade therapy. Cluster3 was intensively correlated with a portal-like signature with the best prognosis. Finally, we identified candidate therapeutic targets and agents for Cluster1 HCC samples.ConclusionThe current study established a novel HCC classification based on liver zonation signature. By classifying HCC into three clusters with non-zonational-like (Cluster1), central-like (Cluster2), and portal-like (Cluster3) features, this study provided new perspectives on the heterogeneity of HCC and shed new light on delivering precision medicine for HCC patients.

Project description:BackgroundSurvival after liver resection of hepatocellular carcinoma (HCC) remains poor because of a high incidence of recurrence. We sought to investigate risk factors, patterns, and long-term prognosis among patients with early and late recurrence after liver resection for hepatitis B virus (HBV)-associated HCC.MethodsData of consecutive patients undergoing curative resection for HBV-associated HCC were analyzed. According to the time to recurrence after surgery, recurrence was divided into early (≤2 years) and late recurrence (>2 years). Characteristics, patterns of initial recurrence, and postrecurrence survival (PRS) were compared between patients with early and late recurrence. Risk factors of early and late recurrence and predictors of PRS were identified by univariable and multivariable Cox regression analyses.ResultsAmong 894 patients, 322 (36.0%) and 282 (31.5%) developed early and late recurrence, respectively. On multivariable analyses, preoperative HBV-DNA >104 copies/mL was associated with both early and late recurrence, whereas postoperative no/irregular antiviral therapy was associated with late recurrence. Compared with patients with late recurrence, patients with early recurrence had a lower proportion of intrahepatic-only recurrence (72.0% vs. 91.1%, p < .001), as well as a lower chance of receiving potentially curative treatments for recurrence (33.9% vs. 50.7%, p < .001) and a worse median PRS (19.1 vs. 37.5 months, p < .001). Multivariable analysis demonstrated that early recurrence was independently associated with worse PRS (hazard ratio, 1.361; 95% confidence interval, 1.094-1.692; p = .006).ConclusionAlthough risk factors associated with early recurrence and late recurrence were different, a high preoperative HBV-DNA load was an independent hepatitis-related risk for both early and late recurrence. Early recurrence was associated with worse postrecurrence survival among patients with recurrence.Implications for practiceLiver resection is the main curative treatment for hepatocellular carcinoma (HCC), but postoperative survival remains poor because of high recurrence rates. This study investigated the risk factors and patterns of early and late recurrence and found that a high preoperative hepatitis B virus (HBV) DNA load was an independent hepatitis-related risk factor for both. Early recurrence was also independently associated with worse postrecurrence survival. These data may provide insights into different biological origin and behavior of early versus late recurrence after resection for HBV-associated HCC, which could be helpful to make individualized treatment decision for recurrent HCC, as well as strategies for surveillance recurrence after resection.

Dataset Information

Classification of early and late stage liver hepatocellular carcinoma patients from their genomics and epigenomics profiles.

Publications

Classification of early and late stage liver hepatocellular carcinoma patients from their genomics and epigenomics profiles.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets