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

CT radiomics-based biomarkers can predict response to immunotherapy in hepatocellular carcinoma

Hepatocellular Carcinoma (HCC) remains a leading cause of cancer deaths. Despite the rise of immunotherapies, many HCC patients don't benefit. There's a clear need for biomarkers to guide treatment decisions. This research aims to identify such biomarkers by combining radiological data and machine learning. We analyzed clinical and CT imaging data of 54 HCC patients undergoing immunotherapy. Radiologic features were examined to develop a model predicting short-term immunotherapy effects. We utilized 9 machine learning and 2 ensemble learning techniques using RapidMiner for model construction. We conducted the validation of the above feature combination using 29 HCC patients who received immunotherapy from another hospital, and tested and validated it using XGBoost combined with a genetic algorithm. In 54 HCC patients, radiomics features varied significantly between those with partial response (PR) and stable disease (SD). Key features in Gray Level Run Length Matrix (GLRLM) and in adjacent tissues' Intensity Direct, Neighborhood Gray Tone Difference Matrix (NGTDM), and Shape correlated with short-term immunotherapy efficacy. Selected feature combinations of 15, 19, and 8/15 features yielded better outcomes. Deep learning, random forest, and naive bayes outperformed other methods, with Bagging being more reliable than Adaboost. In the validation set of 29 HCC patients, the mentioned feature combination also demonstrated favorable performance. Furthermore, we achieved improved results when employing XGBoost in conjunction with a genetic algorithm for testing and validation. The machine learning model built with CT image features derived from GLCM, GLRLM, IntensityDirect, NGTDM, and Shape can accurately forecast the short-term efficacy of immunotherapy for HCC.

Alternative Splicing, RNA Editing, and the Current Limits of Next Generation Sequencing

The advent of next generation sequencing (NGS) has fostered a shift in basic analytic strategies of a gene expression analysis in diverse pathologies for the purposes of research, pharmacology, and personalized medicine. What was once highly focused research on individual signaling pathways or pathway members has, from the time of gene expression arrays, become a global analysis of gene expression that has aided in identifying novel pathway interactions, the discovery of new therapeutic targets, and the establishment of disease-associated profiles for assessing progression, stratification, or a therapeutic response. But there are significant caveats to this analysis that do not allow for the construction of the full picture. The lack of timely updates to publicly available databases and the "hit and miss" deposition of scientific data to these databases relegate a large amount of potentially important data to "garbage", begging the question, "how much are we really missing?" This brief perspective aims to highlight some of the limitations that RNA binding/modifying proteins and RNA processing impose on our current usage of NGS technologies as relating to cancer and how not fully appreciating the limitations of current NGS technology may negatively affect therapeutic strategies in the long run.

Detection of key mRNAs in liver tissue of hepatocellular carcinoma patients based on machine learning and bioinformatics analysis

One methodology extensively used to develop biomarkers is the precise detection of highly responsive genes that can distinguish cancer samples from healthy samples. The purpose of this study was to screen for potential hepatocellular carcinoma (HCC) biomarkers based on non-fusion integrative multi-platform meta-analysis method. The gene expression profiles of liver tissue samples from two microarray platforms were initially analyzed using a meta-analysis based on an empirical Bayesian method to robust discover differentially expressed genes in HCC and non-tumor tissues. Then, using the bioinformatics technique of weighted correlation network analysis, the highly associated prioritized Differentially Expressed Genes (DEGs) were clustered. Co-expression network and topological analysis were utilized to identify sub-clusters and confirm candidate genes. Next, a diagnostic model was developed and validated using a machine learning algorithm. To construct a prognostic model, the Cox proportional hazard regression analysis was applied and validated. We identified three genes as specific biomarkers for the diagnosis of HCC based on accuracy and feasibility. The diagnostic model's area under the curve was 0.931 with confidence interval of 0.923-0.952.•Non-fusion integrative multi-platform meta-analysis method.•Classification methods and biomarkers recognition via machine learning method.•Biomarker validation models.

A Novel Method for Survival Prediction of Hepatocellular Carcinoma Using Feature-Selection Techniques

The World Health Organization (WHO) predicted that 10 million people would have died of cancer by 2020. According to recent studies, liver cancer is the most prevalent cancer worldwide. Hepatocellular carcinoma (HCC) is the leading cause of early-stage liver cancer. However, HCC occurs most frequently in patients with chronic liver conditions (such as cirrhosis). Therefore, it is important to predict liver cancer more explicitly by using machine learning. This study examines the survival prediction of a dataset of HCC based on three strategies. Originally, missing values are estimated using mean, mode, and k-Nearest Neighbor (k-NN). We then compare the different select features using the wrapper and embedded methods. The embedded method employs Least Absolute Shrinkage and Selection Operator (LASSO) and ridge regression in conjunction with Logistic Regression (LR). In the wrapper method, gradient boosting and random forests eliminate features recursively. Classification algorithms for predicting results include k-NN, Random Forest (RF), and Logistic Regression. The experimental results indicate that Recursive Feature Elimination with Gradient Boosting (RFE-GB) produces better results, with a 96.66% accuracy rate and a 95.66% F1-score.

Comprehensive Analysis Identified Mutation-Gene Signature Impacts the Prognosis Through Immune Function in Hepatocellular Carcinoma

Background

Hepatocellular carcinoma (HCC) is a life-threatening and refractory malignancy with poor outcome. Genetic mutations are the hallmark of cancer. Thus far, there is no comprehensive prognostic model constructed by mutation-gene transcriptome in HCC. The prognostic value of mutation-gene signature in HCC remains elusive.

Methods

RNA expression profiles and the corresponding clinical information were recruited from The Cancer Genome Atlas (TCGA) and International Cancer Genome Consortium (ICGC) databases. The least absolute shrinkage and selection operator (LASSO) Cox regression analysis was employed to establish gene signature. Kaplan–Meier curve and time-dependent receiver operating characteristic curve were implemented to evaluate the prognostic value. The Wilcoxon test was performed to analyze the expression of immune checkpoint genes, cell cycle genes, and tumor drug resistance genes in different risk groups. Finally, quantitative real-time PCR (qRT-RCR) and immunohistochemistry (IHC) were performed to validate the mRNA and protein expression between HCC and adjacent nontumorous tissues in an independent cohort.

Results

A prognostic model consisting of five mutated genes was established by LASSO Cox regression analysis. The prognostic model classified patients into high- and low-risk groups. Compared with the low‐risk group, patients in the high‐risk group had significantly worse survival results. The prognostic model can accurately predict the overall survival of HCC patients and predict overall survival more accurately when combined with stage. Furthermore, the immune checkpoint genes, cell cycle genes, and tumor drug resistance genes were higher expressed in the high-risk group compared in the low-risk group. In addition, the expression level of prognostic signature genes was validated in an independent sample cohort, which was consistent with RNA sequencing expression in the TCGA database.

Conclusion

The prediction model of HCC constructed using mutation-related genes is of great significance for clinical decision making and the personalized treatment of patients with HCC.

A Seven-Gene Signature to Predict Prognosis of Patients With Hepatocellular Carcinoma

Purpose: Hepatocellular carcinoma (HCC) is one of the most prevalent malignant diseases worldwide and has a poor prognosis. Gene-based prognostic models have been reported to predict the overall survival of patients with HCC. Unfortunately, most of the genes used in earlier prognostic models lack prospective validation and, thus, cannot be used in clinical practice.

Methods: Candidate genes were selected from GEPIA (Gene Expression Profiling Interactive Analysis), and their associations with patients’ survival were confirmed by RT-PCR using cDNA tissue microarrays established from patients with HCC after radical resection. A multivariate Cox proportion model was used to calculate the coefficient of corresponding gene. The expression of seven genes of interest (MKI67, AR, PLG, DNASE1L3, PTTG1, PPP1R1A, and TTR) with two reference genes was defined to calculate a risk score which determined groups of different risks.

Results: Our risk scoring efficiently classified patients (n = 129) with HCC into a low-, intermediate-, and high-risk group. The three groups showed meaningful distinction of 3-year overall survival rate, i.e., 88.9, 74.5, and 20.6% for the low-, intermediate-, and high-risk group, respectively. The prognostic prediction model of risk scores was subsequently verified using an independent prospective cohort (n = 77) and showed high accuracy.

Conclusion: Our seven-gene signature model performed excellent long-term prediction power and provided crucially guiding therapy for patients who are not a candidate for surgery.