Machine learning-based prognostic modeling of lysosome-related genes for predicting prognosis and immune status of patients with hepatocellular carcinoma

Abstract

Background: Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related deaths worldwide, and lysosomes play an important role in cancer progression as organelles that break down biomolecules such as proteins, nucleic acids, and polysaccharides; however, the molecular mechanisms of lysosome-related genes in hepatocellular carcinoma are not fully understood. Methods：We downloaded hepatocellular carcinoma datasets from the Cancer Genome Atlas(TCGA) and the Gene Expression Omnibus (GEO) as well as lysosome-related gene sets from AIMGO .After univariate Cox screening of the set of lysosome-associated genes differentially expressed in hepatocellular carcinoma and normal tissues, risk models were built by machine learning. Model effects were then assessed using the concordance index (C-index), Kaplan-Meier (K-M) and receiver operating characteristic curves (ROC), and the “GSVA” package was used to explore the biological function and immune microenvironment between the high- and low-risk groups, and the “IMvigor210CoreBiologies” package was used to analyse the response of the high- and low-risk groups to immunotherapy responsiveness, the “pRRophetic”package was used to explore the sensitivity of the high and low-risk groups to chemotherapeutic agents and finally the function of a key gene (RAMP3) was explored at the cellular level. Results ：univariate Cox yielded 46 differentially and prognostically significant lysosome-related genes and risk models were constructed using eight genes (RAMP3,GPLD1,FABP5,CD68,CSPG4,SORT1,CSPG5,CSF3R) derived from machine learning. The C-index and ROC showed that the risk model was a better predictor of clinical outcomes, with the K-M values indicating that the higher risk group had worse clinical outcomes. There were significant differences in biological function, immune microenvironment and responsiveness to immunotherapy and drug sensitivity between the high and low-risk groups. Finally, we found that RAMP3 inhibited the proliferation, migration and invasion of hepatocellular carcinoma cells and correlated with the sensitivity of hepatocellular carcinoma cells to Idarubicin. Conclusion：Lysosome-associated gene risk models built by machine learning can effectively predict patient prognosis and offer new prospects for chemotherapy and immunotherapy in HCC. In addition, cellular-level experiments suggest that RAMP3 may be a new target for the treatment of hepatocellular carcinoma.

Construction and validation of a novel lysosomal signature for hepatocellular carcinoma prognosis, diagnosis, and therapeutic decision-making

“…Previous studies have indicated that lysosomal-related genes may serve as potential targets for cancer therapy 14,16,41 . However, the clinical relevance of lysosomal-related genes in the diagnosis and treatment of primary liver cancer has not been fully elucidated.…”

“…Additionally, we observed that all 8 model genes play a crucial role in the progression and development of tumors through the regulation of lysosomal-related pathways. Recently, a prognostic model of related lysosome-related genes has also been reported 41 . The authors used 8 genes (RAMP3, GPLD1, FABP5, CD68, CSPG4, SORT1, CSPG5, CSF3R) to construct a risk model, and the study showed that the risk model could better predict the clinical outcome, and the higher the risk, the worse the clinical outcome.…”

HDAC2 promotes autophagy-associated HCC malignant progression by transcriptionally activating LAPTM4B

“…These findings reaffirm the significance of HDAC2 in controlling autophagy in HCC tumor cells. Meanwhile, 91 lysosome-associated differentially expressed genes (IFI30 [34] , HYAL2 [35] , HPS6, ARSB, WDR24 [36] , CSF3R [37] , FNIP2 [38] , ARSK and so on) were detected, demonstrating that HDAC2 can significantly enrich to lysosomal pathways (Fig 4B). As lysosomes are the primary executors of autophagy function [39] , we focused downstream targets of HDAC2 on lysosomal genes.…”

Multiple machine learning algorithms identified SLC6A8 as a diagnostic biomarker of the late stage of Hepatocellular carcinoma

“…Genes such as Receptor Activity Modifying Protein 3 (RAMP3) and CD68 Molecule (CD68) have shown signi cantly expressed in HCC based on machine-learning algorithms [13]. Overall, these genes are considered vital candidates for potential diagnostic markers in HCC clinical settings.…”

“…At the same time, bioinformatic analysis and machine learning have emerged as increasingly promising strategies for comprehensive and indepth analysis of large datasets, such as transcriptome sequences, and interdisciplinary collaborations have been instrumental in advancing clinical therapeutic methods [2]. The advent of modern computerassisted medical science has provided signi cant guidance and hope for previously untreatable diseases, such as utilizing the XGBoost algorithm for HCC diagnosis [13]. To meet the demand for early diagnosis, numerous efforts have focused on developing new methods based on deep learning analysis [33].…”