Immune-related biomarkers predict the prognosis and immune response of breast cancer based on bioinformatic analysis and machine learning

Exploring Prognostic Gene Factors in Breast Cancer via Machine Learning

Breast cancer remains the most prevalent cancer in women. To date, its underlying molecular mechanisms have not been fully uncovered. The determination of gene factors is important to improve our understanding on breast cancer, which can correlate the specific gene expression and tumor staging. However, the knowledge in this regard is still far from complete. Thus, this study aimed to explore these knowledge gaps by analyzing existing gene expression profile data from 3149 breast cancer samples, where each sample was represented by the expression of 19,644 genes and classified into Nottingham histological grade (NHG) classes (Grade 1, 2, and 3). To this end, a machine learning-based framework was designed. First, the profile data were analyzed by using seven feature ranking algorithms to evaluate the importance of features (genes). Seven feature lists were generated, each of which sorted features in accordance with feature importance evaluated from a special aspect. Then, the incremental feature selection method was applied to each list to determine essential features for classification and building efficient classifiers. Consequently, overlapping genes, such as AURKA, CBX2, and MYBL2, were deemed as potentially related to breast cancer malignancy and prognosis, indicating that such genes were identified to be important by multiple feature ranking algorithms. In addition, the study formulated classification rules to reflect special gene expression patterns for three NHG classes. Some genes and rules were analyzed and supported by recent literature, providing new references for studying breast cancer.

PLAU and GREM1 are prognostic biomarkers for predicting immune response in lung adenocarcinoma

Lung adenocarcinoma (LUAD) is a common malignant tumor. Identification of biomarkers and understanding their potential functions will facilitate the treatment and diagnosis in LUAD patients. The yellow module (cor = 0.31, P = 2e-6) was selected as the core module based on weighted gene co-expression network analysis (WGCNA) by integrating RNA-seq data and tumor stage. Two upregulated genes (PLAU and GREM1) in yellow module were identified to be biomarkers. Kaplan-Meier curve analysis displayed that high expression levels of them had a poor overall survival (OS). And, their high expression levels revealed higher tumor stage and relapse possibility in LUAD patients, and could be a prognostic parameter. Both biomarkers showed similar immune cell expression profiles in low- and high-expression groups. Strongly positive correlation between both biomarkers and biomarkers of tumor-infiltrating lymphocytes were also clarified in TCGA-LUAD cohort. Importantly, single gene GSEA showed that transcriptional mis-regulation in cancer and microRNAs in cancer were enriched in LUAD patients. Therefore, a miRNA-mRNA-transcription factors (TFs) co-expression regulatory networks was constructed for each biomarker, various miRNAs and TFs were related to PLAU and GREM1. Among which, 6 downstream TFs were overlapped genes for both biomarkers. Notably, 2 of these TFs (FOXF1 and TFAP2A) exhibited significantly abnormal expression levels. Among which, FOXF1 was downregulated and TFAP2A was upregulated in TCGA-LUAD cohort. Both TFs showed a significantly positive correlation with the expression level of PLAU. In conclusion, we identified 2 biomarkers related to immune response and achieved a good accuracy in predicting OS in patients with LUAD.

The integration of multidisciplinary approaches revealed PTGES3 as a novel drug target for breast cancer treatment

BACKGROUND

The main challenge in personalized treatment of breast cancer (BC) is how to integrate massive amounts of computing resources and data. This study aimed to identify a novel molecular target that might be effective for BC prognosis and for targeted therapy by using network-based multidisciplinary approaches.

METHODS

Differentially expressed genes (DEGs) were first identified based on ESTIMATE analysis. A risk model in the TCGA-BRCA cohort was constructed using the risk score of six DEGs and validated in external and clinical in-house cohorts. Subsequently, independent prognostic factors in the internal and external cohorts were evaluated. Cell viability CCK-8 and wound healing assays were performed after PTGES3 siRNA was transiently transfected into the BC cell lines. Drug prediction and molecular docking between PTGES3 and drugs were further analyzed. Cell viability and PTGES3 expression in two BC cell lines after drug treatment were also investigated.

RESULTS

A novel six-gene signature (including APOOL, BNIP3, F2RL2, HINT3, PTGES3 and RTN3) was used to establish a prognostic risk stratification model. The risk score was an independent prognostic factor that was more accurate than clinicopathological risk factors alone in predicting overall survival (OS) in BC patients. A high risk score favored tumor stage/grade but not OS. PTGES3 had the highest hazard ratio among the six genes in the signature, and its mRNA and protein levels significantly increased in BC cell lines. PTGES3 knockdown significantly inhibited BC cell proliferation and migration. Three drugs (gedunin, genistein and diethylstilbestrol) were confirmed to target PTGES3, and genistein and diethylstilbestrol demonstrated stronger binding affinities than did gedunin. Genistein and diethylstilbestrol significantly inhibited BC cell proliferation and reduced the protein and mRNA levels of PTGES3.

CONCLUSIONS

PTGES3 was found to be a novel drug target in a robust six-gene prognostic signature that may serve as a potential therapeutic strategy for BC.

Role of UBE2C in Brain Cancer Invasion and Dissemination

Glioblastoma (GB) and brain metastases (BM) are the most common brain tumors in adults and are invariably associated with a dismal outcome. These highly malignant tumors share common features including increased invasion and migration of the primary or metastatic brain cancer cells, whose triggering mechanisms are largely unknown. Emerging evidence has suggested that the ubiquitin-conjugating enzyme E2C (UBE2C), essential for controlling cell cycle progression, is overexpressed in diverse malignancies, including brain cancer. This review highlights the crucial role of UBE2C in brain tumorigenesis and its association with higher proliferative phenotype and histopathological grade, with autophagy and apoptosis suppression, epithelial-to-mesenchymal transition (EMT), invasion, migration, and dissemination. High expression of UBE2C has been associated with patients' poor prognosis and drug resistance. UBE2C has also been proven as a promising therapeutic target, despite the lack of specific inhibitors. Thus, there is a need to further explore the role of UBE2C in malignant brain cancer and to develop effective targeted therapies for patients with this deadly disease.

Promoting proliferation and tumorigenesis of breast cancer: KCND2's significance as a prognostic factor

In recent years, the potassium voltage-gated channel subfamily D (KCND) channels, particularly KCND2 (also known as Kv4.2), have been suggested to play a role in a variety of cancers, but their role in breast cancer has not yet been revealed. We analyzed RNA sequencing data from The Cancer Genome Atlas database and the Genotype-Tissue Expression database to investigate the differential expression of KCND2 in breast cancer and normal breast tissue. In addition, we leveraged GO and KEGG analysis techniques to gain a better understanding of the potential functional enrichment of 500 genes related to KCND2. Our findings were validated using collected tissue samples and clinical data from hospitals showed that KCND2 is a crucial independent factor in the prognosis of breast cancer patients. The higher the expression of KCND2, the shorter the survival time of breast cancer patients. Colony formation assay confirmed that KCND2 promotes the proliferation of breast cancer cells, whereas transwell assay and wound healing assay verified that KCND2 promoted breast cancer invasion and migration. In addition, 5-Ethynyl-2'-deoxyuridine (EdU) and flow cytometry revealed that KCND2 affected the cycle changes of breast cancer cells and contributed to the G1/S phase transition of breast cancer cells. Overall, our study demonstrates that KCND2 holds a promising potential as a significant target for breast cancer diagnosis and therapy.

Identification of RNA-binding protein YBX3 as an oncogene in clear cell renal cell carcinoma

Y box binding protein 3 (YBX3) is an indispensable factor for protein synthesis, cellular growth, and proliferation, and is intricately involved in the progression of diverse tumor types. The objective of the current study was to investigate the role of YBX3 in the prognosis, immune infiltration, and progression of clear cell renal clear cell carcinoma (ccRCC). The expression level of YBX3 in ccRCC tissues was compared using The Cancer Genome Atlas (TCGA) and analyzed using the Wilcoxon rank sum test. Logistic regression and multivariate Cox analyses were subsequently employed to scrutinize the association between YBX3 expression and the clinicopathological characteristics of patients. The TIMER 2.0 tool was also utilized to quantify the degree of immune cell infiltration of YBX3. Kaplan Meier analysis was performed to assess the correlation between YBX3 and the survival rate. A high expression level of YBX3 was significantly correlated with the tumor pathological stage, histological grade, TNM stage, and the abundance of aDC, pDC, Th1, and Treg immune cells. Higher expression of YBX3 in advanced ccRCC was found to be associated with a lower overall survival rate in the M0, N0, and T2 subgroups. In vitro, after the silencing of YBX3 in A498 cells and overexpression of YBX3 in ACHN cells, cell proliferation, colony formation, migration, invasion, cell cycle assays, and flow cytometric apoptotic analysis were performed to evaluate the role of YBX3 in the progression of ccRCC. YBX3 was found to be intricately associated with the progression and prognosis of ccRCC, and may serve as an effective treatment target for ccRCC or a biomarker for prognosis prediction.

Comprehensive multiomics and in silico approach uncovers prognostic, immunological, and therapeutic roles of ANLN in lung adenocarcinoma

The anillin actin-binding protein (ANLN) is immensely overexpressed in cancers, including lung cancer (LC). Phytocompounds have gained interest due to their broader potential and reduced unwanted effects. Screening numerous compounds presents a challenge, but in silico molecular docking is pragmatic. The present study aims to identify the role of ANLN in lung adenocarcinoma (LUAD), along with identification and interaction analysis of anticancer and ANLN inhibitory phytocompounds followed by molecular dynamics (MD) simulation. Using a systematic approach, we found that ANLN is significantly overexpressed in LUAD and mutated with a frequency of 3.73%. It is linked with advanced stages, clinicopathological parameters, worsening of relapse-free survival (RFS), and overall survival (OS), pinpointing its oncogenic and prognostic potential. High-throughput screening and molecular docking of phytocompounds revealed that kaempferol (flavonoid aglycone) interacts strongly with the active site of ANLN protein via hydrogen bonds, Vander Waals interactions, and acts as a potent inhibitor. Furthermore, we discovered that ANLN expression was found to be significantly higher (p) in LC cells compared to normal cells. This is a propitious and first study to demonstrate ANLN and kaempferol interactions, which might eventually lead to removal of rout from cell cycle regulation posed by ANLN overexpression and allow it to resume normal processes of proliferation. Overall, this approach suggested a plausible biomarker role of ANLN and the combination of molecular docking subsequently led to the identification of contemporary phytocompounds, bearing symbolic anticancer effects. The findings would be advantageous for pharmaceutics but require validation using in vitro and in vivo methods. HIGHLIGHTS: • ANLN is significantly overexpressed in LUAD. • ANLN is implicated in the infiltration of TAMs and altering plasticity of TME. • Kaempferol (potential ANLN inhibitor) shows important interactions with ANLN which could remove the alterations in cell cycle regulation, imposed by ANLN overexpression eventually leading to normal process of cell proliferation.

Glycolysis induces Th2 cell infiltration and significantly affects prognosis and immunotherapy response to lung adenocarcinoma

Glycolysis has a major role in cancer progression and can affect the tumor immune microenvironment, while its specific role in lung adenocarcinoma (LUAD) remains poorly studied. We obtained publicly available data from The Cancer Genome Atlas and Gene Expression Omnibus databases and used R software to analyze the specific role of glycolysis in LUAD. The Single Sample Gene Set Enrichment Analysis (ssGSEA) indicated a correlation between glycolysis and unfavorable clinical outcome, as well as a repression effect on the immunotherapy response of LUAD patients. Pathway enrichment analysis revealed a significant enrichment of MYC targets, epithelial-mesenchymal transition (EMT), hypoxia, G2M checkpoint, and mTORC1 signaling pathways in patients with higher activity of glycolysis. Immune infiltration analysis showed a higher infiltration of M0 and M1 macrophages in patients with elevated activity of glycolysis. Moreover, we developed a prognosis model based on six glycolysis-related genes, including DLGAP5, TOP2A, KIF20A, OIP5, HJURP, and ANLN. Both the training and validation cohorts demonstrated the high efficiency of prognostic prediction in this model, which identified that patients with high risk may have a poorer prognosis and lower sensitivity to immunotherapy. Additionally, we also found that Th2 cell infiltration may predict poorer survival and resistance to immunotherapy. The study indicated that glycolysis is significantly associated with poor prognosis in patients with LUAD and immunotherapy resistance, which might be partly dependent on the Th2 cell infiltration. Additionally, the signature comprised of six genes related to glycolysis showed promising predictive value for LUAD prognosis.