Circulating miRNA Expression Profiles and Machine Learning Models in Association with Response to Irinotecan-Based Treatment in Metastatic Colorectal Cancer

Identification of ASMTL-AS1 and LINC02604 lncRNAs as novel biomarkers for diagnosis of colorectal cancer

PURPOSE

Colorectal cancer is one of the major leading causes of death worldwide, and available treatments for advanced colorectal cancer are not successful. Therefore, early detection of colorectal cancer is essential to improve patient survival, and biomarkers are potential tools to achieve this goal. Considering the key role of lncRNAs in cancers, the aim of this study is to identify lncRNAs involved in colorectal cancer as new potential prognosis biomarkers for CRC.

METHODS

In this observational study, gene expression data obtained from the TCGA database were analyzed, Identification of differentially expressed mRNAs, miRNAs, and lncRNAs was performed, and ceRNA network was drawn. Also, survival analysis of patients was performed in order to identify potential biomarkers related to the diagnosis and prognosis of colon cancer. After confirming the results using the GSE39582 dataset, the expression of target lncRNAs in colorectal tumor tissues was also investigated to confirm the bioinformatic data.

RESULTS

Analysis of the TCGA data showed that the expression of three lncRNAs-SNHG7, ASMTL-AS1, and LINC02604-that had the highest interaction with other miRNAs and mRNAs identified based on the ceRNA network was increased in colorectal cancer. Also, based on the ceRNA network, three microRNAs, hsa-let-7d-5p, hsa-mir-92a-3p, and hsa-mir-423-5p, and eight mRNAs, including CPA4, MSI2, RRM2, IGF2BP1, ONECUT2, HMGA1, SOX4, and SRM, were associated with all three mentioned lncRNAs, the expression of microRNAs was decreased and the expression of mRNAs was increased. By enrichment analysis, it was found that the target lncRNAs are involved in the processes of cell proliferation, apoptosis, and metastasis, indicating their importance in the development and malignancy of colorectal cancer. Furthermore, Kaplan-Meier analysis showed a significant increase in mortality in patients with higher expression levels of these lncRNAs. Analysis of the GSE39582 dataset, and real-time RT-PCR analysis, confirmed our bioinformatic results. Also, ROC analysis showed that SNHG7 was a relatively good promising biomarker (AUC = 0.73, p value = 0.02), while ASMTL-AS1 (AUC = 0.92, p value < 0.0001) and LINC02604 (AUC = 1.00, p value < 0.0001) emerged as excellent diagnostic biomarkers in colorectal cancer.

CONCLUSION

It seems that increased expression of lncRNAs ASMTL-AS1 and LINC02604 can serve as molecular biomarkers for CRC, possibly through the sponge hsa-let-7d-5p, hsa-mir-92a-3p, and hsa-mir-423 5p, which increases target mRNAs, which are effective in the carcinogenesis process.

A miRNA-disease association prediction model based on tree-path global feature extraction and fully connected artificial neural network with multi-head self-attention mechanism

BACKGROUND

MicroRNAs (miRNAs) emerge in various organisms, ranging from viruses to humans, and play crucial regulatory roles within cells, participating in a variety of biological processes. In numerous prediction methods for miRNA-disease associations, the issue of over-dependence on both similarity measurement data and the association matrix still hasn't been improved. In this paper, a miRNA-Disease association prediction model (called TP-MDA) based on tree path global feature extraction and fully connected artificial neural network (FANN) with multi-head self-attention mechanism is proposed. The TP-MDA model utilizes an association tree structure to represent the data relationships, multi-head self-attention mechanism for extracting feature vectors, and fully connected artificial neural network with 5-fold cross-validation for model training.

RESULTS

The experimental results indicate that the TP-MDA model outperforms the other comparative models, AUC is 0.9714. In the case studies of miRNAs associated with colorectal cancer and lung cancer, among the top 15 miRNAs predicted by the model, 12 in colorectal cancer and 15 in lung cancer were validated respectively, the accuracy is as high as 0.9227.

CONCLUSIONS

The model proposed in this paper can accurately predict the miRNA-disease association, and can serve as a valuable reference for data mining and association prediction in the fields of life sciences, biology, and disease genetics, among others.

Investigation of the potential of miRNA candidates as non-invasive biomarkers for the diagnosis and follow-up of colorectal cancer

BACKGROUND

Colorectal cancer (CRC) is the third most common cancer worldwide, and despite advances in treatment, molecular biomarkers are needed for both early diagnosis and prognosis monitoring. It is known that microRNAs (miRNA), one of the epigenetic mechanisms, are effective in the initiation and development of cancer by regulating the activity of tumor suppressors and/or oncogenes. In this study, the potential of the molecules let-7, miRNA125b, and miRNA30a, which are known to play a role in cellular processes, as biomarkers for colorectal cancer and their molecular mechanisms were investigated in this model. The aim was to evaluate the diagnostic, prognostic, and predictive utility of the target miRNAs in colorectal cancer patients.

MATERIAL AND METHODS

The expression changes of miRNAs let-7, miRNA125b, and miRNA30a were investigated by miRNAs isolation and cDNA synthesis from the serum samples of 60 patients diagnosed with CRC or from the serum samples of 20 healthy individuals. The calculation was performed using the quantitative real-time polymerase chain reaction method to determine the expression level. The results were compared with clinical parameters.

RESULT

An 8-fold decrease in the expression of let-7 and miRNA125b and a 60-fold decrease in the expression of miRNA30a were found in the serum samples of patients diagnosed with colorectal cancer (CRC) compared to the healthy group. A decrease in let-7 was observed in 53.3%, miRNA125b in 58.3%, and miRNA30a in 55% of patients. A significant correlation was found between the reduced expression status and the stage, lymph nodes, local recurrence, and metastasis (p < 0.05). The ROC analysis showed that the miRNA30a level could be a diagnostic biomarker for CRC (p < 0.001). No significant impact of target miRNA expression changes on overall disease survival was observed.

CONCLUSION

It is thought that the target miRNA30a can be used for early diagnosis and screening and that the target miRNA let-7, miRNA125b, and miRNA30a can be used as non-invasive biomarkers for disease follow-up, with larger patient studies being conducted on CRC patients.

Genomic and Transcriptomic Research in the Discovery and Application of Colorectal Cancer Circulating Markers

Colorectal cancer (CRC) is the most frequently occurring malignancy in the world. However, the mortality from CRC can be reduced through early diagnostics, selection of the most effective treatment, observation of the therapy success, and the earliest possible diagnosis of recurrences. A comprehensive analysis of genetic and epigenetic factors contributing to the CRC development is needed to refine diagnostic, therapeutic, and preventive strategies and to ensure appropriate decision making in managing specific CRC cases. The liquid biopsy approach utilizing circulating markers has demonstrated its good performance as a tool to detect the changes in the molecular pathways associated with various cancers. In this review, we attempted to brief the main tendencies in the development of circulating DNA and RNA-based markers in CRC such as cancer-associated DNA mutations, DNA methylation changes, and non-coding RNA expression shifts. Attention is devoted to the existing circulating nucleic acid-based CRC markers, the possibility of their application in clinical practice today, and their future improvement. Approaches to the discovery and verification of new markers are described, and the existing problems and potential solutions for them are highlighted.

Characterization of ferroptosis driver gene signature in head and neck squamous cell carcinoma (HNSC)

BACKGROUND

Head and neck squamous cell carcinoma (HNSC), a prevalent malignant tumor with a low survival rate, is often accompanied by ferroptosis, which is a recently-described type ofprogrammed cell death. Investigating the significance of ferroptosis driver genes in HNSC, this study aimed to assess their diagnostic and prognostic values, as well as their impact on treatment and tumor immune function. The results of this investigation provide novel insight into using ferroptosis-related genes as molecular biomarkers as well as precise chemotherapeutic targets for the therapy of HNSC.

METHODOLOGY

A detailed in silico and in vitro experiment-based methodology was adopted to achieve the goals.

RESULTS

A total of 233 ferroptosis driver genes were downloaded from the FerrDB database. After comprehensively analyzing these 233 ferroptosis driver genes by various TCGA databases, RNA-sequencing (RNA-seq), and Reverse Transcription Quantitative Real-Time Polymerase Chain Reaction (RT-qPCR) techniques, TP53 (tumor protein 53), PTEN (Phosphatase and TENsin homolog deleted on chromosome 10), KRAS (Ki-ras2 Kirsten rat sarcoma viral oncogene homolog), and HRAS (Harvey Rat sarcoma virus) were identified as differentially expressed hub genes. Interestingly, these hub genes were found to have significant (P < 0.05) variations in their mRNA and protein expressions and effects on overall survival of the HNSC patients. Moreover, targeted bisulfite-sequencing (bisulfite-seq) analysis revealed that promoter hypomethylation pattern was associated with up-regulation of hub genes (TP53, PTEN, KRAS, and HRAS). In addition to this, hub genes were involved in diverse oncogenic pathways.

CONCLUSION

Since HNSC pathogenesis is a complex process, using ferroptosis driver hub genes (TP53, PTEN, KRAS, and HRAS) as a diagnostic and prognostic tool, and therapeutically targeting those genes through appropriate drugs could bring a milestone change in the drug discovery and management and survival in HNSC.

Expression Analysis of hsa-miR-181a-5p, hsa-miR-143-3p, hsa-miR-132-3p and hsa-miR-23a-3p as Biomarkers in Colorectal Cancer-Relationship to the Body Mass Index

This work aims to investigate the expression levels of four preselected miRNAs previously linked to cancer and/or obesity, with the purpose of finding potential biomarkers in the clinical management of CRC developed by patients showing different BMI values. We analyzed samples from a total of 65 subjects: 43 affected by CRC and 22 without cancer. Serum and both subcutaneous and omental adipose tissues (SAT and OAT) were investigated, as well as tumor and non-tumor colorectal tissues in the case of the CRC patients. The relative expression (2^-∆∆Ct) levels of 4 miRNAs (hsa-miR-181a-5p, hsa-miR-143-3p, has-miR-132-3p and hsa-miR-23a-3p) were measured by RT-qPCR. Serum, SAT and OAT expression levels of these miRNAs showed significant differences between subjects with and without CRC, especially in the group of overweight/obese subjects. In CRC, serum levels of hsa-miR-143-3p clearly correlated with their levels in both SAT and OAT, independently of the BMI group. Moreover, hsa-miR-181a-5p could be considered as a biomarker in CRC patients with BMI ≥ 25 Kg/m² and emerges as a tumor location marker. We conclude that both adiposity and CRC induce changes in the expression of the miRNAs investigated, and hsa-miR-143-3p and hsa-miR-181a-5p expression analysis could be useful in the clinical management of CRC.