Artificial intelligence-driven pan-cancer analysis reveals miRNA signatures for cancer stage prediction

Sathipati, Srinivasulu Yerukala; Tsai, Ming‐Ju; Shukla, Sanjay Kumar; Ho, Shinn-Ying

doi:10.1016/j.xhgg.2023.100190

Cited by 12 publications

(6 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, analysis of miRNA expression concerning clinicopathological factors like grade of tumour, KPS and tumour size confirmed that the level of both miR-362-3p and miR-6721-5p miRNAs exhibited a correlation with glioma grading. Our studies are following the latest observations showing that miR-362-3p, along with miR-3651 and let-71-3p, are the most significant contributors to stage prediction across eight types of cancer, including bladder carcinoma, breast invasive cancer, oesophageal carcinoma, kidney renal clear cell carcinoma, lung adenocarcinoma, stomach adenocarcinoma, and uveal melanoma ( Yerukala Sathipati et al, 2023 ). Similarly, Tito et al also demonstrated that miR-362-3p has diagnostic potential as a part of a signature panel comprising miR-193-3p, miR-572, miR-28-5, and miR-378, with an AUC of 0.801 in stage I of clear cell renal cell carcinoma ( Tito et al, 2021 ).…”

Section: Discussionsupporting

confidence: 61%

Circulating serum miR-362-3p and miR-6721-5p as potential biomarkers for classification patients with adult-type diffuse glioma

Niemira,

Bielska,

Chwialkowska

et al. 2024

Front. Mol. Biosci.

View full text Add to dashboard Cite

According to the fifth edition of the WHO Classification of Tumours of the Central Nervous System (CNS) published in 2021, grade 4 gliomas classification includes IDH-mutant astrocytomas and wild-type IDH glioblastomas. Unfortunately, despite precision oncology development, the prognosis for patients with grade 4 glioma remains poor, indicating an urgent need for better diagnostic and therapeutic strategies. Circulating miRNAs besides being important regulators of cancer development could serve as promising diagnostic biomarkers for patients with grade 4 glioma. Here, we propose a two-miRNA miR-362-3p and miR-6721-5p screening signature for serum for non-invasive classification of identified glioma cases into the highest-grade 4 and lower-grade gliomas. A total of 102 samples were included in this study, comprising 78 grade 4 glioma cases and 24 grade 2–3 glioma subjects. Using the NanoString platform, seven miRNAs were identified as differentially expressed (DE), which was subsequently confirmed via RT-qPCR analysis. Next, numerous combinations of DE miRNAs were employed to develop classification models. The dual panel of miR-362-3p and miR-6721-5p displayed the highest diagnostic value to differentiate grade 4 patients and lower grade cases with an AUC of 0.867. Additionally, this signature also had a high AUC = 0.854 in the verification cohorts by RT-qPCR and an AUC = 0.842 using external data from the GEO public database. The functional annotation analyses of predicted DE miRNA target genes showed their primary involvement in the STAT3 and HIF-1 signalling pathways and the signalling pathway of pluripotency of stem cells and glioblastoma-related pathways. For additional exploration of miRNA expression patterns correlated with glioma, we performed the Weighted Gene-Co Expression Network Analysis (WGCNA). We showed that the modules most associated with glioma grade contained as many as six DE miRNAs. In conclusion, this study presents the first evidence of serum miRNA expression profiling in adult-type diffuse glioma using a classification based on the WHO 2021 guidelines. We expect that the discovered dual miR-362-3p and miR-6721-5p signatures have the potential to be utilised for grading gliomas in clinical applications.

show abstract

Section: Discussionsupporting

confidence: 61%

Circulating serum miR-362-3p and miR-6721-5p as potential biomarkers for classification patients with adult-type diffuse glioma

Niemira,

Bielska,

Chwialkowska

et al. 2024

Front. Mol. Biosci.

View full text Add to dashboard Cite

show abstract

“…We establish the HNSC-Sig method based on SVR incorporated with optimal feature selection algorithm called IBCGA. The optimization technique has been successfully applied in various cancer survival estimations [ 28 , [31] , [32] , [33] , [34] , [35] ]. HNSC-Sig was designed to identify a survival associated miRNA signature as well as estimate the survival time in patients with HNSC.…”

Section: Methodsmentioning

confidence: 99%

Survival associated miRNA signature in patients with head and neck carcinomas

Sathipati

2023

Heliyon

Self Cite

View full text Add to dashboard Cite

“…Artificial intelligence (AI) is a sophisticated technology that recognizes complicated healthcare unit difficulties using mathematically based algorithmic concepts that are akin to those of the human mind. These methods may effectively extract significant characteristics and assist in identifying miRNA signatures, predicting targets, and predicting tumor-specific biomarkers by integrating and evaluating large datasets from various sources ( 115 ). AI can improve the miRNA biomarker finding process.…”

Section: Ai-based Strategies For Mirna Use In Bcmentioning

confidence: 99%

The role of miRNAs as biomarkers in breast cancer

Baylie,

Kasaw,

Getinet

et al. 2024

Front. Oncol.

View full text Add to dashboard Cite

Breast cancer (BC) is the second most common cause of deaths reported in women worldwide, and therefore there is a need to identify BC patients at an early stage as timely diagnosis would help in effective management and appropriate monitoring of patients. This will allow for proper patient monitoring and effective care. However, the absence of a particular biomarker for BC early diagnosis and surveillance makes it difficult to accomplish these objectives. miRNAs have been identified as master regulators of the molecular pathways that are emphasized in various tumors and that lead to the advancement of malignancies. Small, non-coding RNA molecules known as miRNAs target particular mRNAs to control the expression of genes. miRNAs dysregulation has been linked to the start and development of a number of human malignancies, including BC, since there is compelling evidence that miRNAs can function as tumor suppressor genes or oncogenes. The current level of knowledge on the role of miRNAs in BC diagnosis, prognosis, and treatment is presented in this review. miRNAs can regulate the tumorigenesis of BC through targeting PI3K pathway and can be used as prognostic or diagnostic biomarkers for BC therapy. Some miRNAs, like miR-9, miR-10b, and miR-17-5p, are becoming known as biomarkers of BC for diagnosis, prognosis, and therapeutic outcome prediction. Other miRNAs, like miR-30c, miR-187, and miR-339-5p, play significant roles in the regulation of hallmark functions of BC, including invasion, metastasis, proliferation, resting death, apoptosis, and genomic instability. Other miRNAs, such as miR-155 and miR-210, are circulating in bodily fluids and are therefore of interest as novel, conveniently accessible, reasonably priced, non-invasive methods for the customized care of patients with BC.

show abstract

Artificial intelligence-driven pan-cancer analysis reveals miRNA signatures for cancer stage prediction

Cited by 12 publications

References 49 publications

Circulating serum miR-362-3p and miR-6721-5p as potential biomarkers for classification patients with adult-type diffuse glioma

Circulating serum miR-362-3p and miR-6721-5p as potential biomarkers for classification patients with adult-type diffuse glioma

Survival associated miRNA signature in patients with head and neck carcinomas

The role of miRNAs as biomarkers in breast cancer

Contact Info

Product

Resources

About