Revisiting miRNA Association with Melanoma Recurrence and Metastasis from a Machine Learning Point of View.

Aigli Korfiati,Sophia Georgiou,George Sakellaropoulos,Katerina Grafanaki,Constantinos Stathopoulos,George C Kyriakopoulos,Ilias Skeparnias

doi:10.3390/ijms23031299

Abstract

The diagnostic and prognostic value of miRNAs in cutaneous melanoma (CM) has been broadly studied and supported by advanced bioinformatics tools. From early studies using miRNA arrays with several limitations, to the recent NGS-derived miRNA expression profiles, an accurate diagnostic panel of a comprehensive pre-specified set of miRNAs that could aid timely identification of specific cancer stages is still elusive, mainly because of the heterogeneity of the approaches and the samples. Herein, we summarize the existing studies that report several miRNAs as important diagnostic and prognostic biomarkers in CM. Using publicly available NGS data, we analyzed the correlation of specific miRNA expression profiles with the expression signatures of known gene targets. Combining network analytics with machine learning, we developed specific non-linear classification models that could successfully predict CM recurrence and metastasis, based on two newly identified miRNA signatures. Subsequent unbiased analyses and independent test sets (i.e., a dataset not used for training, as a validation cohort) using our prediction models resulted in 73.85% and 82.09% accuracy in predicting CM recurrence and metastasis, respectively. Overall, our approach combines detailed analysis of miRNA profiles with heuristic optimization and machine learning, which facilitates dimensionality reduction and optimization of the prediction models. Our approach provides an improved prediction strategy that could serve as an auxiliary tool towards precision treatment.

Highlights

Cutaneous malignant melanoma (CM) is the most dangerous, heterogeneous and with a strong propensity to metastasize, skin cancer
Subsequent unbiased analysis on an external test set, using the trained classification models, revealed an accuracy of 73.85% and 82.09% in melanoma recurrence and metastasis prediction, respectively
Additional GEO datasets deriving from studies on circulating miRNAs clearly demonstrated the utility of circulating cell-free microRNAs as potential blood biomarkers for stage III and IV CM patients and compared plasma of metastatic patients before and during immune checkpoint blockade (ICB) therapy with normal healthy donor samples [31,32]

Summary

Introduction

Cutaneous malignant melanoma (CM) is the most dangerous, heterogeneous and with a strong propensity to metastasize, skin cancer. Precision medicine approaches exploring-omics data have emerged to identify diagnostic and prognostic biomarkers, which can lead to early disease detection, the better understanding of the underlying biological mechanisms and the application of individualized treatment protocols to patients with CM Towards this goal, several studies have provided information using or combining genomics, transcriptomics and epigenomics analyses [6,7,8]. The current analysis underlines the importance of rationalized integration of data from multiple sources which could provide novel and accurate means to prognosis of CM progression, recurrence and metastasis that could apply for virtually any type of cancer It highlights the importance of machine learning classification models for precision medicine approaches, and the opportunity to identify elusive, so far, important new biomarkers

Datasets and miRNA Signatures from CM Patients

Transcriptomics and Gene Signatures from CM Patients

Specific miRNAs Expression Patterns Regulate Melanoma Related Genes

Findings

Discussion

Conclusions