The promising role of new molecular biomarkers in prostate cancer: from coding and non-coding genes to artificial intelligence approaches

Abstract

Background: Risk stratification or progression in prostate cancer is performed with the support of clinical-pathological data such as the sum of the Gleason score and serum levels PSA. For several decades, methods aimed at the early detection of prostate cancer have included the determination of PSA serum levels. The aim of this systematic review is to provide an overview about recent advances in the discovery of new molecular biomarkers through transcriptomics, genomics and artificial intelligence that are expected to improve clinical management of the prostate cancer patient.

Methods: An exhaustive search was conducted by Pubmed, Google Scholar and Connected Papers using keywords relating to the genetics, genomics and artificial intelligence in prostate cancer, it includes "biomarkers", "non-coding RNAs", "lncRNAs", "microRNAs", "repetitive sequence", "prognosis", "prediction", "whole-genome sequencing", "RNA-Seq", "transcriptome", "machine learning", and "deep learning".

Results: New advances, including the search for changes in novel biomarkers such as mRNAs, microRNAs, lncRNAs, and repetitive sequences, are expected to contribute to an earlier and accurate diagnosis for each patient in the context of precision medicine, thus improving the prognosis and quality of life of patients. We analyze several aspects that are relevant for prostate cancer including its new molecular markers associated with diagnosis, prognosis, and prediction to therapy and how bioinformatic approaches such as machine learning and deep learning can contribute to clinic. Furthermore, we also include current techniques that will allow an earlier diagnosis, such as Spatial Transcriptomics, Exome Sequencing, and Whole-Genome Sequencing.

Conclusion: Transcriptomic and genomic analysis have contributed to generate knowledge in the field of prostate carcinogenesis, new information about coding and non-coding genes as biomarkers has emerged. Synergies created by the implementation of artificial intelligence to analyze and understand sequencing data have allowed the development of clinical strategies that facilitate decision-making and improve personalized management in prostate cancer.

Fig. 1. Landscape of precision medicine and molecular tools in prostate cancer.

A Sample collection. This can be achieved by sampling tissue, blood or even urine (a non-invasive sampling) from the patient and proceeding with a direct detection of the biomarker by in situ hybridization in the tissue sample or a nucleic acid extraction and a molecular assay. B Quantitative PCR (qPCR). This molecular tool can be used to quantify gene expression by determining the amount of a target sequence present in the sample based on fluorescent emission, such as My Prostate Score [119]. C Transcription Mediated Amplification (TMA), PROGENSA is a current test based on a TMA assay. D N-counter. It is a highly multiplexed single-molecule counting system where two probes are used to target the RNA molecule of interest, a capture probe and a reporter probe. Dong et al. used the NanoString nCounter assay to target mRNA transcripts in EVs from PCA cell lines [120]. E RNA Sequencing. The RNA massive sequencing allows analyzing the entire transcriptome and even transcripts yet to be discovered. F There are also genomic panels used for the diagnostic of PCa focused on specific biomarkers, such as the commercial test ExoDx Prostate [121] that detects the expression levels of ERG, PCA3 and SPDEF by qPCR in exosomes from urine samples.

Fig. 2. Basic research towards the discovery of new molecular biomarkers.

There are several sources and molecular approaches for the detection of new biomarkers in PCa. A This can be achieved by using an in vivo model -for which a prostate biopsy should be taken-, a primary culture or a PCa cell line. B Exome sequencing. The DNA samples are first fragmented and then biotinylated oligonucleotide probes -also known as baits- are used to selectively hybridize to target regions in the genome. C Whole-Genome Sequencing. This sequencing technique allows a uniform coverage across the complete genome. D RNA-Seq. RNA samples are synthesized into cDNA once it has been fragmented. Then, adaptors are attached to both ends of each fragment so they can be amplificated by PCR and subsequently sequenced [122]. Within the variants of this technique can be found single-cell RNA-seq, total RNA-seq, targeted RNA-seq, small RNA-seq, spatial transcriptomics, poly-A enrichment, ribosomal RNA depletion, among others. E Illumina next-generation sequencing technology: Individual DNA or cDNA molecules are placed on a flowcell for sequencing by synthesis by using fluorescent labeled nucleotides. PacBio sequencer and Nanopore sequencer can read more than 100 Kb in length of DNA, as well as the disposable sequencer MinION which doesn’t need prior installation [123]. F After a bioinformatic data analysis the results of the sequencing provide new genes as biomarkers candidates in PCa.

Fig. 3. Molecular diagnosis and precision medicine in cancer patient management.

The advantages of using approaches that target these signatures for disease diagnosis can be many. A PCa-stage discrimination. Each clinical-pathological profile will be stratified using the emerging techniques mentioned earlier, the results obtained could determine whether the patient has an indolent cancer or if it is an aggressive one. B Personalized treatment. The molecular diagnosis can also determine which specific treatment the patient should receive according to their molecular profile and the type of PCa they have, such as active surveillance, hormone therapy, surgery, radiation therapy, chemotherapy or immunotherapy (e.g., PD-1 inhibitors, sipuleucel-T vaccine [124]). C Prognosis. Finishing by getting a prognostic overview of the length of time that the patient will be alive or how well will the patient respond to the treatment he has been given. The prognosis can include overall-survival, progression-free survival, biochemical recurrence-free survival, disease-free survival, cancer-specific survival, and metastasis-free survival.

Fig. 4. Artificial intelligence and its application in patient stratification in prostate cancer.

ML is an artificial intelligence approach that can predict a possible outcome in PCa research and improve the patient management. A ML techniques. These algorithms are divided into two main types of learning: supervised learning and unsupervised learning. The former uses pre-determined explicit data, it is the most used in radiology and is based on classification and regression (deep learning, convolutional neural network, random forest, support vector machine, decision tree, logistic regression, among others [125]). The latter uses the features of the training data and doesn’t have a prior division of data in categories, it is based on clustering and dimensional reduction (K-means, hierarchical clustering, among others). B ML applied in PCa management. A recent application of ML is the prediction and analysis of radiomic data. This approach aims to improve the patient stratification and management using imageology, tissue analysis, and molecular data so the clinicians can offer a personalized treatment by differentiating the grade of the disease, stratifying the patients, and determine the therapy response.