The scientists from Unidad de Investigación Biomédica en Cáncer, México, looked at numerous facets of prostate cancer, including new molecular markers linked to diagnosis, prognosis, therapy prediction, and how bioinformatics tools like machine learning and deep learning can help in the clinic. The findings are expected to improve prostate cancer patient clinical care by applying artificial intelligence in prostate cancer research.
The detection of prostate cancer (PCa) incorporates the measurement of PSA serum levels and the digital rectal examination (DRE), contributing to the detection of PCa in the early phases. When it is restricted to the prostate, the localized disease is treated with radical prostatectomy (RP).
Conversely, localized advanced PCa is treated with a surgical procedure, adjuvants, radiotherapy like external beam radiation or brachytherapy; and chemical treatment, for example, Luteinizing-hormone releasing hormone (LHRH) agonists and antagonists, abiraterone, and enzalutamide; while the metastatic infection is generally treated with hormone treatment, for example, apalutamide, and chemotherapy.
Assuming that PCa is diagnosed on time, the therapy can be effective and with insignificant morbidity. To cover the extent of indeterminate discoveries, novel diagnostic biomarkers have been developed, for example, Prostate Health Index (PHI), 4 K score, SelectMDx, ConfirmMDx, and PCA3.
Though few examinations have been performed to investigate different molecular biomarkers, for example, variant V7 of the androgen receptor or inactivation of the PTEN or c-MYC gene, until this point in time, not a single one of them has been endorsed as a prognostic biomarker for use in clinical settings.
At present, there are a couple of molecular prognostic biomarkers in clinical use, for example, OncotypeDX Genomic Prostate Score, Prolaris, ProMark, and Decipher, in light of cancer-related gene panels. These molecular tests guide the urologist to layout the fitting treatment and foresee repeat and progression risks after localized treatment.
Nonetheless, it is essential to continue to look for molecular markers that can help with the early determination and prognosis of the patient as well as the foundation of patient response to various treatments, like new genes, gene fusions, AR variations, and non-coding RNAs.
As of now, there is a serious discussion regarding PSA as a diagnostic, prognostic, and screening tool in PCa, and in this manner, it is particularly critical to zero in on different kinds of molecular markers that can uphold clinical results and decision making for treatment.
Specifically, transcriptome and genomics analysis have been added to create new information in the investigation of PCa and the intracellular signaling pathways that direct prostate carcinogenesis producing new data about its biology.
Artificial intelligence and a portion of its algorithms have been used for clinical application in monitoring, detection, diagnosis, and treatment to create new clinical predictive models for PCa Management.
On the other hand, a few investigations have consolidated histology with genomic information, incorporating omics data with pathological images in PCa, and with the execution of artificial intelligence algorithms, for example, deep learning and machine learning have effectively settled an association between various parts of omics to get clinical prediction models, hence, making an integrative point of view that works with the discovery of new diagnostic, prognostic and remedial molecular biomarkers.
The significance of precision medication and the fusion among sequencing and artificial intelligence is laid out to make synergies that permit the improvement of additionally specific and advanced systems that work with getting relevant clinical methodologies for decision making and personalized management of PCa patients.
Artificial Intelligence in Prostate Cancer Research
Current strategies for the recognition of PCa show constraints in its location, addressing a need to further develop PCa diagnostic and patient stratification with complementary tools. Artificial intelligence (AI) has been demonstrated to be a fundamental execution for clinical findings, and it alludes to the capacity of a computational process to recognize patterns and make decisions that previously required human intellect to accomplish a specific purpose.
Machine learning (ML) is a discipline that shows computers how to assemble models from the huge informational indexes that they are assigned and gain from them. This technologic approach depends on statistical algorithms; a large portion of these algorithms are mathematical models that map the variables (features) of a data sample into a set of results.
Then, these algorithms go through a course of training to have the option to predict the labels by analyzing the features. The types of learning utilized in these models are classified as supervised and unsupervised learning.
Supervised learning utilizes explicit data sets determined by experts. The computer utilizes the programmed algorithms to limit the prediction error, which is estimated by the distinction between the predicted labels and the known labels, such as linear logistic regression and random forest.
Then again, unsupervised learning depends on examples isolated into various classes in light of the training data features, for example, principal component analysis. It has been proposed that ML could work on certain aspects of biomedicine, for example, disease diagnosis, monitoring, anatomical imaging of organs, tissue biopsies, and personalized treatment by utilizing an assortment of molecular and phenotypic data.
Likewise, it has been demonstrated as valuable for its application in the human genome project and advances in cancer research and management.
The evidence in light of clinical investigations that spotlights observing new biomarkers proposes that there is a wide molecular field that lies unexplored and that could be the key to some clinical challenges these days. These markers, like the coding (AR, BRCA2, PTEN, MLH1, CUL3, SPEN) and non-coding genes (PCA3, SChLAP1, HERV-K, and miR-21), the TMPRSS2-ERG gene combination including their subsidiaries, and the androgen receptor variant 7 can be found utilizing genomics, transcriptomics and AI strategies.
Notwithstanding, these are not by any means the only adjustments that can be utilized for the diagnostic, prognostic, and prediction in the management of PCa patients. The clinical evidence referenced in this study suggests the significance of these kinds of molecular markers (coding and non-coding genes) and their roles in the dynamic decision-making cycle for laying out the most efficient treatment for patients experiencing prostate cancer.
Subsequently, it is vital to layout the sorts of actionable mutations, or their combinations, in patients with advanced PCa (locally advanced and metastatic), permitting us to decide the kind of treatment that will positively respond to PCa patients.
In this area, genomic analysis, transcriptome sequencing, and new methodologies like spatial transcriptomics, alongside the clinical-pathological data, could give vital information. Similarly, the utilization of ML algorithms will speed up the identification and discovery of novel molecular biomarkers, and it will lead biomedical examination toward artificial intelligence-based precision medication, so it can improve patient management as well as their quality of life, and, soon, permit a scientific revolution in medication for the administration of the PCa patient.
The new molecular biomarkers referenced here, alongside the novel bioinformatic approaches of AI and sequencing strategies, will improve biomedical research by supplementing PSA tests for screening, stratifying patients, and recognizing new molecular biomarkers for separation of indolent and aggressive disease, prognostic, predictive, and surrogate biomarkers with clinical utility.
Eventually, the fusion between sequencing and AI will create synergies that permit the development of additionally specific and advanced systems that work with acquiring significant clinical methodologies for decision-making and personalized management of PCa patients to battle this worldwide public medical issue in men.
Article Source: Alarcón-Zendejas, A.P., Scavuzzo, A., Jiménez-Ríos, M.A. et al. The promising role of new molecular biomarkers in prostate cancer: from coding and non-coding genes to artificial intelligence approaches. Prostate Cancer Prostatic Dis (2022). https://doi.org/10.1038/s41391-022-00537-2
Tanveen Kaur is a consulting intern at CBIRT, currently, she's pursuing post-graduation in Biotechnology from Shoolini University, Himachal Pradesh. Her interests primarily lay in researching the new advancements in the world of biotechnology and bioinformatics, having a dream of being one of the best researchers.