One of the key benefits of AI in cancer genomics is its ability to analyze and interpret vast amounts of genomic data with remarkable speed and accuracy. This helps researchers identify critical genetic signatures and biomarkers associated with the development, progression, and treatment response of PCa and BCa. For example, AI can classify cancer phenotypes based on gene expression microarray data, and reveal molecular interactions among the genes. Understanding the specifics of genes and signaling pathways alteration could forecast a chemotherapy response and provide some fresh information for new drug development. By detecting patterns and correlations that might otherwise go unnoticed, AI empowers scientists to make novel discoveries that can potentially transform patient care.
The obtained data in the systematic review allowed us to trace the following tendencies. First, AI applications have become more clinically oriented, and they show promising outcomes in various fields, ranging from cancer detection and risk stratification to predicting the correct treatment response. AI possibilities allow us to predict the treatment response of immune checkpoint inhibitors (ICIs) in BCa patients. E.g. tumor mutational burden in predicting the response to PD-L1 blockade in locally advanced and metastatic BCa.Also, article numbers related to disease and relapse prognosis rise with time thus proposing more concrete rules for recurrence calculation. Indeed, one of the main issues for clinicians is how AI can be applied in genomics. However, several algorithms have been developed, and validated, but have not extended beyond experimental studies. Second, we found no data regarding the economic effectiveness or benefits of AI implementation to routine practice. The target groups for each method are not defined, and any comparison of different gene values or methods is currently lacking.
The potential implementation of AI in clinical practice may be in therapy selection for more target effects or precise prognosis in severe cases. In this field, AI may be of help in pre-evaluating the chemotherapy response in CRPCa patients or forecasting response to antiandrogens in CRPCa. For patients with BCa, AI could help to predict immunotherapy response or optimize the follow-up by reducing the frequency of diagnostic cystoscopies. Even if AI is not implemented in routine clinical practice, it may be highly advisable to use it in severe cases of cancer in the research framework.
In some studies, it was demonstrated that detecting signature gene expression for cancer prediction at early stages is possible. However, the implication of AI for cancer genomics screening is equivocal. The current cost of WGS is quite high and depends on the sequencing platform. WGS produces more than 90-150 Gb of data that should be processed and therefore computer resources are required to process a large amount of data. In this field, AI applications may simplify, speed up, and therefore make more affordable the WGS procedure.
On the other hand, if the location of the supposed mutation is known in advance, the geneticist may perform this task routinely without the need to use AI. The cost of the analysis remains the same. AI can be applied to large amounts of data such as genome, exome, or non-coding part of DNA. And so, using AI for screening seems economically impractical for now.
The integration of AI into PCa and BCa genomics research is a testament to the remarkable progress we are witnessing in the field of personalized medicine. By combining the power of AI with the expertise of clinicians and researchers, there is the potential to unlock new frontiers in cancer diagnosis, treatment, and prevention.
Written by: Andrey Bazarkin,1 Andrey Morozov,1 Alexander Androsov,2 Harun Fajkovic,3,4 Juan Gomez Rivas,5 Nirmish Singla,6 Svetlana Koroleva,7 Jeremy Yuen-Chun Teoh,8 Andrei V Zvyagin, D.Sc.,9,10 Shahrokh François Shariat, 3,4,11-14, Bhaskar Somani,15 and Dmitry Enikeev,1,3,4,16
- Institute for Urology and Reproductive Health, Sechenov University, Moscow, Russia
- Department of Pediatric Surgery, Division of Pediatric Urology and Andrology, Sechenov University, Moscow, Russia
- Department of Urology and Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria
- Karl Landsteiner Institute of Urology and Andrology, Vienna, Austria
- Department of Urology, Clinico San Carlos University Hospital, Madrid, Spain
- Brady Urological Institute, School of Medicine, Johns Hopkins Medicine, Baltimore, MD, United States
- Clinical Institute for Children Health named after N.F. Filatov, Sechenov University, Moscow, Russia
- S.H. Ho Urology Centre, Department of Surgery, The Chinese University of Hong Kong, Hong Kong, China
- Institute of Molecular Theranostics, Sechenov University, Moscow, Russia
- Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry of the Russian Academy of Sciences, 117997 Moscow, Russia
- Department of Urology, Weill Cornell Medical College, New York, New York
- Department of Urology, University of Texas Southwestern, Dallas, Texas, USA
- Department of Urology, Second Faculty of Medicine, Charles University, Prague, Czech Republic
- Division of Urology, Department of Special Surgery, Jordan University Hospital, The University of Jordan, Amman, Jordan
- Department of Urology, University Hospital Southampton, Southampton, UK
- Division of Urology, Rabin Medical Center, Petach Tikva, Israel