Genetic and biological drivers of prostate cancer disparities in Black men

Towards equitable AI in oncology

Artificial intelligence (AI) stands at the threshold of revolutionizing clinical oncology, with considerable potential to improve early cancer detection and risk assessment, and to enable more accurate personalized treatment recommendations. However, a notable imbalance exists in the distribution of the benefits of AI, which disproportionately favour those living in specific geographical locations and in specific populations. In this Perspective, we discuss the need to foster the development of equitable AI tools that are both accurate in and accessible to a diverse range of patient populations, including those in low-income to middle-income countries. We also discuss some of the challenges and potential solutions in attaining equitable AI, including addressing the historically limited representation of diverse populations in existing clinical datasets and the use of inadequate clinical validation methods. Additionally, we focus on extant sources of inequity including the type of model approach (such as deep learning, and feature engineering-based methods), the implications of dataset curation strategies, the need for rigorous validation across a variety of populations and settings, and the risk of introducing contextual bias that comes with developing tools predominantly in high-income countries.

Population-Specific gene expression profiles in prostate cancer: insights from Weighted Gene Co-expression Network Analysis (WGCNA)

This study investigates the genetic factors contributing to the disparity in prostate cancer incidence and progression among African American men (AAM) compared to European American men (EAM). The research focuses on employing Weighted Gene Co-expression Network Analysis (WGCNA) on public microarray data obtained from prostate cancer patients. The study employed WGCNA to identify clusters of genes with correlated expression patterns, which were then analyzed for their connection to population backgrounds. Additionally, pathway enrichment analysis was conducted to understand the significance of the identified gene modules in prostate cancer pathways. The Least Absolute Shrinkage and Selection Operator (LASSO) and Correlation-based Feature Selection (CFS) methods were utilized for selection of biomarker genes. The results revealed 353 differentially expressed genes (DEGs) between AAM and EAM. Six significant gene expression modules were identified through WGCNA, showing varying degrees of correlation with prostate cancer. LASSO and CFS methods pinpointed critical genes, as well as six common genes between both approaches, which are indicative of their vital role in the disease. The XGBoost classifier validated these findings, achieving satisfactory prediction accuracy. Genes such as APRT, CCL2, BEX2, MGC26963, and PLAU were identified as key genes significantly associated with cancer progression. In conclusion, the research underlines the importance of incorporating AAM and EAM population diversity in genomic studies, particularly in cancer research. In addition, the study highlights the effectiveness of integrating machine learning techniques with gene expression analysis as a robust methodology for identifying critical genes in cancer research.

Germline Mutations and Ancestry in Prostate Cancer

PURPOSE OF REVIEW

Prostate cancer is the most frequently diagnosed non-cutaneous malignancy of men in the USA; notably, the incidence is higher among men of African, followed by European and Asian ancestry. Germline mutations and, in particular, mutations in DNA damage repair genes (DDRGs) have been implicated in the pathogenesis of prostate cancer. This review intends to discuss the implication of ancestry on prostate cancer, specifically in regard to lack of diversity in genomic and genetic databases and the ability of providers to properly counsel patients on the significance of cancer genetic results.

RECENT FINDINGS

Ancestral differences in prostate cancer-associated DDRG germline mutations are increasingly recognized. Guidelines for treatment by the National Comprehensive Cancer Network® (NCCN®) support germline testing in certain patients, and a myriad of genetic testing panels for DDRG mutations are now available in clinical practice. However, the consensus among providers on what genes and mutations to include in the genetic tests has evolved from experience from men of European ancestry (EA). Gaps in ancestry-informed clinical practice exist in genetic risk assessment, implementation of screening, counseling, guiding recommendations, treatment, and clinical trial enrollment. The lack of diversity in tumor genomic and genetic databases may hinder ancestry-specific disease-predisposing alterations from being discovered and targeted in prostate cancer and, therefore, impede the ability of providers to accurately counsel patients on the significance of cancer genetic test results.