To identify the most robust molecular biomarkers in sperm and seminal plasma for the diagnosis of male infertility, and to evaluate their clinical use.
Accessible studies reporting well-defined (in)fertile populations and semen molecular biomarkers were included in this review.
A systematic search of the literature published in MEDLINE-PubMed and EMBASE databases was performed, following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.
Main outcome measure(s)
The primary outcome was the content, expression, or activity of molecular biomarkers in human semen samples. Only studies reporting a receiver-operating characteristic (ROC) analysis values were included.
Eighty-nine studies were included. Direct evaluation of sperm DNA damage has high potential as a diagnostic biomarker of fertility and assisted reproductive technology outcomes (area under the curve [AUCs] median = 0.67). Regarding strand break-associated chromatin modifications, γH2AX levels show good predictive value for the diagnosis of male infertility (AUCs median = 0.93). Some noncoding ribonucleic acid (RNA) exhibit excellent predictive values; miR-34c-5p in semen is the most well-characterized and robust transcriptomic biomarker (AUCs median = 0.78). While many proteins in semen show fair diagnostic value for sperm quality and fertilizing capacity, the levels of some, such as TEX101, in seminal plasma have an excellent diagnostic potential (AUCs median = 0.69). Although individual metabolites and metabolomic profiles in seminal plasma present good predictive value, the latter seem to be better than the former when inferring sperm quality and fertilizing capacity.
The current review supports that some Omics (e.g., DNA structure and integrity, genomics and epigenomics, transcriptomics, metabolomics, and proteomics) could be considered relevant molecular biomarkers that may help identify infertility etiologies and fertilization prognosis with cost-effective, simple, and accurate diagnosis.