Uncovering hidden spermatogenesis: INPP1 and GTSF1 as non-invasive predictors of sperm retrieval in non-obstructive...

The aetiology of non-obstructive azoospermia (NOA) remains poorly understood, highlighting the need for non-invasive molecular biomarkers to predict the outcome of sperm retrieval. In this study, we combined integrative bioinformatic analysis with experimental validation to investigate two regulatory RNA networks involving INPP1 and GTSF1 in NOA. Analysis of seven GEO transcriptome datasets identified INPP1 and GTSF1 as consistently downregulated genes in testicular tissue of NOA patients. Upstream regulatory elements were predicted using miRNA-circRNA interaction databases, leading to the selection of miR-27a and hsa_circ_0057497 for INPP1 and miR-146a-5p and hsa_circ_0026774 for GTSF1. The expression levels of these six markers were quantified in plasma samples from 60 men with NOA (stratified by both histopathological subtypes - SCOS, MA, HYPO - and sperm retrieval status: NOA⁺ vs. NOA-) and 40 fertile controls. INPP1 and GTSF1 were significantly reduced in NOA patients; in particular, INPP1 showed the lowest expression in SCOS patients, while GTSF1 was most downregulated in the MA and HYPO subgroups. The corresponding miRNAs were upregulated, and circRNAs showed inverse patterns. ROC analysis showed excellent diagnostic performance in distinguishing NOA⁺ from NOA-: AUCs were 0.956 for hsa_circ_0057497, 0.970 for hsa_circ_0026774 and ≥ 0.90 for the combined biomarker model. These results suggest that the INPP1 and GTSF1 RNA axes are perturbed in NOA and reflect both sperm retrieval potential and histological severity.,This study provides a basis for non-invasive molecular diagnostics and more personalized treatment strategies for male infertility.