Whole Exome and Genome Sequencing in the definition of the etiology of Non-Obstructive Azoospermia: from diagnosis to prognosis of Testicular Sperm Extraction outcomes

Azoospermia is the most severe manifestation of male infertility, affecting approximately 1% of men worldwide. According to the integrity of post-testicular structures, azoospermia may be divided into two categories: obstructive (OA) or non-obstructive (NOA). While in case of OA spermatogenesis is typically conserved, men with NOA may exhibit different testicular histology, ranging from complete absence of spermatozoa, to maturation arrest or hypospermatogenesis. The only treatment option for men with NOA to father a biological child relies on TESE (testicular sperm extraction) followed by medically assisted reproductive techniques (i.e. Intracytoplasmic Sperm Injection, ICSI). The TESE success rates vary depending on the testis histology of the patient. In the context of NOA, TESE fails to retrieve spermatozoa from the testis in nearly 50% of the cases, and its success remains unpredictable with the currently available clinical tools. Regarding the etiology, approximately 40% of cases of NOA remain idiopathic following a comprehensive diagnostic workup encompassing genetic and non-genetic tests. Given that more than 2000 genes are overexpressed in the human testes, with more than 450 of them being testis-specific, it is plausible that idiopathic NOA cases may underlie genetic factors that have yet to be identified. Over the past decade, next-generation sequencing (NGS), and especially whole-exome sequencing (WES) has led to the identification of several candidate NOA genes, but their diagnostic value and impact on TESE success remain unclear. To address this knowledge gap, in this thesis we developed a virtual gene panel (VGP) comprising 145 genes with at least limited clinical evidence for NOA or found mutated in the literature in more than one unrelated man with spermatogenic failure (SF). In this retrospective study, we performed WES in a highly selected cohort of 258 men with idiopathic NOA and known TESE outcomes recruited in two centers. The primary objective of the thesis was to assess the clinical validity of our VGP and its prognostic value for TESE success. WES data were crossed with our VGP and variants were filtered to select only rare and predicted as damaging variants in coding and splice-site regions. Following the classification according to the American College of Medical Genetics and Genomics guidelines we selected only variants defined as pathogenic (P) or likely pathogenic (LP), which were validated using gold-standard methods. The WES-derived VGP identified a monogenic cause of NOA in 16.7% of cases, with a higher diagnostic yield in patients with maturation arrest and negative TESE outcomes (24.7%). By integrating our findings with the literature, we identified 17 genes systematically associated with negative TESE outcome and 14 linked with successful sperm retrieval. Notably, TESE-negative genes primarily regulate meiotic progression, with conserved functions across mammals. Among them, MSH4, SYCE1, and TEX11 were found mutated in more than 10 carriers (including our patients and those from literature), suggesting their potential as predictors of TESE failure. After re-evaluating the gene-disease relationship for the 31 mutated genes, all except POTEJ reach at least moderate clinical evidence, supporting their inclusion in NOA diagnostic workup. In addition, four genes in which we identified LP/P variants have been previously described in individuals with premature ovarian insufficiency (POI) and all of them have been reported only in TESE negative subjects. All these genes have at least moderate clinical evidence for both POI and NOA. Moreover, we propose five additional genes mutated in our cohort of NOA patients as novel POI candidates, based on their association with female infertility in knock-out mouse models. To further investigate genetic contributors to NOA, we conducted a whole-genome sequencing (WGS) pilot study on four patients with pure SF (i.e. negative TESE outcomes), where WES failed to detect a genetic defect. For the first time, we identified genetic variants in regulatory elements of candidate NOA genes. Our findings suggest a potential role for LEO1 in spermatogenic impairment and propose a digenic model for NOA in case of genetic defects involving regulatory elements, such as those identified in this study. In conclusion, this thesis emphasizes the advancements in comprehending the genetic basis of idiopathic NOA through NGS-based approaches. Our findings provide an estimate on the diagnostic potential of a NOA-specific VGP and enhance the understanding of genetic factors influencing TESE outcomes. These data offer valuable insights to optimize the clinical management and genetic counseling of patients with SF and their relatives, particularly their younger sisters, for whom an early POI diagnosis could enable preventive oocyte cryopreservation. Finally, our preliminary WGS results pave the way for a deeper understanding of the role of defects in regulatory elements in the aetiology of NOA.