Azoospermia is characterized by the complete absence of spermatozoa in the ejaculate semen, occurring in approximately 1% of all men and 10%–15% of infertile men (1). However, in some cases, intracytoplasmic sperm injection treatment can be performed after obtaining sperm through microdissection testicular sperm extraction (micro-TESE) (2). Among the types of non-obstructive azoospermia (NOA), idiopathic NOA (iNOA) presents a particular clinical challenge given the lack of a clear etiology, absence of effective treatments, and poorest sperm retrieval outcomes (SROs) following micro-TESE (3, 4). Regrettably, micro-TESE is only successful in approximately 50% of patients with NOA (5, 6). Although research has identified various factors related to SRO, such as testicular volume (TV) and hormonal levels [e.g., follicle-stimulating hormone (FSH), luteinizing hormone (LH), and testosterone (T)] (5), reliable molecular biomarkers for predicting SROs following micro-TESE in patients with iNOA remain to be identified.

Testicular biopsy is an important invasive procedure for evaluating testicular spermatogenesis (7). However, whether testicular biopsy is necessary before micro-TESE remains controversial. Because spermatogenic foci are few and scattered in patients with iNOA, biopsied testicular tissues may not reflect spermatogenesis in other parts of the testis. Thus, some studies have pointed out that testicular biopsy only contributes to the pathological classification of the biopsied tissues and has limited value in the prediction of SROs following micro-TESE (8). Meanwhile, open testicular biopsy increases the risk of testicular injury and bleeding (9). However, other studies have confirmed that testicular biopsy is of major predictive significance in the context of micro-TESE outcomes. Mao et al. reported extremely poor SROs (11% success) in patients with iNOA whose biopsy results indicated an absence of sperm in both the intraoperative microscopic examination and the post-pathological examination—a rate far below that observed for patients with iNOA exhibiting identifiable sperm in even just one of the two examinations (10). Cao et al. also found significantly different profiles of PIWI-interacting RNA (piRNA) between successful and failed micro-TESE procedures among patients with NOA, providing insight into strategies for identifying biomarkers that can predict residual spermatogenic loci in this population (11). Our collaborators also reported obvious upregulation of beclin-1 in the Sertoli cell-only syndrome (SCOS) group, which was further identified as an important predictor of SRO failure in micro-TESE (12). Taken together, these findings indicate that different histopathological outcomes and differentially expressed genes in testicular tissues obtained during micro-TESE or testicular biopsy may, in part, reflect the extent of spermatogenesis within the whole testis, suggestive of the potential value for predicting SROs following micro-TESE in patients with NOA.

Circular RNAs (circRNAs) are a new class of endogenous non-coding RNAs that covalently bind to a closed circular structure without a 5′ end cap and a 3′ end poly(A) tail (13). Several studies have confirmed that circRNAs play a pivotal role in regulating gene expression and cell fate (14). Many circular RNAs (circRNAs) are specifically expressed in the testis and seminal plasma of patients with NOA (15) and can be used as predictors of the outcome of micro-TESE (16). However, research regarding the expression and function of circRNAs in spermatogenesis is still in its infancy. Our collaborators have shown that approximately 1,000 circRNAs, including circ_0008045 (circular RNA monoglyceride lipase, circ_MGLL), are differentially dysregulated in testicular tissues between NOA and obstructive azoospermia (17). Previous studies have indicated that the expression of monoglyceride lipase (MGLL), an important downstream factor involved in triglyceride catabolism, is very low in germ cells but is specifically abundant in Sertoli cells, indicating that MGLL is a potential functional regulator of Sertoli cells (18). More importantly, silencing the expression of circ_MGLL results in an increase in the proliferation of Sertoli cells (19). However, many studies have reported that changes in the population or function of Sertoli cells are among the main factors leading to defective spermatogenesis (20). Nevertheless, the clinical relevance of circ_MGLL in patients with iNOA remains to be determined.

In the current study, we discovered that circ_MGLL, a circRNA generated from the circularization of the MGLL gene, was significantly upregulated in the testicular tissues of patients with iNOA in the SRO-failure group. Detailed univariate and multivariate logistic analyses were conducted to further clarify the relationship between circ_MGLL and SRO. Furthermore, stratified analyses were performed to determine whether the correlation between circ_MGLL and SRO remained within different subgroups, thus distinguishing our study from previous research. Finally, receiver operating characteristic (ROC) curve and decision curve analysis (DCA) were performed to analyze the ability of circ_MGLL to predict SROs.

In the current study, we observed significant differences in circ_MGLL expression between successful and failed SRO among patients with iNOA, whereas other variables (including age, BMI, TV, FSH, LH, and T) did not differ between the successful and failed SRO groups. When compared with the level observed in the failure group, the relative expression of circ_MGLL was significantly downregulated in the successful SRO group. Both univariate and multivariate analyses indicated that circ_MGLL expression was negatively associated with SRO in patients with iNOA, even after stratification according to age, TV, BMI, FSH, T, and pathological type (all OR <1). Notably, our results demonstrated that the circ_MGLL-based nomogram model was a relatively accurate predictor with the clinical potential to assist andrologists in screening patients with iNOA suitable for micro-TESE surgery.

Some molecules in the seminal plasma have been identified as predictors of residual spermatogenesis in patients with NOA (16, 22). However, seminal plasma is a mixture secreted by accessory glands such as the prostate, seminal vesicle gland, and ureteral bulbar gland (23). Thus, the specific expression of molecules in the seminal plasma or seminal exosomes may only partly reflect the spermatogenic function of the testis. Furthermore, although some studies have indicated that spermatogenesis varies in different areas of the testis (24), a series of studies have verified that patterns of expression for testis-specific genes or non-coding RNAs are significantly associated with the success of sperm recovery in patients with NOA, emphasizing their potential value as predictors of successful SRO in these patients (25, 26). In our previous collaboration, we observed that circ_MGLL expression is upregulated in patients with NOA when compared with the level observed in patients with obstructive azoospermia (17). In the current study, we further demonstrated that the expression of circ_MGLL is negatively correlated with the SRO in patients with iNOA, and the circ_MGLL-based nomogram had better diagnostic efficiency in predicting the SRO of micro-TESE for these patients. Further increasing the predictive power of this circ_MGLL-based nomogram by integrating histopathological results and the expression of testis-specific molecular biomarkers may help to avoid unnecessary surgery and excessive financial burden in patients with iNOA.

Previously, we reported that the proliferating speed of Sertoli cells was increased, the proportion of proliferating cells and cells at S stage was upregulated, and the apoptosis levels of Sertoli cells were decreased after silencing circ_MGLL expression (19). Furthermore, bioinformatics analysis and luciferase reporter assays showed that circ_MGLL can bind to miR-1228, miR-1233, miR-149, and miR-924 (19). Previous studies had already demonstrated that miR-1288 expression is abundant in the plasma and testis of cynomolgus monkeys, exhibiting associations with apoptosis and cell differentiation (27). Additional studies have indicated that miR-1233 inhibits apoptosis and promotes the proliferation of A549 cells by targeting dual-specificity phosphatase 9 (28). More importantly, miR-149 expression in sperm is correlated with the quality of early embryonic development (29). Previous literatures proved that Sertoli cells provide physical support and stable microenvironments or niche for developing sperm and perform an important role in sustaining spermatogenesis (30, 31). Considering the important functions of circ_MGLL and its targeted miRNAs in cell apoptosis, we speculate that a regulatory network with circ_MGLL as the core would induce apoptosis and the dysfunctions of Sertoli cells in patients with NOA, which may further result in the residual sperm losing nutritional supports and relatively normal microenvironments. Consequently, this dysfunction of Sertoli cells induced by circ_MGLL upregulation will eventually lead to the failure of micro-TESE. However, the mechanism by which circ_MGLL regulates cell proliferation and the functions of Sertoli cells and spermatogenesis should be further explored in future studies.

Notably, our results also indicated that the serum levels of FSH, LH, and T were not correlated with the outcome of sperm collection and could not be used as reliable predictors of SRO in patients with iNOA undergoing micro-TESE. However, findings regarding the correlation between hormonal levels and SRO have been inconsistent. Consistent with our results, some preliminary studies of patients with NOA have reported no significant differences in TV or hormonal levels based on the success or failure of micro-TESE (32, 33). Accordingly, serum hormone levels were not identified as a significant predictor of successful SRO (34). However, some studies have reported significantly higher levels of serum FSH in the failed group than in the successful group. This may be because the abovementioned studies failed to exclude patients with Klinefelter syndrome (KS) and AZF c microdeletions. Patients in these latter two groups may exhibit less remarkable increases in FSH as well as a relatively higher success rate following sperm collection than those with iNOA (5, 35, 36). In contrast, our study excluded patients with KS and AZFc microdeletion, instead focusing on patients with iNOA, which is often associated with significantly elevated FSH levels and low sperm retrieval rates (36, 37). Despite the associations observed in the current study, further research is required to clarify the effects of hormonal levels on SRO in patients with iNOA as well as the mechanisms underlying their actions.

In addition to the retrospective nature of this study, some non-negligible limitations exist—for example, all testicular tissues used for the detection of circ_MGLL were paraffin-embedded testicular tissues obtained during micro-TESE surgery, and we did not examine the testicular biopsy tissue obtained via needle aspiration prior to micro-TESE surgery. Thus, for patients who meet the criteria of testicular needle aspiration, such an analysis may help to verify our findings. Furthermore, although we observed that the circ_MGLL-based nomogram could reflect the focal spermatogenesis of the whole testis and could be used as a predictor for SRO in patients with iNOA, the expression profiles and the function of circ_MGLL in germ cells and testicular tissues should be explored in future studies.

In summary, the detailed data obtained in our study demonstrate that higher circ_MGLL levels are associated with decreased success in sperm retrieval via micro-TESE in patients with iNOA. More importantly, to the best of our knowledge, the present study is the first to establish a competing risk nomogram for calculating the probability of successful SRO, which may aid in screening for micro-TESE eligibility. Although these results provide evidence regarding the significant association between circ_MGLL expression and failure of micro-TESE in patients with iNOA, an additional large-sample multicenter prospective cohort study is required to further validate the efficacy of the circ_MGLL-based nomogram for predicting successful SRO.

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by the Ethics Review Board of the Northwest Women’s and Children’s Hospital (ethical review number: 2021-XBFE-023). The patients/participants provided their written informed consent to participate in this study.

MW designed the research study. SS, TW, and LW performed the research. SS analyzed the data. MW and SS wrote the paper. All authors contributed to the article and approved the submitted version.