Background: Testicular sperm extraction (TESE) is an essential therapeutic tool for the management of male infertility. However, it is an invasive procedure with a success rate up to 50%. To date, no model based on clinical and laboratory parameters is sufficiently powerful to accurately predict the success of sperm retrieval in TESE. Objective: The aim of this study is to compare a wide range of predictive models under similar conditions for TESE outcomes in patients with nonobstructive azoospermia (NOA) to identify the correct mathematical approach to apply, most appropriate study size, and relevance of the input biomarkers. Methods: We analyzed 201 patients who underwent TESE at Tenon Hospital (Assistance Publique-Hôpitaux de Paris, Sorbonne University, Paris), distributed in a retrospective training cohort of 175 patients (January 2012 to April 2021) and a prospective testing cohort (May 2021 to December 2021) of 26 patients. Preoperative data (according to the French standard exploration of male infertility, 16 variables) including urogenital history, hormonal data, genetic data, and TESE outcomes (representing the target variable) were collected. A TESE was considered positive if we obtained sufficient spermatozoa for intracytoplasmic sperm injection. After preprocessing the raw data, 8 machine learning (ML) models were trained and optimized on the retrospective training cohort data set: The hyperparameter tuning was performed by random search. Finally, the prospective testing cohort data set was used for the model evaluation. The metrics used to evaluate and compare the models were the following: sensitivity, specificity, area under the receiver operating characteristic curve (AUC-ROC), and accuracy. The importance of each variable in the model was assessed using the permutation feature importance technique, and the optimal number of patients to include in the study was assessed using the learning curve. Results: The ensemble models, based on decision trees, showed the best performance, especially the random forest model, which yielded the following results: AUC=0.90, sensitivity=100%, and specificity=69.2%. Furthermore, a study size of 120 patients seemed sufficient to properly exploit the preoperative data in the modeling process, since increasing the number of patients beyond 120 during model training did not bring any performance improvement. Furthermore, inhibin B and a history of varicoceles exhibited the highest predictive capacity. Conclusions: An ML algorithm based on an appropriate approach can predict successful sperm retrieval in men with NOA undergoing TESE, with promising performance. However, although this study is consistent with the first step of this process, a subsequent formal prospective multicentric validation study should be undertaken before any clinical applications. As future work, we consider the use of recent and clinically relevant data sets (including seminal plasma biomarkers, especially noncoding RNAs, as markers of residual spermatogenesis in NOA patients) to improve our results even more.

STUDY QUESTION: Can a combination of metabolomic signature and machine learning (ML) models distinguish nonclassic 21-hydroxylase deficiency (NC21OHD) from polycystic ovary syndrome (PCOS) without adrenocorticotrophic hormone (ACTH) testing? SUMMARY ANSWER: A single sampling methodology may be an alternative to the dynamic ACTH test in order to exclude the diagnosis of NC21OHD in the presence of a clinical hyperandrogenic presentation at any time of the menstrual cycle. WHAT IS KNOWN ALREADY: The clinical presentation of patients with NC21OHD is similar with that for other disorders of androgen excess. Currently, cosyntropin stimulation remains the gold standard diagnosis of NC21OHD. STUDY DESIGN, SIZE, DURATION: The study was designed using a bicentric recruitment: an internal training set included 19 women with NC21OHD and 19 controls used for developing the model; a test set included 17 NC21OHD, 72 controls and 266 PCOS patients used to evaluate the performance of the diagnostic strategy thanks to an ML approach. PARTICIPANTS/MATERIALS, SETTING, METHODS: Fifteen steroid species were measured in serum by liquid chromatography–mass spectrometry (LC-MS/MS). This set of 15 steroids (defined as ‘steroidome’) used to map the steroid biosynthesis pathway was the input for our models. MAIN RESULTS AND THE ROLE OF CHANCE: From a single sample, modeling involving metabolic pathway mapping by profiling 15 circulating steroids allowed us to identify perfectly NC21OHD from a confounding PCOS population. The constructed model using baseline LC-MS/MS-acquired steroid fingerprinting successfully excluded all 17 NC21OHDs (sensitivity and specificity of 100%) from 266 PCOS from an external testing cohort of originally 549 women, without the use of ACTH testing. Blood sampling timing during the menstrual cycle phase did not impact the efficiency of our model. LIMITATIONS, REASONS FOR CAUTION: The main limitations were the use of a restricted and fully prospective cohort as well as an analytical issue, as not all laboratories are equipped with mass spectrometers able to routinely measure this panel of 15 steroids. Moreover, the robustness of our model needs to be established with a larger prospective study for definitive validation in clinical practice. WIDER IMPLICATIONS OF THE FINDINGS: This tool makes it possible to propose a new semiology for the management of hyperandrogenism. The model presents better diagnostic performances compared to the current reference strategy. The management of patients may be facilitated by limiting the use of ACTH tests. Finally, the modeling process allows a classification of steroid contributions to rationalize the biomarker approach and highlight some underlying pathophysiological mechanisms. STUDY FUNDING/COMPETING INTEREST(S): This study was supported by ‘Agence Française de Lutte contre le dopage’ and DIM Région Ile de France. This study was supported by the French institutional PHRC 2010-AOR10032 funding source and APHP. All authors declare no competing financial interests. TRIAL REGISTRATION NUMBER: N/A.

Bile acids (BAs) regulate dietary lipid hydrolysis and absorption in the proximal intestine. Several studies have highlighted a determinant role of circulating levels and/or metabolism of BAs in the pathogenesis of major cardiometabolic diseases. Whether changes in BA profiles are causative or are consequence of these diseases remains to be determined. Healthy male volunteers (n = 71) underwent a postprandial exploration following consumption of a hypercaloric high fat typical Western meal providing 1200 kcal. We investigated variations of circulating levels of 28 BA species, together with BA synthesis marker 7α-hydroxy-4-cholesten-3-one (C4) over an approximately diurnal 12 h period. Analysis of BA variations during the postprandial time course revealed two major phenotypes with opposite fluctuations, i.e., circulating levels of each individual species of unconjugated BAs were reduced after meal consumption whereas those of tauro- and glyco-conjugated BAs were increased. By an unbiased classification strategy based on absolute postprandial changes in BA species levels, we classified subjects into three distinct clusters; the two extreme clusters being characterized by the smallest absolute changes in either unconjugated-BAs or conjugated-BAs. Finally, we demonstrated that our clustering based on postprandial changes in BA profiles was associated with specific clinical and biochemical features, including postprandial triglyceride levels, BMI or waist circumference. Altogether, our study reveals that postprandial profiles/patterns of BAs in response to a hypercaloric high fat challenge is associated with healthy or unhealthy metabolic phenotypes that may help in the early identification of subjects at risk of developing metabolic disorders.

We investigated how endogenous ω-3 PUFA production compared to supplementation could improve obesity-related metabolic dysfunction. Fat-1 transgenic mice, able to endogenously convert exogenous ω-6 to ω-3 PUFA, and their wild-type littermates were fed a high-fat diet and daily gavaged with either ω-3 or ω-6 PUFA-rich oil for 12 weeks. The endogenous production of ω-3 PUFA improved glucose intolerance and insulin resistance but did not affect hepatic steatosis. Conversely, ω-3 PUFA supplementation fully prevented hepatic steatosis but failed to improve insulin resistance. Both models increased levels of ω-3 PUFA-containing 2–monoacylglycerol and N-acylethanolamine congeners, and reduced levels of ω-6 PUFA-derived endocannabinoids, in the liver and gastrocnemius muscle, with different impact at molecular species levels, and supplementation of ω-3 PUFAs being more efficacious. Reduced hepatic lipid accumulation was shown to be associated with the endocannabinoidome metabolites EPEA and DHEA, which was causally demonstrated by lower lipid accumulation in oleic acid-treated hepatic cells treated with these metabolites. While both genetic and supplementation models induced a significant fecal enrichment of the beneficial Allobaculum genus, mice gavaged with ω-3 PUFA also displayed additional changes in the gut microbiota functions with significant reduction of fecal levels of the pro-inflammatory molecules lipopolysaccharide and flagellin. By employing multiple factor analysis, we were able to identify that the metabolic improvements induced by ω-3 PUFA were accompanied by a reduced production of the pro-inflammatory cytokine TNFα, and that ω-3 PUFA supplementation had a stronger effect on improving the hepatic fatty acid profile than endogenous ω-3 PUFA. In conclusion, while endogenous ω-3 PUFA production preferably improves glucose tolerance and insulin resistance, oral ω-3 PUFA intake appears to be required to elicit selective changes in hepatic endocannabinoidome signaling that are essential to alleviate high-fat diet-induced hepatic steatosis.

We aimed to develop and evaluate a machine learning model that can stratify infertile/fertile couples on the basis of their bioclinical signature helping the management of couples with unexplained infertility. Fertile and infertile couples were recruited in the ALIFERT cross-sectional case–control multicentric study between September 2009 and December 2013 (NCT01093378). The study group consisted of 97 infertile couples presenting a primary idiopathic infertility (> 12 months) from 4 French infertility centers compared with 100 fertile couples (with a spontaneously conceived child (< 2 years of age) and with time to pregnancy < 12 months) recruited from the healthy population of the areas around the infertility centers. The study group is comprised of 2 independent sets: a development set (n = 136 from 3 centers) serving to train the model and a test set (n = 61 from 1 center) used to provide an unbiased validation of the model. Our results have shown that: (i) a couple-modeling approach was more discriminant than models in which men’s and women’s parameters are considered separately; (ii) the most discriminating variables were anthropometric, or related to the metabolic and oxidative status; (iii) a refined model capable to stratify fertile vs. infertile couples with accuracy 73.8% was proposed after the variables selection (from 80 to 13). These influential factors (anthropometric, antioxidative, and metabolic signatures) are all modifiable by the couple lifestyle. The model proposed takes place in the management of couples with idiopathic infertility, for whom the decision-making tools are scarce. Prospective interventional studies are now needed to validate the model clinical use.Trial registration: NCT01093378 ALIFERT https://clinicaltrials.gov/ct2/show/NCT01093378?term=ALIFERT&rank=1. Registered: March 25, 2010.

Individuals born with intrauterine growth retardation (IUGR) are more prone to cardio-metabolic diseases as adults, and environmental changes during the perinatal period have been identified as potentially crucial factors. We have studied in a preclinical model early-onset molecular alterations present before the development of a clinical phenotype. We used a preclinical mouse model of induced IUGR, in which we modulated the nutrition of the pups during the suckling period, to modify their susceptibility to cardio-metabolic diseases in adulthood. Mice born with IUGR that were overfed (IUGR-O) during lactation rapidly developed obesity, hepatic steatosis and insulin resistance, by three months of age, whereas those subjected to nutrition restriction during lactation (IUGR-R) remained permanently thin and highly sensitive to insulin. Mice born with IUGR and fed normally during lactation (IUGR-N) presented an intermediate phenotype and developed insulin resistance by 12 months of age. Molecular alterations to the insulin signaling pathway with an early onset were observed in the livers of adult IUGR-N mice, nine months before the appearance of insulin resistance. The implication of epigenetic changes was revealed by ChIP sequencing, with both posttranslational H3K4me3 histone modifications and microRNAs involved. These two changes lead to the coherent regulation of insulin signaling, with a decrease in Akt gene transcription associated with an increase in the translation of its inhibitor, Pten. Moreover, we found that the levels of the implicated miRNA19a-3p also decreased in the blood of young adult IUGR mice nine months before the appearance of insulin resistance, suggesting a possible role for this miRNA as an early circulating biomarker of metabolic fate of potential use for precision medicine.