Impact of Cushing’s syndrome on the gonadotrope axis and testicular functions in men

Abstract

Abstract STUDY QUESTION Does Cushing’s syndrome (CS) differently affect the gonadotrope axis and testicular functions (GA/TF) according to the hypercortisolism intensity and underlying etiology? SUMMARY ANSWER Endogenous cortisol excess caused by CS leads to varying degrees of hypogonadotropic hypogonadism (HH) with more severe GA/TF impairment and altered spermatogenesis in men with intense hypercortisolism associated with paraneoplastic/ectopic adrenocorticotrophic hormone (ACTH) secretion (EAS). WHAT IS KNOWN ALREADY CS is very rarely studied in men due to its lower prevalence in men than in women. In a few old reports focusing exclusively on a limited number of men with Cushing’s disease (CD), the occurrence of hypogonadism was reported. However, a detailed assessment of the impact of CS on the GA/TF in a significant series of patients has not been performed. Yet, hypogonadism could worsen CS-associated comorbidities such as osteoporosis and myopathy. To date, the full spectrum of GA/TF impairment in men with CS of different etiologies and intensity remains unknown. STUDY DESIGN, SIZE, DURATION In this monocentric study, 89 men with CS diagnosed at a tertiary endocrine university center (Bicêtre, Paris Saclay) between January 1990 and July 2021 were evaluated and compared to 40 normal men of similar age. PARTICIPANTS/MATERIALS, SETTING, METHODS The CS patient cohort of 89 men included 51 with CD, 29 with EAS and 9 with CS of adrenal origin i.e. (ACTH-independent CS (AI-CS)). They all had frank hypercortisolism, with increased 24 h-urinary-free cortisol (24 h-UFC) in two separate samples. A case–control study was performed focusing on pituitary gonadotrope function and testicular sex steroids and peptides. An additional set of six CS men had an evaluation including semen analysis. In a subgroup of 20 men with available data after CS remission, a longitudinal analysis was conducted to assess the reversibility of GA/TF defects. MAIN RESULTS AND THE ROLE OF CHANCE Compared to controls, men with CS had significantly lower total testosterone (TT), bioavailable TT, and free TT (P < 0.0001). Hypogonadism, defined as serum TT levels <3.0 ng/ml, was present in 83% of men with EAS, in 61% of men with CD, and in 33% of men with AI-CS. Low-normal LH concentrations in the included men with hypercortisolism indicated HH. Serum sex hormone-binding globulin levels were moderately decreased in men with CD (P = 0.01 vs controls). Among the CS men, those with EAS had significantly lower TT, LH, and FSH levels than those with CD or AI-CS. When compared to controls, patients with EAS were the only group exhibiting a significant decrease in both serum FSH (P = 0.002) and the testicular peptides inhibin B (P < 0.0001) and anti-Müllerian hormone (P = 0.003). Serum INSL3 levels were significantly lower in men with CD than in the controls (P = 0.03). Of note, 24 h-UFC and ACTH were inversely and significantly associated with the majority of reproductive hormones including LH, FSH, TT, and inhibin B. Following successful curative therapy, reproductive assessment at a mean of 6.0 ± 4.3 years showed a significant increase in serum TT (P < 0.0001) and plasma LH (P = 0.02) levels, indicating a reversal of HH in 75% of the affected males. Among the six patients with available semen analysis, the two EAS cases exhibited a decrease in Sertoli cell peptides associated with a severe oligozoospermia, which completely normalized following removal of the source of hypercortisolism. LIMITATIONS, REASONS FOR CAUTION The potential bias due to the retrospective design is counteracted by the analysis of the largest male CS cohort to date as well as the use of stringent inclusion and exclusion criteria. Due to the low number of patients with semen analysis in this study, further research is needed to unravel the full spectrum of spermatogenesis defects in men with CS. WIDER IMPLICATIONS OF THE FINDINGS This work reveals the variable spectrum of reproductive impact in men with CS. We demonstrate that GA/TF impairment depends on the intensity of hypercortisolism which in turn is related to the underlying etiology. The causal link between hypercortisolism and GA/TF impairment was attested by its reversibility in most patients after CS remission. The wider implications of our findings lie in the potential generalization to a much commoner entity, iatrogenic CS due to chronic exposure to exogenous glucocorticoids. STUDY FUNDING/COMPETING INTEREST(S) Several research grants were attributed to J.Y.: (i) a grant from Programme Hospitalier de Recherche Clinique (PHRC # P081212 HYPOPROTEO); (ii) a grant from the French Association of Patients with Adrenal Diseases (‘Association surrénales’); and (iii) independent Investigator Research Grants from HRA Pharma, Novartis and Recordati Pharma. A SICPA Foundation grant (Lausanne, Switzerland) allowed protected research time for G.E.P. The above sponsors were not involved in any part of the study. The authors have no competing or other conflicts of interest to declare. TRIAL REGISTRATION NUMBER N/A.