Leydig and Sertoli cell function in individuals with genital ambiguity, 46,XY karyotype, palpable gonads and normal testosterone secretion: a case-control study

Abstract BACKGROUND: Because normal male sexual differentiation is more complex than normal female sexual differentiation, there are more cases of disorders of sex development (DSDs) with 46,XY karyotype that have unclear etiology. However, Leydig and Sertoli cell markers are rarely used in distinguishing such individuals. OBJECTIVES: To evaluate the function of Leydig and Sertoli cells in individuals with genital ambiguity, 46,XY karyotype, palpable gonads and normal testosterone secretion. STUDY DESIGN AND SETTING: Case-control study with 77 patients, including eight with partial androgen insensitivity syndrome, eight with 5α-reductase deficiency type 2 (5ARD2) and 19 with idiopathic 46,XY DSD, and 42 healthy controls, from the Interdisciplinary Study Group for Sex Determination and Differentiation (GIEDDS), at the State University of Campinas (UNICAMP), Campinas, Brazil. METHODS: Baseline levels of gonadotropins, anti-Müllerian hormone (AMH), inhibin B, insulin-like 3 (INSL3), testosterone and dihydrotestosterone in cases, and AMH, inhibin B, and INSL3 levels in controls, were assessed. RESULTS: There was no significant difference in age between cases and controls (P = 0.595). AMH and inhibin B levels were significantly lower in cases than in controls (P = 0.031 and P < 0.001, respectively). INSL3 levels were significantly higher in cases than in controls (P = 0.003). Inhibin B levels were lower in 5ARD2 patients (P = 0.045) and idiopathic patients (P = 0.001), in separate comparisons with the controls. CONCLUSION: According to our findings, we can speculate that inhibin B levels may be used to differentiate among DSD cases.


INTRODUCTION
Ambiguous genitalia are the most complex clinical manifestation of disorders of sex development (DSDs), an umbrella term that is used for congenital conditions characterized by atypical chromosomal, gonadal or anatomical development. 1 Because normal male sexual differentiation involves more genetically determined and hormonal events than those in normal female sexual differentiation, DSDs with 46,XY karyotype present greater etiological complexity.
Among the main etiologies of patients with genital ambiguity, 46,XY karyotype and normal testosterone secretion, partial androgen insensitivity syndrome (PAIS) (OMIM #312300) potentially presents with clinical features indistinguishable from those of other etiologies, particularly those of 5α-reductase type 2 deficiency (5ARD2) (OMIM #264600). 2 Both diagnoses are solely confirmed on the basis of molecular alterations in specific genes, which is a costly procedure and thus conducted at only a few centers. 3 Routine measurement of the function of Leydig cells, including testosterone and dihydrotestosterone (DHT) levels, is not always effective for differentiating PAIS from 5ARD2. 3,4 There is a lack of data regarding insulin-like 3 (INSL3), which is also produced by Leydig cells, in relation to management of DSDs. 4 In some studies, high levels of anti-Müllerian hormone (AMH), a marker for Sertoli cells, were observed in patients with androgen insensitivity. [5][6][7] However, in patients with 5ARD2, low levels have been reported in only two studies. 8,9 Inhibin B is a good marker for Sertoli cells and gonadal function; nevertheless, data on cases of androgen insensitivity and 5ARD2 remain limited. 4 Moreover, the role of these Leydig and Sertoli cell markers in patients with idiopathic 46,XY DSDs and genital ambiguity remains unclear.

OBJECTIVE
The aim of this study was to evaluate the function of Leydig cells (testosterone, dihydrotestosterone [DHT] and INSL3) and Sertoli cells (AMH and inhibin B) in patients with 46,XY DSDs and genital ambiguity.

Patients and control group
The inclusion criteria were palpable gonads (in the scrotal and/or

Clinical evaluation
Upon recruitment, patients were clinically evaluated with regard to the following variables: age (in months), weight (in kg), height (in cm) and body mass index (BMI; in kg/m 2 ). These values were converted to z-scores using the NCHS 2000 data. Moreover, the patients' stage of puberty was assessed, and the patients were then classified as pubertal (Tanner stage ≥ 2) or pre-pubertal (Tanner stage 1). Furthermore, the following data were obtained from medical records: birth weight (in g), birth length (in cm) and features of the genitalia at initial presentation. The grade of masculinization of the genitalia was determined on the basis of the external masculinization score (EMS), in accordance with the method described by Ahmed et al. 11

Laboratory evaluation
Karyotyping was performed at the cytogenetics laboratory at our Testosterone secretion was considered normal in individuals presenting a total increase in testosterone of 1.5 ng/ml after stimulation relative to baseline levels. 3,12 In these cases, only testosterone and DHT levels after hCG administration were evaluated. Only baseline AMH, inhibin B and INSL3 levels were measured in the control group. Blood samples were collected through peripheral vein puncture, and serum was extracted via centrifugation at 2000 ×g for 10 minutes and was stored at -20 °C until evaluation.

Hormonal assays
The following hormonal assays were performed: LH, electro-

Statistical analysis
The data were analyzed using the Statistical Package for the Social were presented as values of the unstandardized coefficients: beta (β) ± standard error and as values of the adjusted explanation coefficient (r 2 ). The significance level was set at 5% (P < 0.05).

Among the 35 patients included in the study, eight had PAIS
(five prepubertal and three pubertal), eight had 5ARD2 (three prepubertal and five pubertal) and 19 were idiopathic (13 prepubertal and six pubertal). The baseline clinical and laboratory data of all the patients included are summarized in  were observed, such that both of these were lower in the idiopathic group only, compared with the PAIS group. In contrast, length at birth was significantly shorter in the idiopathic group than in the 5ARD2 group (P = 0.030) ( Table 1).
AMH levels were inversely proportional to age, with a moderate correlation between cases (r = -0.68; P < 0.0001) and controls (r = -0.83; P < 0.0001). Subgroup analysis revealed that this age-based correlation was strong in the 5ARD2 group (r = -0.95; P < 0.0001) and moderate in the idiopathic group (r = -0.71;  Serum AMH levels were significantly lower in cases than in the control group (P = 0.031). Nonetheless, subgroup comparisons with the control group and among the groups did not reveal any significant differences in AMH levels ( Table 2).
Inhibin B levels could not be evaluated in one case with 5ARD2 (aged 27 years 10 months) and in five cases in the idiopathic subgroup (aged 1 year 5 months, 6 years 6 months, 7 years 3 months, 8 years 2 months and 8 years 4 months). Inhibin B levels were evaluated in all patients in the PAIS and control groups, and no correlation with age was observed, either among cases or controls.
As described above, inhibin B was positively correlated with AMH only in the idiopathic group. Inhibin B levels were significantly lower in cases than in controls (P < 0.001) ( Table 2) that INSL3 levels were significantly higher in cases than in controls (P = 0.003) ( Table 2). Subgroup comparisons among each other and with the control group revealed higher values in the 5ARD2 and PAIS subgroups than in the control group; however, these differences were not significant (Bonferroni test) ( Table 3). However, among pubertal individuals, serum inhibin B levels (Kruskal-Wallis test, P = 0.040) and AMH levels (Kruskal-Wallis test, P = 0.036) were significantly higher in the PAIS group than in the 5ARD2 and idiopathic groups.
In the stepwise multivariate linear regression analysis, we observed that among all the variables analyzed, the weight z-score (β = 0.006 ± 0.003; P = 0.049) significantly explained 12% of the   Inhibin B, which was not indicated as a diagnostic marker in the 2016 DSD consensus, 19 was also evaluated in this study. It has been demonstrated to be an useful parameter in evaluating DSDs, especially among individuals with 46,XY karyotype. 20,21 The clinical significance of inhibin B is particularly evident in cases of cryptorchidism, which is a frequent clinical manifestation in patients with 46,XY DSDs with normal testosterone secretion; and in assessing fertility, which is an important aspect in DSD management. [22][23][24][25][26] Studies on cryptorchidism have revealed an association between inhibin B levels and testicular volume, and have suggested that inhibin B acts as a marker for testicular recovery after treatment for cryptorchidism. 22,23 Some studies evaluating inhibin B levels in patients with infertility have reported that a direct association between this hormone and sperm parameters was observed, which would suggest that inhibin B is also a good marker for spermatogenesis. [24][25][26] Moreover, serum inhibin B levels are low in individuals with 46,XY gonadal dysgenesis, 27 and in some cases, this disorder may present even with normal serum testosterone levels. 28 In such cases, inhibin B could be used for differential diagnosis and in the prognosis for gonadal viability.
From a physiological point of view, inhibin B is useful for assessing 46,XY DSDs because it is present at measurable levels for most of an individual's lifespan. 21,29 In addition, this hormone is already present at measurable levels at birth, even in cord blood samples, and its level increases more rapidly during the first week of life, unlike the levels of the hormones traditionally used to assess genital ambiguities (e.g. testosterone and AMH). [29][30][31] Furthermore, the present study showed that, similar to AMH, inhibin B levels were lower in cases than in controls. No age-related changes were observed in inhibin B levels; however, again similar to AMH, inhibin B levels were higher in pubertal patients in the   36 and secondly, this increment in INSL3 levels is potentially associated with its induction of steroidogenesis in the context of both relative (5ARD2) and partial (PAIS) androgen insufficiency. 32,[37][38][39] Another potential explanation for this increment in INSL3 levels is Leydig cell hyperplasia. This has been already reported among individuals with 5ARD2 and PAIS at puberty, occurring after LH hyperstimulation through a negative feedback mechanism in 5ARD2, with reductions in DHT and in PAIS, owing to testosterone activity. 33 However, further research on INSL3 behavior in 46,XY DSD patients is required in order to elucidate this trend.
Multivariate analysis was performed, and no data that would correlate the findings regarding inhibin B, AMH and INSL3 levels with puberty, age and anthropometric data were found, except for weight and BMI. However, there was low explanatory power for these two variables.
One limitation of this study was the small cohort, which may have influenced the statistical power of some of our findings.
However, it should be noted that it is not easy to recruit individuals with such rare disorders, through molecular diagnosis at a single center. In addition, the lack of evaluation of testicular histology and its correlation with hormonal levels formed another study limitation, thus further restricting the elucidation of cellular patterns. However, gonadal biopsy is not usually performed for management of 5ARD2 and PAIS.

CONCLUSIONS
To our knowledge, our study was the first to report that individuals with 5ARD2 have low levels of inhibin B, thus suggesting that this hormone may be a biochemical marker that can differentiate the diagnosis of 5ARD2 from that of other etiologies with a similar clinical presentation (particularly PAIS). Our study was also the first to report that INSL3 levels were higher in patients with 46,XY karyotype, palpable gonads and normal testosterone secretion. However, further evaluation of each etiological group analyzed is necessary. Lastly, for differentiation of these groups, the AMH levels did not show promising results.