Evaluation of Leydig cell activity using single-dose hCG stimulation in prepubertal children suspected of hypogonadism: experience from a tertiary institution

Article information

Ann Pediatr Endocrinol Metab. 2026;31(2):110-118

Publication date (electronic) : 2026 April 30

doi : https://doi.org/10.6065/apem.2550110.055

Ramya Obbai Manohar, Vani Hebbal Nagarajappa, Meghana Narasimhegowda, Lakshmi Deepika Relangi

Department of Pediatric Endocrinology, Indira Gandhi Institute of Child Health, Bangalore, India

Address for correspondence: Ramya Obbai Manohar Department of Pediatric Endocrinology, Indira Gandhi Institute of Child Health, Bangalore, India Email: ramya.manohar93@gmail.com

Received 2025 April 3; Revised 2025 August 24; Accepted 2025 August 29.

Abstract

Highlights

· Single dose human chorionic gonadotropin stimulation test is effective in assessing Leydig cell function.

Introduction

Testosterone is the primary androgen synthesized by Leydig cells. Testosterone and its metabolite dihydrotestosterone (DHT) are essential for the sex differentiation of male infants and the development of sexual characteristics during puberty [1]. Following the perinatal decline, the hypothalamic-pituitary-gonadal (HPG) axis becomes functional again for the first 3–6 months of life before becoming quiescent until puberty. During childhood, from 6 months to the onset of puberty at 9–14 years, basal gonadal steroids are frequently undetectable in plasma, and gonadal function can only be assessed by stimulating Leydig cells with human chorionic gonadotropin (hCG) [2].

Antimullerian hormone (AMH) and inhibin B levels provide information about the existence of functional testicular tissue. However, determining testosterone following hCG stimulation remains a key step in assessing testicular androgen production, which is useful for evaluating a wide range of disorders affecting male genital differentiation and development, leading to 46,XY disorders of sex development (DSD) and hypogonadism [3,4].

The hormone hCG is a glycoprotein produced by the human placenta. Several protocols for hCG stimulation have been proposed, differing in hCG doses, number of injections, and sequence of blood draws [5,6]. The present study analyzes the testosterone response and the magnitude of increment following a single dose of the hCG stimulation test among children with suspected hypogonadism.

The primary outcome is to assess the testosterone response and the fold of increase in testosterone levels following a single dose of the hCG stimulation test among children with suspected hypogonadism. Secondary outcome is as follows: (1) to assess the age-based testosterone response to the single-dose hCG stimulation test. (2) to compare single-dose and multidose (3-day) hCG stimulation tests among children with suboptimal testosterone response after a single dose of hCG.

Micropenis is a condition in which the penis is more than 2.5 standard deviations shorter than the mean of an age-matched reference without hypospadias or epispadias [7,8]. Cryptorchidism (undescended testis) is a condition in which one or both testes fail to descend into the scrotum [9]. Hypospadias is typically characterized by proximal displacement of the urethral opening, penile curvature, and a ventrally deficient hooded foreskin. Proximal hypospadias is the most severe form and involves the urethral opening in a penoscrotal or perineal location, which requires endocrinological evaluation to exclude disorders of sexual differentiation [10]. Criteria suggesting DSD with a 46,XY karyotype [11] include (1) apparent female genitalia with an enlarged clitoris, posterior labial fusion, or an inguinal/labial mass; (2) apparent male genitalia with bilateral undescended testes, micropenis, isolated perineal hypospadias, or mild hypospadias with undescended testes.

Materials and methods

An observational study was conducted after approval from the institutional review board (IGICH/IRB/014/2023). A retrospective review of the medical records of the prepubertal boys (Tanner stage I) aged 6 months to 18 years old with suspected hypogonadism who were assessed by a single-dose hCG test was performed. The initial study population was 127. After exclusion, the group analyzed included 103 prepubertal children (Fig. 1).

Fig. 1.

Flowchart of the study. hCG, human chorionic gonadotropin.

Patients with previous use of hCG or testosterone, previous genital surgeries, or high basal testosterone levels were excluded. A retrospective review of the case records of 103 children who were subjected to singledose hCG tests between the years 2019 and 2023 was conducted. Data on the demographic details, clinical symptoms, and examination findings, such as stretched penile length, presence or absence of testis, testicular volume, the position of the urethral meatus, and scoring of atypical genitalia by the external genital score (EGS), were described. Initial laboratory data were collected, including basal testosterone, luteinizing hormone (LH), follicle-stimulating hormone (FSH), and karyotype.

Results

The study group included 103 patients. The median age and distribution of age groups are described in Table 1. Seven participants (6.8%) were raised as female initially. The mean stretched penile length, testicular volume, median external genitalia score, and the external genitalia phenotype are described in Table 1.

Table 1.

Demographic and clinical characteristics of study participants.

The androgen response to a single dose of hCG was analyzed. The median basal testosterone was 5.27 (interquartile range [IQR], 1.22–14.35) ng/dL, and the stimulated testosterone was 167.26 (IQR, 62.30–279.15) ng/dL, showing a significant 23.65-fold increment of testosterone (P<0.001). The subgroup analysis based on the phenotype is described in Table 2.

Table 2.

Baseline and stimulated testosterone values (ng/dL) following a single-dose human chorionic gonadotropin test

In age group analyses, children of 6 months to 9 years of age, and >9 years of age had median basal testosterone of 7.00 (IQR, 0.10–14.62) ng/dL and 3.0 (IQR, 2.5–12.0) ng/dL, respectively. The stimulated testosterone levels were 203.22 (IQR, 97.61–300.04) ng/dL, with a 32-fold increase in the 6 months to 9 years age group, and 31.7 (IQR, 8.2–73.9) ng/dL with a 4-fold increase in the >9 years age group, which is statistically significant (P<0.001). To account for the above difference, the absolute versus relative increase in testosterone was examined, considering the variable baseline testosterone levels. A positive correlation was observed in those >9 years of age (Spearman P=0.82, P<0.00001) and a moderate correlation in those <9 years (Spearman P=0.40, P<0.00017), indicating greater baseline variability responsible for increased fold of testosterone in the former age group.

Among the 103 participants, 84 (81.5%) had an adequate testosterone response, whereas 19 (18.4%) revealed a suboptimal response with median stimulated values of 8.20 (IQR, 3.48–29.70) ng/dL. In a 3-day hCG stimulation test on these 19 patients, the testosterone values were 18.4 (IQR, 10.6–64.2) ng/dL, which is statistically significant in comparison with the single-dose hCG test (P<0.001) (Table 3). However, scrutinizing the results based on the adequacy of testosterone response, the 3-day hCG stimulation test revealed an adequate response in only 3 patients, and the remaining 16 participants did not show the expected outcome (Table 4).

Table 3.

Comparison between single-dose versus 3-day human chorionic gonadotropin (hCG) test (n=19)

Table 4.

Testosterone increment based on adequacy of testosterone response

On analyzing the 3-day hCG test results against baseline values, the stimulated testosterone of 253 (IQR, 155.7–270.0) ng/dL was only observed in 3 of the 19 children, which is not statistically significant despite a good testosterone increment (P=0.06). This is probably explained by a smaller sample size (n=3). However, the remaining 16 children who also showed a suboptimal response in the single-dose test revealed stimulated testosterone levels of 17.1 (IQR, 9.35–43.31) ng/dL (Table 4), which was statistically significant. Although there was no adequate testosterone increment in these 16 participants, all showed some increment against the baseline values, which could be the probable explanation for the above significance. Of the 16 participants, 15 children had evidence of hypogonadism either genetically or biochemically (Table 5), and the majority fell under subgroup 1.

Table 5.

Subgroup analysis on phenotype severity, adequacy of testosterone response, and the diagnosis supporting the suboptimal response to both the protocols (n=16)

The T/DHT ratio results were available in only a subset of patients suspected of 46,XY DSD (n=24), with a median of 16.40 (IQR, 9.98–31.00). Among these, 2 cases revealed a genetic diagnosis of SRD5A2 mutation, and 22 cases had a T/DHT ratio >20. All 24 cases revealed a good testosterone response to a single-dose test. Among the entire study population (n=103), only 14 underwent genetic testing, with 9 showing normal results and 5 cases showing pathogenic mutations (2 with SRD5A2, 1 with NR5A1, 1 with MAMLD1, and 1 with IL17RD). The remaining 89 subjects were assessed based on supporting biochemical parameters and karyotype results. Of these, 4 cases had chromosomal mosaicism (45,X[5]/46,XY[15]; 46,XY,t(2;11); 46,X,idic(Y)[12]/45,X[8]; 45,X[2]/46,XY[18]) and 22 cases had biochemically diagnosed 5-alpha reductase deficiency based on the T/DHT ratio. All these participants had adequate testosterone response, which indicates the effectiveness of a single dose in the absolute assessment of Leydig cell function. Subgroup analysis based on the phenotype (Table 5) showed that although a severe phenotype was noted in group 6, the majority (20 subjects) had biochemical/genetic detection of 5α reductase deficiency and partial androgen insensitivity, and they all had adequate testosterone response. In addition, the baseline gonadotropins and GnRH stimulation test among the adolescent population (≥9 years old, n=21) gave a useful insight, with 4 subjects showing hypergonadotropic hypogonadism and 9 showing hypogonadotropic hypogonadism, in 2 of whom genetic study (MAMLD1 and IL1RD mutations) adds to the definitive diagnosis (Table 5).

Discussion

In the initial 3–6 months of life, testosterone and INSL3 are testicular Leydig cell function markers, and AMH and inhibin B reflect testicular Sertoli cell function [23]. However, during the rest of infancy and adolescence, basal testosterone, INSL3, and basal gonadotropins are very low or undetectable, making them ineffective indicators of the pituitary-Leydig cell axis unless stimulation tests using hCG are used [3].

The hCG is a glycopeptide mainly produced by the human placenta. It is usually isolated from the urine of pregnant women. However, recombinant hCG has also become available recently [5,24]. The test is based on the capacity of hCG to stimulate androgen production via activation of the LH-chorionic gonadotropin receptor in Leydig cells because of structural similarity with LH [17]. However, hCG can induce sustained and prolonged stimulation of steroidogenesis in comparison to LH because it is structurally modified by the inclusion of a terminal sialic acid on the carbohydrate chain. This structural modification lengthens the half-life, and an extra 24 amino acids on its carboxy-terminal end increase its biological activity [17,25]. The hCG stimulation test is useful for the evaluation of testicular enzyme deficiencies in hypogonadism, 46 XY DSD, and delayed puberty [2,26].

The age at which hypogonadism develops results in a specific phenotype. Fetal onset hypogonadism results in sex development disorders, whereas insult during the latter half of gestation results in micropenis or cryptorchidism with no atypical genitalia [3,27]. Therefore, our study included children with all the above manifestations. To the best of our knowledge, there is no study evaluating a single-dose hCG test in the context of the Indian population.

Multiple protocols exist for the dosing and duration of hCG and sampling time [5,6]. Of these, the 3-day hCG stimulation test is performed routinely. Therefore, the present study evaluates the efficacy of a single-dose hCG test. In our study, a significant 24-fold increment of testosterone was observed. These results are consistent with those observed in the multidose hCG protocols of several earlier studies [6,28-30]. Various dosages have been described for the single-dose hCG test. One study showed excellent testosterone increment with dosages between 1,500 IU and 6,000 IU, and a single dose of 5,000 IU has exhibited significant Leydig cell stimulation [12,13,31]. This factor is the basis for the dosage used in our study population [14]. A single dose of hCG 5,000 IU/1.7 m² or 100 IU/kg has also been applied in other studies, with similar results [6]. Some studies have documented a much higher testosterone increment of 40- to 70-fold following a single-dose hCG test [32,33]. However, the variables to consider for the differences in the results include sample size, the severity of the external genitalia phenotype, different hCG dosages, and the actual number of hypogonadism cases among the study population. More recently, recombinant hCG has been used and found to be as effective as extracted hCG. Oliveira et al. [24] performed a study with a single dose of 250-μg rhCG in a group of 31 prepubertal boys with undescended testes, and observed a significant rise in testosterone levels from 10 ng/dL to 247.8±135.8 ng/dL.

Subgroup analysis indicates significant interphenotypic heterogeneity (Tables 2 and 5). Groups 2 and 3 showed a greater fold rise in testosterone, consistent with nonspecific or mild genital abnormalities and intact androgen biosynthesis and HPG axis function. Although both groups 4 and 5 had 3 cases of hypogonadism, the possible explanation for the higher increment in group 5 compared to group 4 is due to the low and variable baseline testosterone in the former group. To account for this, the correlation of the absolute and relative percentage increase was analyzed. A moderate positive correlation was found in group 4 (P=0.158), whereas group 5 revealed no significant correlation (P=0.971). Although some individuals in group 5 exhibited disproportionately high fold increases, others remained unresponsive. The high variability and lack of association between baseline and stimulated values in group 5 may reflect underlying heterogeneity in testicular pathology or hypothalamic pituitary axis impairment. In group 1, although being a minor phenotype, 7 participants had low-volume palpable testes and were genetically/biochemically diagnosed with hypogonadism. This data may point to the limitations of phenotypebased grouping in predicting abnormal biochemistry and gonadal function [34,35].

On analyzing the testosterone response by age group, those aged 6 months to 9 years had a 20-fold increase in testosterone, and those aged above 9 years showed only a 6-fold increment. There was a statistical difference between the 2 groups. The low testosterone response in the latter group might be explained by less variation in the baseline testosterone and higher prevalence of biochemically/genetically diagnosed cases of hypogonadism (Table 5). In contrast, low and greater variability in baseline testosterone among the former group can yield a larger percentage increase, even with a moderate absolute increase in testosterone. In addition, phenotype severity also explains the above difference, with the majority of subtle genital abnormalities (groups 1–3) seen in the 6 months to 9 years age group. In a study by Forest [36], age and pubertal-related androgen responses post-hCG were studied among 100 apparently normal children, and no significant age-related changes in the testosterone response were observed during the prepubertal period. In a study by Tapanainen et al. [32], there was a difference in testosterone response among pre- and early pubertal cryptorchid boys following a single dose of the hCG test. The relative response of serum testosterone was 70-fold in prepubertal boys and 6-fold at early puberty.

The definition of an appropriate testosterone response to hCG stimulation varies widely and may depend on the regimen and the age of the child. PosthCG stimulation, testosterone often rises by 2 to 10 times or even 20 times during infancy and 5 to 10 times in childhood. Increments in plasma testosterone are typically between 55 and 245 ng/dL, with peak values between 66 and 250 ng/dL [2]. Ishii et al. [7] performed a retrospective study in 50 boys with micropenis aged 6 months to 18 years of age who were assessed by hCG test with a dosage of 3,000 IU/m² for 3 consecutive days. It was observed that 34 spontaneously developed puberty (group 1) and 16 did not (group 2). Based on this, a cutoff testosterone level of 110 ng/dL post-hCG was assigned for patients who required further evaluation and hormone replacement treatment. Hence, the present study used a cutoff value to define the adequacy of testosterone response based on the above considerations. However, it is important to note that the above threshold value (110 ng/dL) was validated in a study following a 3-day hCG stimulation test rather than a single-dose test, and extrapolating such a threshold might limit the reliability of this cutoff. If a blunted response is obtained to a single dose of hCG, multidose hCG tests might help in establishing the diagnosis of hypogonadism [30]. A 3-day hCG test was incorporated in our unit protocol.

The maximal testosterone values are seen 72 to 120 hours after the first injection, and the first injection induces a progressive and modest rise in testosterone levels [28]. This is the basis of interpreting testosterone at 72 hours following a single dose of hCG. The repeated injection of hCG was ineffective in modifying the testosterone levels for approximately 3 to 5 days; hence, the rationale for single-dose or alternate-day injections used in some studies [28,30]. However, pathologic conditions such as hypogonadotropic hypogonadism or dysgenetic testis may require prolonged hCG stimulation.

A search of the literature yielded a few studies comparing single and multidose (3-day) hCG testing. In a related trial, 25 children suspected of having hypogonadism were tested using 2 protocols (single dosage versus alternate 3-day regimen), with a 1-month delay between the 2 tests. The dosage for a single-dose test was 5,000 IU/1.7 m², while for a multidose test, it was 1,500 IU. There was a substantial difference detected, with the repeated injection test resulting in a larger increase in testosterone levels. Given the aforementioned findings, Kauschansky et al. [30] proposed that the outcome of a single test can reliably forecast the outcome of repeated injection tests. In another study by Kolon and Miller [6], a 35-fold increase in testosterone levels was seen with a single hCG dose of 5,000 IU/1.72 m2 among 60 prepubertal children with undervirilization. No significant differences were observed when the single dose was compared with a 3-day hCG test performed in 17 children with a 20-fold increase in testosterone. In a similar study by Kardelen et al. [33], a 40-fold testosterone increment was observed with a single dose of 5,000 IU/m2 among 18 children with features of undervirilization. Correspondingly, a 40.1-fold increment was observed over 3 consecutive days with 1,500 IU/m2 hCG regimens in 18 cases with a molecular diagnosis of androgen insensitivity syndrome or 5-alpha reductase deficiency, suggesting no difference between the 2 protocols.

We observed a suboptimal testosterone response with a single-dose test in 3 cases, but, ultimately, a good response was observed in a 3-day test. The reason for the false negative response observed with the former protocol was uncertain. Possible explanations could include errors in the administration/delivery of the appropriate doses and the potency of hCG used in the single-dose hCG stimulation test. As the current study was a retros pective observational study, the above factors could not be ascertained. Moreover, it is reasonable to assume that some midchildhood testes have populations of responsive cells that are smaller or cells that are less responsive to stimulation during the short-term hCG stimulation testing [37]. The above factor might also explain our findings with a single-dose test.

Although there was a statistical difference between the 2 protocols in the 19 participants evaluated in the current investigation, extrapolating these results from such a small sample size is questionable. Furthermore, 15 of the 16 individuals with poor testosterone responses to both single and 3-day tests had clinical or biochemical evidence of hypogonadism, showing that a single dose could replace the 3-day hCG test in the initial assessment of Leydig cell activity.

The limitations in this study include a cross-sectional design, nonuniformity in the duration between the single dose and the 3-day test, a lack of follow-up data on spontaneous puberty progression, inaccessibility of molecular diagnosis details in all hypogonadism cases, and a small sample size that is not representative of the population. There is a clear need for research with a bigger sample size and follow-up for receiver operating characteristic analysis and validation of results after a single-dose hCG test. Nonetheless, the current study emphasizes the usefulness of the single-dose test, particularly in the context of the Indian population.

In conclusion, a single-dose hCG stimulation test is useful in assessing Leydig cell activity. A significant 24-fold increment of testosterone was observed following a single-dose test, but it is worth considering that baseline testosterone variability affects the age-based difference in testosterone response. Comparable testosterone responses observed with both the protocols suggest that a single-dose hCG test could be an alternative to a 3-day hCG test in the initial evaluation of hypogonadism.

Article information Continued