Constitutional delay of puberty: presentation and inheritance pattern in 48 familial cases

Delayed puberty is diagnosed when there is no testicular enlargement after 14 years in boys or breast development after 13 years in girls. The constitutional delay of puberty (CDP) can be diagnosed only after underlying conditions have been ruled out and when the pubertal delay is followed by spontaneous pubertal development [1]. Boys are affected much more frequently than girls, while the reverse is true in idiopathic central precocious puberty [2].

The age at onset and the progression of puberty are driven by a complex interaction between genetic, metabolic and environmental factors. The age at onset of puberty is often homogeneous within the family and concordant between twins [3]. Kaprio et al. [4] analyzed the age at menarche in monozygotic twins and showed that 74 % of the variance is due to genetic factors and 26 % to environmental factors. The frequency of the familial forms has been reported at 56–80 % in CPD [5] while they represent 41 % in our series of 493 girls with central precocious puberty [6]. Two studies analyzed the inheritance of CDP [7, 8]. The studies [7–10] suggested an autosomal dominant mode, without the influence of obesity.

We evaluated 48 patients seen consecutively by the same senior pediatric endocrinologist for CDP with a familial component. The purpose of this study is to describe the presentation and the mode of inheritance of CDP in a series of familial cases.

Fourteen girls and thirty-four boys were included.

Mode of inheritance

None of the families reported a history of consanguinity. Pedigrees of the 48 families are represented in Fig. 1 and familial clustering in Fig. 2. A total of 128 cases of CDP were identified, of whom 60 were affected girls and 68 affected boys. The male to female ratio for the probands was 2.4:1. It was 1:1.37 for the first degree relatives.

Of the 48 probands, 46 (96 %, 14 girls and 32 boys) had at least one affected 1^st degree relatives and 2 (4 %, cases 27 and 28) had only 2^nd degree relatives affected. There was no family in whom only siblings were affected.

In the girls, 11 families (79 %) exhibited exclusive maternal inheritance, 1 (7 %) paternal inheritance and 2 (14 %) presented both maternal and paternal inheritance. In the boys, 14 families (41 %) exhibited exclusive maternal inheritance, 12 (35 %) paternal inheritance and 8 (24 %) presented both maternal and paternal inheritance.

Of the 36 unilineal families, CDP was verified in three generations in only 3 of the patients (families 11, 26 and 28).

In the boys with bilineal inheritance, the ages at onset of puberty (Fig. 3) and at evaluation were higher than in those with unilineal inheritance, but the difference was not significant (Table 2). They had similar plasma testosterone concentrations suggesting that these differences are not due to a difference in their pubertal development stages.

	Unilineal inheritance (n = 26)		Bilineal inheritance (n = 8)		p
	n	m ± SD	n	m ± SD
Age at onset, years	22	15.25 ± 0.35	7	16 ± 1.41	0.06
Age at evaluation, years	26	15.1 ± 0.42	8	16.05 ± 2.47	0.12
Testosterone, ng/mL	22	0.38 ± 0.35	6	0.32 ± 0.4	0.31

The results of this study suggest that despite various inheritance patterns among the pedigrees, there is evidence of autosomal dominant inheritance in CDP, with either complete or incomplete penetrance. In girls exclusive maternal inheritance seems to be the major mode of inheritance whereas for boys the mode of inheritance was almost equally maternal, paternal or bilineal. The results also suggest a more severe phenotype for boys with bilineal inheritance, but the difference did not reach statistical significance perhaps because of the small sample size.

Previous data, such as studies of twins, suggested that genetic factors account for 50–80 % of all variations in pubertal timing [4, 20]. Two extended studies addressed the segregation of CDP in families [7, 8] and showed an inheritance pattern compatible with autosomal dominant inheritance with complete or incomplete penetrance. Our study confirms this hypothesis and is the first to present all of the pedigrees included. We deliberately chose to include only familial CDP and no sporadic CDP, contrary to the previous studies, to focus on the mode of inheritance. Reproducing all of the pedigrees highlighted that 11/12 girls with CDP with unilineal inheritance had a mother who also had late puberty, while Wehkalampi et al. [7] reported a proportion of 5/12. We found no explanation for this discrepancy. The greater age at menstruation in the girls’ mothers than the boys’ mothers is probably linked to this quasi exclusive transmission by the mothers of the girls. Families with only siblings affected were rare: none in our series and 2 % in that reported by Wehkalampi et al. [7].

Our study is the first to suggest that there might be a more severe phenotype in boys with bilineal inheritance than in those with unilineal inheritance. The ages at onset of puberty and at evaluation were indeed higher in the boys with bilineal inheritance with a comparable testosterone level for boys with unilineal or bilineal inheritance, suggesting that this difference is not due to a difference in their pubertal development stages. This result has to be interpreted carefully regarding the small number of probands with bilineal inheritance analyzed.

In previous reports, the male to female ratio among CDP has ranged from 2:1 to 5:1 [5, 21, 22]. In our study, this ratio was 2.4:1 for the probands, but this male preponderance disappeared when calculated for the first degree relatives (1:1.37). In their study, Wehkalampi et al. [7] found this ratio at 1.2:1 . They suggested that the marked overrepresentation of boys with CDP at pediatric clinics may be biased. The authors noted that this potential biased overrepresentation could be linked to psychosocial reasons with boys being more interested in athletic performance and physical characteristics around puberty. Another explanation given by the authors is the trend toward earlier puberty in girls [23, 24], which has also been reported for boys [25].

Many studies of girls have demonstrated a link between more advanced breast development and higher BMI [26, 27] and that obese girls achieve menarche earlier than lean ones [28]. In boys, the association between pubertal development and adiposity has been less consistently reported with studies finding a positive association [25], lack of association [29] or negative association [30]. Children and adolescents with delayed puberty are typically relatively lean and this is true for the majority of our case series. However, it is possible that reporting the BMI SDS to BA, and not to chronological age, reduce the frequency of the lean patients as puberty itself results in increases in BMI. Some of the probands were overweight, especially the boys, which was already reported in CDP [5]. A recent genome-wide association study reported that BMI-increasing alleles correlated with earlier breast development in girls, whereas in boys some BMI-increasing alleles are associated with earlier and others with delayed sexual development [31]. This finding is consistent with epidemiological data and confirms that some mechanisms regulating pubertal onset in males and females might be different. These data are also consistent with our finding of different modes of inheritance in CDP for boys and girls; however, the high percentages of mothers that reached menarche after 14 years of age in both sexes suggest that if the low BMI has a role, it is additional to that of the familial factor. Alternatively, there could be a gene-environment interaction such that genetics could produce greater susceptibility to having pubertal timing affected by BMI.

What controls puberty remains largely unknown. The genome-wide association studies have shown that variants of more than 30 genes are associated with the age of menarche in humans [32]. Lomniczi et al. recently reported that the timing of female puberty was under the regulatory control of an epigenetic mechanism of transcriptional repression [33]. All of these reports confirm that the timing of puberty is controlled by many genes. This hypothesis is supported by the greater severity of the phenotype in boys with CDP with a bilineal family history, likely due to inheriting different puberty timing genes from both parents. This hypothesis should be confirmed in a larger cohort.

We acknowledge that our study has some limitations. We have no extended pedigrees. This may bias ascertainment/inclusion towards pedigrees with autosomal dominant inheritance. Relying on a parental report for the onset of puberty is not very accurate as many parents fail to notice when the puberty actually starts; however, even recalled information is valuable; studies suggest that even decades after the event, 75–90 % of women remember their age of menarche and 50 % of men remember the timing of their pubertal growth spurt within a year [34, 35]. Despite a 10-year inclusion period, the number of families included is much lower than for precocious and advanced puberty [36]. To try to confirm our data, further multicenter studies on larger cohorts should be conducted.

This study confirms the autosomal dominant inheritance pattern in CDP in a monocentric pediatric cohort and the need for further genetic studies. Our data are also consistent with other studies suggesting that the mechanisms underlying the onset of puberty are different in girls and boys, and different modes of transmission in boys and girls were revealed. The greater severity of the phenotype in boys with a bilineal family history is likely due to inheriting different puberty timing genes from both parents. Future research should be directed at identifying such genes.